{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "eval_steps": 500, "global_step": 7566, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00013217023526301876, "grad_norm": 6.957700714032043, "learning_rate": 0.0, "loss": 1.0983, "mean_token_accuracy": 0.7192040085792542, "num_tokens": 65536.0, "step": 1 }, { "epoch": 0.0002643404705260375, "grad_norm": 7.1641672338970865, "learning_rate": 1.3210039630118892e-08, "loss": 1.1494, "mean_token_accuracy": 0.7012574672698975, "num_tokens": 131072.0, "step": 2 }, { "epoch": 0.0003965107057890563, "grad_norm": 6.8095072604849705, "learning_rate": 2.6420079260237783e-08, "loss": 1.0964, "mean_token_accuracy": 0.7189750671386719, "num_tokens": 196608.0, "step": 3 }, { "epoch": 0.000528680941052075, "grad_norm": 6.347712549380873, "learning_rate": 3.9630118890356674e-08, "loss": 1.053, "mean_token_accuracy": 0.7296117544174194, "num_tokens": 262144.0, "step": 4 }, { "epoch": 0.0006608511763150939, "grad_norm": 6.421823322800878, "learning_rate": 5.284015852047557e-08, "loss": 1.0693, "mean_token_accuracy": 0.7282230257987976, "num_tokens": 327680.0, "step": 5 }, { "epoch": 0.0007930214115781126, "grad_norm": 7.505086399864066, "learning_rate": 6.605019815059445e-08, "loss": 1.1291, "mean_token_accuracy": 0.7070870399475098, "num_tokens": 393216.0, "step": 6 }, { "epoch": 0.0009251916468411313, "grad_norm": 7.471167856269708, "learning_rate": 7.926023778071335e-08, "loss": 1.1172, "mean_token_accuracy": 0.7096965909004211, "num_tokens": 458752.0, "step": 7 }, { "epoch": 0.00105736188210415, "grad_norm": 6.3763571120938725, "learning_rate": 9.247027741083225e-08, "loss": 1.0567, "mean_token_accuracy": 0.7274295091629028, "num_tokens": 524288.0, "step": 8 }, { "epoch": 0.0011895321173671688, "grad_norm": 6.648417263026968, "learning_rate": 1.0568031704095113e-07, "loss": 1.108, "mean_token_accuracy": 0.7143816351890564, "num_tokens": 589824.0, "step": 9 }, { "epoch": 0.0013217023526301878, "grad_norm": 7.7373278081337595, "learning_rate": 1.1889035667107001e-07, "loss": 1.2152, "mean_token_accuracy": 0.6888353228569031, "num_tokens": 655360.0, "step": 10 }, { "epoch": 0.0014538725878932065, "grad_norm": 6.615297236440097, "learning_rate": 1.321003963011889e-07, "loss": 1.0899, "mean_token_accuracy": 0.7199517488479614, "num_tokens": 720896.0, "step": 11 }, { "epoch": 0.0015860428231562252, "grad_norm": 8.144764086900624, "learning_rate": 1.4531043593130781e-07, "loss": 1.1564, "mean_token_accuracy": 0.6984189748764038, "num_tokens": 786432.0, "step": 12 }, { "epoch": 0.001718213058419244, "grad_norm": 6.970620047800608, "learning_rate": 1.585204755614267e-07, "loss": 1.1247, "mean_token_accuracy": 0.7095134854316711, "num_tokens": 851968.0, "step": 13 }, { "epoch": 0.0018503832936822627, "grad_norm": 6.364672535544319, "learning_rate": 1.717305151915456e-07, "loss": 1.0948, "mean_token_accuracy": 0.7188377380371094, "num_tokens": 917504.0, "step": 14 }, { "epoch": 0.0019825535289452814, "grad_norm": 6.6392520741193115, "learning_rate": 1.849405548216645e-07, "loss": 1.1362, "mean_token_accuracy": 0.7114973664283752, "num_tokens": 983040.0, "step": 15 }, { "epoch": 0.0021147237642083, "grad_norm": 7.0539462144113765, "learning_rate": 1.9815059445178336e-07, "loss": 1.1366, "mean_token_accuracy": 0.7088115215301514, "num_tokens": 1048576.0, "step": 16 }, { "epoch": 0.002246893999471319, "grad_norm": 6.573898584405452, "learning_rate": 2.1136063408190227e-07, "loss": 1.1159, "mean_token_accuracy": 0.7138780355453491, "num_tokens": 1114112.0, "step": 17 }, { "epoch": 0.0023790642347343376, "grad_norm": 6.9753962714477735, "learning_rate": 2.2457067371202117e-07, "loss": 1.1236, "mean_token_accuracy": 0.7106885313987732, "num_tokens": 1179648.0, "step": 18 }, { "epoch": 0.0025112344699973568, "grad_norm": 6.420970725687045, "learning_rate": 2.3778071334214003e-07, "loss": 1.1217, "mean_token_accuracy": 0.7096965909004211, "num_tokens": 1245184.0, "step": 19 }, { "epoch": 0.0026434047052603755, "grad_norm": 7.190737085921202, "learning_rate": 2.5099075297225896e-07, "loss": 1.1344, "mean_token_accuracy": 0.7049505710601807, "num_tokens": 1310720.0, "step": 20 }, { "epoch": 0.0027755749405233942, "grad_norm": 6.618450148060287, "learning_rate": 2.642007926023778e-07, "loss": 1.1232, "mean_token_accuracy": 0.7096202969551086, "num_tokens": 1376256.0, "step": 21 }, { "epoch": 0.002907745175786413, "grad_norm": 6.661187211154599, "learning_rate": 2.774108322324967e-07, "loss": 1.1099, "mean_token_accuracy": 0.7140153646469116, "num_tokens": 1441792.0, "step": 22 }, { "epoch": 0.0030399154110494317, "grad_norm": 6.345009601233029, "learning_rate": 2.9062087186261563e-07, "loss": 1.0761, "mean_token_accuracy": 0.7224240303039551, "num_tokens": 1507328.0, "step": 23 }, { "epoch": 0.0031720856463124504, "grad_norm": 6.692214391165073, "learning_rate": 3.0383091149273453e-07, "loss": 1.1303, "mean_token_accuracy": 0.706415593624115, "num_tokens": 1572864.0, "step": 24 }, { "epoch": 0.003304255881575469, "grad_norm": 6.859221073925105, "learning_rate": 3.170409511228534e-07, "loss": 1.0964, "mean_token_accuracy": 0.7123672366142273, "num_tokens": 1638400.0, "step": 25 }, { "epoch": 0.003436426116838488, "grad_norm": 6.137929317346025, "learning_rate": 3.302509907529723e-07, "loss": 1.0901, "mean_token_accuracy": 0.7146716117858887, "num_tokens": 1703936.0, "step": 26 }, { "epoch": 0.0035685963521015066, "grad_norm": 6.22433789902895, "learning_rate": 3.434610303830912e-07, "loss": 1.1297, "mean_token_accuracy": 0.7058662176132202, "num_tokens": 1769472.0, "step": 27 }, { "epoch": 0.0037007665873645254, "grad_norm": 6.00276604313404, "learning_rate": 3.5667107001321006e-07, "loss": 1.1621, "mean_token_accuracy": 0.6918874382972717, "num_tokens": 1835008.0, "step": 28 }, { "epoch": 0.003832936822627544, "grad_norm": 5.511226091344694, "learning_rate": 3.69881109643329e-07, "loss": 1.0854, "mean_token_accuracy": 0.7167317867279053, "num_tokens": 1900544.0, "step": 29 }, { "epoch": 0.003965107057890563, "grad_norm": 5.994356961141411, "learning_rate": 3.8309114927344787e-07, "loss": 1.0953, "mean_token_accuracy": 0.7111005783081055, "num_tokens": 1966080.0, "step": 30 }, { "epoch": 0.004097277293153582, "grad_norm": 6.067553529025381, "learning_rate": 3.963011889035667e-07, "loss": 1.1199, "mean_token_accuracy": 0.7103680968284607, "num_tokens": 2031616.0, "step": 31 }, { "epoch": 0.0042294475284166, "grad_norm": 6.218810066437411, "learning_rate": 4.095112285336857e-07, "loss": 1.1463, "mean_token_accuracy": 0.7032413482666016, "num_tokens": 2097152.0, "step": 32 }, { "epoch": 0.0043616177636796195, "grad_norm": 5.842525746365554, "learning_rate": 4.2272126816380454e-07, "loss": 1.0766, "mean_token_accuracy": 0.7208826541900635, "num_tokens": 2162688.0, "step": 33 }, { "epoch": 0.004493787998942638, "grad_norm": 5.620671372295368, "learning_rate": 4.359313077939234e-07, "loss": 1.1391, "mean_token_accuracy": 0.6971065998077393, "num_tokens": 2228224.0, "step": 34 }, { "epoch": 0.004625958234205657, "grad_norm": 5.583696434132944, "learning_rate": 4.4914134742404235e-07, "loss": 1.1067, "mean_token_accuracy": 0.710749626159668, "num_tokens": 2293760.0, "step": 35 }, { "epoch": 0.004758128469468675, "grad_norm": 4.688088516753559, "learning_rate": 4.623513870541612e-07, "loss": 1.1175, "mean_token_accuracy": 0.7068734169006348, "num_tokens": 2359296.0, "step": 36 }, { "epoch": 0.004890298704731694, "grad_norm": 3.944155727288372, "learning_rate": 4.7556142668428006e-07, "loss": 1.0424, "mean_token_accuracy": 0.7219356894493103, "num_tokens": 2424832.0, "step": 37 }, { "epoch": 0.0050224689399947136, "grad_norm": 3.4088892906599106, "learning_rate": 4.88771466314399e-07, "loss": 0.9575, "mean_token_accuracy": 0.7496489882469177, "num_tokens": 2490368.0, "step": 38 }, { "epoch": 0.005154639175257732, "grad_norm": 3.2618638709763745, "learning_rate": 5.019815059445179e-07, "loss": 1.0604, "mean_token_accuracy": 0.7096813321113586, "num_tokens": 2555904.0, "step": 39 }, { "epoch": 0.005286809410520751, "grad_norm": 3.712266007679404, "learning_rate": 5.151915455746368e-07, "loss": 1.0714, "mean_token_accuracy": 0.7127792835235596, "num_tokens": 2621440.0, "step": 40 }, { "epoch": 0.005418979645783769, "grad_norm": 2.85827477809128, "learning_rate": 5.284015852047556e-07, "loss": 0.9977, "mean_token_accuracy": 0.7294286489486694, "num_tokens": 2686976.0, "step": 41 }, { "epoch": 0.0055511498810467885, "grad_norm": 3.1073956750772584, "learning_rate": 5.416116248348746e-07, "loss": 1.0333, "mean_token_accuracy": 0.7258729338645935, "num_tokens": 2752512.0, "step": 42 }, { "epoch": 0.005683320116309807, "grad_norm": 3.190817507950437, "learning_rate": 5.548216644649934e-07, "loss": 1.0319, "mean_token_accuracy": 0.7236448526382446, "num_tokens": 2818048.0, "step": 43 }, { "epoch": 0.005815490351572826, "grad_norm": 3.034654849473094, "learning_rate": 5.680317040951124e-07, "loss": 1.08, "mean_token_accuracy": 0.7105359435081482, "num_tokens": 2883584.0, "step": 44 }, { "epoch": 0.005947660586835844, "grad_norm": 2.9224782406362246, "learning_rate": 5.812417437252313e-07, "loss": 0.9883, "mean_token_accuracy": 0.7297338247299194, "num_tokens": 2949120.0, "step": 45 }, { "epoch": 0.006079830822098863, "grad_norm": 2.5273790436134735, "learning_rate": 5.944517833553501e-07, "loss": 0.9925, "mean_token_accuracy": 0.731488823890686, "num_tokens": 3014656.0, "step": 46 }, { "epoch": 0.006212001057361882, "grad_norm": 2.7242956629213118, "learning_rate": 6.076618229854691e-07, "loss": 1.0685, "mean_token_accuracy": 0.7128403186798096, "num_tokens": 3080192.0, "step": 47 }, { "epoch": 0.006344171292624901, "grad_norm": 2.8802758070784376, "learning_rate": 6.208718626155879e-07, "loss": 1.0775, "mean_token_accuracy": 0.7116041779518127, "num_tokens": 3145728.0, "step": 48 }, { "epoch": 0.00647634152788792, "grad_norm": 2.299955792020621, "learning_rate": 6.340819022457068e-07, "loss": 0.9791, "mean_token_accuracy": 0.7343120574951172, "num_tokens": 3211264.0, "step": 49 }, { "epoch": 0.006608511763150938, "grad_norm": 1.8273130934916917, "learning_rate": 6.472919418758257e-07, "loss": 1.0571, "mean_token_accuracy": 0.706476628780365, "num_tokens": 3276800.0, "step": 50 }, { "epoch": 0.0067406819984139575, "grad_norm": 1.6886931474260805, "learning_rate": 6.605019815059446e-07, "loss": 0.9893, "mean_token_accuracy": 0.7291234135627747, "num_tokens": 3342336.0, "step": 51 }, { "epoch": 0.006872852233676976, "grad_norm": 1.6370475395820292, "learning_rate": 6.737120211360634e-07, "loss": 1.0628, "mean_token_accuracy": 0.7128403186798096, "num_tokens": 3407872.0, "step": 52 }, { "epoch": 0.007005022468939995, "grad_norm": 1.4536353189160003, "learning_rate": 6.869220607661824e-07, "loss": 0.995, "mean_token_accuracy": 0.7282230257987976, "num_tokens": 3473408.0, "step": 53 }, { "epoch": 0.007137192704203013, "grad_norm": 1.5100616157585254, "learning_rate": 7.001321003963013e-07, "loss": 1.0327, "mean_token_accuracy": 0.716350257396698, "num_tokens": 3538944.0, "step": 54 }, { "epoch": 0.0072693629394660324, "grad_norm": 1.4388359377055655, "learning_rate": 7.133421400264201e-07, "loss": 1.0409, "mean_token_accuracy": 0.7122451663017273, "num_tokens": 3604480.0, "step": 55 }, { "epoch": 0.007401533174729051, "grad_norm": 1.3596538775172875, "learning_rate": 7.265521796565391e-07, "loss": 0.9287, "mean_token_accuracy": 0.7452234029769897, "num_tokens": 3670016.0, "step": 56 }, { "epoch": 0.00753370340999207, "grad_norm": 1.4056840831280957, "learning_rate": 7.39762219286658e-07, "loss": 0.9769, "mean_token_accuracy": 0.7321755290031433, "num_tokens": 3735552.0, "step": 57 }, { "epoch": 0.007665873645255088, "grad_norm": 1.3660244371219448, "learning_rate": 7.529722589167768e-07, "loss": 0.9772, "mean_token_accuracy": 0.7328165173530579, "num_tokens": 3801088.0, "step": 58 }, { "epoch": 0.007798043880518107, "grad_norm": 1.2873708030814357, "learning_rate": 7.661822985468957e-07, "loss": 0.948, "mean_token_accuracy": 0.7344036102294922, "num_tokens": 3866624.0, "step": 59 }, { "epoch": 0.007930214115781126, "grad_norm": 1.3001906700079386, "learning_rate": 7.793923381770147e-07, "loss": 0.9852, "mean_token_accuracy": 0.7268496155738831, "num_tokens": 3932160.0, "step": 60 }, { "epoch": 0.008062384351044146, "grad_norm": 1.370849879948316, "learning_rate": 7.926023778071334e-07, "loss": 1.054, "mean_token_accuracy": 0.7053778767585754, "num_tokens": 3997696.0, "step": 61 }, { "epoch": 0.008194554586307164, "grad_norm": 1.2274046770147693, "learning_rate": 8.058124174372524e-07, "loss": 0.9613, "mean_token_accuracy": 0.7323892116546631, "num_tokens": 4063232.0, "step": 62 }, { "epoch": 0.008326724821570182, "grad_norm": 1.1633431521899247, "learning_rate": 8.190224570673714e-07, "loss": 0.8995, "mean_token_accuracy": 0.7525485157966614, "num_tokens": 4128768.0, "step": 63 }, { "epoch": 0.0084588950568332, "grad_norm": 1.1737035757402159, "learning_rate": 8.322324966974901e-07, "loss": 0.9435, "mean_token_accuracy": 0.7386155724525452, "num_tokens": 4194304.0, "step": 64 }, { "epoch": 0.00859106529209622, "grad_norm": 1.2164647609947612, "learning_rate": 8.454425363276091e-07, "loss": 0.9722, "mean_token_accuracy": 0.7283298969268799, "num_tokens": 4259840.0, "step": 65 }, { "epoch": 0.008723235527359239, "grad_norm": 1.1271297598243555, "learning_rate": 8.58652575957728e-07, "loss": 0.9849, "mean_token_accuracy": 0.7222256064414978, "num_tokens": 4325376.0, "step": 66 }, { "epoch": 0.008855405762622257, "grad_norm": 1.1324507469468128, "learning_rate": 8.718626155878468e-07, "loss": 0.9801, "mean_token_accuracy": 0.7276278734207153, "num_tokens": 4390912.0, "step": 67 }, { "epoch": 0.008987575997885276, "grad_norm": 1.1133189033725213, "learning_rate": 8.850726552179657e-07, "loss": 0.9473, "mean_token_accuracy": 0.7363722324371338, "num_tokens": 4456448.0, "step": 68 }, { "epoch": 0.009119746233148296, "grad_norm": 1.0435466150540147, "learning_rate": 8.982826948480847e-07, "loss": 0.941, "mean_token_accuracy": 0.7331064343452454, "num_tokens": 4521984.0, "step": 69 }, { "epoch": 0.009251916468411314, "grad_norm": 1.1049484316105527, "learning_rate": 9.114927344782034e-07, "loss": 0.9997, "mean_token_accuracy": 0.7185019850730896, "num_tokens": 4587520.0, "step": 70 }, { "epoch": 0.009384086703674332, "grad_norm": 1.0140348771137748, "learning_rate": 9.247027741083224e-07, "loss": 0.9475, "mean_token_accuracy": 0.7320229411125183, "num_tokens": 4653056.0, "step": 71 }, { "epoch": 0.00951625693893735, "grad_norm": 0.9342213186334061, "learning_rate": 9.379128137384414e-07, "loss": 0.8563, "mean_token_accuracy": 0.7617201805114746, "num_tokens": 4718592.0, "step": 72 }, { "epoch": 0.00964842717420037, "grad_norm": 0.9361677240088865, "learning_rate": 9.511228533685601e-07, "loss": 0.8835, "mean_token_accuracy": 0.7502442002296448, "num_tokens": 4784128.0, "step": 73 }, { "epoch": 0.009780597409463389, "grad_norm": 0.9736155278924413, "learning_rate": 9.64332892998679e-07, "loss": 0.9569, "mean_token_accuracy": 0.726284921169281, "num_tokens": 4849664.0, "step": 74 }, { "epoch": 0.009912767644726407, "grad_norm": 0.9837630705056091, "learning_rate": 9.77542932628798e-07, "loss": 0.9593, "mean_token_accuracy": 0.7262086272239685, "num_tokens": 4915200.0, "step": 75 }, { "epoch": 0.010044937879989427, "grad_norm": 0.8946493094277936, "learning_rate": 9.907529722589168e-07, "loss": 0.8638, "mean_token_accuracy": 0.7511292695999146, "num_tokens": 4980736.0, "step": 76 }, { "epoch": 0.010177108115252445, "grad_norm": 0.9441929300417938, "learning_rate": 1.0039630118890358e-06, "loss": 0.9864, "mean_token_accuracy": 0.7187004089355469, "num_tokens": 5046272.0, "step": 77 }, { "epoch": 0.010309278350515464, "grad_norm": 0.9082959440881991, "learning_rate": 1.0171730515191547e-06, "loss": 0.9093, "mean_token_accuracy": 0.7393480539321899, "num_tokens": 5111808.0, "step": 78 }, { "epoch": 0.010441448585778482, "grad_norm": 0.8525971904448233, "learning_rate": 1.0303830911492736e-06, "loss": 0.9037, "mean_token_accuracy": 0.7359907031059265, "num_tokens": 5177344.0, "step": 79 }, { "epoch": 0.010573618821041502, "grad_norm": 0.8406455973339398, "learning_rate": 1.0435931307793924e-06, "loss": 0.8803, "mean_token_accuracy": 0.749847412109375, "num_tokens": 5242880.0, "step": 80 }, { "epoch": 0.01070578905630452, "grad_norm": 0.8608474494064567, "learning_rate": 1.0568031704095113e-06, "loss": 0.9418, "mean_token_accuracy": 0.7286198139190674, "num_tokens": 5308416.0, "step": 81 }, { "epoch": 0.010837959291567539, "grad_norm": 0.8541023753737392, "learning_rate": 1.0700132100396301e-06, "loss": 0.9473, "mean_token_accuracy": 0.7230802178382874, "num_tokens": 5373952.0, "step": 82 }, { "epoch": 0.010970129526830557, "grad_norm": 0.8732108060019078, "learning_rate": 1.0832232496697492e-06, "loss": 0.94, "mean_token_accuracy": 0.7307868599891663, "num_tokens": 5439488.0, "step": 83 }, { "epoch": 0.011102299762093577, "grad_norm": 0.8103886777629647, "learning_rate": 1.096433289299868e-06, "loss": 0.8806, "mean_token_accuracy": 0.7434532046318054, "num_tokens": 5505024.0, "step": 84 }, { "epoch": 0.011234469997356595, "grad_norm": 0.8457127504662689, "learning_rate": 1.1096433289299869e-06, "loss": 0.8914, "mean_token_accuracy": 0.7421102523803711, "num_tokens": 5570560.0, "step": 85 }, { "epoch": 0.011366640232619614, "grad_norm": 0.8311653447548312, "learning_rate": 1.1228533685601057e-06, "loss": 0.934, "mean_token_accuracy": 0.7344341278076172, "num_tokens": 5636096.0, "step": 86 }, { "epoch": 0.011498810467882634, "grad_norm": 0.8430781567300121, "learning_rate": 1.1360634081902248e-06, "loss": 0.905, "mean_token_accuracy": 0.7408893704414368, "num_tokens": 5701632.0, "step": 87 }, { "epoch": 0.011630980703145652, "grad_norm": 0.8376456510339795, "learning_rate": 1.1492734478203434e-06, "loss": 0.9564, "mean_token_accuracy": 0.7225155830383301, "num_tokens": 5767168.0, "step": 88 }, { "epoch": 0.01176315093840867, "grad_norm": 0.7847804139810242, "learning_rate": 1.1624834874504625e-06, "loss": 0.936, "mean_token_accuracy": 0.731214165687561, "num_tokens": 5832704.0, "step": 89 }, { "epoch": 0.011895321173671689, "grad_norm": 0.7931066437831544, "learning_rate": 1.1756935270805814e-06, "loss": 0.9144, "mean_token_accuracy": 0.7337474226951599, "num_tokens": 5898240.0, "step": 90 }, { "epoch": 0.012027491408934709, "grad_norm": 0.7678763695649288, "learning_rate": 1.1889035667107002e-06, "loss": 0.8798, "mean_token_accuracy": 0.7453454732894897, "num_tokens": 5963776.0, "step": 91 }, { "epoch": 0.012159661644197727, "grad_norm": 0.8252597838926503, "learning_rate": 1.202113606340819e-06, "loss": 0.909, "mean_token_accuracy": 0.7366469502449036, "num_tokens": 6029312.0, "step": 92 }, { "epoch": 0.012291831879460745, "grad_norm": 0.7710129639070472, "learning_rate": 1.2153236459709381e-06, "loss": 0.8477, "mean_token_accuracy": 0.7530826330184937, "num_tokens": 6094848.0, "step": 93 }, { "epoch": 0.012424002114723763, "grad_norm": 0.8163925541175688, "learning_rate": 1.2285336856010568e-06, "loss": 0.8986, "mean_token_accuracy": 0.7384018898010254, "num_tokens": 6160384.0, "step": 94 }, { "epoch": 0.012556172349986783, "grad_norm": 0.8253101441918174, "learning_rate": 1.2417437252311758e-06, "loss": 0.9133, "mean_token_accuracy": 0.7352887392044067, "num_tokens": 6225920.0, "step": 95 }, { "epoch": 0.012688342585249802, "grad_norm": 0.765668665869954, "learning_rate": 1.2549537648612947e-06, "loss": 0.8728, "mean_token_accuracy": 0.7450860738754272, "num_tokens": 6291456.0, "step": 96 }, { "epoch": 0.01282051282051282, "grad_norm": 0.7800072717525395, "learning_rate": 1.2681638044914136e-06, "loss": 0.929, "mean_token_accuracy": 0.7280246615409851, "num_tokens": 6356992.0, "step": 97 }, { "epoch": 0.01295268305577584, "grad_norm": 0.7834134475051282, "learning_rate": 1.2813738441215326e-06, "loss": 0.9254, "mean_token_accuracy": 0.7311988472938538, "num_tokens": 6422528.0, "step": 98 }, { "epoch": 0.013084853291038858, "grad_norm": 0.7347659712555972, "learning_rate": 1.2945838837516515e-06, "loss": 0.8523, "mean_token_accuracy": 0.7514039874076843, "num_tokens": 6488064.0, "step": 99 }, { "epoch": 0.013217023526301877, "grad_norm": 0.8259396129218242, "learning_rate": 1.3077939233817701e-06, "loss": 0.9136, "mean_token_accuracy": 0.731305718421936, "num_tokens": 6553600.0, "step": 100 }, { "epoch": 0.013349193761564895, "grad_norm": 0.696184875743928, "learning_rate": 1.3210039630118892e-06, "loss": 0.8067, "mean_token_accuracy": 0.7654132843017578, "num_tokens": 6619136.0, "step": 101 }, { "epoch": 0.013481363996827915, "grad_norm": 0.7409308148014699, "learning_rate": 1.334214002642008e-06, "loss": 0.889, "mean_token_accuracy": 0.7436057925224304, "num_tokens": 6684672.0, "step": 102 }, { "epoch": 0.013613534232090933, "grad_norm": 0.7778519466298204, "learning_rate": 1.3474240422721269e-06, "loss": 0.944, "mean_token_accuracy": 0.725888192653656, "num_tokens": 6750208.0, "step": 103 }, { "epoch": 0.013745704467353952, "grad_norm": 0.7788415223735142, "learning_rate": 1.360634081902246e-06, "loss": 0.9368, "mean_token_accuracy": 0.7246825695037842, "num_tokens": 6815744.0, "step": 104 }, { "epoch": 0.01387787470261697, "grad_norm": 0.733464533735661, "learning_rate": 1.3738441215323648e-06, "loss": 0.9066, "mean_token_accuracy": 0.7374709844589233, "num_tokens": 6881280.0, "step": 105 }, { "epoch": 0.01401004493787999, "grad_norm": 0.7633797658675768, "learning_rate": 1.3870541611624834e-06, "loss": 0.8931, "mean_token_accuracy": 0.7331369519233704, "num_tokens": 6946816.0, "step": 106 }, { "epoch": 0.014142215173143008, "grad_norm": 0.7267903751935816, "learning_rate": 1.4002642007926025e-06, "loss": 0.9005, "mean_token_accuracy": 0.7335489988327026, "num_tokens": 7012352.0, "step": 107 }, { "epoch": 0.014274385408406027, "grad_norm": 0.7643446738040165, "learning_rate": 1.4134742404227214e-06, "loss": 0.8908, "mean_token_accuracy": 0.7413777112960815, "num_tokens": 7077888.0, "step": 108 }, { "epoch": 0.014406555643669047, "grad_norm": 0.6772256860008208, "learning_rate": 1.4266842800528402e-06, "loss": 0.8463, "mean_token_accuracy": 0.7537236213684082, "num_tokens": 7143424.0, "step": 109 }, { "epoch": 0.014538725878932065, "grad_norm": 0.7609062026270402, "learning_rate": 1.4398943196829593e-06, "loss": 0.8842, "mean_token_accuracy": 0.7386766076087952, "num_tokens": 7208960.0, "step": 110 }, { "epoch": 0.014670896114195083, "grad_norm": 0.7225536354591765, "learning_rate": 1.4531043593130781e-06, "loss": 0.88, "mean_token_accuracy": 0.7402331829071045, "num_tokens": 7274496.0, "step": 111 }, { "epoch": 0.014803066349458101, "grad_norm": 0.7780220914766722, "learning_rate": 1.4663143989431968e-06, "loss": 0.8615, "mean_token_accuracy": 0.7509766817092896, "num_tokens": 7340032.0, "step": 112 }, { "epoch": 0.014935236584721122, "grad_norm": 0.7584207358035956, "learning_rate": 1.479524438573316e-06, "loss": 0.8626, "mean_token_accuracy": 0.7442772388458252, "num_tokens": 7405568.0, "step": 113 }, { "epoch": 0.01506740681998414, "grad_norm": 0.7163184678477569, "learning_rate": 1.4927344782034347e-06, "loss": 0.8717, "mean_token_accuracy": 0.7387834191322327, "num_tokens": 7471104.0, "step": 114 }, { "epoch": 0.015199577055247158, "grad_norm": 0.730280213325568, "learning_rate": 1.5059445178335536e-06, "loss": 0.8406, "mean_token_accuracy": 0.7508546113967896, "num_tokens": 7536640.0, "step": 115 }, { "epoch": 0.015331747290510176, "grad_norm": 0.7054418606824134, "learning_rate": 1.5191545574636726e-06, "loss": 0.8733, "mean_token_accuracy": 0.7415303587913513, "num_tokens": 7602176.0, "step": 116 }, { "epoch": 0.015463917525773196, "grad_norm": 0.7344535005657318, "learning_rate": 1.5323645970937915e-06, "loss": 0.8746, "mean_token_accuracy": 0.7372115850448608, "num_tokens": 7667712.0, "step": 117 }, { "epoch": 0.015596087761036215, "grad_norm": 0.7250817490786946, "learning_rate": 1.5455746367239103e-06, "loss": 0.8756, "mean_token_accuracy": 0.7432852983474731, "num_tokens": 7733248.0, "step": 118 }, { "epoch": 0.015728257996299233, "grad_norm": 0.7693856296766783, "learning_rate": 1.5587846763540294e-06, "loss": 0.9179, "mean_token_accuracy": 0.7280399203300476, "num_tokens": 7798784.0, "step": 119 }, { "epoch": 0.01586042823156225, "grad_norm": 0.7409731580889363, "learning_rate": 1.571994715984148e-06, "loss": 0.8644, "mean_token_accuracy": 0.7407978177070618, "num_tokens": 7864320.0, "step": 120 }, { "epoch": 0.01599259846682527, "grad_norm": 0.7091728049779311, "learning_rate": 1.5852047556142669e-06, "loss": 0.8192, "mean_token_accuracy": 0.7562568783760071, "num_tokens": 7929856.0, "step": 121 }, { "epoch": 0.01612476870208829, "grad_norm": 0.7645451723592865, "learning_rate": 1.598414795244386e-06, "loss": 0.8394, "mean_token_accuracy": 0.7513887286186218, "num_tokens": 7995392.0, "step": 122 }, { "epoch": 0.01625693893735131, "grad_norm": 0.7424407841223357, "learning_rate": 1.6116248348745048e-06, "loss": 0.8706, "mean_token_accuracy": 0.7409809827804565, "num_tokens": 8060928.0, "step": 123 }, { "epoch": 0.016389109172614328, "grad_norm": 0.757890891705412, "learning_rate": 1.6248348745046237e-06, "loss": 0.8886, "mean_token_accuracy": 0.7355023622512817, "num_tokens": 8126464.0, "step": 124 }, { "epoch": 0.016521279407877346, "grad_norm": 0.7853282693829975, "learning_rate": 1.6380449141347427e-06, "loss": 0.937, "mean_token_accuracy": 0.720806360244751, "num_tokens": 8192000.0, "step": 125 }, { "epoch": 0.016653449643140365, "grad_norm": 0.7249840691226063, "learning_rate": 1.6512549537648614e-06, "loss": 0.8445, "mean_token_accuracy": 0.7489928007125854, "num_tokens": 8257536.0, "step": 126 }, { "epoch": 0.016785619878403383, "grad_norm": 0.7161037903754257, "learning_rate": 1.6644649933949802e-06, "loss": 0.8827, "mean_token_accuracy": 0.7360517382621765, "num_tokens": 8323072.0, "step": 127 }, { "epoch": 0.0169177901136664, "grad_norm": 0.709625479934526, "learning_rate": 1.6776750330250993e-06, "loss": 0.9001, "mean_token_accuracy": 0.7337626814842224, "num_tokens": 8388608.0, "step": 128 }, { "epoch": 0.01704996034892942, "grad_norm": 0.717977471732701, "learning_rate": 1.6908850726552181e-06, "loss": 0.8957, "mean_token_accuracy": 0.7344493865966797, "num_tokens": 8454144.0, "step": 129 }, { "epoch": 0.01718213058419244, "grad_norm": 0.7611425621021748, "learning_rate": 1.704095112285337e-06, "loss": 0.8821, "mean_token_accuracy": 0.7405994534492493, "num_tokens": 8519680.0, "step": 130 }, { "epoch": 0.01731430081945546, "grad_norm": 0.6872765931282789, "learning_rate": 1.717305151915456e-06, "loss": 0.8004, "mean_token_accuracy": 0.7653064131736755, "num_tokens": 8585216.0, "step": 131 }, { "epoch": 0.017446471054718478, "grad_norm": 0.6916386075663681, "learning_rate": 1.7305151915455747e-06, "loss": 0.8314, "mean_token_accuracy": 0.7548986673355103, "num_tokens": 8650752.0, "step": 132 }, { "epoch": 0.017578641289981496, "grad_norm": 0.7201869596362779, "learning_rate": 1.7437252311756936e-06, "loss": 0.8309, "mean_token_accuracy": 0.7539525032043457, "num_tokens": 8716288.0, "step": 133 }, { "epoch": 0.017710811525244514, "grad_norm": 0.7428115741343573, "learning_rate": 1.7569352708058126e-06, "loss": 0.848, "mean_token_accuracy": 0.7475125193595886, "num_tokens": 8781824.0, "step": 134 }, { "epoch": 0.017842981760507533, "grad_norm": 0.7354543020678891, "learning_rate": 1.7701453104359315e-06, "loss": 0.8064, "mean_token_accuracy": 0.7611860632896423, "num_tokens": 8847360.0, "step": 135 }, { "epoch": 0.01797515199577055, "grad_norm": 0.7173735624310742, "learning_rate": 1.7833553500660503e-06, "loss": 0.8643, "mean_token_accuracy": 0.7417134642601013, "num_tokens": 8912896.0, "step": 136 }, { "epoch": 0.018107322231033573, "grad_norm": 0.7039234038441876, "learning_rate": 1.7965653896961694e-06, "loss": 0.7757, "mean_token_accuracy": 0.7678702473640442, "num_tokens": 8978432.0, "step": 137 }, { "epoch": 0.01823949246629659, "grad_norm": 0.7524570417419832, "learning_rate": 1.809775429326288e-06, "loss": 0.8415, "mean_token_accuracy": 0.7514650225639343, "num_tokens": 9043968.0, "step": 138 }, { "epoch": 0.01837166270155961, "grad_norm": 0.7341665927461095, "learning_rate": 1.8229854689564069e-06, "loss": 0.8575, "mean_token_accuracy": 0.7472835779190063, "num_tokens": 9109504.0, "step": 139 }, { "epoch": 0.018503832936822628, "grad_norm": 0.7034133841500246, "learning_rate": 1.836195508586526e-06, "loss": 0.8095, "mean_token_accuracy": 0.7555548548698425, "num_tokens": 9175040.0, "step": 140 }, { "epoch": 0.018636003172085646, "grad_norm": 0.757168652500877, "learning_rate": 1.8494055482166448e-06, "loss": 0.8715, "mean_token_accuracy": 0.7405079007148743, "num_tokens": 9240576.0, "step": 141 }, { "epoch": 0.018768173407348664, "grad_norm": 0.7451104664175263, "learning_rate": 1.8626155878467637e-06, "loss": 0.8649, "mean_token_accuracy": 0.7421102523803711, "num_tokens": 9306112.0, "step": 142 }, { "epoch": 0.018900343642611683, "grad_norm": 0.7482632623511752, "learning_rate": 1.8758256274768827e-06, "loss": 0.8249, "mean_token_accuracy": 0.7554938197135925, "num_tokens": 9371648.0, "step": 143 }, { "epoch": 0.0190325138778747, "grad_norm": 0.7169049812067287, "learning_rate": 1.8890356671070016e-06, "loss": 0.8391, "mean_token_accuracy": 0.7514497637748718, "num_tokens": 9437184.0, "step": 144 }, { "epoch": 0.019164684113137723, "grad_norm": 0.7840028002580148, "learning_rate": 1.9022457067371202e-06, "loss": 0.8632, "mean_token_accuracy": 0.7407978177070618, "num_tokens": 9502720.0, "step": 145 }, { "epoch": 0.01929685434840074, "grad_norm": 0.751976730811151, "learning_rate": 1.9154557463672395e-06, "loss": 0.8197, "mean_token_accuracy": 0.7525332570075989, "num_tokens": 9568256.0, "step": 146 }, { "epoch": 0.01942902458366376, "grad_norm": 0.7728936585269056, "learning_rate": 1.928665785997358e-06, "loss": 0.8385, "mean_token_accuracy": 0.7471462488174438, "num_tokens": 9633792.0, "step": 147 }, { "epoch": 0.019561194818926778, "grad_norm": 0.7296169074733622, "learning_rate": 1.9418758256274768e-06, "loss": 0.7878, "mean_token_accuracy": 0.7613539099693298, "num_tokens": 9699328.0, "step": 148 }, { "epoch": 0.019693365054189796, "grad_norm": 0.7437287829533681, "learning_rate": 1.955085865257596e-06, "loss": 0.8466, "mean_token_accuracy": 0.7513887286186218, "num_tokens": 9764864.0, "step": 149 }, { "epoch": 0.019825535289452814, "grad_norm": 0.7300176221580168, "learning_rate": 1.968295904887715e-06, "loss": 0.8194, "mean_token_accuracy": 0.7548681497573853, "num_tokens": 9830400.0, "step": 150 }, { "epoch": 0.019957705524715832, "grad_norm": 0.6855477463238256, "learning_rate": 1.9815059445178336e-06, "loss": 0.8287, "mean_token_accuracy": 0.7540440559387207, "num_tokens": 9895936.0, "step": 151 }, { "epoch": 0.020089875759978854, "grad_norm": 0.7160434127250255, "learning_rate": 1.9947159841479526e-06, "loss": 0.8026, "mean_token_accuracy": 0.7601941227912903, "num_tokens": 9961472.0, "step": 152 }, { "epoch": 0.020222045995241873, "grad_norm": 0.756620779427381, "learning_rate": 2.0079260237780717e-06, "loss": 0.8662, "mean_token_accuracy": 0.7397295832633972, "num_tokens": 10027008.0, "step": 153 }, { "epoch": 0.02035421623050489, "grad_norm": 0.7161792888717843, "learning_rate": 2.0211360634081903e-06, "loss": 0.8189, "mean_token_accuracy": 0.7562568783760071, "num_tokens": 10092544.0, "step": 154 }, { "epoch": 0.02048638646576791, "grad_norm": 0.7223239903986668, "learning_rate": 2.0343461030383094e-06, "loss": 0.7906, "mean_token_accuracy": 0.7672140002250671, "num_tokens": 10158080.0, "step": 155 }, { "epoch": 0.020618556701030927, "grad_norm": 0.7295178067082708, "learning_rate": 2.047556142668428e-06, "loss": 0.7749, "mean_token_accuracy": 0.7662373185157776, "num_tokens": 10223616.0, "step": 156 }, { "epoch": 0.020750726936293946, "grad_norm": 0.7033934649323509, "learning_rate": 2.060766182298547e-06, "loss": 0.7906, "mean_token_accuracy": 0.7612013220787048, "num_tokens": 10289152.0, "step": 157 }, { "epoch": 0.020882897171556964, "grad_norm": 0.7714248017323682, "learning_rate": 2.073976221928666e-06, "loss": 0.8809, "mean_token_accuracy": 0.7308631539344788, "num_tokens": 10354688.0, "step": 158 }, { "epoch": 0.021015067406819986, "grad_norm": 0.7267939427010185, "learning_rate": 2.087186261558785e-06, "loss": 0.8413, "mean_token_accuracy": 0.7505951523780823, "num_tokens": 10420224.0, "step": 159 }, { "epoch": 0.021147237642083004, "grad_norm": 0.6926840522099964, "learning_rate": 2.1003963011889035e-06, "loss": 0.7986, "mean_token_accuracy": 0.7612928748130798, "num_tokens": 10485760.0, "step": 160 }, { "epoch": 0.021279407877346022, "grad_norm": 0.6803683006365302, "learning_rate": 2.1136063408190225e-06, "loss": 0.7373, "mean_token_accuracy": 0.7798956036567688, "num_tokens": 10551296.0, "step": 161 }, { "epoch": 0.02141157811260904, "grad_norm": 0.7301445690185864, "learning_rate": 2.1268163804491416e-06, "loss": 0.8019, "mean_token_accuracy": 0.758851170539856, "num_tokens": 10616832.0, "step": 162 }, { "epoch": 0.02154374834787206, "grad_norm": 0.7608866124324192, "learning_rate": 2.1400264200792602e-06, "loss": 0.8337, "mean_token_accuracy": 0.7478482723236084, "num_tokens": 10682368.0, "step": 163 }, { "epoch": 0.021675918583135077, "grad_norm": 0.7237344170367529, "learning_rate": 2.1532364597093793e-06, "loss": 0.7998, "mean_token_accuracy": 0.7569283246994019, "num_tokens": 10747904.0, "step": 164 }, { "epoch": 0.021808088818398096, "grad_norm": 0.7392523985813007, "learning_rate": 2.1664464993394984e-06, "loss": 0.7699, "mean_token_accuracy": 0.7671834826469421, "num_tokens": 10813440.0, "step": 165 }, { "epoch": 0.021940259053661114, "grad_norm": 0.6696173473328371, "learning_rate": 2.179656538969617e-06, "loss": 0.7985, "mean_token_accuracy": 0.7582255005836487, "num_tokens": 10878976.0, "step": 166 }, { "epoch": 0.022072429288924136, "grad_norm": 0.7138637661283654, "learning_rate": 2.192866578599736e-06, "loss": 0.7973, "mean_token_accuracy": 0.7612013220787048, "num_tokens": 10944512.0, "step": 167 }, { "epoch": 0.022204599524187154, "grad_norm": 0.7550641944545804, "learning_rate": 2.2060766182298547e-06, "loss": 0.8538, "mean_token_accuracy": 0.7430258989334106, "num_tokens": 11010048.0, "step": 168 }, { "epoch": 0.022336769759450172, "grad_norm": 0.7688619154383091, "learning_rate": 2.2192866578599738e-06, "loss": 0.8761, "mean_token_accuracy": 0.7371047735214233, "num_tokens": 11075584.0, "step": 169 }, { "epoch": 0.02246893999471319, "grad_norm": 0.7730721194852236, "learning_rate": 2.232496697490093e-06, "loss": 0.8163, "mean_token_accuracy": 0.7537236213684082, "num_tokens": 11141120.0, "step": 170 }, { "epoch": 0.02260111022997621, "grad_norm": 0.7489275808059612, "learning_rate": 2.2457067371202115e-06, "loss": 0.8573, "mean_token_accuracy": 0.7428579926490784, "num_tokens": 11206656.0, "step": 171 }, { "epoch": 0.022733280465239227, "grad_norm": 0.759385860381698, "learning_rate": 2.2589167767503305e-06, "loss": 0.8114, "mean_token_accuracy": 0.7543492913246155, "num_tokens": 11272192.0, "step": 172 }, { "epoch": 0.022865450700502245, "grad_norm": 0.7305776370029441, "learning_rate": 2.2721268163804496e-06, "loss": 0.8583, "mean_token_accuracy": 0.7429496049880981, "num_tokens": 11337728.0, "step": 173 }, { "epoch": 0.022997620935765267, "grad_norm": 0.7146889374434929, "learning_rate": 2.2853368560105682e-06, "loss": 0.7946, "mean_token_accuracy": 0.7610639929771423, "num_tokens": 11403264.0, "step": 174 }, { "epoch": 0.023129791171028286, "grad_norm": 0.7561430706192427, "learning_rate": 2.298546895640687e-06, "loss": 0.8276, "mean_token_accuracy": 0.7516023516654968, "num_tokens": 11468800.0, "step": 175 }, { "epoch": 0.023261961406291304, "grad_norm": 0.765564717455702, "learning_rate": 2.311756935270806e-06, "loss": 0.8678, "mean_token_accuracy": 0.7350903153419495, "num_tokens": 11534336.0, "step": 176 }, { "epoch": 0.023394131641554322, "grad_norm": 0.7390946820804956, "learning_rate": 2.324966974900925e-06, "loss": 0.8138, "mean_token_accuracy": 0.7526400685310364, "num_tokens": 11599872.0, "step": 177 }, { "epoch": 0.02352630187681734, "grad_norm": 0.7075473430340745, "learning_rate": 2.3381770145310437e-06, "loss": 0.7686, "mean_token_accuracy": 0.773989737033844, "num_tokens": 11665408.0, "step": 178 }, { "epoch": 0.02365847211208036, "grad_norm": 0.7325726585721395, "learning_rate": 2.3513870541611627e-06, "loss": 0.7877, "mean_token_accuracy": 0.7652453780174255, "num_tokens": 11730944.0, "step": 179 }, { "epoch": 0.023790642347343377, "grad_norm": 0.7099949221444268, "learning_rate": 2.3645970937912814e-06, "loss": 0.7896, "mean_token_accuracy": 0.7602856755256653, "num_tokens": 11796480.0, "step": 180 }, { "epoch": 0.023922812582606395, "grad_norm": 0.7271040479289158, "learning_rate": 2.3778071334214004e-06, "loss": 0.8143, "mean_token_accuracy": 0.7537541389465332, "num_tokens": 11862016.0, "step": 181 }, { "epoch": 0.024054982817869417, "grad_norm": 0.7244938914523293, "learning_rate": 2.3910171730515195e-06, "loss": 0.7953, "mean_token_accuracy": 0.7637498378753662, "num_tokens": 11927552.0, "step": 182 }, { "epoch": 0.024187153053132435, "grad_norm": 0.692347393450686, "learning_rate": 2.404227212681638e-06, "loss": 0.7725, "mean_token_accuracy": 0.7665730714797974, "num_tokens": 11993088.0, "step": 183 }, { "epoch": 0.024319323288395454, "grad_norm": 0.7626537046166831, "learning_rate": 2.417437252311757e-06, "loss": 0.7536, "mean_token_accuracy": 0.7713801860809326, "num_tokens": 12058624.0, "step": 184 }, { "epoch": 0.024451493523658472, "grad_norm": 0.7236166663175728, "learning_rate": 2.4306472919418763e-06, "loss": 0.8052, "mean_token_accuracy": 0.7583628296852112, "num_tokens": 12124160.0, "step": 185 }, { "epoch": 0.02458366375892149, "grad_norm": 0.7467941027216998, "learning_rate": 2.443857331571995e-06, "loss": 0.792, "mean_token_accuracy": 0.758881688117981, "num_tokens": 12189696.0, "step": 186 }, { "epoch": 0.02471583399418451, "grad_norm": 0.7510723021233771, "learning_rate": 2.4570673712021136e-06, "loss": 0.8331, "mean_token_accuracy": 0.7464748024940491, "num_tokens": 12255232.0, "step": 187 }, { "epoch": 0.024848004229447527, "grad_norm": 0.8190386784303135, "learning_rate": 2.4702774108322326e-06, "loss": 0.7918, "mean_token_accuracy": 0.7612623572349548, "num_tokens": 12320768.0, "step": 188 }, { "epoch": 0.02498017446471055, "grad_norm": 0.7226337462447449, "learning_rate": 2.4834874504623517e-06, "loss": 0.7586, "mean_token_accuracy": 0.7695794105529785, "num_tokens": 12386304.0, "step": 189 }, { "epoch": 0.025112344699973567, "grad_norm": 0.782511039355095, "learning_rate": 2.4966974900924703e-06, "loss": 0.8048, "mean_token_accuracy": 0.7548071146011353, "num_tokens": 12451840.0, "step": 190 }, { "epoch": 0.025244514935236585, "grad_norm": 0.73799158866485, "learning_rate": 2.5099075297225894e-06, "loss": 0.8012, "mean_token_accuracy": 0.7567757368087769, "num_tokens": 12517376.0, "step": 191 }, { "epoch": 0.025376685170499604, "grad_norm": 0.7129513532555495, "learning_rate": 2.523117569352708e-06, "loss": 0.8076, "mean_token_accuracy": 0.7566841840744019, "num_tokens": 12582912.0, "step": 192 }, { "epoch": 0.025508855405762622, "grad_norm": 0.7399902223014233, "learning_rate": 2.536327608982827e-06, "loss": 0.8226, "mean_token_accuracy": 0.7526248097419739, "num_tokens": 12648448.0, "step": 193 }, { "epoch": 0.02564102564102564, "grad_norm": 0.7732238089258137, "learning_rate": 2.5495376486129457e-06, "loss": 0.7736, "mean_token_accuracy": 0.7655353546142578, "num_tokens": 12713984.0, "step": 194 }, { "epoch": 0.02577319587628866, "grad_norm": 0.7766725449949912, "learning_rate": 2.5627476882430652e-06, "loss": 0.7817, "mean_token_accuracy": 0.7632309794425964, "num_tokens": 12779520.0, "step": 195 }, { "epoch": 0.02590536611155168, "grad_norm": 0.6966504055642281, "learning_rate": 2.575957727873184e-06, "loss": 0.7675, "mean_token_accuracy": 0.7668324708938599, "num_tokens": 12845056.0, "step": 196 }, { "epoch": 0.0260375363468147, "grad_norm": 0.8004707627981364, "learning_rate": 2.589167767503303e-06, "loss": 0.7956, "mean_token_accuracy": 0.7596447467803955, "num_tokens": 12910592.0, "step": 197 }, { "epoch": 0.026169706582077717, "grad_norm": 0.7800020209616247, "learning_rate": 2.6023778071334216e-06, "loss": 0.7696, "mean_token_accuracy": 0.7634751796722412, "num_tokens": 12976128.0, "step": 198 }, { "epoch": 0.026301876817340735, "grad_norm": 0.7392216791931434, "learning_rate": 2.6155878467635402e-06, "loss": 0.7822, "mean_token_accuracy": 0.7671987414360046, "num_tokens": 13041664.0, "step": 199 }, { "epoch": 0.026434047052603753, "grad_norm": 0.7494411643230846, "learning_rate": 2.6287978863936593e-06, "loss": 0.7805, "mean_token_accuracy": 0.7617049217224121, "num_tokens": 13107200.0, "step": 200 }, { "epoch": 0.02656621728786677, "grad_norm": 0.7367926384927534, "learning_rate": 2.6420079260237784e-06, "loss": 0.806, "mean_token_accuracy": 0.7534336447715759, "num_tokens": 13172736.0, "step": 201 }, { "epoch": 0.02669838752312979, "grad_norm": 0.7095891168305133, "learning_rate": 2.6552179656538974e-06, "loss": 0.767, "mean_token_accuracy": 0.7671377062797546, "num_tokens": 13238272.0, "step": 202 }, { "epoch": 0.02683055775839281, "grad_norm": 0.7342382484019445, "learning_rate": 2.668428005284016e-06, "loss": 0.77, "mean_token_accuracy": 0.7702508568763733, "num_tokens": 13303808.0, "step": 203 }, { "epoch": 0.02696272799365583, "grad_norm": 0.6856032201011079, "learning_rate": 2.681638044914135e-06, "loss": 0.761, "mean_token_accuracy": 0.7726467847824097, "num_tokens": 13369344.0, "step": 204 }, { "epoch": 0.02709489822891885, "grad_norm": 0.7386335937637796, "learning_rate": 2.6948480845442538e-06, "loss": 0.7765, "mean_token_accuracy": 0.7630783915519714, "num_tokens": 13434880.0, "step": 205 }, { "epoch": 0.027227068464181867, "grad_norm": 0.7857040795781662, "learning_rate": 2.7080581241743724e-06, "loss": 0.7986, "mean_token_accuracy": 0.7572335600852966, "num_tokens": 13500416.0, "step": 206 }, { "epoch": 0.027359238699444885, "grad_norm": 0.8042749309956072, "learning_rate": 2.721268163804492e-06, "loss": 0.8933, "mean_token_accuracy": 0.7255829572677612, "num_tokens": 13565952.0, "step": 207 }, { "epoch": 0.027491408934707903, "grad_norm": 0.7884984846085789, "learning_rate": 2.7344782034346105e-06, "loss": 0.8, "mean_token_accuracy": 0.7513734698295593, "num_tokens": 13631488.0, "step": 208 }, { "epoch": 0.02762357916997092, "grad_norm": 0.8173438302461916, "learning_rate": 2.7476882430647296e-06, "loss": 0.8103, "mean_token_accuracy": 0.7516328692436218, "num_tokens": 13697024.0, "step": 209 }, { "epoch": 0.02775574940523394, "grad_norm": 0.7491783948329647, "learning_rate": 2.7608982826948483e-06, "loss": 0.7987, "mean_token_accuracy": 0.7549902200698853, "num_tokens": 13762560.0, "step": 210 }, { "epoch": 0.02788791964049696, "grad_norm": 0.7993251624251987, "learning_rate": 2.774108322324967e-06, "loss": 0.8227, "mean_token_accuracy": 0.7496184706687927, "num_tokens": 13828096.0, "step": 211 }, { "epoch": 0.02802008987575998, "grad_norm": 0.7412940564728008, "learning_rate": 2.787318361955086e-06, "loss": 0.7842, "mean_token_accuracy": 0.7608197927474976, "num_tokens": 13893632.0, "step": 212 }, { "epoch": 0.028152260111022998, "grad_norm": 0.7778621669671549, "learning_rate": 2.800528401585205e-06, "loss": 0.7714, "mean_token_accuracy": 0.7626663446426392, "num_tokens": 13959168.0, "step": 213 }, { "epoch": 0.028284430346286017, "grad_norm": 0.7494896164725238, "learning_rate": 2.813738441215324e-06, "loss": 0.7952, "mean_token_accuracy": 0.7571877837181091, "num_tokens": 14024704.0, "step": 214 }, { "epoch": 0.028416600581549035, "grad_norm": 0.8293792649180755, "learning_rate": 2.8269484808454427e-06, "loss": 0.7937, "mean_token_accuracy": 0.758729100227356, "num_tokens": 14090240.0, "step": 215 }, { "epoch": 0.028548770816812053, "grad_norm": 0.7967839171526191, "learning_rate": 2.840158520475562e-06, "loss": 0.803, "mean_token_accuracy": 0.7561653256416321, "num_tokens": 14155776.0, "step": 216 }, { "epoch": 0.02868094105207507, "grad_norm": 0.7698044670136173, "learning_rate": 2.8533685601056804e-06, "loss": 0.8302, "mean_token_accuracy": 0.7507019639015198, "num_tokens": 14221312.0, "step": 217 }, { "epoch": 0.028813111287338093, "grad_norm": 0.8314135816686421, "learning_rate": 2.866578599735799e-06, "loss": 0.8302, "mean_token_accuracy": 0.7464442849159241, "num_tokens": 14286848.0, "step": 218 }, { "epoch": 0.02894528152260111, "grad_norm": 0.7631844030754845, "learning_rate": 2.8797886393659186e-06, "loss": 0.7495, "mean_token_accuracy": 0.7717617154121399, "num_tokens": 14352384.0, "step": 219 }, { "epoch": 0.02907745175786413, "grad_norm": 0.7587201067062704, "learning_rate": 2.8929986789960372e-06, "loss": 0.8187, "mean_token_accuracy": 0.7503967881202698, "num_tokens": 14417920.0, "step": 220 }, { "epoch": 0.029209621993127148, "grad_norm": 0.8249983948928367, "learning_rate": 2.9062087186261563e-06, "loss": 0.8157, "mean_token_accuracy": 0.7501373291015625, "num_tokens": 14483456.0, "step": 221 }, { "epoch": 0.029341792228390166, "grad_norm": 0.7608513828539731, "learning_rate": 2.919418758256275e-06, "loss": 0.7778, "mean_token_accuracy": 0.7628494501113892, "num_tokens": 14548992.0, "step": 222 }, { "epoch": 0.029473962463653185, "grad_norm": 0.7315677069441571, "learning_rate": 2.9326287978863936e-06, "loss": 0.7316, "mean_token_accuracy": 0.7739744782447815, "num_tokens": 14614528.0, "step": 223 }, { "epoch": 0.029606132698916203, "grad_norm": 0.886143655639606, "learning_rate": 2.9458388375165126e-06, "loss": 0.856, "mean_token_accuracy": 0.7354718446731567, "num_tokens": 14680064.0, "step": 224 }, { "epoch": 0.02973830293417922, "grad_norm": 0.7799622807493779, "learning_rate": 2.959048877146632e-06, "loss": 0.7807, "mean_token_accuracy": 0.758637547492981, "num_tokens": 14745600.0, "step": 225 }, { "epoch": 0.029870473169442243, "grad_norm": 0.7872482293279625, "learning_rate": 2.9722589167767508e-06, "loss": 0.7681, "mean_token_accuracy": 0.7635362148284912, "num_tokens": 14811136.0, "step": 226 }, { "epoch": 0.03000264340470526, "grad_norm": 0.7727062092535665, "learning_rate": 2.9854689564068694e-06, "loss": 0.7454, "mean_token_accuracy": 0.7714259624481201, "num_tokens": 14876672.0, "step": 227 }, { "epoch": 0.03013481363996828, "grad_norm": 0.9284326278866231, "learning_rate": 2.9986789960369885e-06, "loss": 0.8229, "mean_token_accuracy": 0.7516633868217468, "num_tokens": 14942208.0, "step": 228 }, { "epoch": 0.030266983875231298, "grad_norm": 0.8318704629856479, "learning_rate": 3.011889035667107e-06, "loss": 0.8108, "mean_token_accuracy": 0.751083493232727, "num_tokens": 15007744.0, "step": 229 }, { "epoch": 0.030399154110494316, "grad_norm": 0.8153947456727185, "learning_rate": 3.0250990752972257e-06, "loss": 0.8347, "mean_token_accuracy": 0.7486723065376282, "num_tokens": 15073280.0, "step": 230 }, { "epoch": 0.030531324345757335, "grad_norm": 0.7818867415806732, "learning_rate": 3.0383091149273452e-06, "loss": 0.7699, "mean_token_accuracy": 0.7646044492721558, "num_tokens": 15138816.0, "step": 231 }, { "epoch": 0.030663494581020353, "grad_norm": 0.8213369799217602, "learning_rate": 3.051519154557464e-06, "loss": 0.8126, "mean_token_accuracy": 0.7500762939453125, "num_tokens": 15204352.0, "step": 232 }, { "epoch": 0.030795664816283375, "grad_norm": 0.7468218351932058, "learning_rate": 3.064729194187583e-06, "loss": 0.6947, "mean_token_accuracy": 0.7859998941421509, "num_tokens": 15269888.0, "step": 233 }, { "epoch": 0.030927835051546393, "grad_norm": 0.813319229197614, "learning_rate": 3.0779392338177016e-06, "loss": 0.829, "mean_token_accuracy": 0.7482144832611084, "num_tokens": 15335424.0, "step": 234 }, { "epoch": 0.03106000528680941, "grad_norm": 0.7615909587388535, "learning_rate": 3.0911492734478207e-06, "loss": 0.7843, "mean_token_accuracy": 0.7608503103256226, "num_tokens": 15400960.0, "step": 235 }, { "epoch": 0.03119217552207243, "grad_norm": 0.7802406753894858, "learning_rate": 3.1043593130779393e-06, "loss": 0.7488, "mean_token_accuracy": 0.7706018686294556, "num_tokens": 15466496.0, "step": 236 }, { "epoch": 0.03132434575733545, "grad_norm": 0.8225929038621069, "learning_rate": 3.1175693527080588e-06, "loss": 0.8405, "mean_token_accuracy": 0.7418965697288513, "num_tokens": 15532032.0, "step": 237 }, { "epoch": 0.031456515992598466, "grad_norm": 0.7491735931144728, "learning_rate": 3.1307793923381774e-06, "loss": 0.7734, "mean_token_accuracy": 0.7630326151847839, "num_tokens": 15597568.0, "step": 238 }, { "epoch": 0.031588686227861484, "grad_norm": 0.7343188087094665, "learning_rate": 3.143989431968296e-06, "loss": 0.7429, "mean_token_accuracy": 0.7754089832305908, "num_tokens": 15663104.0, "step": 239 }, { "epoch": 0.0317208564631245, "grad_norm": 0.7659153382456473, "learning_rate": 3.157199471598415e-06, "loss": 0.8062, "mean_token_accuracy": 0.7547155618667603, "num_tokens": 15728640.0, "step": 240 }, { "epoch": 0.03185302669838752, "grad_norm": 0.8047577272847064, "learning_rate": 3.1704095112285338e-06, "loss": 0.8046, "mean_token_accuracy": 0.7516481280326843, "num_tokens": 15794176.0, "step": 241 }, { "epoch": 0.03198519693365054, "grad_norm": 0.731082618376276, "learning_rate": 3.1836195508586524e-06, "loss": 0.7653, "mean_token_accuracy": 0.7666035890579224, "num_tokens": 15859712.0, "step": 242 }, { "epoch": 0.03211736716891356, "grad_norm": 0.7647451039688387, "learning_rate": 3.196829590488772e-06, "loss": 0.7707, "mean_token_accuracy": 0.7668324708938599, "num_tokens": 15925248.0, "step": 243 }, { "epoch": 0.03224953740417658, "grad_norm": 0.788879499478051, "learning_rate": 3.2100396301188905e-06, "loss": 0.7828, "mean_token_accuracy": 0.7588664293289185, "num_tokens": 15990784.0, "step": 244 }, { "epoch": 0.0323817076394396, "grad_norm": 0.8636566367710148, "learning_rate": 3.2232496697490096e-06, "loss": 0.864, "mean_token_accuracy": 0.7368605732917786, "num_tokens": 16056320.0, "step": 245 }, { "epoch": 0.03251387787470262, "grad_norm": 0.825325913983544, "learning_rate": 3.2364597093791283e-06, "loss": 0.7704, "mean_token_accuracy": 0.7668172121047974, "num_tokens": 16121856.0, "step": 246 }, { "epoch": 0.03264604810996564, "grad_norm": 0.7914401211055546, "learning_rate": 3.2496697490092473e-06, "loss": 0.7752, "mean_token_accuracy": 0.7657947540283203, "num_tokens": 16187392.0, "step": 247 }, { "epoch": 0.032778218345228656, "grad_norm": 0.8027576313873162, "learning_rate": 3.262879788639366e-06, "loss": 0.7961, "mean_token_accuracy": 0.7575998306274414, "num_tokens": 16252928.0, "step": 248 }, { "epoch": 0.032910388580491674, "grad_norm": 0.7952435693627449, "learning_rate": 3.2760898282694855e-06, "loss": 0.7998, "mean_token_accuracy": 0.7556617259979248, "num_tokens": 16318464.0, "step": 249 }, { "epoch": 0.03304255881575469, "grad_norm": 0.8051664422889488, "learning_rate": 3.289299867899604e-06, "loss": 0.7978, "mean_token_accuracy": 0.7572945952415466, "num_tokens": 16384000.0, "step": 250 }, { "epoch": 0.03317472905101771, "grad_norm": 0.7868129025179686, "learning_rate": 3.3025099075297227e-06, "loss": 0.7851, "mean_token_accuracy": 0.7589579820632935, "num_tokens": 16449536.0, "step": 251 }, { "epoch": 0.03330689928628073, "grad_norm": 0.7954351935241238, "learning_rate": 3.315719947159842e-06, "loss": 0.7821, "mean_token_accuracy": 0.7594463229179382, "num_tokens": 16515072.0, "step": 252 }, { "epoch": 0.03343906952154375, "grad_norm": 0.8191292971746249, "learning_rate": 3.3289299867899604e-06, "loss": 0.8229, "mean_token_accuracy": 0.7464442849159241, "num_tokens": 16580608.0, "step": 253 }, { "epoch": 0.033571239756806766, "grad_norm": 0.8293448838395114, "learning_rate": 3.342140026420079e-06, "loss": 0.8023, "mean_token_accuracy": 0.7550512552261353, "num_tokens": 16646144.0, "step": 254 }, { "epoch": 0.033703409992069784, "grad_norm": 0.8257046815827183, "learning_rate": 3.3553500660501986e-06, "loss": 0.8425, "mean_token_accuracy": 0.7445977330207825, "num_tokens": 16711680.0, "step": 255 }, { "epoch": 0.0338355802273328, "grad_norm": 0.7835744834205397, "learning_rate": 3.3685601056803176e-06, "loss": 0.8239, "mean_token_accuracy": 0.7463527321815491, "num_tokens": 16777216.0, "step": 256 }, { "epoch": 0.03396775046259582, "grad_norm": 0.8118731489646601, "learning_rate": 3.3817701453104363e-06, "loss": 0.7661, "mean_token_accuracy": 0.7631089091300964, "num_tokens": 16842752.0, "step": 257 }, { "epoch": 0.03409992069785884, "grad_norm": 0.8134648560447796, "learning_rate": 3.394980184940555e-06, "loss": 0.8032, "mean_token_accuracy": 0.7540898323059082, "num_tokens": 16908288.0, "step": 258 }, { "epoch": 0.034232090933121864, "grad_norm": 0.7485261751792613, "learning_rate": 3.408190224570674e-06, "loss": 0.7689, "mean_token_accuracy": 0.7659474015235901, "num_tokens": 16973824.0, "step": 259 }, { "epoch": 0.03436426116838488, "grad_norm": 0.7503113103184508, "learning_rate": 3.4214002642007926e-06, "loss": 0.7676, "mean_token_accuracy": 0.7672902941703796, "num_tokens": 17039360.0, "step": 260 }, { "epoch": 0.0344964314036479, "grad_norm": 0.7118092730357078, "learning_rate": 3.434610303830912e-06, "loss": 0.7206, "mean_token_accuracy": 0.7767061591148376, "num_tokens": 17104896.0, "step": 261 }, { "epoch": 0.03462860163891092, "grad_norm": 0.776050869021872, "learning_rate": 3.4478203434610308e-06, "loss": 0.7591, "mean_token_accuracy": 0.7679618000984192, "num_tokens": 17170432.0, "step": 262 }, { "epoch": 0.03476077187417394, "grad_norm": 0.8173065698441644, "learning_rate": 3.4610303830911494e-06, "loss": 0.7495, "mean_token_accuracy": 0.7705255746841431, "num_tokens": 17235968.0, "step": 263 }, { "epoch": 0.034892942109436956, "grad_norm": 0.8312055916006524, "learning_rate": 3.4742404227212685e-06, "loss": 0.769, "mean_token_accuracy": 0.7613233923912048, "num_tokens": 17301504.0, "step": 264 }, { "epoch": 0.035025112344699974, "grad_norm": 0.802789717544756, "learning_rate": 3.487450462351387e-06, "loss": 0.8, "mean_token_accuracy": 0.7529758214950562, "num_tokens": 17367040.0, "step": 265 }, { "epoch": 0.03515728257996299, "grad_norm": 0.8467904216580417, "learning_rate": 3.500660501981506e-06, "loss": 0.8279, "mean_token_accuracy": 0.7460322380065918, "num_tokens": 17432576.0, "step": 266 }, { "epoch": 0.03528945281522601, "grad_norm": 0.7263767900447556, "learning_rate": 3.5138705416116252e-06, "loss": 0.7736, "mean_token_accuracy": 0.7648791074752808, "num_tokens": 17498112.0, "step": 267 }, { "epoch": 0.03542162305048903, "grad_norm": 0.8524931844140684, "learning_rate": 3.5270805812417443e-06, "loss": 0.799, "mean_token_accuracy": 0.7560431957244873, "num_tokens": 17563648.0, "step": 268 }, { "epoch": 0.03555379328575205, "grad_norm": 0.7912877977679494, "learning_rate": 3.540290620871863e-06, "loss": 0.7627, "mean_token_accuracy": 0.7661610245704651, "num_tokens": 17629184.0, "step": 269 }, { "epoch": 0.035685963521015066, "grad_norm": 0.7914778002908024, "learning_rate": 3.5535006605019816e-06, "loss": 0.7494, "mean_token_accuracy": 0.7691368460655212, "num_tokens": 17694720.0, "step": 270 }, { "epoch": 0.035818133756278084, "grad_norm": 0.8324048422784303, "learning_rate": 3.5667107001321007e-06, "loss": 0.7946, "mean_token_accuracy": 0.75529545545578, "num_tokens": 17760256.0, "step": 271 }, { "epoch": 0.0359503039915411, "grad_norm": 0.80358437819994, "learning_rate": 3.5799207397622193e-06, "loss": 0.7579, "mean_token_accuracy": 0.7649402022361755, "num_tokens": 17825792.0, "step": 272 }, { "epoch": 0.03608247422680412, "grad_norm": 0.7457434995534333, "learning_rate": 3.5931307793923388e-06, "loss": 0.7369, "mean_token_accuracy": 0.7720211148262024, "num_tokens": 17891328.0, "step": 273 }, { "epoch": 0.036214644462067146, "grad_norm": 0.7605445325932457, "learning_rate": 3.6063408190224574e-06, "loss": 0.7958, "mean_token_accuracy": 0.7579202651977539, "num_tokens": 17956864.0, "step": 274 }, { "epoch": 0.036346814697330164, "grad_norm": 0.7997125651710348, "learning_rate": 3.619550858652576e-06, "loss": 0.7745, "mean_token_accuracy": 0.7619643807411194, "num_tokens": 18022400.0, "step": 275 }, { "epoch": 0.03647898493259318, "grad_norm": 0.8375163195592626, "learning_rate": 3.632760898282695e-06, "loss": 0.8248, "mean_token_accuracy": 0.7419576644897461, "num_tokens": 18087936.0, "step": 276 }, { "epoch": 0.0366111551678562, "grad_norm": 0.737882558386852, "learning_rate": 3.6459709379128138e-06, "loss": 0.7236, "mean_token_accuracy": 0.779498815536499, "num_tokens": 18153472.0, "step": 277 }, { "epoch": 0.03674332540311922, "grad_norm": 0.8082147847763389, "learning_rate": 3.659180977542933e-06, "loss": 0.7659, "mean_token_accuracy": 0.7673666477203369, "num_tokens": 18219008.0, "step": 278 }, { "epoch": 0.03687549563838224, "grad_norm": 0.8357458469206095, "learning_rate": 3.672391017173052e-06, "loss": 0.8334, "mean_token_accuracy": 0.7431174516677856, "num_tokens": 18284544.0, "step": 279 }, { "epoch": 0.037007665873645255, "grad_norm": 0.7738832077420589, "learning_rate": 3.685601056803171e-06, "loss": 0.7489, "mean_token_accuracy": 0.7691978812217712, "num_tokens": 18350080.0, "step": 280 }, { "epoch": 0.037139836108908274, "grad_norm": 0.7448950230466799, "learning_rate": 3.6988110964332896e-06, "loss": 0.6904, "mean_token_accuracy": 0.7872207164764404, "num_tokens": 18415616.0, "step": 281 }, { "epoch": 0.03727200634417129, "grad_norm": 0.7541355940986763, "learning_rate": 3.7120211360634083e-06, "loss": 0.7922, "mean_token_accuracy": 0.7564705014228821, "num_tokens": 18481152.0, "step": 282 }, { "epoch": 0.03740417657943431, "grad_norm": 0.8046762892170491, "learning_rate": 3.7252311756935273e-06, "loss": 0.7819, "mean_token_accuracy": 0.7605298757553101, "num_tokens": 18546688.0, "step": 283 }, { "epoch": 0.03753634681469733, "grad_norm": 0.7519925702199056, "learning_rate": 3.738441215323646e-06, "loss": 0.6779, "mean_token_accuracy": 0.79355388879776, "num_tokens": 18612224.0, "step": 284 }, { "epoch": 0.03766851704996035, "grad_norm": 0.8069577803910146, "learning_rate": 3.7516512549537655e-06, "loss": 0.7362, "mean_token_accuracy": 0.7709071040153503, "num_tokens": 18677760.0, "step": 285 }, { "epoch": 0.037800687285223365, "grad_norm": 0.7954811826334626, "learning_rate": 3.764861294583884e-06, "loss": 0.761, "mean_token_accuracy": 0.7645281553268433, "num_tokens": 18743296.0, "step": 286 }, { "epoch": 0.037932857520486384, "grad_norm": 0.7634335030030928, "learning_rate": 3.778071334214003e-06, "loss": 0.7273, "mean_token_accuracy": 0.7760651707649231, "num_tokens": 18808832.0, "step": 287 }, { "epoch": 0.0380650277557494, "grad_norm": 0.7932369164813577, "learning_rate": 3.791281373844122e-06, "loss": 0.7483, "mean_token_accuracy": 0.7709528803825378, "num_tokens": 18874368.0, "step": 288 }, { "epoch": 0.03819719799101243, "grad_norm": 0.8589644421736623, "learning_rate": 3.8044914134742404e-06, "loss": 0.7886, "mean_token_accuracy": 0.7590190172195435, "num_tokens": 18939904.0, "step": 289 }, { "epoch": 0.038329368226275445, "grad_norm": 0.8032151044010292, "learning_rate": 3.8177014531043595e-06, "loss": 0.7932, "mean_token_accuracy": 0.7566231489181519, "num_tokens": 19005440.0, "step": 290 }, { "epoch": 0.038461538461538464, "grad_norm": 0.8010758735120755, "learning_rate": 3.830911492734479e-06, "loss": 0.7438, "mean_token_accuracy": 0.7710444331169128, "num_tokens": 19070976.0, "step": 291 }, { "epoch": 0.03859370869680148, "grad_norm": 0.821093625466762, "learning_rate": 3.844121532364598e-06, "loss": 0.7685, "mean_token_accuracy": 0.7618880271911621, "num_tokens": 19136512.0, "step": 292 }, { "epoch": 0.0387258789320645, "grad_norm": 0.8600298111603055, "learning_rate": 3.857331571994716e-06, "loss": 0.7465, "mean_token_accuracy": 0.769686222076416, "num_tokens": 19202048.0, "step": 293 }, { "epoch": 0.03885804916732752, "grad_norm": 0.8699801634158648, "learning_rate": 3.870541611624835e-06, "loss": 0.803, "mean_token_accuracy": 0.7535557150840759, "num_tokens": 19267584.0, "step": 294 }, { "epoch": 0.03899021940259054, "grad_norm": 0.7565806721543882, "learning_rate": 3.8837516512549536e-06, "loss": 0.697, "mean_token_accuracy": 0.7820473909378052, "num_tokens": 19333120.0, "step": 295 }, { "epoch": 0.039122389637853555, "grad_norm": 0.8451752711228975, "learning_rate": 3.896961690885073e-06, "loss": 0.7954, "mean_token_accuracy": 0.7590343356132507, "num_tokens": 19398656.0, "step": 296 }, { "epoch": 0.03925455987311657, "grad_norm": 0.7858913864520309, "learning_rate": 3.910171730515192e-06, "loss": 0.7661, "mean_token_accuracy": 0.7638719081878662, "num_tokens": 19464192.0, "step": 297 }, { "epoch": 0.03938673010837959, "grad_norm": 0.7960011239384529, "learning_rate": 3.923381770145311e-06, "loss": 0.7901, "mean_token_accuracy": 0.7573708891868591, "num_tokens": 19529728.0, "step": 298 }, { "epoch": 0.03951890034364261, "grad_norm": 0.8203552824856194, "learning_rate": 3.93659180977543e-06, "loss": 0.7837, "mean_token_accuracy": 0.7592784762382507, "num_tokens": 19595264.0, "step": 299 }, { "epoch": 0.03965107057890563, "grad_norm": 0.7617552409352167, "learning_rate": 3.9498018494055485e-06, "loss": 0.7549, "mean_token_accuracy": 0.7704797983169556, "num_tokens": 19660800.0, "step": 300 }, { "epoch": 0.03978324081416865, "grad_norm": 0.7744668759991761, "learning_rate": 3.963011889035667e-06, "loss": 0.7334, "mean_token_accuracy": 0.7743560075759888, "num_tokens": 19726336.0, "step": 301 }, { "epoch": 0.039915411049431665, "grad_norm": 0.810766790517258, "learning_rate": 3.976221928665786e-06, "loss": 0.7757, "mean_token_accuracy": 0.7606824636459351, "num_tokens": 19791872.0, "step": 302 }, { "epoch": 0.04004758128469469, "grad_norm": 0.8348352682205219, "learning_rate": 3.989431968295905e-06, "loss": 0.7608, "mean_token_accuracy": 0.7663441300392151, "num_tokens": 19857408.0, "step": 303 }, { "epoch": 0.04017975151995771, "grad_norm": 0.7923842195606253, "learning_rate": 4.002642007926024e-06, "loss": 0.8103, "mean_token_accuracy": 0.7527469396591187, "num_tokens": 19922944.0, "step": 304 }, { "epoch": 0.04031192175522073, "grad_norm": 0.8506819643585403, "learning_rate": 4.015852047556143e-06, "loss": 0.821, "mean_token_accuracy": 0.750839352607727, "num_tokens": 19988480.0, "step": 305 }, { "epoch": 0.040444091990483745, "grad_norm": 0.8201201119504129, "learning_rate": 4.029062087186262e-06, "loss": 0.7591, "mean_token_accuracy": 0.7660084366798401, "num_tokens": 20054016.0, "step": 306 }, { "epoch": 0.04057626222574676, "grad_norm": 0.8350274851038256, "learning_rate": 4.042272126816381e-06, "loss": 0.7575, "mean_token_accuracy": 0.7639787793159485, "num_tokens": 20119552.0, "step": 307 }, { "epoch": 0.04070843246100978, "grad_norm": 0.7407879878607747, "learning_rate": 4.0554821664465e-06, "loss": 0.7624, "mean_token_accuracy": 0.7669698596000671, "num_tokens": 20185088.0, "step": 308 }, { "epoch": 0.0408406026962728, "grad_norm": 0.8593153940583875, "learning_rate": 4.068692206076619e-06, "loss": 0.7674, "mean_token_accuracy": 0.7631241679191589, "num_tokens": 20250624.0, "step": 309 }, { "epoch": 0.04097277293153582, "grad_norm": 0.7642521367921372, "learning_rate": 4.0819022457067374e-06, "loss": 0.7853, "mean_token_accuracy": 0.7610487341880798, "num_tokens": 20316160.0, "step": 310 }, { "epoch": 0.04110494316679884, "grad_norm": 0.7930646566889225, "learning_rate": 4.095112285336856e-06, "loss": 0.7865, "mean_token_accuracy": 0.758942723274231, "num_tokens": 20381696.0, "step": 311 }, { "epoch": 0.041237113402061855, "grad_norm": 0.8535947898193181, "learning_rate": 4.1083223249669756e-06, "loss": 0.793, "mean_token_accuracy": 0.755142867565155, "num_tokens": 20447232.0, "step": 312 }, { "epoch": 0.04136928363732487, "grad_norm": 0.831024900230466, "learning_rate": 4.121532364597094e-06, "loss": 0.7471, "mean_token_accuracy": 0.7705713510513306, "num_tokens": 20512768.0, "step": 313 }, { "epoch": 0.04150145387258789, "grad_norm": 0.8418394640752138, "learning_rate": 4.134742404227214e-06, "loss": 0.7584, "mean_token_accuracy": 0.7681143879890442, "num_tokens": 20578304.0, "step": 314 }, { "epoch": 0.04163362410785091, "grad_norm": 0.8334835994584365, "learning_rate": 4.147952443857332e-06, "loss": 0.7634, "mean_token_accuracy": 0.7637803554534912, "num_tokens": 20643840.0, "step": 315 }, { "epoch": 0.04176579434311393, "grad_norm": 0.8122306424676199, "learning_rate": 4.161162483487451e-06, "loss": 0.7814, "mean_token_accuracy": 0.7608045339584351, "num_tokens": 20709376.0, "step": 316 }, { "epoch": 0.041897964578376946, "grad_norm": 0.8285204145293741, "learning_rate": 4.17437252311757e-06, "loss": 0.7323, "mean_token_accuracy": 0.7748748660087585, "num_tokens": 20774912.0, "step": 317 }, { "epoch": 0.04203013481363997, "grad_norm": 0.8493327317390713, "learning_rate": 4.187582562747688e-06, "loss": 0.809, "mean_token_accuracy": 0.750869870185852, "num_tokens": 20840448.0, "step": 318 }, { "epoch": 0.04216230504890299, "grad_norm": 0.8727262296489746, "learning_rate": 4.200792602377807e-06, "loss": 0.8038, "mean_token_accuracy": 0.7517855167388916, "num_tokens": 20905984.0, "step": 319 }, { "epoch": 0.04229447528416601, "grad_norm": 0.8579294625256043, "learning_rate": 4.214002642007926e-06, "loss": 0.7537, "mean_token_accuracy": 0.7702661752700806, "num_tokens": 20971520.0, "step": 320 }, { "epoch": 0.042426645519429027, "grad_norm": 0.8286790525035163, "learning_rate": 4.227212681638045e-06, "loss": 0.7195, "mean_token_accuracy": 0.7799413800239563, "num_tokens": 21037056.0, "step": 321 }, { "epoch": 0.042558815754692045, "grad_norm": 0.8503892110484725, "learning_rate": 4.2404227212681645e-06, "loss": 0.8131, "mean_token_accuracy": 0.7507325410842896, "num_tokens": 21102592.0, "step": 322 }, { "epoch": 0.04269098598995506, "grad_norm": 0.8611085525731758, "learning_rate": 4.253632760898283e-06, "loss": 0.8082, "mean_token_accuracy": 0.7513582110404968, "num_tokens": 21168128.0, "step": 323 }, { "epoch": 0.04282315622521808, "grad_norm": 0.8134939114077576, "learning_rate": 4.266842800528402e-06, "loss": 0.7176, "mean_token_accuracy": 0.7768740057945251, "num_tokens": 21233664.0, "step": 324 }, { "epoch": 0.0429553264604811, "grad_norm": 0.7966307893631472, "learning_rate": 4.2800528401585204e-06, "loss": 0.6981, "mean_token_accuracy": 0.784824788570404, "num_tokens": 21299200.0, "step": 325 }, { "epoch": 0.04308749669574412, "grad_norm": 0.7620776996811601, "learning_rate": 4.29326287978864e-06, "loss": 0.733, "mean_token_accuracy": 0.7726162672042847, "num_tokens": 21364736.0, "step": 326 }, { "epoch": 0.043219666931007136, "grad_norm": 0.8213464371711094, "learning_rate": 4.306472919418759e-06, "loss": 0.7748, "mean_token_accuracy": 0.7580270767211914, "num_tokens": 21430272.0, "step": 327 }, { "epoch": 0.043351837166270155, "grad_norm": 0.7951694154071558, "learning_rate": 4.319682959048877e-06, "loss": 0.7707, "mean_token_accuracy": 0.7612928748130798, "num_tokens": 21495808.0, "step": 328 }, { "epoch": 0.04348400740153317, "grad_norm": 0.839440425143574, "learning_rate": 4.332892998678997e-06, "loss": 0.7865, "mean_token_accuracy": 0.7550054788589478, "num_tokens": 21561344.0, "step": 329 }, { "epoch": 0.04361617763679619, "grad_norm": 0.8216745932265515, "learning_rate": 4.346103038309115e-06, "loss": 0.7754, "mean_token_accuracy": 0.7603314518928528, "num_tokens": 21626880.0, "step": 330 }, { "epoch": 0.04374834787205921, "grad_norm": 0.8030483480434127, "learning_rate": 4.359313077939234e-06, "loss": 0.7419, "mean_token_accuracy": 0.7715480327606201, "num_tokens": 21692416.0, "step": 331 }, { "epoch": 0.04388051810732223, "grad_norm": 0.8208128498656216, "learning_rate": 4.3725231175693535e-06, "loss": 0.7704, "mean_token_accuracy": 0.7628952264785767, "num_tokens": 21757952.0, "step": 332 }, { "epoch": 0.04401268834258525, "grad_norm": 0.765913894270719, "learning_rate": 4.385733157199472e-06, "loss": 0.7393, "mean_token_accuracy": 0.7719295620918274, "num_tokens": 21823488.0, "step": 333 }, { "epoch": 0.04414485857784827, "grad_norm": 0.8254741936625415, "learning_rate": 4.398943196829591e-06, "loss": 0.8096, "mean_token_accuracy": 0.7482755184173584, "num_tokens": 21889024.0, "step": 334 }, { "epoch": 0.04427702881311129, "grad_norm": 0.8876120460370389, "learning_rate": 4.412153236459709e-06, "loss": 0.8549, "mean_token_accuracy": 0.7370895147323608, "num_tokens": 21954560.0, "step": 335 }, { "epoch": 0.04440919904837431, "grad_norm": 0.812164422091014, "learning_rate": 4.425363276089829e-06, "loss": 0.7478, "mean_token_accuracy": 0.769716739654541, "num_tokens": 22020096.0, "step": 336 }, { "epoch": 0.044541369283637326, "grad_norm": 0.7720524884619422, "learning_rate": 4.4385733157199475e-06, "loss": 0.6853, "mean_token_accuracy": 0.7860304117202759, "num_tokens": 22085632.0, "step": 337 }, { "epoch": 0.044673539518900345, "grad_norm": 0.8150231008640392, "learning_rate": 4.451783355350067e-06, "loss": 0.7983, "mean_token_accuracy": 0.7550970315933228, "num_tokens": 22151168.0, "step": 338 }, { "epoch": 0.04480570975416336, "grad_norm": 0.7748678756868259, "learning_rate": 4.464993394980186e-06, "loss": 0.6987, "mean_token_accuracy": 0.782810389995575, "num_tokens": 22216704.0, "step": 339 }, { "epoch": 0.04493787998942638, "grad_norm": 0.8041543542119833, "learning_rate": 4.478203434610304e-06, "loss": 0.7378, "mean_token_accuracy": 0.7727689146995544, "num_tokens": 22282240.0, "step": 340 }, { "epoch": 0.0450700502246894, "grad_norm": 0.8161970748893086, "learning_rate": 4.491413474240423e-06, "loss": 0.7592, "mean_token_accuracy": 0.7667561769485474, "num_tokens": 22347776.0, "step": 341 }, { "epoch": 0.04520222045995242, "grad_norm": 0.8168495971138188, "learning_rate": 4.504623513870542e-06, "loss": 0.7794, "mean_token_accuracy": 0.7613691687583923, "num_tokens": 22413312.0, "step": 342 }, { "epoch": 0.045334390695215436, "grad_norm": 0.8036867279831807, "learning_rate": 4.517833553500661e-06, "loss": 0.7377, "mean_token_accuracy": 0.7698388695716858, "num_tokens": 22478848.0, "step": 343 }, { "epoch": 0.045466560930478454, "grad_norm": 0.8152795645996423, "learning_rate": 4.53104359313078e-06, "loss": 0.7387, "mean_token_accuracy": 0.7706629037857056, "num_tokens": 22544384.0, "step": 344 }, { "epoch": 0.04559873116574147, "grad_norm": 0.8308068350740329, "learning_rate": 4.544253632760899e-06, "loss": 0.746, "mean_token_accuracy": 0.7691978812217712, "num_tokens": 22609920.0, "step": 345 }, { "epoch": 0.04573090140100449, "grad_norm": 0.765666589214006, "learning_rate": 4.557463672391018e-06, "loss": 0.7288, "mean_token_accuracy": 0.774859607219696, "num_tokens": 22675456.0, "step": 346 }, { "epoch": 0.04586307163626751, "grad_norm": 0.868658894333656, "learning_rate": 4.5706737120211365e-06, "loss": 0.7787, "mean_token_accuracy": 0.7579813003540039, "num_tokens": 22740992.0, "step": 347 }, { "epoch": 0.045995241871530534, "grad_norm": 0.8116544175175842, "learning_rate": 4.583883751651255e-06, "loss": 0.7444, "mean_token_accuracy": 0.7721126675605774, "num_tokens": 22806528.0, "step": 348 }, { "epoch": 0.04612741210679355, "grad_norm": 0.880266916814189, "learning_rate": 4.597093791281374e-06, "loss": 0.7349, "mean_token_accuracy": 0.7749359011650085, "num_tokens": 22872064.0, "step": 349 }, { "epoch": 0.04625958234205657, "grad_norm": 0.7422024161311556, "learning_rate": 4.610303830911493e-06, "loss": 0.6878, "mean_token_accuracy": 0.7873122692108154, "num_tokens": 22937600.0, "step": 350 }, { "epoch": 0.04639175257731959, "grad_norm": 0.9281909716981132, "learning_rate": 4.623513870541612e-06, "loss": 0.7453, "mean_token_accuracy": 0.769625186920166, "num_tokens": 23003136.0, "step": 351 }, { "epoch": 0.04652392281258261, "grad_norm": 0.8720314985821085, "learning_rate": 4.6367239101717306e-06, "loss": 0.694, "mean_token_accuracy": 0.7842601537704468, "num_tokens": 23068672.0, "step": 352 }, { "epoch": 0.046656093047845626, "grad_norm": 0.7925575940566519, "learning_rate": 4.64993394980185e-06, "loss": 0.7073, "mean_token_accuracy": 0.782871425151825, "num_tokens": 23134208.0, "step": 353 }, { "epoch": 0.046788263283108644, "grad_norm": 0.8387168983330182, "learning_rate": 4.663143989431969e-06, "loss": 0.7387, "mean_token_accuracy": 0.769564151763916, "num_tokens": 23199744.0, "step": 354 }, { "epoch": 0.04692043351837166, "grad_norm": 0.8646449279376699, "learning_rate": 4.676354029062087e-06, "loss": 0.7867, "mean_token_accuracy": 0.7530826330184937, "num_tokens": 23265280.0, "step": 355 }, { "epoch": 0.04705260375363468, "grad_norm": 0.8424204302539733, "learning_rate": 4.689564068692207e-06, "loss": 0.7204, "mean_token_accuracy": 0.773898184299469, "num_tokens": 23330816.0, "step": 356 }, { "epoch": 0.0471847739888977, "grad_norm": 0.8445931914018319, "learning_rate": 4.7027741083223255e-06, "loss": 0.7651, "mean_token_accuracy": 0.7664052248001099, "num_tokens": 23396352.0, "step": 357 }, { "epoch": 0.04731694422416072, "grad_norm": 0.8078596370906337, "learning_rate": 4.715984147952444e-06, "loss": 0.7045, "mean_token_accuracy": 0.7815590500831604, "num_tokens": 23461888.0, "step": 358 }, { "epoch": 0.047449114459423736, "grad_norm": 0.8488977667046452, "learning_rate": 4.729194187582563e-06, "loss": 0.7485, "mean_token_accuracy": 0.7676260471343994, "num_tokens": 23527424.0, "step": 359 }, { "epoch": 0.047581284694686754, "grad_norm": 0.845450039982898, "learning_rate": 4.742404227212682e-06, "loss": 0.7565, "mean_token_accuracy": 0.7680380940437317, "num_tokens": 23592960.0, "step": 360 }, { "epoch": 0.04771345492994977, "grad_norm": 0.7644552684916214, "learning_rate": 4.755614266842801e-06, "loss": 0.6755, "mean_token_accuracy": 0.792561948299408, "num_tokens": 23658496.0, "step": 361 }, { "epoch": 0.04784562516521279, "grad_norm": 0.8104869634256515, "learning_rate": 4.76882430647292e-06, "loss": 0.742, "mean_token_accuracy": 0.7688469290733337, "num_tokens": 23724032.0, "step": 362 }, { "epoch": 0.047977795400475816, "grad_norm": 0.7839027206489168, "learning_rate": 4.782034346103039e-06, "loss": 0.7335, "mean_token_accuracy": 0.7746459245681763, "num_tokens": 23789568.0, "step": 363 }, { "epoch": 0.048109965635738834, "grad_norm": 0.8267151455182563, "learning_rate": 4.795244385733158e-06, "loss": 0.7279, "mean_token_accuracy": 0.7729977965354919, "num_tokens": 23855104.0, "step": 364 }, { "epoch": 0.04824213587100185, "grad_norm": 0.7931530778237736, "learning_rate": 4.808454425363276e-06, "loss": 0.6906, "mean_token_accuracy": 0.788853645324707, "num_tokens": 23920640.0, "step": 365 }, { "epoch": 0.04837430610626487, "grad_norm": 0.8799404307776028, "learning_rate": 4.821664464993395e-06, "loss": 0.7463, "mean_token_accuracy": 0.7689842581748962, "num_tokens": 23986176.0, "step": 366 }, { "epoch": 0.04850647634152789, "grad_norm": 0.8178568681545735, "learning_rate": 4.834874504623514e-06, "loss": 0.7131, "mean_token_accuracy": 0.7789646983146667, "num_tokens": 24051712.0, "step": 367 }, { "epoch": 0.04863864657679091, "grad_norm": 0.8476321126694517, "learning_rate": 4.848084544253633e-06, "loss": 0.7263, "mean_token_accuracy": 0.7729215025901794, "num_tokens": 24117248.0, "step": 368 }, { "epoch": 0.048770816812053926, "grad_norm": 0.8154680039026881, "learning_rate": 4.8612945838837526e-06, "loss": 0.7307, "mean_token_accuracy": 0.7726315259933472, "num_tokens": 24182784.0, "step": 369 }, { "epoch": 0.048902987047316944, "grad_norm": 0.8234731661992672, "learning_rate": 4.874504623513871e-06, "loss": 0.8041, "mean_token_accuracy": 0.7538609504699707, "num_tokens": 24248320.0, "step": 370 }, { "epoch": 0.04903515728257996, "grad_norm": 0.883781765172588, "learning_rate": 4.88771466314399e-06, "loss": 0.7421, "mean_token_accuracy": 0.7664357423782349, "num_tokens": 24313856.0, "step": 371 }, { "epoch": 0.04916732751784298, "grad_norm": 0.8536063978020105, "learning_rate": 4.9009247027741085e-06, "loss": 0.6988, "mean_token_accuracy": 0.7807502150535583, "num_tokens": 24379392.0, "step": 372 }, { "epoch": 0.049299497753106, "grad_norm": 0.7540634084830988, "learning_rate": 4.914134742404227e-06, "loss": 0.6725, "mean_token_accuracy": 0.7919820547103882, "num_tokens": 24444928.0, "step": 373 }, { "epoch": 0.04943166798836902, "grad_norm": 0.8083450524640311, "learning_rate": 4.927344782034347e-06, "loss": 0.76, "mean_token_accuracy": 0.7617354393005371, "num_tokens": 24510464.0, "step": 374 }, { "epoch": 0.049563838223632035, "grad_norm": 0.8711024289549478, "learning_rate": 4.940554821664465e-06, "loss": 0.7722, "mean_token_accuracy": 0.7590953707695007, "num_tokens": 24576000.0, "step": 375 }, { "epoch": 0.049696008458895054, "grad_norm": 0.8297197307892307, "learning_rate": 4.953764861294585e-06, "loss": 0.7365, "mean_token_accuracy": 0.7729825377464294, "num_tokens": 24641536.0, "step": 376 }, { "epoch": 0.04982817869415808, "grad_norm": 0.9190559602708935, "learning_rate": 4.966974900924703e-06, "loss": 0.8069, "mean_token_accuracy": 0.7458949089050293, "num_tokens": 24707072.0, "step": 377 }, { "epoch": 0.0499603489294211, "grad_norm": 0.8174662550181419, "learning_rate": 4.980184940554822e-06, "loss": 0.7621, "mean_token_accuracy": 0.7605451345443726, "num_tokens": 24772608.0, "step": 378 }, { "epoch": 0.050092519164684116, "grad_norm": 0.788788214652604, "learning_rate": 4.993394980184941e-06, "loss": 0.7025, "mean_token_accuracy": 0.7788578867912292, "num_tokens": 24838144.0, "step": 379 }, { "epoch": 0.050224689399947134, "grad_norm": 0.8943032990306043, "learning_rate": 5.006605019815059e-06, "loss": 0.7896, "mean_token_accuracy": 0.7537999153137207, "num_tokens": 24903680.0, "step": 380 }, { "epoch": 0.05035685963521015, "grad_norm": 0.7962417586098434, "learning_rate": 5.019815059445179e-06, "loss": 0.7482, "mean_token_accuracy": 0.7694268226623535, "num_tokens": 24969216.0, "step": 381 }, { "epoch": 0.05048902987047317, "grad_norm": 0.7787215372913533, "learning_rate": 5.0330250990752974e-06, "loss": 0.7064, "mean_token_accuracy": 0.7787510752677917, "num_tokens": 25034752.0, "step": 382 }, { "epoch": 0.05062120010573619, "grad_norm": 0.8388662268116762, "learning_rate": 5.046235138705416e-06, "loss": 0.7741, "mean_token_accuracy": 0.7590190172195435, "num_tokens": 25100288.0, "step": 383 }, { "epoch": 0.05075337034099921, "grad_norm": 0.8077542768918237, "learning_rate": 5.0594451783355356e-06, "loss": 0.7212, "mean_token_accuracy": 0.7763551473617554, "num_tokens": 25165824.0, "step": 384 }, { "epoch": 0.050885540576262225, "grad_norm": 0.8729830435913907, "learning_rate": 5.072655217965654e-06, "loss": 0.7543, "mean_token_accuracy": 0.7634599208831787, "num_tokens": 25231360.0, "step": 385 }, { "epoch": 0.051017710811525244, "grad_norm": 0.855027553761253, "learning_rate": 5.085865257595774e-06, "loss": 0.7512, "mean_token_accuracy": 0.764024555683136, "num_tokens": 25296896.0, "step": 386 }, { "epoch": 0.05114988104678826, "grad_norm": 0.908186950302715, "learning_rate": 5.0990752972258915e-06, "loss": 0.7635, "mean_token_accuracy": 0.7636277675628662, "num_tokens": 25362432.0, "step": 387 }, { "epoch": 0.05128205128205128, "grad_norm": 0.9147056202753256, "learning_rate": 5.112285336856011e-06, "loss": 0.8294, "mean_token_accuracy": 0.7418355345726013, "num_tokens": 25427968.0, "step": 388 }, { "epoch": 0.0514142215173143, "grad_norm": 0.8169617352583163, "learning_rate": 5.1254953764861305e-06, "loss": 0.7136, "mean_token_accuracy": 0.779285192489624, "num_tokens": 25493504.0, "step": 389 }, { "epoch": 0.05154639175257732, "grad_norm": 0.8177295481534437, "learning_rate": 5.138705416116248e-06, "loss": 0.7481, "mean_token_accuracy": 0.7690758109092712, "num_tokens": 25559040.0, "step": 390 }, { "epoch": 0.051678561987840335, "grad_norm": 0.7649579660987758, "learning_rate": 5.151915455746368e-06, "loss": 0.6764, "mean_token_accuracy": 0.7894335389137268, "num_tokens": 25624576.0, "step": 391 }, { "epoch": 0.05181073222310336, "grad_norm": 0.7960263847219581, "learning_rate": 5.165125495376486e-06, "loss": 0.6942, "mean_token_accuracy": 0.7849469184875488, "num_tokens": 25690112.0, "step": 392 }, { "epoch": 0.05194290245836638, "grad_norm": 0.8672495033350632, "learning_rate": 5.178335535006606e-06, "loss": 0.7373, "mean_token_accuracy": 0.7717158794403076, "num_tokens": 25755648.0, "step": 393 }, { "epoch": 0.0520750726936294, "grad_norm": 0.7417904926697516, "learning_rate": 5.191545574636724e-06, "loss": 0.705, "mean_token_accuracy": 0.7809943556785583, "num_tokens": 25821184.0, "step": 394 }, { "epoch": 0.052207242928892415, "grad_norm": 0.7841262854278931, "learning_rate": 5.204755614266843e-06, "loss": 0.707, "mean_token_accuracy": 0.7816506028175354, "num_tokens": 25886720.0, "step": 395 }, { "epoch": 0.052339413164155434, "grad_norm": 0.7864141475938489, "learning_rate": 5.217965653896963e-06, "loss": 0.7388, "mean_token_accuracy": 0.7709681391716003, "num_tokens": 25952256.0, "step": 396 }, { "epoch": 0.05247158339941845, "grad_norm": 0.8644070495960862, "learning_rate": 5.2311756935270805e-06, "loss": 0.7413, "mean_token_accuracy": 0.7689079642295837, "num_tokens": 26017792.0, "step": 397 }, { "epoch": 0.05260375363468147, "grad_norm": 0.8238542366787851, "learning_rate": 5.2443857331572e-06, "loss": 0.7353, "mean_token_accuracy": 0.7706018686294556, "num_tokens": 26083328.0, "step": 398 }, { "epoch": 0.05273592386994449, "grad_norm": 0.7934797058458861, "learning_rate": 5.257595772787319e-06, "loss": 0.7142, "mean_token_accuracy": 0.778171181678772, "num_tokens": 26148864.0, "step": 399 }, { "epoch": 0.05286809410520751, "grad_norm": 0.8255772941050236, "learning_rate": 5.270805812417438e-06, "loss": 0.7675, "mean_token_accuracy": 0.7655048370361328, "num_tokens": 26214400.0, "step": 400 }, { "epoch": 0.053000264340470525, "grad_norm": 0.8133959792691043, "learning_rate": 5.284015852047557e-06, "loss": 0.7093, "mean_token_accuracy": 0.7800635099411011, "num_tokens": 26279936.0, "step": 401 }, { "epoch": 0.05313243457573354, "grad_norm": 0.8482117456791908, "learning_rate": 5.297225891677675e-06, "loss": 0.7247, "mean_token_accuracy": 0.773959219455719, "num_tokens": 26345472.0, "step": 402 }, { "epoch": 0.05326460481099656, "grad_norm": 0.8261648648563853, "learning_rate": 5.310435931307795e-06, "loss": 0.7638, "mean_token_accuracy": 0.7618575096130371, "num_tokens": 26411008.0, "step": 403 }, { "epoch": 0.05339677504625958, "grad_norm": 0.8093045708068551, "learning_rate": 5.323645970937913e-06, "loss": 0.7078, "mean_token_accuracy": 0.7796972393989563, "num_tokens": 26476544.0, "step": 404 }, { "epoch": 0.0535289452815226, "grad_norm": 0.8463526417826424, "learning_rate": 5.336856010568032e-06, "loss": 0.8107, "mean_token_accuracy": 0.7501220703125, "num_tokens": 26542080.0, "step": 405 }, { "epoch": 0.05366111551678562, "grad_norm": 0.8591591203105969, "learning_rate": 5.350066050198151e-06, "loss": 0.7308, "mean_token_accuracy": 0.7746306657791138, "num_tokens": 26607616.0, "step": 406 }, { "epoch": 0.05379328575204864, "grad_norm": 0.8401932003934568, "learning_rate": 5.36327608982827e-06, "loss": 0.7262, "mean_token_accuracy": 0.7747222781181335, "num_tokens": 26673152.0, "step": 407 }, { "epoch": 0.05392545598731166, "grad_norm": 0.8744408036569079, "learning_rate": 5.376486129458389e-06, "loss": 0.7854, "mean_token_accuracy": 0.7579202651977539, "num_tokens": 26738688.0, "step": 408 }, { "epoch": 0.05405762622257468, "grad_norm": 0.8319454134207498, "learning_rate": 5.3896961690885075e-06, "loss": 0.7445, "mean_token_accuracy": 0.7688164114952087, "num_tokens": 26804224.0, "step": 409 }, { "epoch": 0.0541897964578377, "grad_norm": 0.8430064262452798, "learning_rate": 5.402906208718627e-06, "loss": 0.7995, "mean_token_accuracy": 0.7493895888328552, "num_tokens": 26869760.0, "step": 410 }, { "epoch": 0.054321966693100715, "grad_norm": 0.8625405318456912, "learning_rate": 5.416116248348745e-06, "loss": 0.7267, "mean_token_accuracy": 0.7731503844261169, "num_tokens": 26935296.0, "step": 411 }, { "epoch": 0.05445413692836373, "grad_norm": 0.8468518752796168, "learning_rate": 5.429326287978864e-06, "loss": 0.7581, "mean_token_accuracy": 0.7635514736175537, "num_tokens": 27000832.0, "step": 412 }, { "epoch": 0.05458630716362675, "grad_norm": 0.8236696867012405, "learning_rate": 5.442536327608984e-06, "loss": 0.7143, "mean_token_accuracy": 0.7759278416633606, "num_tokens": 27066368.0, "step": 413 }, { "epoch": 0.05471847739888977, "grad_norm": 0.7931068037218557, "learning_rate": 5.455746367239102e-06, "loss": 0.6959, "mean_token_accuracy": 0.783497154712677, "num_tokens": 27131904.0, "step": 414 }, { "epoch": 0.05485064763415279, "grad_norm": 0.8317450136013517, "learning_rate": 5.468956406869221e-06, "loss": 0.7195, "mean_token_accuracy": 0.7768282294273376, "num_tokens": 27197440.0, "step": 415 }, { "epoch": 0.054982817869415807, "grad_norm": 0.8178403663362296, "learning_rate": 5.48216644649934e-06, "loss": 0.739, "mean_token_accuracy": 0.769289493560791, "num_tokens": 27262976.0, "step": 416 }, { "epoch": 0.055114988104678825, "grad_norm": 0.901135376681888, "learning_rate": 5.495376486129459e-06, "loss": 0.7552, "mean_token_accuracy": 0.7635972499847412, "num_tokens": 27328512.0, "step": 417 }, { "epoch": 0.05524715833994184, "grad_norm": 0.8017759583426457, "learning_rate": 5.508586525759577e-06, "loss": 0.6849, "mean_token_accuracy": 0.7871444225311279, "num_tokens": 27394048.0, "step": 418 }, { "epoch": 0.05537932857520486, "grad_norm": 0.7439681230552563, "learning_rate": 5.5217965653896965e-06, "loss": 0.6641, "mean_token_accuracy": 0.7982236742973328, "num_tokens": 27459584.0, "step": 419 }, { "epoch": 0.05551149881046788, "grad_norm": 0.8378129295018933, "learning_rate": 5.535006605019816e-06, "loss": 0.7626, "mean_token_accuracy": 0.7625290155410767, "num_tokens": 27525120.0, "step": 420 }, { "epoch": 0.0556436690457309, "grad_norm": 0.8813531965115541, "learning_rate": 5.548216644649934e-06, "loss": 0.7128, "mean_token_accuracy": 0.7726773023605347, "num_tokens": 27590656.0, "step": 421 }, { "epoch": 0.05577583928099392, "grad_norm": 0.8772065095474026, "learning_rate": 5.561426684280053e-06, "loss": 0.8099, "mean_token_accuracy": 0.7506256699562073, "num_tokens": 27656192.0, "step": 422 }, { "epoch": 0.05590800951625694, "grad_norm": 0.8037611398860487, "learning_rate": 5.574636723910172e-06, "loss": 0.7215, "mean_token_accuracy": 0.7775149345397949, "num_tokens": 27721728.0, "step": 423 }, { "epoch": 0.05604017975151996, "grad_norm": 0.7930191303087958, "learning_rate": 5.587846763540291e-06, "loss": 0.635, "mean_token_accuracy": 0.803305447101593, "num_tokens": 27787264.0, "step": 424 }, { "epoch": 0.05617234998678298, "grad_norm": 0.8924072755343615, "learning_rate": 5.60105680317041e-06, "loss": 0.7134, "mean_token_accuracy": 0.7774386405944824, "num_tokens": 27852800.0, "step": 425 }, { "epoch": 0.056304520222045996, "grad_norm": 0.9028409113052922, "learning_rate": 5.614266842800529e-06, "loss": 0.7817, "mean_token_accuracy": 0.7571420073509216, "num_tokens": 27918336.0, "step": 426 }, { "epoch": 0.056436690457309015, "grad_norm": 0.8443937173648478, "learning_rate": 5.627476882430648e-06, "loss": 0.7204, "mean_token_accuracy": 0.7755768299102783, "num_tokens": 27983872.0, "step": 427 }, { "epoch": 0.05656886069257203, "grad_norm": 0.8418508487659485, "learning_rate": 5.640686922060766e-06, "loss": 0.7683, "mean_token_accuracy": 0.7626968622207642, "num_tokens": 28049408.0, "step": 428 }, { "epoch": 0.05670103092783505, "grad_norm": 0.8401787460121175, "learning_rate": 5.6538969616908855e-06, "loss": 0.7565, "mean_token_accuracy": 0.7651843428611755, "num_tokens": 28114944.0, "step": 429 }, { "epoch": 0.05683320116309807, "grad_norm": 0.8447684940724468, "learning_rate": 5.667107001321004e-06, "loss": 0.7316, "mean_token_accuracy": 0.7694268226623535, "num_tokens": 28180480.0, "step": 430 }, { "epoch": 0.05696537139836109, "grad_norm": 0.815267052170806, "learning_rate": 5.680317040951124e-06, "loss": 0.7314, "mean_token_accuracy": 0.775073230266571, "num_tokens": 28246016.0, "step": 431 }, { "epoch": 0.057097541633624106, "grad_norm": 0.8549071879036857, "learning_rate": 5.693527080581242e-06, "loss": 0.7561, "mean_token_accuracy": 0.7642839550971985, "num_tokens": 28311552.0, "step": 432 }, { "epoch": 0.057229711868887125, "grad_norm": 0.7963408394077313, "learning_rate": 5.706737120211361e-06, "loss": 0.7077, "mean_token_accuracy": 0.7841991186141968, "num_tokens": 28377088.0, "step": 433 }, { "epoch": 0.05736188210415014, "grad_norm": 0.8609156831778504, "learning_rate": 5.71994715984148e-06, "loss": 0.7535, "mean_token_accuracy": 0.7654895782470703, "num_tokens": 28442624.0, "step": 434 }, { "epoch": 0.05749405233941316, "grad_norm": 0.8358823402511072, "learning_rate": 5.733157199471598e-06, "loss": 0.7738, "mean_token_accuracy": 0.7624374032020569, "num_tokens": 28508160.0, "step": 435 }, { "epoch": 0.057626222574676186, "grad_norm": 0.8968508271747427, "learning_rate": 5.746367239101718e-06, "loss": 0.7392, "mean_token_accuracy": 0.7735929489135742, "num_tokens": 28573696.0, "step": 436 }, { "epoch": 0.057758392809939205, "grad_norm": 0.8669378349313499, "learning_rate": 5.759577278731837e-06, "loss": 0.7371, "mean_token_accuracy": 0.769472599029541, "num_tokens": 28639232.0, "step": 437 }, { "epoch": 0.05789056304520222, "grad_norm": 0.8172904745979976, "learning_rate": 5.772787318361956e-06, "loss": 0.6621, "mean_token_accuracy": 0.7948510646820068, "num_tokens": 28704768.0, "step": 438 }, { "epoch": 0.05802273328046524, "grad_norm": 0.8970439168423383, "learning_rate": 5.7859973579920744e-06, "loss": 0.7676, "mean_token_accuracy": 0.7612623572349548, "num_tokens": 28770304.0, "step": 439 }, { "epoch": 0.05815490351572826, "grad_norm": 0.8264360572238322, "learning_rate": 5.799207397622193e-06, "loss": 0.7481, "mean_token_accuracy": 0.7669393420219421, "num_tokens": 28835840.0, "step": 440 }, { "epoch": 0.05828707375099128, "grad_norm": 0.8905918423477456, "learning_rate": 5.8124174372523126e-06, "loss": 0.7794, "mean_token_accuracy": 0.7583933472633362, "num_tokens": 28901376.0, "step": 441 }, { "epoch": 0.058419243986254296, "grad_norm": 0.802200423857209, "learning_rate": 5.82562747688243e-06, "loss": 0.6884, "mean_token_accuracy": 0.784794270992279, "num_tokens": 28966912.0, "step": 442 }, { "epoch": 0.058551414221517314, "grad_norm": 0.872485261785001, "learning_rate": 5.83883751651255e-06, "loss": 0.7713, "mean_token_accuracy": 0.7584696412086487, "num_tokens": 29032448.0, "step": 443 }, { "epoch": 0.05868358445678033, "grad_norm": 0.8222034928254408, "learning_rate": 5.852047556142669e-06, "loss": 0.7243, "mean_token_accuracy": 0.7779574990272522, "num_tokens": 29097984.0, "step": 444 }, { "epoch": 0.05881575469204335, "grad_norm": 0.8123746553982175, "learning_rate": 5.865257595772787e-06, "loss": 0.7286, "mean_token_accuracy": 0.7730588316917419, "num_tokens": 29163520.0, "step": 445 }, { "epoch": 0.05894792492730637, "grad_norm": 0.8861670815957295, "learning_rate": 5.878467635402907e-06, "loss": 0.7096, "mean_token_accuracy": 0.778415322303772, "num_tokens": 29229056.0, "step": 446 }, { "epoch": 0.05908009516256939, "grad_norm": 0.8454495284994684, "learning_rate": 5.891677675033025e-06, "loss": 0.716, "mean_token_accuracy": 0.7753326892852783, "num_tokens": 29294592.0, "step": 447 }, { "epoch": 0.059212265397832406, "grad_norm": 0.8573829196537468, "learning_rate": 5.904887714663145e-06, "loss": 0.7142, "mean_token_accuracy": 0.779346227645874, "num_tokens": 29360128.0, "step": 448 }, { "epoch": 0.059344435633095424, "grad_norm": 0.8149731867764495, "learning_rate": 5.918097754293264e-06, "loss": 0.6899, "mean_token_accuracy": 0.7838175892829895, "num_tokens": 29425664.0, "step": 449 }, { "epoch": 0.05947660586835844, "grad_norm": 0.8583521110789449, "learning_rate": 5.931307793923382e-06, "loss": 0.7468, "mean_token_accuracy": 0.7667256593704224, "num_tokens": 29491200.0, "step": 450 }, { "epoch": 0.05960877610362147, "grad_norm": 0.871938013117209, "learning_rate": 5.9445178335535015e-06, "loss": 0.7483, "mean_token_accuracy": 0.7637803554534912, "num_tokens": 29556736.0, "step": 451 }, { "epoch": 0.059740946338884486, "grad_norm": 0.9031568950684211, "learning_rate": 5.957727873183619e-06, "loss": 0.7354, "mean_token_accuracy": 0.7704187631607056, "num_tokens": 29622272.0, "step": 452 }, { "epoch": 0.059873116574147504, "grad_norm": 0.8242640210663928, "learning_rate": 5.970937912813739e-06, "loss": 0.7584, "mean_token_accuracy": 0.7635972499847412, "num_tokens": 29687808.0, "step": 453 }, { "epoch": 0.06000528680941052, "grad_norm": 0.8247062935405328, "learning_rate": 5.9841479524438574e-06, "loss": 0.711, "mean_token_accuracy": 0.7802618741989136, "num_tokens": 29753344.0, "step": 454 }, { "epoch": 0.06013745704467354, "grad_norm": 0.8355840544828255, "learning_rate": 5.997357992073977e-06, "loss": 0.7209, "mean_token_accuracy": 0.7766756415367126, "num_tokens": 29818880.0, "step": 455 }, { "epoch": 0.06026962727993656, "grad_norm": 0.8318710197817185, "learning_rate": 6.0105680317040956e-06, "loss": 0.6861, "mean_token_accuracy": 0.7860456705093384, "num_tokens": 29884416.0, "step": 456 }, { "epoch": 0.06040179751519958, "grad_norm": 0.9596985006495974, "learning_rate": 6.023778071334214e-06, "loss": 0.7574, "mean_token_accuracy": 0.7617049217224121, "num_tokens": 29949952.0, "step": 457 }, { "epoch": 0.060533967750462596, "grad_norm": 0.8672064341834685, "learning_rate": 6.036988110964334e-06, "loss": 0.7321, "mean_token_accuracy": 0.775134265422821, "num_tokens": 30015488.0, "step": 458 }, { "epoch": 0.060666137985725614, "grad_norm": 0.8215191562300795, "learning_rate": 6.0501981505944515e-06, "loss": 0.7247, "mean_token_accuracy": 0.7717922329902649, "num_tokens": 30081024.0, "step": 459 }, { "epoch": 0.06079830822098863, "grad_norm": 0.842728668708872, "learning_rate": 6.063408190224571e-06, "loss": 0.7273, "mean_token_accuracy": 0.7706629037857056, "num_tokens": 30146560.0, "step": 460 }, { "epoch": 0.06093047845625165, "grad_norm": 0.8321088409082924, "learning_rate": 6.0766182298546905e-06, "loss": 0.7212, "mean_token_accuracy": 0.7764772176742554, "num_tokens": 30212096.0, "step": 461 }, { "epoch": 0.06106264869151467, "grad_norm": 0.8337073004634691, "learning_rate": 6.089828269484809e-06, "loss": 0.7286, "mean_token_accuracy": 0.7719295620918274, "num_tokens": 30277632.0, "step": 462 }, { "epoch": 0.06119481892677769, "grad_norm": 0.8308493878809249, "learning_rate": 6.103038309114928e-06, "loss": 0.7268, "mean_token_accuracy": 0.7753937244415283, "num_tokens": 30343168.0, "step": 463 }, { "epoch": 0.061326989162040706, "grad_norm": 0.8086135457731484, "learning_rate": 6.116248348745046e-06, "loss": 0.6991, "mean_token_accuracy": 0.7824899554252625, "num_tokens": 30408704.0, "step": 464 }, { "epoch": 0.061459159397303724, "grad_norm": 0.8551861374624733, "learning_rate": 6.129458388375166e-06, "loss": 0.7077, "mean_token_accuracy": 0.7785984873771667, "num_tokens": 30474240.0, "step": 465 }, { "epoch": 0.06159132963256675, "grad_norm": 0.9303105710777785, "learning_rate": 6.142668428005285e-06, "loss": 0.7448, "mean_token_accuracy": 0.7659779191017151, "num_tokens": 30539776.0, "step": 466 }, { "epoch": 0.06172349986782977, "grad_norm": 0.8020935397837817, "learning_rate": 6.155878467635403e-06, "loss": 0.7055, "mean_token_accuracy": 0.7766298651695251, "num_tokens": 30605312.0, "step": 467 }, { "epoch": 0.061855670103092786, "grad_norm": 0.8123332489944033, "learning_rate": 6.169088507265523e-06, "loss": 0.708, "mean_token_accuracy": 0.7792394161224365, "num_tokens": 30670848.0, "step": 468 }, { "epoch": 0.061987840338355804, "grad_norm": 0.9054271462501108, "learning_rate": 6.182298546895641e-06, "loss": 0.8004, "mean_token_accuracy": 0.7459712028503418, "num_tokens": 30736384.0, "step": 469 }, { "epoch": 0.06212001057361882, "grad_norm": 0.8148327669953519, "learning_rate": 6.19550858652576e-06, "loss": 0.7697, "mean_token_accuracy": 0.7593242526054382, "num_tokens": 30801920.0, "step": 470 }, { "epoch": 0.06225218080888184, "grad_norm": 0.8384366903514612, "learning_rate": 6.208718626155879e-06, "loss": 0.7055, "mean_token_accuracy": 0.7797735333442688, "num_tokens": 30867456.0, "step": 471 }, { "epoch": 0.06238435104414486, "grad_norm": 0.7920556053101887, "learning_rate": 6.221928665785998e-06, "loss": 0.7111, "mean_token_accuracy": 0.7795599102973938, "num_tokens": 30932992.0, "step": 472 }, { "epoch": 0.06251652127940788, "grad_norm": 0.8851069581810109, "learning_rate": 6.2351387054161176e-06, "loss": 0.7273, "mean_token_accuracy": 0.7742797136306763, "num_tokens": 30998528.0, "step": 473 }, { "epoch": 0.0626486915146709, "grad_norm": 0.8191190676127398, "learning_rate": 6.248348745046235e-06, "loss": 0.7625, "mean_token_accuracy": 0.7601178288459778, "num_tokens": 31064064.0, "step": 474 }, { "epoch": 0.06278086174993391, "grad_norm": 0.8305902065049863, "learning_rate": 6.261558784676355e-06, "loss": 0.732, "mean_token_accuracy": 0.7718685269355774, "num_tokens": 31129600.0, "step": 475 }, { "epoch": 0.06291303198519693, "grad_norm": 0.8303561576440054, "learning_rate": 6.274768824306473e-06, "loss": 0.698, "mean_token_accuracy": 0.7843517065048218, "num_tokens": 31195136.0, "step": 476 }, { "epoch": 0.06304520222045995, "grad_norm": 0.928287215775691, "learning_rate": 6.287978863936592e-06, "loss": 0.7225, "mean_token_accuracy": 0.7722042202949524, "num_tokens": 31260672.0, "step": 477 }, { "epoch": 0.06317737245572297, "grad_norm": 0.7801884285244164, "learning_rate": 6.301188903566712e-06, "loss": 0.7087, "mean_token_accuracy": 0.7819863557815552, "num_tokens": 31326208.0, "step": 478 }, { "epoch": 0.06330954269098599, "grad_norm": 0.8245273202764889, "learning_rate": 6.31439894319683e-06, "loss": 0.7371, "mean_token_accuracy": 0.7715785503387451, "num_tokens": 31391744.0, "step": 479 }, { "epoch": 0.063441712926249, "grad_norm": 0.7944464841325167, "learning_rate": 6.32760898282695e-06, "loss": 0.6906, "mean_token_accuracy": 0.783619225025177, "num_tokens": 31457280.0, "step": 480 }, { "epoch": 0.06357388316151202, "grad_norm": 0.9065494840465366, "learning_rate": 6.3408190224570676e-06, "loss": 0.7352, "mean_token_accuracy": 0.7709528803825378, "num_tokens": 31522816.0, "step": 481 }, { "epoch": 0.06370605339677504, "grad_norm": 0.8603281428158277, "learning_rate": 6.354029062087187e-06, "loss": 0.6454, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 31588352.0, "step": 482 }, { "epoch": 0.06383822363203806, "grad_norm": 0.8550435111506404, "learning_rate": 6.367239101717305e-06, "loss": 0.6978, "mean_token_accuracy": 0.7814216613769531, "num_tokens": 31653888.0, "step": 483 }, { "epoch": 0.06397039386730108, "grad_norm": 0.8547121110107044, "learning_rate": 6.380449141347424e-06, "loss": 0.6985, "mean_token_accuracy": 0.778262734413147, "num_tokens": 31719424.0, "step": 484 }, { "epoch": 0.0641025641025641, "grad_norm": 0.8682307008968082, "learning_rate": 6.393659180977544e-06, "loss": 0.7315, "mean_token_accuracy": 0.7746001482009888, "num_tokens": 31784960.0, "step": 485 }, { "epoch": 0.06423473433782712, "grad_norm": 0.9704314197447439, "learning_rate": 6.4068692206076625e-06, "loss": 0.7106, "mean_token_accuracy": 0.7768434882164001, "num_tokens": 31850496.0, "step": 486 }, { "epoch": 0.06436690457309013, "grad_norm": 0.859075086800721, "learning_rate": 6.420079260237781e-06, "loss": 0.6899, "mean_token_accuracy": 0.7845653891563416, "num_tokens": 31916032.0, "step": 487 }, { "epoch": 0.06449907480835317, "grad_norm": 0.8481498535439828, "learning_rate": 6.4332892998679e-06, "loss": 0.7828, "mean_token_accuracy": 0.7573556303977966, "num_tokens": 31981568.0, "step": 488 }, { "epoch": 0.06463124504361618, "grad_norm": 0.8895913641146433, "learning_rate": 6.446499339498019e-06, "loss": 0.7706, "mean_token_accuracy": 0.755264937877655, "num_tokens": 32047104.0, "step": 489 }, { "epoch": 0.0647634152788792, "grad_norm": 0.8944655317819636, "learning_rate": 6.459709379128139e-06, "loss": 0.7374, "mean_token_accuracy": 0.7680075764656067, "num_tokens": 32112640.0, "step": 490 }, { "epoch": 0.06489558551414222, "grad_norm": 0.907821262810655, "learning_rate": 6.4729194187582565e-06, "loss": 0.729, "mean_token_accuracy": 0.7698541283607483, "num_tokens": 32178176.0, "step": 491 }, { "epoch": 0.06502775574940524, "grad_norm": 0.8990320982328557, "learning_rate": 6.486129458388376e-06, "loss": 0.7405, "mean_token_accuracy": 0.7705289721488953, "num_tokens": 32243007.0, "step": 492 }, { "epoch": 0.06515992598466826, "grad_norm": 0.8793567125163431, "learning_rate": 6.499339498018495e-06, "loss": 0.7021, "mean_token_accuracy": 0.7803534269332886, "num_tokens": 32308543.0, "step": 493 }, { "epoch": 0.06529209621993128, "grad_norm": 0.8100524795593216, "learning_rate": 6.512549537648613e-06, "loss": 0.7155, "mean_token_accuracy": 0.7773165702819824, "num_tokens": 32374079.0, "step": 494 }, { "epoch": 0.0654242664551943, "grad_norm": 0.8330636158067326, "learning_rate": 6.525759577278732e-06, "loss": 0.7584, "mean_token_accuracy": 0.764085590839386, "num_tokens": 32439615.0, "step": 495 }, { "epoch": 0.06555643669045731, "grad_norm": 0.9252329544554811, "learning_rate": 6.538969616908851e-06, "loss": 0.7626, "mean_token_accuracy": 0.7645739316940308, "num_tokens": 32505151.0, "step": 496 }, { "epoch": 0.06568860692572033, "grad_norm": 0.8806139278424625, "learning_rate": 6.552179656538971e-06, "loss": 0.7506, "mean_token_accuracy": 0.7625290155410767, "num_tokens": 32570687.0, "step": 497 }, { "epoch": 0.06582077716098335, "grad_norm": 0.8949745159278039, "learning_rate": 6.565389696169089e-06, "loss": 0.7519, "mean_token_accuracy": 0.7662068009376526, "num_tokens": 32636223.0, "step": 498 }, { "epoch": 0.06595294739624637, "grad_norm": 0.8071974661194145, "learning_rate": 6.578599735799208e-06, "loss": 0.6755, "mean_token_accuracy": 0.7895403504371643, "num_tokens": 32701759.0, "step": 499 }, { "epoch": 0.06608511763150939, "grad_norm": 0.8281226339872482, "learning_rate": 6.591809775429327e-06, "loss": 0.7133, "mean_token_accuracy": 0.7760346531867981, "num_tokens": 32767295.0, "step": 500 }, { "epoch": 0.0662172878667724, "grad_norm": 0.8561655100760255, "learning_rate": 6.6050198150594455e-06, "loss": 0.7124, "mean_token_accuracy": 0.7788578867912292, "num_tokens": 32832831.0, "step": 501 }, { "epoch": 0.06634945810203542, "grad_norm": 0.85775697997907, "learning_rate": 6.618229854689565e-06, "loss": 0.717, "mean_token_accuracy": 0.7784610986709595, "num_tokens": 32898367.0, "step": 502 }, { "epoch": 0.06648162833729844, "grad_norm": 0.8307686961924842, "learning_rate": 6.631439894319684e-06, "loss": 0.7226, "mean_token_accuracy": 0.7735776901245117, "num_tokens": 32963903.0, "step": 503 }, { "epoch": 0.06661379857256146, "grad_norm": 0.8822178571070023, "learning_rate": 6.644649933949803e-06, "loss": 0.7392, "mean_token_accuracy": 0.7673666477203369, "num_tokens": 33029439.0, "step": 504 }, { "epoch": 0.06674596880782448, "grad_norm": 0.8612414357604269, "learning_rate": 6.657859973579921e-06, "loss": 0.7307, "mean_token_accuracy": 0.7717464566230774, "num_tokens": 33094975.0, "step": 505 }, { "epoch": 0.0668781390430875, "grad_norm": 0.951321297983899, "learning_rate": 6.67107001321004e-06, "loss": 0.764, "mean_token_accuracy": 0.7666646242141724, "num_tokens": 33160511.0, "step": 506 }, { "epoch": 0.06701030927835051, "grad_norm": 0.9573599715034806, "learning_rate": 6.684280052840158e-06, "loss": 0.7777, "mean_token_accuracy": 0.7590953707695007, "num_tokens": 33226047.0, "step": 507 }, { "epoch": 0.06714247951361353, "grad_norm": 0.8169385581010877, "learning_rate": 6.697490092470278e-06, "loss": 0.6854, "mean_token_accuracy": 0.7855115532875061, "num_tokens": 33291583.0, "step": 508 }, { "epoch": 0.06727464974887655, "grad_norm": 0.9402206811339456, "learning_rate": 6.710700132100397e-06, "loss": 0.7392, "mean_token_accuracy": 0.7682669758796692, "num_tokens": 33357119.0, "step": 509 }, { "epoch": 0.06740681998413957, "grad_norm": 0.7869756223515435, "learning_rate": 6.723910171730516e-06, "loss": 0.7066, "mean_token_accuracy": 0.7799871563911438, "num_tokens": 33422655.0, "step": 510 }, { "epoch": 0.06753899021940259, "grad_norm": 0.8969915393867541, "learning_rate": 6.737120211360635e-06, "loss": 0.717, "mean_token_accuracy": 0.7711970210075378, "num_tokens": 33488191.0, "step": 511 }, { "epoch": 0.0676711604546656, "grad_norm": 0.8669165474295265, "learning_rate": 6.750330250990753e-06, "loss": 0.7319, "mean_token_accuracy": 0.7709376215934753, "num_tokens": 33553727.0, "step": 512 }, { "epoch": 0.06780333068992862, "grad_norm": 0.8020745678087539, "learning_rate": 6.7635402906208726e-06, "loss": 0.7053, "mean_token_accuracy": 0.7811927795410156, "num_tokens": 33619263.0, "step": 513 }, { "epoch": 0.06793550092519164, "grad_norm": 0.9025885916709108, "learning_rate": 6.776750330250992e-06, "loss": 0.7123, "mean_token_accuracy": 0.7770418524742126, "num_tokens": 33684799.0, "step": 514 }, { "epoch": 0.06806767116045466, "grad_norm": 0.9137515553587665, "learning_rate": 6.78996036988111e-06, "loss": 0.7126, "mean_token_accuracy": 0.7774081230163574, "num_tokens": 33750335.0, "step": 515 }, { "epoch": 0.06819984139571768, "grad_norm": 0.9124967973036683, "learning_rate": 6.803170409511229e-06, "loss": 0.7336, "mean_token_accuracy": 0.7715938091278076, "num_tokens": 33815871.0, "step": 516 }, { "epoch": 0.0683320116309807, "grad_norm": 0.7549753019450375, "learning_rate": 6.816380449141348e-06, "loss": 0.6407, "mean_token_accuracy": 0.7968959808349609, "num_tokens": 33881407.0, "step": 517 }, { "epoch": 0.06846418186624373, "grad_norm": 0.95099721434128, "learning_rate": 6.829590488771467e-06, "loss": 0.7852, "mean_token_accuracy": 0.755509078502655, "num_tokens": 33946943.0, "step": 518 }, { "epoch": 0.06859635210150675, "grad_norm": 0.9968586672709484, "learning_rate": 6.842800528401585e-06, "loss": 0.7584, "mean_token_accuracy": 0.7630783915519714, "num_tokens": 34012479.0, "step": 519 }, { "epoch": 0.06872852233676977, "grad_norm": 0.8617140334427917, "learning_rate": 6.856010568031705e-06, "loss": 0.6766, "mean_token_accuracy": 0.7866560816764832, "num_tokens": 34078015.0, "step": 520 }, { "epoch": 0.06886069257203278, "grad_norm": 0.8585260309686527, "learning_rate": 6.869220607661824e-06, "loss": 0.6931, "mean_token_accuracy": 0.783832848072052, "num_tokens": 34143551.0, "step": 521 }, { "epoch": 0.0689928628072958, "grad_norm": 0.9057668228899581, "learning_rate": 6.882430647291942e-06, "loss": 0.6665, "mean_token_accuracy": 0.7913563847541809, "num_tokens": 34209087.0, "step": 522 }, { "epoch": 0.06912503304255882, "grad_norm": 0.8259505394675988, "learning_rate": 6.8956406869220615e-06, "loss": 0.7352, "mean_token_accuracy": 0.7706476449966431, "num_tokens": 34274623.0, "step": 523 }, { "epoch": 0.06925720327782184, "grad_norm": 0.9079550207693405, "learning_rate": 6.90885072655218e-06, "loss": 0.7324, "mean_token_accuracy": 0.7700524926185608, "num_tokens": 34340159.0, "step": 524 }, { "epoch": 0.06938937351308486, "grad_norm": 0.960120943737717, "learning_rate": 6.922060766182299e-06, "loss": 0.7571, "mean_token_accuracy": 0.7628341913223267, "num_tokens": 34405695.0, "step": 525 }, { "epoch": 0.06952154374834787, "grad_norm": 0.9197098405385515, "learning_rate": 6.935270805812418e-06, "loss": 0.7153, "mean_token_accuracy": 0.7773928642272949, "num_tokens": 34471231.0, "step": 526 }, { "epoch": 0.0696537139836109, "grad_norm": 0.846754995308951, "learning_rate": 6.948480845442537e-06, "loss": 0.7211, "mean_token_accuracy": 0.7714869976043701, "num_tokens": 34536767.0, "step": 527 }, { "epoch": 0.06978588421887391, "grad_norm": 0.8409605490303296, "learning_rate": 6.9616908850726564e-06, "loss": 0.6486, "mean_token_accuracy": 0.797613263130188, "num_tokens": 34602303.0, "step": 528 }, { "epoch": 0.06991805445413693, "grad_norm": 0.8339167486455162, "learning_rate": 6.974900924702774e-06, "loss": 0.6789, "mean_token_accuracy": 0.7900134325027466, "num_tokens": 34667839.0, "step": 529 }, { "epoch": 0.07005022468939995, "grad_norm": 0.8982102926170575, "learning_rate": 6.988110964332894e-06, "loss": 0.7145, "mean_token_accuracy": 0.7762788534164429, "num_tokens": 34733375.0, "step": 530 }, { "epoch": 0.07018239492466297, "grad_norm": 0.9565920703936324, "learning_rate": 7.001321003963012e-06, "loss": 0.7733, "mean_token_accuracy": 0.7567299604415894, "num_tokens": 34798911.0, "step": 531 }, { "epoch": 0.07031456515992598, "grad_norm": 0.8636680030924074, "learning_rate": 7.014531043593131e-06, "loss": 0.7284, "mean_token_accuracy": 0.7743712663650513, "num_tokens": 34864447.0, "step": 532 }, { "epoch": 0.070446735395189, "grad_norm": 0.8230151040630274, "learning_rate": 7.0277410832232505e-06, "loss": 0.7573, "mean_token_accuracy": 0.7638413906097412, "num_tokens": 34929983.0, "step": 533 }, { "epoch": 0.07057890563045202, "grad_norm": 0.9135581594239227, "learning_rate": 7.040951122853369e-06, "loss": 0.7793, "mean_token_accuracy": 0.7550054788589478, "num_tokens": 34995519.0, "step": 534 }, { "epoch": 0.07071107586571504, "grad_norm": 0.9534681697021468, "learning_rate": 7.054161162483489e-06, "loss": 0.7593, "mean_token_accuracy": 0.7615218162536621, "num_tokens": 35061055.0, "step": 535 }, { "epoch": 0.07084324610097806, "grad_norm": 0.9673069884733553, "learning_rate": 7.067371202113606e-06, "loss": 0.7308, "mean_token_accuracy": 0.7723110914230347, "num_tokens": 35126591.0, "step": 536 }, { "epoch": 0.07097541633624108, "grad_norm": 1.0113764535804923, "learning_rate": 7.080581241743726e-06, "loss": 0.7452, "mean_token_accuracy": 0.7664204835891724, "num_tokens": 35192127.0, "step": 537 }, { "epoch": 0.0711075865715041, "grad_norm": 1.4432393310032265, "learning_rate": 7.093791281373845e-06, "loss": 0.7261, "mean_token_accuracy": 0.7723568677902222, "num_tokens": 35257663.0, "step": 538 }, { "epoch": 0.07123975680676711, "grad_norm": 0.8974505690177144, "learning_rate": 7.107001321003963e-06, "loss": 0.7027, "mean_token_accuracy": 0.7805213332176208, "num_tokens": 35323199.0, "step": 539 }, { "epoch": 0.07137192704203013, "grad_norm": 1.0445652588674534, "learning_rate": 7.120211360634083e-06, "loss": 0.7604, "mean_token_accuracy": 0.7606824636459351, "num_tokens": 35388735.0, "step": 540 }, { "epoch": 0.07150409727729315, "grad_norm": 0.8963541763057167, "learning_rate": 7.133421400264201e-06, "loss": 0.7363, "mean_token_accuracy": 0.7717922329902649, "num_tokens": 35454271.0, "step": 541 }, { "epoch": 0.07163626751255617, "grad_norm": 0.8643859418809366, "learning_rate": 7.146631439894321e-06, "loss": 0.7481, "mean_token_accuracy": 0.7665730714797974, "num_tokens": 35519807.0, "step": 542 }, { "epoch": 0.07176843774781919, "grad_norm": 0.8604224811618412, "learning_rate": 7.159841479524439e-06, "loss": 0.6938, "mean_token_accuracy": 0.7831003665924072, "num_tokens": 35585343.0, "step": 543 }, { "epoch": 0.0719006079830822, "grad_norm": 0.8998537383154747, "learning_rate": 7.173051519154558e-06, "loss": 0.7002, "mean_token_accuracy": 0.7814064025878906, "num_tokens": 35650879.0, "step": 544 }, { "epoch": 0.07203277821834522, "grad_norm": 0.9086850112757792, "learning_rate": 7.1862615587846776e-06, "loss": 0.7557, "mean_token_accuracy": 0.7660084366798401, "num_tokens": 35716415.0, "step": 545 }, { "epoch": 0.07216494845360824, "grad_norm": 0.8446986389358476, "learning_rate": 7.199471598414795e-06, "loss": 0.6845, "mean_token_accuracy": 0.783558189868927, "num_tokens": 35781951.0, "step": 546 }, { "epoch": 0.07229711868887127, "grad_norm": 0.9292815178337758, "learning_rate": 7.212681638044915e-06, "loss": 0.7224, "mean_token_accuracy": 0.774081289768219, "num_tokens": 35847487.0, "step": 547 }, { "epoch": 0.07242928892413429, "grad_norm": 0.8974451332363738, "learning_rate": 7.2258916776750335e-06, "loss": 0.7377, "mean_token_accuracy": 0.7679312825202942, "num_tokens": 35913023.0, "step": 548 }, { "epoch": 0.07256145915939731, "grad_norm": 0.9413592987335078, "learning_rate": 7.239101717305152e-06, "loss": 0.7741, "mean_token_accuracy": 0.7586833238601685, "num_tokens": 35978559.0, "step": 549 }, { "epoch": 0.07269362939466033, "grad_norm": 0.9836939169092944, "learning_rate": 7.252311756935272e-06, "loss": 0.7633, "mean_token_accuracy": 0.7594768404960632, "num_tokens": 36044095.0, "step": 550 }, { "epoch": 0.07282579962992335, "grad_norm": 0.8654460469600876, "learning_rate": 7.26552179656539e-06, "loss": 0.6844, "mean_token_accuracy": 0.7895861268043518, "num_tokens": 36109631.0, "step": 551 }, { "epoch": 0.07295796986518636, "grad_norm": 0.9145406740747142, "learning_rate": 7.27873183619551e-06, "loss": 0.7158, "mean_token_accuracy": 0.7771029472351074, "num_tokens": 36175167.0, "step": 552 }, { "epoch": 0.07309014010044938, "grad_norm": 0.8941584061951514, "learning_rate": 7.2919418758256276e-06, "loss": 0.7205, "mean_token_accuracy": 0.7732572555541992, "num_tokens": 36240703.0, "step": 553 }, { "epoch": 0.0732223103357124, "grad_norm": 0.824965932601898, "learning_rate": 7.305151915455747e-06, "loss": 0.6825, "mean_token_accuracy": 0.7852826118469238, "num_tokens": 36306239.0, "step": 554 }, { "epoch": 0.07335448057097542, "grad_norm": 0.8728184022553787, "learning_rate": 7.318361955085866e-06, "loss": 0.7111, "mean_token_accuracy": 0.7775607109069824, "num_tokens": 36371775.0, "step": 555 }, { "epoch": 0.07348665080623844, "grad_norm": 0.9238618515729171, "learning_rate": 7.331571994715984e-06, "loss": 0.7185, "mean_token_accuracy": 0.7717006206512451, "num_tokens": 36437311.0, "step": 556 }, { "epoch": 0.07361882104150146, "grad_norm": 1.0279498714213355, "learning_rate": 7.344782034346104e-06, "loss": 0.7742, "mean_token_accuracy": 0.7568062543869019, "num_tokens": 36502847.0, "step": 557 }, { "epoch": 0.07375099127676447, "grad_norm": 0.9451901336404311, "learning_rate": 7.3579920739762225e-06, "loss": 0.7576, "mean_token_accuracy": 0.7635667324066162, "num_tokens": 36568383.0, "step": 558 }, { "epoch": 0.07388316151202749, "grad_norm": 0.9144896927745266, "learning_rate": 7.371202113606342e-06, "loss": 0.7388, "mean_token_accuracy": 0.7674276828765869, "num_tokens": 36633919.0, "step": 559 }, { "epoch": 0.07401533174729051, "grad_norm": 0.8771010076146156, "learning_rate": 7.38441215323646e-06, "loss": 0.763, "mean_token_accuracy": 0.7620711922645569, "num_tokens": 36699455.0, "step": 560 }, { "epoch": 0.07414750198255353, "grad_norm": 0.8865133147092913, "learning_rate": 7.397622192866579e-06, "loss": 0.7055, "mean_token_accuracy": 0.7799261212348938, "num_tokens": 36764991.0, "step": 561 }, { "epoch": 0.07427967221781655, "grad_norm": 0.840694642390859, "learning_rate": 7.410832232496699e-06, "loss": 0.6636, "mean_token_accuracy": 0.7912189960479736, "num_tokens": 36830527.0, "step": 562 }, { "epoch": 0.07441184245307957, "grad_norm": 0.982653191657337, "learning_rate": 7.4240422721268165e-06, "loss": 0.712, "mean_token_accuracy": 0.7787968516349792, "num_tokens": 36896063.0, "step": 563 }, { "epoch": 0.07454401268834258, "grad_norm": 0.9392785374402683, "learning_rate": 7.437252311756936e-06, "loss": 0.7416, "mean_token_accuracy": 0.7685874700546265, "num_tokens": 36961599.0, "step": 564 }, { "epoch": 0.0746761829236056, "grad_norm": 0.8103161032594277, "learning_rate": 7.450462351387055e-06, "loss": 0.6594, "mean_token_accuracy": 0.7956598997116089, "num_tokens": 37027135.0, "step": 565 }, { "epoch": 0.07480835315886862, "grad_norm": 0.8165712222928677, "learning_rate": 7.463672391017174e-06, "loss": 0.6094, "mean_token_accuracy": 0.8101574778556824, "num_tokens": 37092671.0, "step": 566 }, { "epoch": 0.07494052339413164, "grad_norm": 1.0247286381362133, "learning_rate": 7.476882430647292e-06, "loss": 0.742, "mean_token_accuracy": 0.7668324708938599, "num_tokens": 37158207.0, "step": 567 }, { "epoch": 0.07507269362939466, "grad_norm": 0.8946589192744797, "learning_rate": 7.490092470277411e-06, "loss": 0.7234, "mean_token_accuracy": 0.7729367613792419, "num_tokens": 37223743.0, "step": 568 }, { "epoch": 0.07520486386465768, "grad_norm": 0.8889651821230654, "learning_rate": 7.503302509907531e-06, "loss": 0.7342, "mean_token_accuracy": 0.7705103158950806, "num_tokens": 37289279.0, "step": 569 }, { "epoch": 0.0753370340999207, "grad_norm": 0.9529915389962619, "learning_rate": 7.516512549537649e-06, "loss": 0.712, "mean_token_accuracy": 0.7729825377464294, "num_tokens": 37354815.0, "step": 570 }, { "epoch": 0.07546920433518371, "grad_norm": 0.866690353706755, "learning_rate": 7.529722589167768e-06, "loss": 0.7214, "mean_token_accuracy": 0.7730588316917419, "num_tokens": 37420351.0, "step": 571 }, { "epoch": 0.07560137457044673, "grad_norm": 0.9789671695574619, "learning_rate": 7.542932628797887e-06, "loss": 0.7239, "mean_token_accuracy": 0.769381046295166, "num_tokens": 37485887.0, "step": 572 }, { "epoch": 0.07573354480570975, "grad_norm": 0.8795218393323827, "learning_rate": 7.556142668428006e-06, "loss": 0.7208, "mean_token_accuracy": 0.774111807346344, "num_tokens": 37551423.0, "step": 573 }, { "epoch": 0.07586571504097277, "grad_norm": 0.88888134465908, "learning_rate": 7.569352708058125e-06, "loss": 0.7017, "mean_token_accuracy": 0.7783085107803345, "num_tokens": 37616959.0, "step": 574 }, { "epoch": 0.07599788527623579, "grad_norm": 0.9112273619391054, "learning_rate": 7.582562747688244e-06, "loss": 0.7584, "mean_token_accuracy": 0.7615218162536621, "num_tokens": 37682495.0, "step": 575 }, { "epoch": 0.0761300555114988, "grad_norm": 0.8772804564535248, "learning_rate": 7.595772787318363e-06, "loss": 0.7167, "mean_token_accuracy": 0.7758343815803528, "num_tokens": 37746652.0, "step": 576 }, { "epoch": 0.07626222574676184, "grad_norm": 0.9139183896880267, "learning_rate": 7.608982826948481e-06, "loss": 0.7587, "mean_token_accuracy": 0.7624679803848267, "num_tokens": 37812188.0, "step": 577 }, { "epoch": 0.07639439598202485, "grad_norm": 0.8491005575380884, "learning_rate": 7.6221928665786e-06, "loss": 0.7146, "mean_token_accuracy": 0.7776980996131897, "num_tokens": 37877724.0, "step": 578 }, { "epoch": 0.07652656621728787, "grad_norm": 0.9308193164797792, "learning_rate": 7.635402906208719e-06, "loss": 0.7358, "mean_token_accuracy": 0.7698236107826233, "num_tokens": 37943260.0, "step": 579 }, { "epoch": 0.07665873645255089, "grad_norm": 0.8561060954528057, "learning_rate": 7.648612945838839e-06, "loss": 0.7303, "mean_token_accuracy": 0.7709681391716003, "num_tokens": 38008796.0, "step": 580 }, { "epoch": 0.07679090668781391, "grad_norm": 0.8920783704448746, "learning_rate": 7.661822985468958e-06, "loss": 0.687, "mean_token_accuracy": 0.7819100022315979, "num_tokens": 38074332.0, "step": 581 }, { "epoch": 0.07692307692307693, "grad_norm": 0.9153234182253237, "learning_rate": 7.675033025099076e-06, "loss": 0.7307, "mean_token_accuracy": 0.7701898217201233, "num_tokens": 38139868.0, "step": 582 }, { "epoch": 0.07705524715833995, "grad_norm": 0.9007467085815235, "learning_rate": 7.688243064729195e-06, "loss": 0.7757, "mean_token_accuracy": 0.758942723274231, "num_tokens": 38205404.0, "step": 583 }, { "epoch": 0.07718741739360296, "grad_norm": 0.862487369030809, "learning_rate": 7.701453104359313e-06, "loss": 0.6986, "mean_token_accuracy": 0.778262734413147, "num_tokens": 38270940.0, "step": 584 }, { "epoch": 0.07731958762886598, "grad_norm": 0.9219322994883185, "learning_rate": 7.714663143989433e-06, "loss": 0.7027, "mean_token_accuracy": 0.7771182060241699, "num_tokens": 38336476.0, "step": 585 }, { "epoch": 0.077451757864129, "grad_norm": 0.8586460534702335, "learning_rate": 7.727873183619552e-06, "loss": 0.6667, "mean_token_accuracy": 0.7877700924873352, "num_tokens": 38402012.0, "step": 586 }, { "epoch": 0.07758392809939202, "grad_norm": 0.9258116423208996, "learning_rate": 7.74108322324967e-06, "loss": 0.6999, "mean_token_accuracy": 0.779346227645874, "num_tokens": 38467548.0, "step": 587 }, { "epoch": 0.07771609833465504, "grad_norm": 0.8980592714881938, "learning_rate": 7.75429326287979e-06, "loss": 0.727, "mean_token_accuracy": 0.7673208117485046, "num_tokens": 38533084.0, "step": 588 }, { "epoch": 0.07784826856991806, "grad_norm": 0.793058539253305, "learning_rate": 7.767503302509907e-06, "loss": 0.6315, "mean_token_accuracy": 0.8005585670471191, "num_tokens": 38598620.0, "step": 589 }, { "epoch": 0.07798043880518107, "grad_norm": 0.8499887110666704, "learning_rate": 7.780713342140027e-06, "loss": 0.6943, "mean_token_accuracy": 0.7849164009094238, "num_tokens": 38664156.0, "step": 590 }, { "epoch": 0.07811260904044409, "grad_norm": 0.8900124655109088, "learning_rate": 7.793923381770146e-06, "loss": 0.6803, "mean_token_accuracy": 0.7873886227607727, "num_tokens": 38729692.0, "step": 591 }, { "epoch": 0.07824477927570711, "grad_norm": 0.9028271965856384, "learning_rate": 7.807133421400266e-06, "loss": 0.728, "mean_token_accuracy": 0.7710749506950378, "num_tokens": 38795228.0, "step": 592 }, { "epoch": 0.07837694951097013, "grad_norm": 1.0142441202488046, "learning_rate": 7.820343461030383e-06, "loss": 0.7528, "mean_token_accuracy": 0.7634904384613037, "num_tokens": 38860764.0, "step": 593 }, { "epoch": 0.07850911974623315, "grad_norm": 0.8819524989670187, "learning_rate": 7.833553500660503e-06, "loss": 0.7136, "mean_token_accuracy": 0.7740965485572815, "num_tokens": 38926300.0, "step": 594 }, { "epoch": 0.07864128998149617, "grad_norm": 0.902605277691991, "learning_rate": 7.846763540290622e-06, "loss": 0.7247, "mean_token_accuracy": 0.7726773023605347, "num_tokens": 38991836.0, "step": 595 }, { "epoch": 0.07877346021675918, "grad_norm": 0.9681945864702138, "learning_rate": 7.85997357992074e-06, "loss": 0.7173, "mean_token_accuracy": 0.7740965485572815, "num_tokens": 39057372.0, "step": 596 }, { "epoch": 0.0789056304520222, "grad_norm": 0.9201650934536937, "learning_rate": 7.87318361955086e-06, "loss": 0.666, "mean_token_accuracy": 0.7897844910621643, "num_tokens": 39122908.0, "step": 597 }, { "epoch": 0.07903780068728522, "grad_norm": 0.9143373994140365, "learning_rate": 7.886393659180979e-06, "loss": 0.7471, "mean_token_accuracy": 0.764055073261261, "num_tokens": 39188444.0, "step": 598 }, { "epoch": 0.07916997092254824, "grad_norm": 0.9089495436695441, "learning_rate": 7.899603698811097e-06, "loss": 0.6765, "mean_token_accuracy": 0.7849926948547363, "num_tokens": 39253980.0, "step": 599 }, { "epoch": 0.07930214115781126, "grad_norm": 0.9294342177221803, "learning_rate": 7.912813738441216e-06, "loss": 0.7275, "mean_token_accuracy": 0.7705713510513306, "num_tokens": 39319516.0, "step": 600 }, { "epoch": 0.07943431139307428, "grad_norm": 0.9139241746172192, "learning_rate": 7.926023778071334e-06, "loss": 0.7141, "mean_token_accuracy": 0.7741575837135315, "num_tokens": 39385052.0, "step": 601 }, { "epoch": 0.0795664816283373, "grad_norm": 0.8462100167604867, "learning_rate": 7.939233817701454e-06, "loss": 0.7419, "mean_token_accuracy": 0.7720058560371399, "num_tokens": 39450588.0, "step": 602 }, { "epoch": 0.07969865186360031, "grad_norm": 0.8680144106236815, "learning_rate": 7.952443857331572e-06, "loss": 0.7179, "mean_token_accuracy": 0.7762025594711304, "num_tokens": 39516124.0, "step": 603 }, { "epoch": 0.07983082209886333, "grad_norm": 0.9208432692864699, "learning_rate": 7.965653896961691e-06, "loss": 0.7318, "mean_token_accuracy": 0.7697778344154358, "num_tokens": 39581660.0, "step": 604 }, { "epoch": 0.07996299233412635, "grad_norm": 0.9468431390109421, "learning_rate": 7.97886393659181e-06, "loss": 0.7418, "mean_token_accuracy": 0.7648638486862183, "num_tokens": 39647196.0, "step": 605 }, { "epoch": 0.08009516256938938, "grad_norm": 0.8392096876567526, "learning_rate": 7.992073976221928e-06, "loss": 0.65, "mean_token_accuracy": 0.7948052883148193, "num_tokens": 39712732.0, "step": 606 }, { "epoch": 0.0802273328046524, "grad_norm": 0.8840618309255123, "learning_rate": 8.005284015852048e-06, "loss": 0.7274, "mean_token_accuracy": 0.7699151635169983, "num_tokens": 39778268.0, "step": 607 }, { "epoch": 0.08035950303991542, "grad_norm": 0.920225633239678, "learning_rate": 8.018494055482167e-06, "loss": 0.7028, "mean_token_accuracy": 0.778171181678772, "num_tokens": 39843804.0, "step": 608 }, { "epoch": 0.08049167327517844, "grad_norm": 0.9441401877984895, "learning_rate": 8.031704095112287e-06, "loss": 0.7432, "mean_token_accuracy": 0.7658100128173828, "num_tokens": 39909340.0, "step": 609 }, { "epoch": 0.08062384351044145, "grad_norm": 0.9727958123865195, "learning_rate": 8.044914134742405e-06, "loss": 0.7532, "mean_token_accuracy": 0.764055073261261, "num_tokens": 39974876.0, "step": 610 }, { "epoch": 0.08075601374570447, "grad_norm": 0.8639159966236043, "learning_rate": 8.058124174372524e-06, "loss": 0.6752, "mean_token_accuracy": 0.789097785949707, "num_tokens": 40040412.0, "step": 611 }, { "epoch": 0.08088818398096749, "grad_norm": 0.8932204258944219, "learning_rate": 8.071334214002644e-06, "loss": 0.6702, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 40105948.0, "step": 612 }, { "epoch": 0.08102035421623051, "grad_norm": 0.8151665439327641, "learning_rate": 8.084544253632761e-06, "loss": 0.652, "mean_token_accuracy": 0.7930350303649902, "num_tokens": 40171484.0, "step": 613 }, { "epoch": 0.08115252445149353, "grad_norm": 0.8632537733400508, "learning_rate": 8.09775429326288e-06, "loss": 0.6349, "mean_token_accuracy": 0.79897141456604, "num_tokens": 40237020.0, "step": 614 }, { "epoch": 0.08128469468675655, "grad_norm": 0.9112801962245389, "learning_rate": 8.110964332893e-06, "loss": 0.7317, "mean_token_accuracy": 0.7689232230186462, "num_tokens": 40302556.0, "step": 615 }, { "epoch": 0.08141686492201956, "grad_norm": 0.9436209991621812, "learning_rate": 8.124174372523118e-06, "loss": 0.7503, "mean_token_accuracy": 0.7655963897705078, "num_tokens": 40368092.0, "step": 616 }, { "epoch": 0.08154903515728258, "grad_norm": 0.9397199460609994, "learning_rate": 8.137384412153238e-06, "loss": 0.7238, "mean_token_accuracy": 0.7710902094841003, "num_tokens": 40433628.0, "step": 617 }, { "epoch": 0.0816812053925456, "grad_norm": 0.8366382966885092, "learning_rate": 8.150594451783355e-06, "loss": 0.6031, "mean_token_accuracy": 0.807669997215271, "num_tokens": 40499164.0, "step": 618 }, { "epoch": 0.08181337562780862, "grad_norm": 0.8580979609724044, "learning_rate": 8.163804491413475e-06, "loss": 0.6046, "mean_token_accuracy": 0.8100201487541199, "num_tokens": 40564700.0, "step": 619 }, { "epoch": 0.08194554586307164, "grad_norm": 0.9058085960171806, "learning_rate": 8.177014531043593e-06, "loss": 0.6615, "mean_token_accuracy": 0.7922720313072205, "num_tokens": 40630236.0, "step": 620 }, { "epoch": 0.08207771609833465, "grad_norm": 0.8912284885871566, "learning_rate": 8.190224570673712e-06, "loss": 0.6594, "mean_token_accuracy": 0.7972622513771057, "num_tokens": 40695772.0, "step": 621 }, { "epoch": 0.08220988633359767, "grad_norm": 0.8277811925273747, "learning_rate": 8.203434610303832e-06, "loss": 0.703, "mean_token_accuracy": 0.7795904278755188, "num_tokens": 40761308.0, "step": 622 }, { "epoch": 0.08234205656886069, "grad_norm": 0.8908104096131122, "learning_rate": 8.216644649933951e-06, "loss": 0.7055, "mean_token_accuracy": 0.7788731455802917, "num_tokens": 40826844.0, "step": 623 }, { "epoch": 0.08247422680412371, "grad_norm": 0.9086484462772032, "learning_rate": 8.229854689564069e-06, "loss": 0.7256, "mean_token_accuracy": 0.7706018686294556, "num_tokens": 40892380.0, "step": 624 }, { "epoch": 0.08260639703938673, "grad_norm": 0.8805391817739724, "learning_rate": 8.243064729194188e-06, "loss": 0.701, "mean_token_accuracy": 0.7796514630317688, "num_tokens": 40957916.0, "step": 625 }, { "epoch": 0.08273856727464975, "grad_norm": 0.8866057795658359, "learning_rate": 8.256274768824308e-06, "loss": 0.708, "mean_token_accuracy": 0.7771029472351074, "num_tokens": 41023452.0, "step": 626 }, { "epoch": 0.08287073750991276, "grad_norm": 0.9202412099051084, "learning_rate": 8.269484808454427e-06, "loss": 0.7365, "mean_token_accuracy": 0.768419623374939, "num_tokens": 41088988.0, "step": 627 }, { "epoch": 0.08300290774517578, "grad_norm": 0.9078083849274796, "learning_rate": 8.282694848084545e-06, "loss": 0.7222, "mean_token_accuracy": 0.7723568677902222, "num_tokens": 41154524.0, "step": 628 }, { "epoch": 0.0831350779804388, "grad_norm": 0.9509132611008836, "learning_rate": 8.295904887714665e-06, "loss": 0.7478, "mean_token_accuracy": 0.7687095403671265, "num_tokens": 41220060.0, "step": 629 }, { "epoch": 0.08326724821570182, "grad_norm": 0.9278103898687908, "learning_rate": 8.309114927344782e-06, "loss": 0.7252, "mean_token_accuracy": 0.7762483358383179, "num_tokens": 41285596.0, "step": 630 }, { "epoch": 0.08339941845096484, "grad_norm": 0.8942933745725473, "learning_rate": 8.322324966974902e-06, "loss": 0.7205, "mean_token_accuracy": 0.7729520201683044, "num_tokens": 41351132.0, "step": 631 }, { "epoch": 0.08353158868622786, "grad_norm": 0.8288100068940627, "learning_rate": 8.33553500660502e-06, "loss": 0.6946, "mean_token_accuracy": 0.7813606262207031, "num_tokens": 41416668.0, "step": 632 }, { "epoch": 0.08366375892149087, "grad_norm": 0.9470753236867425, "learning_rate": 8.34874504623514e-06, "loss": 0.6739, "mean_token_accuracy": 0.7883958220481873, "num_tokens": 41482204.0, "step": 633 }, { "epoch": 0.08379592915675389, "grad_norm": 0.8826788586721321, "learning_rate": 8.361955085865259e-06, "loss": 0.675, "mean_token_accuracy": 0.7885484099388123, "num_tokens": 41547740.0, "step": 634 }, { "epoch": 0.08392809939201691, "grad_norm": 0.9296276987736091, "learning_rate": 8.375165125495377e-06, "loss": 0.6895, "mean_token_accuracy": 0.7821999788284302, "num_tokens": 41613276.0, "step": 635 }, { "epoch": 0.08406026962727994, "grad_norm": 0.9741772500873176, "learning_rate": 8.388375165125496e-06, "loss": 0.7456, "mean_token_accuracy": 0.7662830948829651, "num_tokens": 41678812.0, "step": 636 }, { "epoch": 0.08419243986254296, "grad_norm": 0.9128332125015621, "learning_rate": 8.401585204755614e-06, "loss": 0.7364, "mean_token_accuracy": 0.769411563873291, "num_tokens": 41744348.0, "step": 637 }, { "epoch": 0.08432461009780598, "grad_norm": 0.9259617889009492, "learning_rate": 8.414795244385733e-06, "loss": 0.7265, "mean_token_accuracy": 0.7740965485572815, "num_tokens": 41809884.0, "step": 638 }, { "epoch": 0.084456780333069, "grad_norm": 0.9486391758084334, "learning_rate": 8.428005284015853e-06, "loss": 0.7363, "mean_token_accuracy": 0.7672292590141296, "num_tokens": 41875420.0, "step": 639 }, { "epoch": 0.08458895056833202, "grad_norm": 0.8397963670569345, "learning_rate": 8.441215323645972e-06, "loss": 0.6671, "mean_token_accuracy": 0.7894945740699768, "num_tokens": 41940956.0, "step": 640 }, { "epoch": 0.08472112080359503, "grad_norm": 0.8718681032480261, "learning_rate": 8.45442536327609e-06, "loss": 0.703, "mean_token_accuracy": 0.7810554504394531, "num_tokens": 42006492.0, "step": 641 }, { "epoch": 0.08485329103885805, "grad_norm": 0.9118178823363788, "learning_rate": 8.46763540290621e-06, "loss": 0.6963, "mean_token_accuracy": 0.782871425151825, "num_tokens": 42072028.0, "step": 642 }, { "epoch": 0.08498546127412107, "grad_norm": 0.8522403517370312, "learning_rate": 8.480845442536329e-06, "loss": 0.6739, "mean_token_accuracy": 0.7873886227607727, "num_tokens": 42137564.0, "step": 643 }, { "epoch": 0.08511763150938409, "grad_norm": 1.0265911604684996, "learning_rate": 8.494055482166447e-06, "loss": 0.7829, "mean_token_accuracy": 0.7535099387168884, "num_tokens": 42203100.0, "step": 644 }, { "epoch": 0.08524980174464711, "grad_norm": 0.9458454389253749, "learning_rate": 8.507265521796566e-06, "loss": 0.7108, "mean_token_accuracy": 0.7773165702819824, "num_tokens": 42268636.0, "step": 645 }, { "epoch": 0.08538197197991013, "grad_norm": 0.9725821236520416, "learning_rate": 8.520475561426686e-06, "loss": 0.7344, "mean_token_accuracy": 0.7684348821640015, "num_tokens": 42334172.0, "step": 646 }, { "epoch": 0.08551414221517314, "grad_norm": 0.9431877993555253, "learning_rate": 8.533685601056804e-06, "loss": 0.7251, "mean_token_accuracy": 0.7720363736152649, "num_tokens": 42399708.0, "step": 647 }, { "epoch": 0.08564631245043616, "grad_norm": 0.9334982796149303, "learning_rate": 8.546895640686923e-06, "loss": 0.703, "mean_token_accuracy": 0.7766298651695251, "num_tokens": 42465244.0, "step": 648 }, { "epoch": 0.08577848268569918, "grad_norm": 0.9588233403400039, "learning_rate": 8.560105680317041e-06, "loss": 0.696, "mean_token_accuracy": 0.7803229093551636, "num_tokens": 42530780.0, "step": 649 }, { "epoch": 0.0859106529209622, "grad_norm": 0.9869886419214788, "learning_rate": 8.57331571994716e-06, "loss": 0.7368, "mean_token_accuracy": 0.7713344097137451, "num_tokens": 42596316.0, "step": 650 }, { "epoch": 0.08604282315622522, "grad_norm": 0.9038928518003985, "learning_rate": 8.58652575957728e-06, "loss": 0.6815, "mean_token_accuracy": 0.7857404351234436, "num_tokens": 42661852.0, "step": 651 }, { "epoch": 0.08617499339148824, "grad_norm": 0.9181572903628505, "learning_rate": 8.599735799207398e-06, "loss": 0.7297, "mean_token_accuracy": 0.7675192356109619, "num_tokens": 42727388.0, "step": 652 }, { "epoch": 0.08630716362675125, "grad_norm": 0.8851916555942252, "learning_rate": 8.612945838837517e-06, "loss": 0.6641, "mean_token_accuracy": 0.7905780673027039, "num_tokens": 42792924.0, "step": 653 }, { "epoch": 0.08643933386201427, "grad_norm": 0.9362668134639162, "learning_rate": 8.626155878467637e-06, "loss": 0.6927, "mean_token_accuracy": 0.7820321321487427, "num_tokens": 42858460.0, "step": 654 }, { "epoch": 0.08657150409727729, "grad_norm": 0.979239384359082, "learning_rate": 8.639365918097754e-06, "loss": 0.7734, "mean_token_accuracy": 0.7574624419212341, "num_tokens": 42923996.0, "step": 655 }, { "epoch": 0.08670367433254031, "grad_norm": 0.9441975776089329, "learning_rate": 8.652575957727874e-06, "loss": 0.6842, "mean_token_accuracy": 0.7809790968894958, "num_tokens": 42989532.0, "step": 656 }, { "epoch": 0.08683584456780333, "grad_norm": 0.9623236292343107, "learning_rate": 8.665785997357993e-06, "loss": 0.7238, "mean_token_accuracy": 0.7712886333465576, "num_tokens": 43055068.0, "step": 657 }, { "epoch": 0.08696801480306635, "grad_norm": 0.799055999823451, "learning_rate": 8.678996036988113e-06, "loss": 0.6407, "mean_token_accuracy": 0.7983152270317078, "num_tokens": 43120604.0, "step": 658 }, { "epoch": 0.08710018503832936, "grad_norm": 0.9300620601028979, "learning_rate": 8.69220607661823e-06, "loss": 0.7052, "mean_token_accuracy": 0.7744323015213013, "num_tokens": 43186140.0, "step": 659 }, { "epoch": 0.08723235527359238, "grad_norm": 0.8734424711174437, "learning_rate": 8.70541611624835e-06, "loss": 0.6764, "mean_token_accuracy": 0.789250373840332, "num_tokens": 43251676.0, "step": 660 }, { "epoch": 0.0873645255088554, "grad_norm": 0.9528220384998026, "learning_rate": 8.718626155878468e-06, "loss": 0.7389, "mean_token_accuracy": 0.7696709632873535, "num_tokens": 43317212.0, "step": 661 }, { "epoch": 0.08749669574411842, "grad_norm": 0.8250260589795071, "learning_rate": 8.731836195508587e-06, "loss": 0.6507, "mean_token_accuracy": 0.7965602278709412, "num_tokens": 43382748.0, "step": 662 }, { "epoch": 0.08762886597938144, "grad_norm": 0.8848120985584734, "learning_rate": 8.745046235138707e-06, "loss": 0.7264, "mean_token_accuracy": 0.7718532681465149, "num_tokens": 43448284.0, "step": 663 }, { "epoch": 0.08776103621464446, "grad_norm": 0.8675321255063198, "learning_rate": 8.758256274768825e-06, "loss": 0.7001, "mean_token_accuracy": 0.779254674911499, "num_tokens": 43513820.0, "step": 664 }, { "epoch": 0.08789320644990749, "grad_norm": 0.8623704175280138, "learning_rate": 8.771466314398944e-06, "loss": 0.6861, "mean_token_accuracy": 0.7833750247955322, "num_tokens": 43579356.0, "step": 665 }, { "epoch": 0.0880253766851705, "grad_norm": 0.9114140404077363, "learning_rate": 8.784676354029062e-06, "loss": 0.6825, "mean_token_accuracy": 0.7833445072174072, "num_tokens": 43644892.0, "step": 666 }, { "epoch": 0.08815754692043352, "grad_norm": 0.9340879698240353, "learning_rate": 8.797886393659182e-06, "loss": 0.7086, "mean_token_accuracy": 0.7803839445114136, "num_tokens": 43710428.0, "step": 667 }, { "epoch": 0.08828971715569654, "grad_norm": 0.8584843662901755, "learning_rate": 8.8110964332893e-06, "loss": 0.671, "mean_token_accuracy": 0.789280891418457, "num_tokens": 43775964.0, "step": 668 }, { "epoch": 0.08842188739095956, "grad_norm": 0.9183205942704075, "learning_rate": 8.824306472919419e-06, "loss": 0.6893, "mean_token_accuracy": 0.7792394161224365, "num_tokens": 43841500.0, "step": 669 }, { "epoch": 0.08855405762622258, "grad_norm": 0.8929088409325849, "learning_rate": 8.837516512549538e-06, "loss": 0.6529, "mean_token_accuracy": 0.7895708680152893, "num_tokens": 43907036.0, "step": 670 }, { "epoch": 0.0886862278614856, "grad_norm": 0.9959435019376949, "learning_rate": 8.850726552179658e-06, "loss": 0.7674, "mean_token_accuracy": 0.7589579820632935, "num_tokens": 43972572.0, "step": 671 }, { "epoch": 0.08881839809674862, "grad_norm": 0.8913607307676132, "learning_rate": 8.863936591809776e-06, "loss": 0.7196, "mean_token_accuracy": 0.7740660309791565, "num_tokens": 44038108.0, "step": 672 }, { "epoch": 0.08895056833201163, "grad_norm": 1.0062832144237732, "learning_rate": 8.877146631439895e-06, "loss": 0.7404, "mean_token_accuracy": 0.7659931778907776, "num_tokens": 44103644.0, "step": 673 }, { "epoch": 0.08908273856727465, "grad_norm": 1.0304663425672973, "learning_rate": 8.890356671070015e-06, "loss": 0.7355, "mean_token_accuracy": 0.7674429416656494, "num_tokens": 44169180.0, "step": 674 }, { "epoch": 0.08921490880253767, "grad_norm": 0.8769166657649058, "learning_rate": 8.903566710700134e-06, "loss": 0.6753, "mean_token_accuracy": 0.7896319031715393, "num_tokens": 44234716.0, "step": 675 }, { "epoch": 0.08934707903780069, "grad_norm": 0.9546513978603935, "learning_rate": 8.916776750330252e-06, "loss": 0.7387, "mean_token_accuracy": 0.7662220597267151, "num_tokens": 44300252.0, "step": 676 }, { "epoch": 0.08947924927306371, "grad_norm": 0.8359316172031233, "learning_rate": 8.929986789960371e-06, "loss": 0.6318, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 44365788.0, "step": 677 }, { "epoch": 0.08961141950832673, "grad_norm": 1.0339596845979173, "learning_rate": 8.943196829590489e-06, "loss": 0.7169, "mean_token_accuracy": 0.7754547595977783, "num_tokens": 44431324.0, "step": 678 }, { "epoch": 0.08974358974358974, "grad_norm": 1.0831503085946088, "learning_rate": 8.956406869220609e-06, "loss": 0.7214, "mean_token_accuracy": 0.7717769742012024, "num_tokens": 44496860.0, "step": 679 }, { "epoch": 0.08987575997885276, "grad_norm": 0.9289800071356433, "learning_rate": 8.969616908850726e-06, "loss": 0.6983, "mean_token_accuracy": 0.7795751690864563, "num_tokens": 44562396.0, "step": 680 }, { "epoch": 0.09000793021411578, "grad_norm": 0.9275729998273565, "learning_rate": 8.982826948480846e-06, "loss": 0.6747, "mean_token_accuracy": 0.7870681285858154, "num_tokens": 44627932.0, "step": 681 }, { "epoch": 0.0901401004493788, "grad_norm": 0.9525966875799872, "learning_rate": 8.996036988110965e-06, "loss": 0.7132, "mean_token_accuracy": 0.7800635099411011, "num_tokens": 44693468.0, "step": 682 }, { "epoch": 0.09027227068464182, "grad_norm": 1.007616329044657, "learning_rate": 9.009247027741083e-06, "loss": 0.7122, "mean_token_accuracy": 0.775073230266571, "num_tokens": 44759004.0, "step": 683 }, { "epoch": 0.09040444091990484, "grad_norm": 1.0094951572236746, "learning_rate": 9.022457067371203e-06, "loss": 0.7797, "mean_token_accuracy": 0.7547155618667603, "num_tokens": 44824540.0, "step": 684 }, { "epoch": 0.09053661115516785, "grad_norm": 0.7810798425126518, "learning_rate": 9.035667107001322e-06, "loss": 0.6569, "mean_token_accuracy": 0.7909901142120361, "num_tokens": 44890076.0, "step": 685 }, { "epoch": 0.09066878139043087, "grad_norm": 0.8635804541028439, "learning_rate": 9.04887714663144e-06, "loss": 0.6365, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 44955612.0, "step": 686 }, { "epoch": 0.09080095162569389, "grad_norm": 0.9890023910800286, "learning_rate": 9.06208718626156e-06, "loss": 0.6541, "mean_token_accuracy": 0.7905170321464539, "num_tokens": 45021148.0, "step": 687 }, { "epoch": 0.09093312186095691, "grad_norm": 1.027632273767516, "learning_rate": 9.075297225891679e-06, "loss": 0.7563, "mean_token_accuracy": 0.7618117332458496, "num_tokens": 45086684.0, "step": 688 }, { "epoch": 0.09106529209621993, "grad_norm": 0.9306587355996779, "learning_rate": 9.088507265521798e-06, "loss": 0.7114, "mean_token_accuracy": 0.7713954448699951, "num_tokens": 45152220.0, "step": 689 }, { "epoch": 0.09119746233148295, "grad_norm": 0.8928169324682294, "learning_rate": 9.101717305151916e-06, "loss": 0.6991, "mean_token_accuracy": 0.784580647945404, "num_tokens": 45217756.0, "step": 690 }, { "epoch": 0.09132963256674596, "grad_norm": 0.9011703747697934, "learning_rate": 9.114927344782036e-06, "loss": 0.6966, "mean_token_accuracy": 0.7775149345397949, "num_tokens": 45283292.0, "step": 691 }, { "epoch": 0.09146180280200898, "grad_norm": 1.016460519927283, "learning_rate": 9.128137384412154e-06, "loss": 0.7195, "mean_token_accuracy": 0.7705255746841431, "num_tokens": 45348828.0, "step": 692 }, { "epoch": 0.091593973037272, "grad_norm": 0.9854065671121898, "learning_rate": 9.141347424042273e-06, "loss": 0.7246, "mean_token_accuracy": 0.7712733745574951, "num_tokens": 45414364.0, "step": 693 }, { "epoch": 0.09172614327253502, "grad_norm": 0.953122464003638, "learning_rate": 9.154557463672392e-06, "loss": 0.7412, "mean_token_accuracy": 0.7649402022361755, "num_tokens": 45479900.0, "step": 694 }, { "epoch": 0.09185831350779805, "grad_norm": 0.9425012355143312, "learning_rate": 9.16776750330251e-06, "loss": 0.6704, "mean_token_accuracy": 0.7893114686012268, "num_tokens": 45545436.0, "step": 695 }, { "epoch": 0.09199048374306107, "grad_norm": 0.9235498069937826, "learning_rate": 9.18097754293263e-06, "loss": 0.7005, "mean_token_accuracy": 0.7787358164787292, "num_tokens": 45610972.0, "step": 696 }, { "epoch": 0.09212265397832409, "grad_norm": 0.9082344107284032, "learning_rate": 9.194187582562748e-06, "loss": 0.6909, "mean_token_accuracy": 0.7821847200393677, "num_tokens": 45676508.0, "step": 697 }, { "epoch": 0.0922548242135871, "grad_norm": 0.907650418429372, "learning_rate": 9.207397622192867e-06, "loss": 0.6852, "mean_token_accuracy": 0.7845959067344666, "num_tokens": 45742044.0, "step": 698 }, { "epoch": 0.09238699444885012, "grad_norm": 1.002476781805203, "learning_rate": 9.220607661822987e-06, "loss": 0.7323, "mean_token_accuracy": 0.7702661752700806, "num_tokens": 45807580.0, "step": 699 }, { "epoch": 0.09251916468411314, "grad_norm": 0.9042132725128953, "learning_rate": 9.233817701453104e-06, "loss": 0.6448, "mean_token_accuracy": 0.7973843216896057, "num_tokens": 45873116.0, "step": 700 }, { "epoch": 0.09265133491937616, "grad_norm": 1.0368746426118989, "learning_rate": 9.247027741083224e-06, "loss": 0.7678, "mean_token_accuracy": 0.7561195492744446, "num_tokens": 45938652.0, "step": 701 }, { "epoch": 0.09278350515463918, "grad_norm": 0.8261434782062146, "learning_rate": 9.260237780713343e-06, "loss": 0.6876, "mean_token_accuracy": 0.7839549779891968, "num_tokens": 46004188.0, "step": 702 }, { "epoch": 0.0929156753899022, "grad_norm": 0.9224786230883056, "learning_rate": 9.273447820343461e-06, "loss": 0.7788, "mean_token_accuracy": 0.7572488188743591, "num_tokens": 46069724.0, "step": 703 }, { "epoch": 0.09304784562516522, "grad_norm": 0.8735028343286196, "learning_rate": 9.28665785997358e-06, "loss": 0.643, "mean_token_accuracy": 0.7969570159912109, "num_tokens": 46135260.0, "step": 704 }, { "epoch": 0.09318001586042823, "grad_norm": 1.0793180507177433, "learning_rate": 9.2998678996037e-06, "loss": 0.7244, "mean_token_accuracy": 0.7695183753967285, "num_tokens": 46200796.0, "step": 705 }, { "epoch": 0.09331218609569125, "grad_norm": 0.9800538972425604, "learning_rate": 9.31307793923382e-06, "loss": 0.6773, "mean_token_accuracy": 0.7853894233703613, "num_tokens": 46266332.0, "step": 706 }, { "epoch": 0.09344435633095427, "grad_norm": 0.8755494541942357, "learning_rate": 9.326287978863937e-06, "loss": 0.6764, "mean_token_accuracy": 0.7901049852371216, "num_tokens": 46331868.0, "step": 707 }, { "epoch": 0.09357652656621729, "grad_norm": 0.8892130212590553, "learning_rate": 9.339498018494057e-06, "loss": 0.6208, "mean_token_accuracy": 0.8037480115890503, "num_tokens": 46397404.0, "step": 708 }, { "epoch": 0.0937086968014803, "grad_norm": 0.9887768301984661, "learning_rate": 9.352708058124175e-06, "loss": 0.6964, "mean_token_accuracy": 0.7805060148239136, "num_tokens": 46462940.0, "step": 709 }, { "epoch": 0.09384086703674333, "grad_norm": 0.9993112282819103, "learning_rate": 9.365918097754294e-06, "loss": 0.715, "mean_token_accuracy": 0.774981677532196, "num_tokens": 46528476.0, "step": 710 }, { "epoch": 0.09397303727200634, "grad_norm": 0.9835583686240726, "learning_rate": 9.379128137384414e-06, "loss": 0.7581, "mean_token_accuracy": 0.7596752643585205, "num_tokens": 46594012.0, "step": 711 }, { "epoch": 0.09410520750726936, "grad_norm": 0.8493840431200598, "learning_rate": 9.392338177014531e-06, "loss": 0.6502, "mean_token_accuracy": 0.7956751585006714, "num_tokens": 46659548.0, "step": 712 }, { "epoch": 0.09423737774253238, "grad_norm": 0.9853194145617228, "learning_rate": 9.405548216644651e-06, "loss": 0.7203, "mean_token_accuracy": 0.7758362889289856, "num_tokens": 46725084.0, "step": 713 }, { "epoch": 0.0943695479777954, "grad_norm": 0.9769939998967881, "learning_rate": 9.418758256274769e-06, "loss": 0.6925, "mean_token_accuracy": 0.7825815081596375, "num_tokens": 46790620.0, "step": 714 }, { "epoch": 0.09450171821305842, "grad_norm": 0.9674821484532731, "learning_rate": 9.431968295904888e-06, "loss": 0.746, "mean_token_accuracy": 0.7626205682754517, "num_tokens": 46856156.0, "step": 715 }, { "epoch": 0.09463388844832143, "grad_norm": 0.7890664204064178, "learning_rate": 9.445178335535008e-06, "loss": 0.6548, "mean_token_accuracy": 0.7942711710929871, "num_tokens": 46921692.0, "step": 716 }, { "epoch": 0.09476605868358445, "grad_norm": 0.8198363379509263, "learning_rate": 9.458388375165125e-06, "loss": 0.6347, "mean_token_accuracy": 0.8018556833267212, "num_tokens": 46987228.0, "step": 717 }, { "epoch": 0.09489822891884747, "grad_norm": 0.8902148782694503, "learning_rate": 9.471598414795245e-06, "loss": 0.6574, "mean_token_accuracy": 0.7941642999649048, "num_tokens": 47052764.0, "step": 718 }, { "epoch": 0.09503039915411049, "grad_norm": 0.9176090027478911, "learning_rate": 9.484808454425364e-06, "loss": 0.6653, "mean_token_accuracy": 0.7917073369026184, "num_tokens": 47118300.0, "step": 719 }, { "epoch": 0.09516256938937351, "grad_norm": 0.9520031383700688, "learning_rate": 9.498018494055484e-06, "loss": 0.7561, "mean_token_accuracy": 0.7647417783737183, "num_tokens": 47183836.0, "step": 720 }, { "epoch": 0.09529473962463653, "grad_norm": 0.9533157329431929, "learning_rate": 9.511228533685602e-06, "loss": 0.686, "mean_token_accuracy": 0.7828561663627625, "num_tokens": 47249372.0, "step": 721 }, { "epoch": 0.09542690985989954, "grad_norm": 1.033332789698973, "learning_rate": 9.524438573315721e-06, "loss": 0.7732, "mean_token_accuracy": 0.7574929594993591, "num_tokens": 47314908.0, "step": 722 }, { "epoch": 0.09555908009516256, "grad_norm": 1.0427924704530043, "learning_rate": 9.53764861294584e-06, "loss": 0.6918, "mean_token_accuracy": 0.7805060148239136, "num_tokens": 47380444.0, "step": 723 }, { "epoch": 0.09569125033042558, "grad_norm": 1.0796605781091755, "learning_rate": 9.550858652575959e-06, "loss": 0.7637, "mean_token_accuracy": 0.7606824636459351, "num_tokens": 47445980.0, "step": 724 }, { "epoch": 0.09582342056568861, "grad_norm": 0.8792114928465437, "learning_rate": 9.564068692206078e-06, "loss": 0.6909, "mean_token_accuracy": 0.784458577632904, "num_tokens": 47511516.0, "step": 725 }, { "epoch": 0.09595559080095163, "grad_norm": 0.8664194384692256, "learning_rate": 9.577278731836196e-06, "loss": 0.6835, "mean_token_accuracy": 0.7819710373878479, "num_tokens": 47577052.0, "step": 726 }, { "epoch": 0.09608776103621465, "grad_norm": 0.9338705253193075, "learning_rate": 9.590488771466315e-06, "loss": 0.6828, "mean_token_accuracy": 0.7852978706359863, "num_tokens": 47642588.0, "step": 727 }, { "epoch": 0.09621993127147767, "grad_norm": 0.9131503258321092, "learning_rate": 9.603698811096433e-06, "loss": 0.7106, "mean_token_accuracy": 0.775103747844696, "num_tokens": 47708124.0, "step": 728 }, { "epoch": 0.09635210150674069, "grad_norm": 0.9011507784108407, "learning_rate": 9.616908850726553e-06, "loss": 0.747, "mean_token_accuracy": 0.7637040615081787, "num_tokens": 47773660.0, "step": 729 }, { "epoch": 0.0964842717420037, "grad_norm": 0.8992709630170962, "learning_rate": 9.630118890356672e-06, "loss": 0.7383, "mean_token_accuracy": 0.7688011527061462, "num_tokens": 47839196.0, "step": 730 }, { "epoch": 0.09661644197726672, "grad_norm": 0.9247466180595372, "learning_rate": 9.64332892998679e-06, "loss": 0.729, "mean_token_accuracy": 0.7692283987998962, "num_tokens": 47904732.0, "step": 731 }, { "epoch": 0.09674861221252974, "grad_norm": 1.0621546739866514, "learning_rate": 9.65653896961691e-06, "loss": 0.7402, "mean_token_accuracy": 0.7660694718360901, "num_tokens": 47970268.0, "step": 732 }, { "epoch": 0.09688078244779276, "grad_norm": 0.8593659190253683, "learning_rate": 9.669749009247029e-06, "loss": 0.6832, "mean_token_accuracy": 0.783863365650177, "num_tokens": 48035804.0, "step": 733 }, { "epoch": 0.09701295268305578, "grad_norm": 0.8639328478180783, "learning_rate": 9.682959048877147e-06, "loss": 0.656, "mean_token_accuracy": 0.7900592088699341, "num_tokens": 48101340.0, "step": 734 }, { "epoch": 0.0971451229183188, "grad_norm": 0.8884532799459596, "learning_rate": 9.696169088507266e-06, "loss": 0.7174, "mean_token_accuracy": 0.7774844169616699, "num_tokens": 48166876.0, "step": 735 }, { "epoch": 0.09727729315358181, "grad_norm": 0.8683034489771582, "learning_rate": 9.709379128137386e-06, "loss": 0.6514, "mean_token_accuracy": 0.7970027923583984, "num_tokens": 48232412.0, "step": 736 }, { "epoch": 0.09740946338884483, "grad_norm": 0.9401700354881638, "learning_rate": 9.722589167767505e-06, "loss": 0.7275, "mean_token_accuracy": 0.7714564800262451, "num_tokens": 48297948.0, "step": 737 }, { "epoch": 0.09754163362410785, "grad_norm": 0.8827242819247497, "learning_rate": 9.735799207397623e-06, "loss": 0.651, "mean_token_accuracy": 0.7916920781135559, "num_tokens": 48363484.0, "step": 738 }, { "epoch": 0.09767380385937087, "grad_norm": 0.8830694637040988, "learning_rate": 9.749009247027742e-06, "loss": 0.6776, "mean_token_accuracy": 0.7891740798950195, "num_tokens": 48429020.0, "step": 739 }, { "epoch": 0.09780597409463389, "grad_norm": 1.0370488659835977, "learning_rate": 9.76221928665786e-06, "loss": 0.6642, "mean_token_accuracy": 0.7905780673027039, "num_tokens": 48494556.0, "step": 740 }, { "epoch": 0.0979381443298969, "grad_norm": 0.8600996198959471, "learning_rate": 9.77542932628798e-06, "loss": 0.6528, "mean_token_accuracy": 0.7958124876022339, "num_tokens": 48560092.0, "step": 741 }, { "epoch": 0.09807031456515992, "grad_norm": 0.8629864382861568, "learning_rate": 9.7886393659181e-06, "loss": 0.698, "mean_token_accuracy": 0.7823220491409302, "num_tokens": 48625628.0, "step": 742 }, { "epoch": 0.09820248480042294, "grad_norm": 1.0014845048678342, "learning_rate": 9.801849405548217e-06, "loss": 0.6938, "mean_token_accuracy": 0.7804144620895386, "num_tokens": 48691164.0, "step": 743 }, { "epoch": 0.09833465503568596, "grad_norm": 0.9169069015175981, "learning_rate": 9.815059445178336e-06, "loss": 0.6841, "mean_token_accuracy": 0.7862135171890259, "num_tokens": 48756700.0, "step": 744 }, { "epoch": 0.09846682527094898, "grad_norm": 1.007247983913248, "learning_rate": 9.828269484808454e-06, "loss": 0.758, "mean_token_accuracy": 0.7602246403694153, "num_tokens": 48822236.0, "step": 745 }, { "epoch": 0.098598995506212, "grad_norm": 0.8824989385133425, "learning_rate": 9.841479524438574e-06, "loss": 0.7135, "mean_token_accuracy": 0.774050772190094, "num_tokens": 48887772.0, "step": 746 }, { "epoch": 0.09873116574147502, "grad_norm": 0.9297697267885442, "learning_rate": 9.854689564068693e-06, "loss": 0.7593, "mean_token_accuracy": 0.759660005569458, "num_tokens": 48953308.0, "step": 747 }, { "epoch": 0.09886333597673803, "grad_norm": 0.9081666394528274, "learning_rate": 9.867899603698811e-06, "loss": 0.7179, "mean_token_accuracy": 0.7717922329902649, "num_tokens": 49018844.0, "step": 748 }, { "epoch": 0.09899550621200105, "grad_norm": 0.9340654834743484, "learning_rate": 9.88110964332893e-06, "loss": 0.7108, "mean_token_accuracy": 0.7751953601837158, "num_tokens": 49084380.0, "step": 749 }, { "epoch": 0.09912767644726407, "grad_norm": 0.9612286882108003, "learning_rate": 9.89431968295905e-06, "loss": 0.7337, "mean_token_accuracy": 0.7660236954689026, "num_tokens": 49149916.0, "step": 750 }, { "epoch": 0.09925984668252709, "grad_norm": 0.9452550565565518, "learning_rate": 9.90752972258917e-06, "loss": 0.6957, "mean_token_accuracy": 0.783558189868927, "num_tokens": 49215452.0, "step": 751 }, { "epoch": 0.09939201691779011, "grad_norm": 0.8888474896386511, "learning_rate": 9.920739762219287e-06, "loss": 0.66, "mean_token_accuracy": 0.7921956777572632, "num_tokens": 49280988.0, "step": 752 }, { "epoch": 0.09952418715305313, "grad_norm": 1.042198198231858, "learning_rate": 9.933949801849407e-06, "loss": 0.6981, "mean_token_accuracy": 0.778354287147522, "num_tokens": 49346524.0, "step": 753 }, { "epoch": 0.09965635738831616, "grad_norm": 0.9810508072384632, "learning_rate": 9.947159841479526e-06, "loss": 0.7242, "mean_token_accuracy": 0.7744628190994263, "num_tokens": 49412060.0, "step": 754 }, { "epoch": 0.09978852762357918, "grad_norm": 0.9314540366562482, "learning_rate": 9.960369881109644e-06, "loss": 0.6912, "mean_token_accuracy": 0.7829172015190125, "num_tokens": 49477596.0, "step": 755 }, { "epoch": 0.0999206978588422, "grad_norm": 0.9662428288943913, "learning_rate": 9.973579920739764e-06, "loss": 0.704, "mean_token_accuracy": 0.7758210301399231, "num_tokens": 49543132.0, "step": 756 }, { "epoch": 0.10005286809410521, "grad_norm": 0.9313529747537305, "learning_rate": 9.986789960369881e-06, "loss": 0.6568, "mean_token_accuracy": 0.7931113243103027, "num_tokens": 49608668.0, "step": 757 }, { "epoch": 0.10018503832936823, "grad_norm": 1.0014809464335597, "learning_rate": 1e-05, "loss": 0.7462, "mean_token_accuracy": 0.763963520526886, "num_tokens": 49674204.0, "step": 758 }, { "epoch": 0.10031720856463125, "grad_norm": 0.9430340835849269, "learning_rate": 9.999999521021978e-06, "loss": 0.6822, "mean_token_accuracy": 0.7861982583999634, "num_tokens": 49739740.0, "step": 759 }, { "epoch": 0.10044937879989427, "grad_norm": 1.0203813811846874, "learning_rate": 9.99999808408801e-06, "loss": 0.7561, "mean_token_accuracy": 0.7611402869224548, "num_tokens": 49805276.0, "step": 760 }, { "epoch": 0.10058154903515729, "grad_norm": 0.982559502776188, "learning_rate": 9.999995689198402e-06, "loss": 0.7009, "mean_token_accuracy": 0.7778964638710022, "num_tokens": 49870812.0, "step": 761 }, { "epoch": 0.1007137192704203, "grad_norm": 0.9373688626067402, "learning_rate": 9.999992336353665e-06, "loss": 0.6773, "mean_token_accuracy": 0.7871901988983154, "num_tokens": 49936348.0, "step": 762 }, { "epoch": 0.10084588950568332, "grad_norm": 0.8649253731422805, "learning_rate": 9.999988025554514e-06, "loss": 0.6864, "mean_token_accuracy": 0.7826425433158875, "num_tokens": 50001884.0, "step": 763 }, { "epoch": 0.10097805974094634, "grad_norm": 0.9164884605175503, "learning_rate": 9.999982756801866e-06, "loss": 0.6813, "mean_token_accuracy": 0.7804449796676636, "num_tokens": 50067420.0, "step": 764 }, { "epoch": 0.10111022997620936, "grad_norm": 0.92140376282921, "learning_rate": 9.99997653009684e-06, "loss": 0.7044, "mean_token_accuracy": 0.7756531834602356, "num_tokens": 50132956.0, "step": 765 }, { "epoch": 0.10124240021147238, "grad_norm": 0.9674548788760647, "learning_rate": 9.999969345440764e-06, "loss": 0.6992, "mean_token_accuracy": 0.7774233818054199, "num_tokens": 50198492.0, "step": 766 }, { "epoch": 0.1013745704467354, "grad_norm": 1.061472237365588, "learning_rate": 9.999961202835167e-06, "loss": 0.7751, "mean_token_accuracy": 0.7587443590164185, "num_tokens": 50264028.0, "step": 767 }, { "epoch": 0.10150674068199841, "grad_norm": 1.0718050527019864, "learning_rate": 9.999952102281784e-06, "loss": 0.7291, "mean_token_accuracy": 0.7700524926185608, "num_tokens": 50329564.0, "step": 768 }, { "epoch": 0.10163891091726143, "grad_norm": 0.9774679052359275, "learning_rate": 9.999942043782547e-06, "loss": 0.6746, "mean_token_accuracy": 0.7849469184875488, "num_tokens": 50395100.0, "step": 769 }, { "epoch": 0.10177108115252445, "grad_norm": 1.0987481570317323, "learning_rate": 9.999931027339603e-06, "loss": 0.695, "mean_token_accuracy": 0.7805518507957458, "num_tokens": 50460636.0, "step": 770 }, { "epoch": 0.10190325138778747, "grad_norm": 0.9173513176937407, "learning_rate": 9.999919052955296e-06, "loss": 0.6933, "mean_token_accuracy": 0.7797430157661438, "num_tokens": 50526172.0, "step": 771 }, { "epoch": 0.10203542162305049, "grad_norm": 0.9356961531994727, "learning_rate": 9.999906120632174e-06, "loss": 0.6996, "mean_token_accuracy": 0.778476357460022, "num_tokens": 50591708.0, "step": 772 }, { "epoch": 0.1021675918583135, "grad_norm": 0.8586536103441378, "learning_rate": 9.999892230372989e-06, "loss": 0.6524, "mean_token_accuracy": 0.7929587364196777, "num_tokens": 50657244.0, "step": 773 }, { "epoch": 0.10229976209357652, "grad_norm": 1.0464520642122521, "learning_rate": 9.999877382180701e-06, "loss": 0.6819, "mean_token_accuracy": 0.7834360599517822, "num_tokens": 50722780.0, "step": 774 }, { "epoch": 0.10243193232883954, "grad_norm": 1.0148467036225517, "learning_rate": 9.999861576058467e-06, "loss": 0.7023, "mean_token_accuracy": 0.7780948877334595, "num_tokens": 50788316.0, "step": 775 }, { "epoch": 0.10256410256410256, "grad_norm": 0.940518356891561, "learning_rate": 9.999844812009655e-06, "loss": 0.6961, "mean_token_accuracy": 0.7796667218208313, "num_tokens": 50853852.0, "step": 776 }, { "epoch": 0.10269627279936558, "grad_norm": 0.9875405193923856, "learning_rate": 9.999827090037831e-06, "loss": 0.7308, "mean_token_accuracy": 0.7690910696983337, "num_tokens": 50919388.0, "step": 777 }, { "epoch": 0.1028284430346286, "grad_norm": 0.9450213294380596, "learning_rate": 9.99980841014677e-06, "loss": 0.6813, "mean_token_accuracy": 0.7801550626754761, "num_tokens": 50984924.0, "step": 778 }, { "epoch": 0.10296061326989162, "grad_norm": 1.0427888364624014, "learning_rate": 9.999788772340448e-06, "loss": 0.7585, "mean_token_accuracy": 0.7576150894165039, "num_tokens": 51050460.0, "step": 779 }, { "epoch": 0.10309278350515463, "grad_norm": 1.0454403511119628, "learning_rate": 9.999768176623045e-06, "loss": 0.7071, "mean_token_accuracy": 0.7777286171913147, "num_tokens": 51115996.0, "step": 780 }, { "epoch": 0.10322495374041765, "grad_norm": 1.0255545423734915, "learning_rate": 9.999746622998946e-06, "loss": 0.7119, "mean_token_accuracy": 0.775012195110321, "num_tokens": 51181532.0, "step": 781 }, { "epoch": 0.10335712397568067, "grad_norm": 1.163451661645414, "learning_rate": 9.99972411147274e-06, "loss": 0.7054, "mean_token_accuracy": 0.7772707939147949, "num_tokens": 51247068.0, "step": 782 }, { "epoch": 0.10348929421094369, "grad_norm": 1.0051393853660504, "learning_rate": 9.999700642049216e-06, "loss": 0.6388, "mean_token_accuracy": 0.8001770377159119, "num_tokens": 51312604.0, "step": 783 }, { "epoch": 0.10362146444620672, "grad_norm": 1.1091026537005302, "learning_rate": 9.999676214733372e-06, "loss": 0.7127, "mean_token_accuracy": 0.7747222781181335, "num_tokens": 51378140.0, "step": 784 }, { "epoch": 0.10375363468146974, "grad_norm": 1.0405449447275483, "learning_rate": 9.999650829530409e-06, "loss": 0.701, "mean_token_accuracy": 0.7786595225334167, "num_tokens": 51443676.0, "step": 785 }, { "epoch": 0.10388580491673276, "grad_norm": 0.9636452845660398, "learning_rate": 9.999624486445732e-06, "loss": 0.6845, "mean_token_accuracy": 0.783619225025177, "num_tokens": 51509212.0, "step": 786 }, { "epoch": 0.10401797515199578, "grad_norm": 1.0508220221650595, "learning_rate": 9.999597185484945e-06, "loss": 0.6557, "mean_token_accuracy": 0.7916768193244934, "num_tokens": 51574748.0, "step": 787 }, { "epoch": 0.1041501453872588, "grad_norm": 1.0857027915374344, "learning_rate": 9.999568926653864e-06, "loss": 0.7351, "mean_token_accuracy": 0.7666951417922974, "num_tokens": 51640284.0, "step": 788 }, { "epoch": 0.10428231562252181, "grad_norm": 0.8597793822268793, "learning_rate": 9.999539709958501e-06, "loss": 0.659, "mean_token_accuracy": 0.7898761034011841, "num_tokens": 51705820.0, "step": 789 }, { "epoch": 0.10441448585778483, "grad_norm": 0.9414718347748537, "learning_rate": 9.999509535405078e-06, "loss": 0.7049, "mean_token_accuracy": 0.7771334648132324, "num_tokens": 51771356.0, "step": 790 }, { "epoch": 0.10454665609304785, "grad_norm": 0.9591808744615186, "learning_rate": 9.999478403000017e-06, "loss": 0.7399, "mean_token_accuracy": 0.7639024257659912, "num_tokens": 51836892.0, "step": 791 }, { "epoch": 0.10467882632831087, "grad_norm": 0.8200431658427246, "learning_rate": 9.999446312749948e-06, "loss": 0.6573, "mean_token_accuracy": 0.792592465877533, "num_tokens": 51902428.0, "step": 792 }, { "epoch": 0.10481099656357389, "grad_norm": 0.913394319013771, "learning_rate": 9.9994132646617e-06, "loss": 0.7164, "mean_token_accuracy": 0.7718074917793274, "num_tokens": 51967964.0, "step": 793 }, { "epoch": 0.1049431667988369, "grad_norm": 1.036265999860957, "learning_rate": 9.99937925874231e-06, "loss": 0.7478, "mean_token_accuracy": 0.7620101571083069, "num_tokens": 52033500.0, "step": 794 }, { "epoch": 0.10507533703409992, "grad_norm": 1.1944388343368708, "learning_rate": 9.999344294999015e-06, "loss": 0.7767, "mean_token_accuracy": 0.7543187737464905, "num_tokens": 52099036.0, "step": 795 }, { "epoch": 0.10520750726936294, "grad_norm": 1.0682747666973347, "learning_rate": 9.999308373439258e-06, "loss": 0.7187, "mean_token_accuracy": 0.7696099281311035, "num_tokens": 52164572.0, "step": 796 }, { "epoch": 0.10533967750462596, "grad_norm": 0.9669417891534334, "learning_rate": 9.99927149407069e-06, "loss": 0.715, "mean_token_accuracy": 0.7719753384590149, "num_tokens": 52230108.0, "step": 797 }, { "epoch": 0.10547184773988898, "grad_norm": 0.9764642578196258, "learning_rate": 9.999233656901157e-06, "loss": 0.7537, "mean_token_accuracy": 0.7619643807411194, "num_tokens": 52295644.0, "step": 798 }, { "epoch": 0.105604017975152, "grad_norm": 0.9453582653957044, "learning_rate": 9.999194861938714e-06, "loss": 0.6752, "mean_token_accuracy": 0.7860456705093384, "num_tokens": 52361180.0, "step": 799 }, { "epoch": 0.10573618821041501, "grad_norm": 0.849547840596546, "learning_rate": 9.999155109191625e-06, "loss": 0.6568, "mean_token_accuracy": 0.7913411259651184, "num_tokens": 52426716.0, "step": 800 }, { "epoch": 0.10586835844567803, "grad_norm": 0.9348660562652361, "learning_rate": 9.999114398668348e-06, "loss": 0.7022, "mean_token_accuracy": 0.7777438759803772, "num_tokens": 52492252.0, "step": 801 }, { "epoch": 0.10600052868094105, "grad_norm": 0.7887642054039251, "learning_rate": 9.99907273037755e-06, "loss": 0.6086, "mean_token_accuracy": 0.8078073263168335, "num_tokens": 52557788.0, "step": 802 }, { "epoch": 0.10613269891620407, "grad_norm": 1.049108793645844, "learning_rate": 9.999030104328098e-06, "loss": 0.7175, "mean_token_accuracy": 0.7750579714775085, "num_tokens": 52623324.0, "step": 803 }, { "epoch": 0.10626486915146709, "grad_norm": 0.9366084484596768, "learning_rate": 9.998986520529073e-06, "loss": 0.6992, "mean_token_accuracy": 0.779346227645874, "num_tokens": 52688860.0, "step": 804 }, { "epoch": 0.1063970393867301, "grad_norm": 0.9833211972694238, "learning_rate": 9.998941978989751e-06, "loss": 0.7494, "mean_token_accuracy": 0.7624221444129944, "num_tokens": 52754396.0, "step": 805 }, { "epoch": 0.10652920962199312, "grad_norm": 0.9072400740208796, "learning_rate": 9.998896479719611e-06, "loss": 0.6784, "mean_token_accuracy": 0.7832376956939697, "num_tokens": 52819932.0, "step": 806 }, { "epoch": 0.10666137985725614, "grad_norm": 0.8434848258323817, "learning_rate": 9.99885002272834e-06, "loss": 0.6861, "mean_token_accuracy": 0.7792088985443115, "num_tokens": 52885468.0, "step": 807 }, { "epoch": 0.10679355009251916, "grad_norm": 1.0222562130785922, "learning_rate": 9.99880260802583e-06, "loss": 0.7759, "mean_token_accuracy": 0.7532423734664917, "num_tokens": 52947699.0, "step": 808 }, { "epoch": 0.10692572032778218, "grad_norm": 1.1131237280666386, "learning_rate": 9.998754235622171e-06, "loss": 0.7576, "mean_token_accuracy": 0.7607129812240601, "num_tokens": 53013235.0, "step": 809 }, { "epoch": 0.1070578905630452, "grad_norm": 0.9836697091907768, "learning_rate": 9.998704905527666e-06, "loss": 0.6608, "mean_token_accuracy": 0.7888689041137695, "num_tokens": 53078771.0, "step": 810 }, { "epoch": 0.10719006079830821, "grad_norm": 0.909118908158412, "learning_rate": 9.99865461775281e-06, "loss": 0.7456, "mean_token_accuracy": 0.7639177441596985, "num_tokens": 53144307.0, "step": 811 }, { "epoch": 0.10732223103357123, "grad_norm": 0.9444877200843256, "learning_rate": 9.998603372308312e-06, "loss": 0.6854, "mean_token_accuracy": 0.7820626497268677, "num_tokens": 53209843.0, "step": 812 }, { "epoch": 0.10745440126883427, "grad_norm": 0.8949543039579927, "learning_rate": 9.998551169205079e-06, "loss": 0.7041, "mean_token_accuracy": 0.779437780380249, "num_tokens": 53275379.0, "step": 813 }, { "epoch": 0.10758657150409728, "grad_norm": 0.971852638328109, "learning_rate": 9.998498008454228e-06, "loss": 0.7382, "mean_token_accuracy": 0.7656726837158203, "num_tokens": 53340915.0, "step": 814 }, { "epoch": 0.1077187417393603, "grad_norm": 0.9368949272731937, "learning_rate": 9.99844389006707e-06, "loss": 0.721, "mean_token_accuracy": 0.7728757262229919, "num_tokens": 53406451.0, "step": 815 }, { "epoch": 0.10785091197462332, "grad_norm": 0.9235024682924179, "learning_rate": 9.998388814055128e-06, "loss": 0.667, "mean_token_accuracy": 0.7889909744262695, "num_tokens": 53471987.0, "step": 816 }, { "epoch": 0.10798308220988634, "grad_norm": 1.039756218673928, "learning_rate": 9.998332780430128e-06, "loss": 0.7634, "mean_token_accuracy": 0.7605756521224976, "num_tokens": 53537523.0, "step": 817 }, { "epoch": 0.10811525244514936, "grad_norm": 0.9620164652432289, "learning_rate": 9.998275789203998e-06, "loss": 0.688, "mean_token_accuracy": 0.78265780210495, "num_tokens": 53603059.0, "step": 818 }, { "epoch": 0.10824742268041238, "grad_norm": 0.9282576932444709, "learning_rate": 9.99821784038887e-06, "loss": 0.6947, "mean_token_accuracy": 0.7822457551956177, "num_tokens": 53668595.0, "step": 819 }, { "epoch": 0.1083795929156754, "grad_norm": 0.8638011932130402, "learning_rate": 9.998158933997078e-06, "loss": 0.65, "mean_token_accuracy": 0.7931265830993652, "num_tokens": 53734131.0, "step": 820 }, { "epoch": 0.10851176315093841, "grad_norm": 1.0038225527421831, "learning_rate": 9.998099070041167e-06, "loss": 0.7085, "mean_token_accuracy": 0.7760193943977356, "num_tokens": 53799667.0, "step": 821 }, { "epoch": 0.10864393338620143, "grad_norm": 0.9589252367027318, "learning_rate": 9.998038248533875e-06, "loss": 0.7096, "mean_token_accuracy": 0.7768434882164001, "num_tokens": 53865203.0, "step": 822 }, { "epoch": 0.10877610362146445, "grad_norm": 0.9931717965096524, "learning_rate": 9.997976469488153e-06, "loss": 0.6822, "mean_token_accuracy": 0.783497154712677, "num_tokens": 53930739.0, "step": 823 }, { "epoch": 0.10890827385672747, "grad_norm": 0.9521003385853303, "learning_rate": 9.99791373291715e-06, "loss": 0.7456, "mean_token_accuracy": 0.7618727684020996, "num_tokens": 53996275.0, "step": 824 }, { "epoch": 0.10904044409199048, "grad_norm": 1.0120205662747463, "learning_rate": 9.997850038834226e-06, "loss": 0.6675, "mean_token_accuracy": 0.7864577174186707, "num_tokens": 54061811.0, "step": 825 }, { "epoch": 0.1091726143272535, "grad_norm": 0.9066329203115422, "learning_rate": 9.997785387252935e-06, "loss": 0.7194, "mean_token_accuracy": 0.7719600796699524, "num_tokens": 54127347.0, "step": 826 }, { "epoch": 0.10930478456251652, "grad_norm": 0.9644939174847917, "learning_rate": 9.997719778187042e-06, "loss": 0.7523, "mean_token_accuracy": 0.7592326998710632, "num_tokens": 54192883.0, "step": 827 }, { "epoch": 0.10943695479777954, "grad_norm": 0.8352333732072047, "learning_rate": 9.997653211650515e-06, "loss": 0.6553, "mean_token_accuracy": 0.7932639718055725, "num_tokens": 54258419.0, "step": 828 }, { "epoch": 0.10956912503304256, "grad_norm": 1.0013775511099468, "learning_rate": 9.997585687657523e-06, "loss": 0.6635, "mean_token_accuracy": 0.7876785397529602, "num_tokens": 54323955.0, "step": 829 }, { "epoch": 0.10970129526830558, "grad_norm": 1.0413257412200287, "learning_rate": 9.99751720622244e-06, "loss": 0.7264, "mean_token_accuracy": 0.7716548442840576, "num_tokens": 54389491.0, "step": 830 }, { "epoch": 0.1098334655035686, "grad_norm": 0.9668747151904575, "learning_rate": 9.997447767359846e-06, "loss": 0.6878, "mean_token_accuracy": 0.783558189868927, "num_tokens": 54455027.0, "step": 831 }, { "epoch": 0.10996563573883161, "grad_norm": 1.028564018371081, "learning_rate": 9.997377371084522e-06, "loss": 0.7323, "mean_token_accuracy": 0.7657947540283203, "num_tokens": 54520563.0, "step": 832 }, { "epoch": 0.11009780597409463, "grad_norm": 1.0258673112735959, "learning_rate": 9.997306017411455e-06, "loss": 0.7421, "mean_token_accuracy": 0.7644671201705933, "num_tokens": 54586099.0, "step": 833 }, { "epoch": 0.11022997620935765, "grad_norm": 0.944806147065333, "learning_rate": 9.997233706355833e-06, "loss": 0.6851, "mean_token_accuracy": 0.7830851078033447, "num_tokens": 54651635.0, "step": 834 }, { "epoch": 0.11036214644462067, "grad_norm": 0.9660036106713333, "learning_rate": 9.997160437933051e-06, "loss": 0.7063, "mean_token_accuracy": 0.7746459245681763, "num_tokens": 54717171.0, "step": 835 }, { "epoch": 0.11049431667988369, "grad_norm": 0.867073563302699, "learning_rate": 9.997086212158705e-06, "loss": 0.6752, "mean_token_accuracy": 0.784428060054779, "num_tokens": 54782707.0, "step": 836 }, { "epoch": 0.1106264869151467, "grad_norm": 1.0226059030444568, "learning_rate": 9.997011029048598e-06, "loss": 0.701, "mean_token_accuracy": 0.7761873006820679, "num_tokens": 54848243.0, "step": 837 }, { "epoch": 0.11075865715040972, "grad_norm": 0.9668023227900814, "learning_rate": 9.996934888618732e-06, "loss": 0.7332, "mean_token_accuracy": 0.7701898217201233, "num_tokens": 54913779.0, "step": 838 }, { "epoch": 0.11089082738567274, "grad_norm": 0.8988865064257372, "learning_rate": 9.99685779088532e-06, "loss": 0.6655, "mean_token_accuracy": 0.7859693765640259, "num_tokens": 54979315.0, "step": 839 }, { "epoch": 0.11102299762093576, "grad_norm": 1.0351219975452641, "learning_rate": 9.996779735864769e-06, "loss": 0.6614, "mean_token_accuracy": 0.7897844910621643, "num_tokens": 55044851.0, "step": 840 }, { "epoch": 0.11115516785619878, "grad_norm": 0.9523718301833014, "learning_rate": 9.9967007235737e-06, "loss": 0.6861, "mean_token_accuracy": 0.7812232971191406, "num_tokens": 55110387.0, "step": 841 }, { "epoch": 0.1112873380914618, "grad_norm": 0.8998057433332352, "learning_rate": 9.996620754028931e-06, "loss": 0.7088, "mean_token_accuracy": 0.778201699256897, "num_tokens": 55175923.0, "step": 842 }, { "epoch": 0.11141950832672483, "grad_norm": 0.907618772434734, "learning_rate": 9.996539827247486e-06, "loss": 0.7039, "mean_token_accuracy": 0.7755463123321533, "num_tokens": 55241459.0, "step": 843 }, { "epoch": 0.11155167856198785, "grad_norm": 0.8480347421124249, "learning_rate": 9.996457943246592e-06, "loss": 0.705, "mean_token_accuracy": 0.7784305810928345, "num_tokens": 55306995.0, "step": 844 }, { "epoch": 0.11168384879725086, "grad_norm": 0.9037112220201505, "learning_rate": 9.996375102043683e-06, "loss": 0.6864, "mean_token_accuracy": 0.7802618741989136, "num_tokens": 55372531.0, "step": 845 }, { "epoch": 0.11181601903251388, "grad_norm": 0.8734307122610332, "learning_rate": 9.996291303656392e-06, "loss": 0.6997, "mean_token_accuracy": 0.778323769569397, "num_tokens": 55438067.0, "step": 846 }, { "epoch": 0.1119481892677769, "grad_norm": 0.8212643452822223, "learning_rate": 9.996206548102556e-06, "loss": 0.6506, "mean_token_accuracy": 0.7944542765617371, "num_tokens": 55503603.0, "step": 847 }, { "epoch": 0.11208035950303992, "grad_norm": 0.8256631430003405, "learning_rate": 9.996120835400222e-06, "loss": 0.63, "mean_token_accuracy": 0.7977811098098755, "num_tokens": 55569139.0, "step": 848 }, { "epoch": 0.11221252973830294, "grad_norm": 0.918818423451415, "learning_rate": 9.996034165567635e-06, "loss": 0.6777, "mean_token_accuracy": 0.7831156253814697, "num_tokens": 55634675.0, "step": 849 }, { "epoch": 0.11234469997356596, "grad_norm": 0.9621811300485074, "learning_rate": 9.995946538623243e-06, "loss": 0.6875, "mean_token_accuracy": 0.7852978706359863, "num_tokens": 55700211.0, "step": 850 }, { "epoch": 0.11247687020882897, "grad_norm": 0.8556945353457956, "learning_rate": 9.995857954585702e-06, "loss": 0.6786, "mean_token_accuracy": 0.784489095211029, "num_tokens": 55765747.0, "step": 851 }, { "epoch": 0.11260904044409199, "grad_norm": 0.8901788793276064, "learning_rate": 9.995768413473869e-06, "loss": 0.6575, "mean_token_accuracy": 0.7918447256088257, "num_tokens": 55831283.0, "step": 852 }, { "epoch": 0.11274121067935501, "grad_norm": 0.9680345995600157, "learning_rate": 9.995677915306805e-06, "loss": 0.7311, "mean_token_accuracy": 0.7670919299125671, "num_tokens": 55896819.0, "step": 853 }, { "epoch": 0.11287338091461803, "grad_norm": 0.9404163691382692, "learning_rate": 9.995586460103779e-06, "loss": 0.7674, "mean_token_accuracy": 0.7610792517662048, "num_tokens": 55962355.0, "step": 854 }, { "epoch": 0.11300555114988105, "grad_norm": 0.903379085911433, "learning_rate": 9.995494047884253e-06, "loss": 0.6657, "mean_token_accuracy": 0.7879074811935425, "num_tokens": 56027891.0, "step": 855 }, { "epoch": 0.11313772138514407, "grad_norm": 1.0060337890054836, "learning_rate": 9.995400678667904e-06, "loss": 0.7631, "mean_token_accuracy": 0.7580576539039612, "num_tokens": 56093427.0, "step": 856 }, { "epoch": 0.11326989162040708, "grad_norm": 0.8858640199203629, "learning_rate": 9.995306352474608e-06, "loss": 0.6602, "mean_token_accuracy": 0.7884873747825623, "num_tokens": 56158963.0, "step": 857 }, { "epoch": 0.1134020618556701, "grad_norm": 1.0282708360318873, "learning_rate": 9.995211069324446e-06, "loss": 0.7413, "mean_token_accuracy": 0.7637650966644287, "num_tokens": 56224499.0, "step": 858 }, { "epoch": 0.11353423209093312, "grad_norm": 0.9835742300711621, "learning_rate": 9.995114829237702e-06, "loss": 0.7475, "mean_token_accuracy": 0.7602704167366028, "num_tokens": 56290035.0, "step": 859 }, { "epoch": 0.11366640232619614, "grad_norm": 0.9389487528171833, "learning_rate": 9.995017632234858e-06, "loss": 0.7079, "mean_token_accuracy": 0.7746917605400085, "num_tokens": 56355571.0, "step": 860 }, { "epoch": 0.11379857256145916, "grad_norm": 0.9557709794271193, "learning_rate": 9.99491947833661e-06, "loss": 0.7167, "mean_token_accuracy": 0.7722805738449097, "num_tokens": 56421107.0, "step": 861 }, { "epoch": 0.11393074279672218, "grad_norm": 0.9155564882093563, "learning_rate": 9.994820367563855e-06, "loss": 0.6999, "mean_token_accuracy": 0.7796514630317688, "num_tokens": 56486643.0, "step": 862 }, { "epoch": 0.1140629130319852, "grad_norm": 0.9234931305826509, "learning_rate": 9.994720299937686e-06, "loss": 0.7094, "mean_token_accuracy": 0.7763398885726929, "num_tokens": 56552179.0, "step": 863 }, { "epoch": 0.11419508326724821, "grad_norm": 0.8995387128851795, "learning_rate": 9.99461927547941e-06, "loss": 0.7425, "mean_token_accuracy": 0.7668324708938599, "num_tokens": 56617715.0, "step": 864 }, { "epoch": 0.11432725350251123, "grad_norm": 0.8889530099096715, "learning_rate": 9.994517294210532e-06, "loss": 0.6493, "mean_token_accuracy": 0.7931723594665527, "num_tokens": 56683251.0, "step": 865 }, { "epoch": 0.11445942373777425, "grad_norm": 0.8942246411231233, "learning_rate": 9.994414356152759e-06, "loss": 0.6782, "mean_token_accuracy": 0.7864729762077332, "num_tokens": 56748787.0, "step": 866 }, { "epoch": 0.11459159397303727, "grad_norm": 0.95180070935939, "learning_rate": 9.994310461328007e-06, "loss": 0.6952, "mean_token_accuracy": 0.7807654738426208, "num_tokens": 56814323.0, "step": 867 }, { "epoch": 0.11472376420830029, "grad_norm": 0.952185637676156, "learning_rate": 9.994205609758392e-06, "loss": 0.7111, "mean_token_accuracy": 0.7730740904808044, "num_tokens": 56879859.0, "step": 868 }, { "epoch": 0.1148559344435633, "grad_norm": 1.0216560721238308, "learning_rate": 9.994099801466234e-06, "loss": 0.7393, "mean_token_accuracy": 0.765047013759613, "num_tokens": 56945395.0, "step": 869 }, { "epoch": 0.11498810467882632, "grad_norm": 0.9372859065479991, "learning_rate": 9.99399303647406e-06, "loss": 0.6798, "mean_token_accuracy": 0.7822610139846802, "num_tokens": 57010931.0, "step": 870 }, { "epoch": 0.11512027491408934, "grad_norm": 0.8668952910209262, "learning_rate": 9.993885314804596e-06, "loss": 0.6656, "mean_token_accuracy": 0.7883347868919373, "num_tokens": 57076467.0, "step": 871 }, { "epoch": 0.11525244514935237, "grad_norm": 1.0387812713948474, "learning_rate": 9.993776636480774e-06, "loss": 0.7535, "mean_token_accuracy": 0.7608045339584351, "num_tokens": 57142003.0, "step": 872 }, { "epoch": 0.11538461538461539, "grad_norm": 0.9173675510065151, "learning_rate": 9.993667001525727e-06, "loss": 0.6924, "mean_token_accuracy": 0.7818794846534729, "num_tokens": 57207539.0, "step": 873 }, { "epoch": 0.11551678561987841, "grad_norm": 0.8383367189789802, "learning_rate": 9.9935564099628e-06, "loss": 0.6227, "mean_token_accuracy": 0.8040074706077576, "num_tokens": 57273075.0, "step": 874 }, { "epoch": 0.11564895585514143, "grad_norm": 0.9395578852213122, "learning_rate": 9.993444861815529e-06, "loss": 0.6931, "mean_token_accuracy": 0.7845043540000916, "num_tokens": 57338611.0, "step": 875 }, { "epoch": 0.11578112609040445, "grad_norm": 0.9340266482328976, "learning_rate": 9.993332357107665e-06, "loss": 0.7297, "mean_token_accuracy": 0.7719905972480774, "num_tokens": 57404147.0, "step": 876 }, { "epoch": 0.11591329632566746, "grad_norm": 0.9437587044745644, "learning_rate": 9.993218895863155e-06, "loss": 0.6958, "mean_token_accuracy": 0.7773470878601074, "num_tokens": 57469683.0, "step": 877 }, { "epoch": 0.11604546656093048, "grad_norm": 0.9608546314555536, "learning_rate": 9.993104478106157e-06, "loss": 0.6648, "mean_token_accuracy": 0.787861704826355, "num_tokens": 57535219.0, "step": 878 }, { "epoch": 0.1161776367961935, "grad_norm": 0.880921291364097, "learning_rate": 9.992989103861019e-06, "loss": 0.7074, "mean_token_accuracy": 0.7759431004524231, "num_tokens": 57600755.0, "step": 879 }, { "epoch": 0.11630980703145652, "grad_norm": 0.861641756500726, "learning_rate": 9.992872773152312e-06, "loss": 0.6182, "mean_token_accuracy": 0.8070290684700012, "num_tokens": 57666291.0, "step": 880 }, { "epoch": 0.11644197726671954, "grad_norm": 0.9754829480869345, "learning_rate": 9.992755486004797e-06, "loss": 0.6831, "mean_token_accuracy": 0.7807959914207458, "num_tokens": 57731827.0, "step": 881 }, { "epoch": 0.11657414750198256, "grad_norm": 0.867212191150789, "learning_rate": 9.992637242443436e-06, "loss": 0.6586, "mean_token_accuracy": 0.7891130447387695, "num_tokens": 57797363.0, "step": 882 }, { "epoch": 0.11670631773724557, "grad_norm": 0.8858946319538707, "learning_rate": 9.99251804249341e-06, "loss": 0.6784, "mean_token_accuracy": 0.7829782962799072, "num_tokens": 57862899.0, "step": 883 }, { "epoch": 0.11683848797250859, "grad_norm": 0.8722676774926225, "learning_rate": 9.992397886180089e-06, "loss": 0.6422, "mean_token_accuracy": 0.7965144515037537, "num_tokens": 57928435.0, "step": 884 }, { "epoch": 0.11697065820777161, "grad_norm": 1.0906197812303131, "learning_rate": 9.992276773529051e-06, "loss": 0.723, "mean_token_accuracy": 0.7719753384590149, "num_tokens": 57993971.0, "step": 885 }, { "epoch": 0.11710282844303463, "grad_norm": 0.9119308642924936, "learning_rate": 9.992154704566081e-06, "loss": 0.6533, "mean_token_accuracy": 0.7970790863037109, "num_tokens": 58059507.0, "step": 886 }, { "epoch": 0.11723499867829765, "grad_norm": 0.9106905431395972, "learning_rate": 9.992031679317164e-06, "loss": 0.694, "mean_token_accuracy": 0.7808265089988708, "num_tokens": 58125043.0, "step": 887 }, { "epoch": 0.11736716891356067, "grad_norm": 0.9221750635411498, "learning_rate": 9.99190769780849e-06, "loss": 0.6329, "mean_token_accuracy": 0.7994750142097473, "num_tokens": 58190579.0, "step": 888 }, { "epoch": 0.11749933914882368, "grad_norm": 0.9137786744310805, "learning_rate": 9.99178276006645e-06, "loss": 0.6523, "mean_token_accuracy": 0.7942406535148621, "num_tokens": 58256115.0, "step": 889 }, { "epoch": 0.1176315093840867, "grad_norm": 0.9488628731409298, "learning_rate": 9.991656866117642e-06, "loss": 0.6728, "mean_token_accuracy": 0.783588707447052, "num_tokens": 58321651.0, "step": 890 }, { "epoch": 0.11776367961934972, "grad_norm": 0.9559427553289405, "learning_rate": 9.991530015988868e-06, "loss": 0.7184, "mean_token_accuracy": 0.7728757262229919, "num_tokens": 58387187.0, "step": 891 }, { "epoch": 0.11789584985461274, "grad_norm": 0.9995999769480649, "learning_rate": 9.991402209707129e-06, "loss": 0.6948, "mean_token_accuracy": 0.7796819806098938, "num_tokens": 58452723.0, "step": 892 }, { "epoch": 0.11802802008987576, "grad_norm": 0.9645381502260573, "learning_rate": 9.991273447299632e-06, "loss": 0.7074, "mean_token_accuracy": 0.7745238542556763, "num_tokens": 58518259.0, "step": 893 }, { "epoch": 0.11816019032513878, "grad_norm": 0.9223338674987897, "learning_rate": 9.991143728793791e-06, "loss": 0.6846, "mean_token_accuracy": 0.7867171168327332, "num_tokens": 58583795.0, "step": 894 }, { "epoch": 0.1182923605604018, "grad_norm": 0.9388080310726603, "learning_rate": 9.991013054217217e-06, "loss": 0.7205, "mean_token_accuracy": 0.7691521048545837, "num_tokens": 58649331.0, "step": 895 }, { "epoch": 0.11842453079566481, "grad_norm": 0.8867886409134217, "learning_rate": 9.99088142359773e-06, "loss": 0.6856, "mean_token_accuracy": 0.7815743088722229, "num_tokens": 58714867.0, "step": 896 }, { "epoch": 0.11855670103092783, "grad_norm": 0.8990568953642215, "learning_rate": 9.99074883696335e-06, "loss": 0.6465, "mean_token_accuracy": 0.792439877986908, "num_tokens": 58780403.0, "step": 897 }, { "epoch": 0.11868887126619085, "grad_norm": 0.8478829720958001, "learning_rate": 9.990615294342306e-06, "loss": 0.6851, "mean_token_accuracy": 0.7842296361923218, "num_tokens": 58845939.0, "step": 898 }, { "epoch": 0.11882104150145387, "grad_norm": 1.0089900852077847, "learning_rate": 9.99048079576302e-06, "loss": 0.6908, "mean_token_accuracy": 0.779529333114624, "num_tokens": 58911475.0, "step": 899 }, { "epoch": 0.11895321173671689, "grad_norm": 0.9919234547448456, "learning_rate": 9.990345341254126e-06, "loss": 0.662, "mean_token_accuracy": 0.7867476344108582, "num_tokens": 58977011.0, "step": 900 }, { "epoch": 0.1190853819719799, "grad_norm": 0.8712888380509383, "learning_rate": 9.990208930844464e-06, "loss": 0.6987, "mean_token_accuracy": 0.7795140743255615, "num_tokens": 59042547.0, "step": 901 }, { "epoch": 0.11921755220724294, "grad_norm": 0.9579832137301596, "learning_rate": 9.990071564563067e-06, "loss": 0.7135, "mean_token_accuracy": 0.7717922329902649, "num_tokens": 59108083.0, "step": 902 }, { "epoch": 0.11934972244250595, "grad_norm": 1.1044232349910548, "learning_rate": 9.98993324243918e-06, "loss": 0.6748, "mean_token_accuracy": 0.7824899554252625, "num_tokens": 59173619.0, "step": 903 }, { "epoch": 0.11948189267776897, "grad_norm": 0.9865849815400511, "learning_rate": 9.989793964502251e-06, "loss": 0.6654, "mean_token_accuracy": 0.7870833873748779, "num_tokens": 59239155.0, "step": 904 }, { "epoch": 0.11961406291303199, "grad_norm": 0.8754623461263403, "learning_rate": 9.989653730781924e-06, "loss": 0.6428, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 59304691.0, "step": 905 }, { "epoch": 0.11974623314829501, "grad_norm": 0.8894933171387212, "learning_rate": 9.989512541308056e-06, "loss": 0.6728, "mean_token_accuracy": 0.7856793999671936, "num_tokens": 59370227.0, "step": 906 }, { "epoch": 0.11987840338355803, "grad_norm": 0.9901173135814125, "learning_rate": 9.989370396110703e-06, "loss": 0.684, "mean_token_accuracy": 0.782871425151825, "num_tokens": 59435763.0, "step": 907 }, { "epoch": 0.12001057361882105, "grad_norm": 0.9298988709560575, "learning_rate": 9.989227295220123e-06, "loss": 0.6622, "mean_token_accuracy": 0.7873122692108154, "num_tokens": 59501299.0, "step": 908 }, { "epoch": 0.12014274385408406, "grad_norm": 0.9447884941925063, "learning_rate": 9.989083238666782e-06, "loss": 0.7242, "mean_token_accuracy": 0.7677481174468994, "num_tokens": 59566835.0, "step": 909 }, { "epoch": 0.12027491408934708, "grad_norm": 0.9081142747345232, "learning_rate": 9.988938226481342e-06, "loss": 0.6612, "mean_token_accuracy": 0.7900592088699341, "num_tokens": 59632371.0, "step": 910 }, { "epoch": 0.1204070843246101, "grad_norm": 1.0547624929397803, "learning_rate": 9.988792258694677e-06, "loss": 0.7014, "mean_token_accuracy": 0.7766756415367126, "num_tokens": 59697907.0, "step": 911 }, { "epoch": 0.12053925455987312, "grad_norm": 1.0039868615552447, "learning_rate": 9.98864533533786e-06, "loss": 0.7114, "mean_token_accuracy": 0.7745848894119263, "num_tokens": 59763443.0, "step": 912 }, { "epoch": 0.12067142479513614, "grad_norm": 0.9852799902864808, "learning_rate": 9.988497456442167e-06, "loss": 0.714, "mean_token_accuracy": 0.7720516324043274, "num_tokens": 59828979.0, "step": 913 }, { "epoch": 0.12080359503039916, "grad_norm": 0.9385337169021991, "learning_rate": 9.988348622039079e-06, "loss": 0.6717, "mean_token_accuracy": 0.7857404351234436, "num_tokens": 59894515.0, "step": 914 }, { "epoch": 0.12093576526566217, "grad_norm": 0.9521004447014574, "learning_rate": 9.988198832160278e-06, "loss": 0.6473, "mean_token_accuracy": 0.7928061485290527, "num_tokens": 59960051.0, "step": 915 }, { "epoch": 0.12106793550092519, "grad_norm": 0.969083719400874, "learning_rate": 9.988048086837655e-06, "loss": 0.6809, "mean_token_accuracy": 0.7864729762077332, "num_tokens": 60025587.0, "step": 916 }, { "epoch": 0.12120010573618821, "grad_norm": 0.9195510720987833, "learning_rate": 9.987896386103295e-06, "loss": 0.6204, "mean_token_accuracy": 0.8054572343826294, "num_tokens": 60091123.0, "step": 917 }, { "epoch": 0.12133227597145123, "grad_norm": 0.8450266038480477, "learning_rate": 9.987743729989496e-06, "loss": 0.6793, "mean_token_accuracy": 0.7822610139846802, "num_tokens": 60156659.0, "step": 918 }, { "epoch": 0.12146444620671425, "grad_norm": 0.923007560860558, "learning_rate": 9.987590118528754e-06, "loss": 0.7089, "mean_token_accuracy": 0.7738524079322815, "num_tokens": 60222195.0, "step": 919 }, { "epoch": 0.12159661644197726, "grad_norm": 0.902583141393507, "learning_rate": 9.987435551753771e-06, "loss": 0.685, "mean_token_accuracy": 0.7762778401374817, "num_tokens": 60283614.0, "step": 920 }, { "epoch": 0.12172878667724028, "grad_norm": 0.8755593920192344, "learning_rate": 9.98728002969745e-06, "loss": 0.647, "mean_token_accuracy": 0.7920430898666382, "num_tokens": 60349150.0, "step": 921 }, { "epoch": 0.1218609569125033, "grad_norm": 0.9541454143588372, "learning_rate": 9.987123552392897e-06, "loss": 0.6387, "mean_token_accuracy": 0.7957056760787964, "num_tokens": 60414686.0, "step": 922 }, { "epoch": 0.12199312714776632, "grad_norm": 1.002403460403259, "learning_rate": 9.986966119873426e-06, "loss": 0.7346, "mean_token_accuracy": 0.7657794952392578, "num_tokens": 60480222.0, "step": 923 }, { "epoch": 0.12212529738302934, "grad_norm": 0.9426688071570617, "learning_rate": 9.986807732172547e-06, "loss": 0.6446, "mean_token_accuracy": 0.7945916056632996, "num_tokens": 60545758.0, "step": 924 }, { "epoch": 0.12225746761829236, "grad_norm": 1.0225958433981603, "learning_rate": 9.986648389323981e-06, "loss": 0.7031, "mean_token_accuracy": 0.7776218056678772, "num_tokens": 60611294.0, "step": 925 }, { "epoch": 0.12238963785355537, "grad_norm": 1.0250087037131155, "learning_rate": 9.986488091361647e-06, "loss": 0.6952, "mean_token_accuracy": 0.7804907560348511, "num_tokens": 60676830.0, "step": 926 }, { "epoch": 0.12252180808881839, "grad_norm": 0.9032038558143899, "learning_rate": 9.986326838319669e-06, "loss": 0.6763, "mean_token_accuracy": 0.7856946587562561, "num_tokens": 60742366.0, "step": 927 }, { "epoch": 0.12265397832408141, "grad_norm": 1.0823495816711548, "learning_rate": 9.986164630232375e-06, "loss": 0.7339, "mean_token_accuracy": 0.7673055529594421, "num_tokens": 60807902.0, "step": 928 }, { "epoch": 0.12278614855934443, "grad_norm": 0.9494094797697544, "learning_rate": 9.986001467134296e-06, "loss": 0.7376, "mean_token_accuracy": 0.7675497531890869, "num_tokens": 60873438.0, "step": 929 }, { "epoch": 0.12291831879460745, "grad_norm": 0.8627790303448464, "learning_rate": 9.985837349060166e-06, "loss": 0.6664, "mean_token_accuracy": 0.7858319878578186, "num_tokens": 60938974.0, "step": 930 }, { "epoch": 0.12305048902987048, "grad_norm": 1.0973531160857062, "learning_rate": 9.98567227604492e-06, "loss": 0.7385, "mean_token_accuracy": 0.7671987414360046, "num_tokens": 61004510.0, "step": 931 }, { "epoch": 0.1231826592651335, "grad_norm": 0.8853175123170495, "learning_rate": 9.985506248123702e-06, "loss": 0.6804, "mean_token_accuracy": 0.7836344838142395, "num_tokens": 61070046.0, "step": 932 }, { "epoch": 0.12331482950039652, "grad_norm": 1.0289596913696817, "learning_rate": 9.985339265331855e-06, "loss": 0.7467, "mean_token_accuracy": 0.7626663446426392, "num_tokens": 61135582.0, "step": 933 }, { "epoch": 0.12344699973565953, "grad_norm": 0.9710350766291331, "learning_rate": 9.985171327704926e-06, "loss": 0.7377, "mean_token_accuracy": 0.7665425539016724, "num_tokens": 61201118.0, "step": 934 }, { "epoch": 0.12357916997092255, "grad_norm": 0.9403788888528369, "learning_rate": 9.985002435278663e-06, "loss": 0.7266, "mean_token_accuracy": 0.7687400579452515, "num_tokens": 61266654.0, "step": 935 }, { "epoch": 0.12371134020618557, "grad_norm": 0.9081524291068247, "learning_rate": 9.984832588089023e-06, "loss": 0.6812, "mean_token_accuracy": 0.7854352593421936, "num_tokens": 61332190.0, "step": 936 }, { "epoch": 0.12384351044144859, "grad_norm": 0.9561679695103382, "learning_rate": 9.98466178617216e-06, "loss": 0.6681, "mean_token_accuracy": 0.7835734486579895, "num_tokens": 61397726.0, "step": 937 }, { "epoch": 0.12397568067671161, "grad_norm": 0.8953070599534535, "learning_rate": 9.984490029564438e-06, "loss": 0.6285, "mean_token_accuracy": 0.7960718870162964, "num_tokens": 61463262.0, "step": 938 }, { "epoch": 0.12410785091197463, "grad_norm": 1.002482176826035, "learning_rate": 9.984317318302415e-06, "loss": 0.6743, "mean_token_accuracy": 0.7821999788284302, "num_tokens": 61528798.0, "step": 939 }, { "epoch": 0.12424002114723764, "grad_norm": 0.8897718671881556, "learning_rate": 9.984143652422864e-06, "loss": 0.6905, "mean_token_accuracy": 0.779285192489624, "num_tokens": 61594334.0, "step": 940 }, { "epoch": 0.12437219138250066, "grad_norm": 0.9480685181767987, "learning_rate": 9.983969031962749e-06, "loss": 0.7009, "mean_token_accuracy": 0.7775454521179199, "num_tokens": 61659870.0, "step": 941 }, { "epoch": 0.12450436161776368, "grad_norm": 0.8895504652440092, "learning_rate": 9.983793456959247e-06, "loss": 0.6754, "mean_token_accuracy": 0.7808570265769958, "num_tokens": 61725406.0, "step": 942 }, { "epoch": 0.1246365318530267, "grad_norm": 0.9479886110736038, "learning_rate": 9.983616927449733e-06, "loss": 0.7205, "mean_token_accuracy": 0.7711054682731628, "num_tokens": 61790942.0, "step": 943 }, { "epoch": 0.12476870208828972, "grad_norm": 1.0128379242922825, "learning_rate": 9.983439443471786e-06, "loss": 0.7078, "mean_token_accuracy": 0.7763551473617554, "num_tokens": 61856478.0, "step": 944 }, { "epoch": 0.12490087232355274, "grad_norm": 0.9279821369026368, "learning_rate": 9.98326100506319e-06, "loss": 0.7324, "mean_token_accuracy": 0.7667409181594849, "num_tokens": 61922014.0, "step": 945 }, { "epoch": 0.12503304255881575, "grad_norm": 0.8843651053362263, "learning_rate": 9.983081612261927e-06, "loss": 0.6619, "mean_token_accuracy": 0.7877548336982727, "num_tokens": 61987550.0, "step": 946 }, { "epoch": 0.12516521279407877, "grad_norm": 0.9246097213854958, "learning_rate": 9.982901265106193e-06, "loss": 0.7139, "mean_token_accuracy": 0.7731198668479919, "num_tokens": 62053086.0, "step": 947 }, { "epoch": 0.1252973830293418, "grad_norm": 0.9154746252499102, "learning_rate": 9.982719963634373e-06, "loss": 0.6986, "mean_token_accuracy": 0.7778659462928772, "num_tokens": 62118622.0, "step": 948 }, { "epoch": 0.1254295532646048, "grad_norm": 0.8922775999284589, "learning_rate": 9.982537707885064e-06, "loss": 0.6644, "mean_token_accuracy": 0.7926992774009705, "num_tokens": 62184158.0, "step": 949 }, { "epoch": 0.12556172349986783, "grad_norm": 0.906960272858688, "learning_rate": 9.982354497897068e-06, "loss": 0.6809, "mean_token_accuracy": 0.783588707447052, "num_tokens": 62249694.0, "step": 950 }, { "epoch": 0.12569389373513085, "grad_norm": 0.9832039122550026, "learning_rate": 9.982170333709384e-06, "loss": 0.6723, "mean_token_accuracy": 0.784611165523529, "num_tokens": 62315230.0, "step": 951 }, { "epoch": 0.12582606397039386, "grad_norm": 0.9152105428156156, "learning_rate": 9.981985215361214e-06, "loss": 0.6541, "mean_token_accuracy": 0.7889909744262695, "num_tokens": 62380766.0, "step": 952 }, { "epoch": 0.12595823420565688, "grad_norm": 0.9600460196765886, "learning_rate": 9.981799142891972e-06, "loss": 0.716, "mean_token_accuracy": 0.7750884890556335, "num_tokens": 62446302.0, "step": 953 }, { "epoch": 0.1260904044409199, "grad_norm": 0.9007002791104399, "learning_rate": 9.981612116341267e-06, "loss": 0.6648, "mean_token_accuracy": 0.7898455858230591, "num_tokens": 62511838.0, "step": 954 }, { "epoch": 0.12622257467618292, "grad_norm": 0.8494179057391127, "learning_rate": 9.981424135748908e-06, "loss": 0.6668, "mean_token_accuracy": 0.7884721159934998, "num_tokens": 62577374.0, "step": 955 }, { "epoch": 0.12635474491144594, "grad_norm": 0.9595709358747062, "learning_rate": 9.981235201154918e-06, "loss": 0.6829, "mean_token_accuracy": 0.7815743088722229, "num_tokens": 62642910.0, "step": 956 }, { "epoch": 0.12648691514670896, "grad_norm": 0.932101823962142, "learning_rate": 9.981045312599515e-06, "loss": 0.6747, "mean_token_accuracy": 0.7863203287124634, "num_tokens": 62708446.0, "step": 957 }, { "epoch": 0.12661908538197197, "grad_norm": 0.9117187189192303, "learning_rate": 9.980854470123122e-06, "loss": 0.6882, "mean_token_accuracy": 0.7816506028175354, "num_tokens": 62773982.0, "step": 958 }, { "epoch": 0.126751255617235, "grad_norm": 0.8365262645202238, "learning_rate": 9.980662673766366e-06, "loss": 0.6068, "mean_token_accuracy": 0.807822585105896, "num_tokens": 62839518.0, "step": 959 }, { "epoch": 0.126883425852498, "grad_norm": 0.9573553143690938, "learning_rate": 9.980469923570076e-06, "loss": 0.6662, "mean_token_accuracy": 0.78801429271698, "num_tokens": 62905054.0, "step": 960 }, { "epoch": 0.12701559608776103, "grad_norm": 1.0216074437108345, "learning_rate": 9.980276219575286e-06, "loss": 0.7476, "mean_token_accuracy": 0.7608655691146851, "num_tokens": 62970590.0, "step": 961 }, { "epoch": 0.12714776632302405, "grad_norm": 0.9815355140084292, "learning_rate": 9.98008156182323e-06, "loss": 0.666, "mean_token_accuracy": 0.7865034937858582, "num_tokens": 63036126.0, "step": 962 }, { "epoch": 0.12727993655828707, "grad_norm": 0.9143585725234424, "learning_rate": 9.979885950355344e-06, "loss": 0.6947, "mean_token_accuracy": 0.7788273692131042, "num_tokens": 63101662.0, "step": 963 }, { "epoch": 0.12741210679355008, "grad_norm": 0.9414908022839665, "learning_rate": 9.979689385213275e-06, "loss": 0.6677, "mean_token_accuracy": 0.7913411259651184, "num_tokens": 63167198.0, "step": 964 }, { "epoch": 0.1275442770288131, "grad_norm": 0.9718640550727926, "learning_rate": 9.979491866438864e-06, "loss": 0.6947, "mean_token_accuracy": 0.7785984873771667, "num_tokens": 63232734.0, "step": 965 }, { "epoch": 0.12767644726407612, "grad_norm": 0.8587235221563922, "learning_rate": 9.979293394074161e-06, "loss": 0.6746, "mean_token_accuracy": 0.7850384712219238, "num_tokens": 63298270.0, "step": 966 }, { "epoch": 0.12780861749933914, "grad_norm": 0.9871523630067054, "learning_rate": 9.979093968161415e-06, "loss": 0.7118, "mean_token_accuracy": 0.7731962203979492, "num_tokens": 63363806.0, "step": 967 }, { "epoch": 0.12794078773460216, "grad_norm": 0.9970624203293192, "learning_rate": 9.978893588743079e-06, "loss": 0.6914, "mean_token_accuracy": 0.7802160978317261, "num_tokens": 63429342.0, "step": 968 }, { "epoch": 0.12807295796986518, "grad_norm": 0.9337359626366359, "learning_rate": 9.97869225586181e-06, "loss": 0.6061, "mean_token_accuracy": 0.8057318925857544, "num_tokens": 63494878.0, "step": 969 }, { "epoch": 0.1282051282051282, "grad_norm": 0.9212579135458363, "learning_rate": 9.97848996956047e-06, "loss": 0.6931, "mean_token_accuracy": 0.778537392616272, "num_tokens": 63560414.0, "step": 970 }, { "epoch": 0.1283372984403912, "grad_norm": 0.9049517834386506, "learning_rate": 9.978286729882118e-06, "loss": 0.6469, "mean_token_accuracy": 0.7940269708633423, "num_tokens": 63625950.0, "step": 971 }, { "epoch": 0.12846946867565423, "grad_norm": 1.0590383721574919, "learning_rate": 9.978082536870024e-06, "loss": 0.7275, "mean_token_accuracy": 0.7683738470077515, "num_tokens": 63691486.0, "step": 972 }, { "epoch": 0.12860163891091725, "grad_norm": 0.9455764398096234, "learning_rate": 9.97787739056765e-06, "loss": 0.6355, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 63757022.0, "step": 973 }, { "epoch": 0.12873380914618027, "grad_norm": 1.0193827871158871, "learning_rate": 9.977671291018672e-06, "loss": 0.6638, "mean_token_accuracy": 0.7885941863059998, "num_tokens": 63822558.0, "step": 974 }, { "epoch": 0.12886597938144329, "grad_norm": 0.877101426384349, "learning_rate": 9.977464238266963e-06, "loss": 0.6629, "mean_token_accuracy": 0.7895098328590393, "num_tokens": 63888094.0, "step": 975 }, { "epoch": 0.12899814961670633, "grad_norm": 1.0037643045070037, "learning_rate": 9.9772562323566e-06, "loss": 0.6988, "mean_token_accuracy": 0.7766146063804626, "num_tokens": 63953630.0, "step": 976 }, { "epoch": 0.12913031985196935, "grad_norm": 0.8853418242175783, "learning_rate": 9.977047273331865e-06, "loss": 0.6653, "mean_token_accuracy": 0.7904102206230164, "num_tokens": 64019166.0, "step": 977 }, { "epoch": 0.12926249008723237, "grad_norm": 1.0772923562458603, "learning_rate": 9.97683736123724e-06, "loss": 0.7016, "mean_token_accuracy": 0.773867666721344, "num_tokens": 64084702.0, "step": 978 }, { "epoch": 0.1293946603224954, "grad_norm": 1.0139448331360623, "learning_rate": 9.97662649611741e-06, "loss": 0.76, "mean_token_accuracy": 0.7578592300415039, "num_tokens": 64150238.0, "step": 979 }, { "epoch": 0.1295268305577584, "grad_norm": 0.9358137874227191, "learning_rate": 9.976414678017262e-06, "loss": 0.7146, "mean_token_accuracy": 0.774737536907196, "num_tokens": 64215774.0, "step": 980 }, { "epoch": 0.12965900079302142, "grad_norm": 0.9780090780431341, "learning_rate": 9.976201906981891e-06, "loss": 0.6994, "mean_token_accuracy": 0.7769350409507751, "num_tokens": 64281310.0, "step": 981 }, { "epoch": 0.12979117102828444, "grad_norm": 0.9940297557274543, "learning_rate": 9.975988183056593e-06, "loss": 0.7058, "mean_token_accuracy": 0.7743712663650513, "num_tokens": 64346846.0, "step": 982 }, { "epoch": 0.12992334126354746, "grad_norm": 0.8766567993202298, "learning_rate": 9.97577350628686e-06, "loss": 0.667, "mean_token_accuracy": 0.7852368354797363, "num_tokens": 64412382.0, "step": 983 }, { "epoch": 0.13005551149881048, "grad_norm": 0.8947374968480177, "learning_rate": 9.975557876718393e-06, "loss": 0.66, "mean_token_accuracy": 0.7886247038841248, "num_tokens": 64477918.0, "step": 984 }, { "epoch": 0.1301876817340735, "grad_norm": 1.2233268325525508, "learning_rate": 9.9753412943971e-06, "loss": 0.7325, "mean_token_accuracy": 0.7689689993858337, "num_tokens": 64543454.0, "step": 985 }, { "epoch": 0.13031985196933651, "grad_norm": 1.018689042565179, "learning_rate": 9.975123759369084e-06, "loss": 0.7106, "mean_token_accuracy": 0.7733640670776367, "num_tokens": 64608990.0, "step": 986 }, { "epoch": 0.13045202220459953, "grad_norm": 0.9751959123251318, "learning_rate": 9.974905271680653e-06, "loss": 0.6519, "mean_token_accuracy": 0.7925772070884705, "num_tokens": 64674526.0, "step": 987 }, { "epoch": 0.13058419243986255, "grad_norm": 1.0191307945256143, "learning_rate": 9.974685831378318e-06, "loss": 0.68, "mean_token_accuracy": 0.7824136018753052, "num_tokens": 64740062.0, "step": 988 }, { "epoch": 0.13071636267512557, "grad_norm": 0.9734364720451695, "learning_rate": 9.974465438508794e-06, "loss": 0.6486, "mean_token_accuracy": 0.7905628085136414, "num_tokens": 64805598.0, "step": 989 }, { "epoch": 0.1308485329103886, "grad_norm": 0.8859262497091723, "learning_rate": 9.974244093119e-06, "loss": 0.6415, "mean_token_accuracy": 0.7953699231147766, "num_tokens": 64871134.0, "step": 990 }, { "epoch": 0.1309807031456516, "grad_norm": 1.0889977980053904, "learning_rate": 9.974021795256053e-06, "loss": 0.7099, "mean_token_accuracy": 0.7754395008087158, "num_tokens": 64936670.0, "step": 991 }, { "epoch": 0.13111287338091462, "grad_norm": 0.8535525903870504, "learning_rate": 9.973798544967277e-06, "loss": 0.6332, "mean_token_accuracy": 0.7971554398536682, "num_tokens": 65002206.0, "step": 992 }, { "epoch": 0.13124504361617764, "grad_norm": 0.8756330201737383, "learning_rate": 9.973574342300198e-06, "loss": 0.644, "mean_token_accuracy": 0.7963160872459412, "num_tokens": 65067742.0, "step": 993 }, { "epoch": 0.13137721385144066, "grad_norm": 0.8888252811174543, "learning_rate": 9.973349187302541e-06, "loss": 0.6878, "mean_token_accuracy": 0.7802313566207886, "num_tokens": 65133278.0, "step": 994 }, { "epoch": 0.13150938408670368, "grad_norm": 0.820512880655112, "learning_rate": 9.97312308002224e-06, "loss": 0.5447, "mean_token_accuracy": 0.8267763257026672, "num_tokens": 65198814.0, "step": 995 }, { "epoch": 0.1316415543219667, "grad_norm": 0.9133793400372747, "learning_rate": 9.972896020507425e-06, "loss": 0.6817, "mean_token_accuracy": 0.7849621772766113, "num_tokens": 65264350.0, "step": 996 }, { "epoch": 0.13177372455722972, "grad_norm": 1.1112319367561379, "learning_rate": 9.972668008806438e-06, "loss": 0.7221, "mean_token_accuracy": 0.7699151635169983, "num_tokens": 65329886.0, "step": 997 }, { "epoch": 0.13190589479249273, "grad_norm": 0.9900928014934821, "learning_rate": 9.972439044967813e-06, "loss": 0.7369, "mean_token_accuracy": 0.7636735439300537, "num_tokens": 65395422.0, "step": 998 }, { "epoch": 0.13203806502775575, "grad_norm": 0.9411058190916453, "learning_rate": 9.972209129040293e-06, "loss": 0.6605, "mean_token_accuracy": 0.788044810295105, "num_tokens": 65460958.0, "step": 999 }, { "epoch": 0.13217023526301877, "grad_norm": 0.895888253802484, "learning_rate": 9.971978261072824e-06, "loss": 0.6731, "mean_token_accuracy": 0.7878311276435852, "num_tokens": 65526494.0, "step": 1000 }, { "epoch": 0.1323024054982818, "grad_norm": 1.0902689350771437, "learning_rate": 9.97174644111455e-06, "loss": 0.7171, "mean_token_accuracy": 0.7725247144699097, "num_tokens": 65592030.0, "step": 1001 }, { "epoch": 0.1324345757335448, "grad_norm": 1.0523546924747915, "learning_rate": 9.971513669214819e-06, "loss": 0.6749, "mean_token_accuracy": 0.7832834720611572, "num_tokens": 65657566.0, "step": 1002 }, { "epoch": 0.13256674596880783, "grad_norm": 1.0419510775789709, "learning_rate": 9.971279945423191e-06, "loss": 0.7028, "mean_token_accuracy": 0.7762483358383179, "num_tokens": 65723102.0, "step": 1003 }, { "epoch": 0.13269891620407084, "grad_norm": 0.9511858904861961, "learning_rate": 9.971045269789414e-06, "loss": 0.6927, "mean_token_accuracy": 0.7789494395256042, "num_tokens": 65788638.0, "step": 1004 }, { "epoch": 0.13283108643933386, "grad_norm": 0.9768900937927131, "learning_rate": 9.970809642363449e-06, "loss": 0.6796, "mean_token_accuracy": 0.7851452827453613, "num_tokens": 65854174.0, "step": 1005 }, { "epoch": 0.13296325667459688, "grad_norm": 0.9700812356655776, "learning_rate": 9.970573063195455e-06, "loss": 0.6896, "mean_token_accuracy": 0.782627284526825, "num_tokens": 65919710.0, "step": 1006 }, { "epoch": 0.1330954269098599, "grad_norm": 0.9622502222005809, "learning_rate": 9.970335532335793e-06, "loss": 0.6389, "mean_token_accuracy": 0.7989256381988525, "num_tokens": 65985246.0, "step": 1007 }, { "epoch": 0.13322759714512292, "grad_norm": 1.0048587825059812, "learning_rate": 9.97009704983503e-06, "loss": 0.6624, "mean_token_accuracy": 0.7901965379714966, "num_tokens": 66050782.0, "step": 1008 }, { "epoch": 0.13335976738038594, "grad_norm": 1.0840263469801044, "learning_rate": 9.969857615743936e-06, "loss": 0.6433, "mean_token_accuracy": 0.7936301827430725, "num_tokens": 66116318.0, "step": 1009 }, { "epoch": 0.13349193761564895, "grad_norm": 1.100085223970143, "learning_rate": 9.969617230113478e-06, "loss": 0.7095, "mean_token_accuracy": 0.7739744782447815, "num_tokens": 66181854.0, "step": 1010 }, { "epoch": 0.13362410785091197, "grad_norm": 0.9727718941515714, "learning_rate": 9.969375892994832e-06, "loss": 0.6667, "mean_token_accuracy": 0.7893572449684143, "num_tokens": 66247390.0, "step": 1011 }, { "epoch": 0.133756278086175, "grad_norm": 0.9044191974826781, "learning_rate": 9.969133604439372e-06, "loss": 0.671, "mean_token_accuracy": 0.7876785397529602, "num_tokens": 66312926.0, "step": 1012 }, { "epoch": 0.133888448321438, "grad_norm": 0.9631479676027578, "learning_rate": 9.968890364498675e-06, "loss": 0.7117, "mean_token_accuracy": 0.7755920886993408, "num_tokens": 66378462.0, "step": 1013 }, { "epoch": 0.13402061855670103, "grad_norm": 1.0098592628471563, "learning_rate": 9.968646173224523e-06, "loss": 0.6919, "mean_token_accuracy": 0.7778049111366272, "num_tokens": 66443998.0, "step": 1014 }, { "epoch": 0.13415278879196404, "grad_norm": 1.1051431815606423, "learning_rate": 9.9684010306689e-06, "loss": 0.6769, "mean_token_accuracy": 0.7862135171890259, "num_tokens": 66509534.0, "step": 1015 }, { "epoch": 0.13428495902722706, "grad_norm": 0.9588636348933702, "learning_rate": 9.968154936883991e-06, "loss": 0.7107, "mean_token_accuracy": 0.7736387252807617, "num_tokens": 66575070.0, "step": 1016 }, { "epoch": 0.13441712926249008, "grad_norm": 0.8772178367199948, "learning_rate": 9.967907891922186e-06, "loss": 0.6168, "mean_token_accuracy": 0.7995360493659973, "num_tokens": 66640606.0, "step": 1017 }, { "epoch": 0.1345492994977531, "grad_norm": 0.9354573894693121, "learning_rate": 9.96765989583607e-06, "loss": 0.7, "mean_token_accuracy": 0.7753784656524658, "num_tokens": 66706142.0, "step": 1018 }, { "epoch": 0.13468146973301612, "grad_norm": 1.0146713540206533, "learning_rate": 9.967410948678445e-06, "loss": 0.6432, "mean_token_accuracy": 0.7946678996086121, "num_tokens": 66771678.0, "step": 1019 }, { "epoch": 0.13481363996827914, "grad_norm": 0.9407949218280257, "learning_rate": 9.967161050502298e-06, "loss": 0.6726, "mean_token_accuracy": 0.7859541177749634, "num_tokens": 66837214.0, "step": 1020 }, { "epoch": 0.13494581020354215, "grad_norm": 0.9815467412955742, "learning_rate": 9.966910201360835e-06, "loss": 0.6499, "mean_token_accuracy": 0.7938896417617798, "num_tokens": 66902750.0, "step": 1021 }, { "epoch": 0.13507798043880517, "grad_norm": 0.9729466138115945, "learning_rate": 9.966658401307448e-06, "loss": 0.6846, "mean_token_accuracy": 0.7810096740722656, "num_tokens": 66968286.0, "step": 1022 }, { "epoch": 0.1352101506740682, "grad_norm": 0.8310575383514612, "learning_rate": 9.966405650395748e-06, "loss": 0.6427, "mean_token_accuracy": 0.7964839339256287, "num_tokens": 67033822.0, "step": 1023 }, { "epoch": 0.1353423209093312, "grad_norm": 0.8595723106828415, "learning_rate": 9.966151948679534e-06, "loss": 0.6304, "mean_token_accuracy": 0.7963160872459412, "num_tokens": 67099358.0, "step": 1024 }, { "epoch": 0.13547449114459423, "grad_norm": 0.8683271071646329, "learning_rate": 9.965897296212818e-06, "loss": 0.6237, "mean_token_accuracy": 0.7999480962753296, "num_tokens": 67164894.0, "step": 1025 }, { "epoch": 0.13560666137985725, "grad_norm": 1.1392655215593825, "learning_rate": 9.96564169304981e-06, "loss": 0.7267, "mean_token_accuracy": 0.765138566493988, "num_tokens": 67230430.0, "step": 1026 }, { "epoch": 0.13573883161512026, "grad_norm": 0.9179146433501987, "learning_rate": 9.96538513924492e-06, "loss": 0.6516, "mean_token_accuracy": 0.7922109365463257, "num_tokens": 67295966.0, "step": 1027 }, { "epoch": 0.13587100185038328, "grad_norm": 1.069768171737084, "learning_rate": 9.965127634852764e-06, "loss": 0.6853, "mean_token_accuracy": 0.7800940275192261, "num_tokens": 67361502.0, "step": 1028 }, { "epoch": 0.1360031720856463, "grad_norm": 0.8853531920703698, "learning_rate": 9.96486917992816e-06, "loss": 0.6368, "mean_token_accuracy": 0.7967281341552734, "num_tokens": 67427038.0, "step": 1029 }, { "epoch": 0.13613534232090932, "grad_norm": 0.9835424281872198, "learning_rate": 9.96460977452613e-06, "loss": 0.6942, "mean_token_accuracy": 0.7797887921333313, "num_tokens": 67492574.0, "step": 1030 }, { "epoch": 0.13626751255617234, "grad_norm": 0.8268860036598262, "learning_rate": 9.96434941870189e-06, "loss": 0.5992, "mean_token_accuracy": 0.8102185130119324, "num_tokens": 67558110.0, "step": 1031 }, { "epoch": 0.13639968279143536, "grad_norm": 0.9534704228063442, "learning_rate": 9.964088112510869e-06, "loss": 0.6748, "mean_token_accuracy": 0.7825509905815125, "num_tokens": 67623646.0, "step": 1032 }, { "epoch": 0.13653185302669837, "grad_norm": 0.8365973925145966, "learning_rate": 9.963825856008692e-06, "loss": 0.6457, "mean_token_accuracy": 0.7915852665901184, "num_tokens": 67689182.0, "step": 1033 }, { "epoch": 0.1366640232619614, "grad_norm": 0.863571038951319, "learning_rate": 9.963562649251188e-06, "loss": 0.6523, "mean_token_accuracy": 0.7959803342819214, "num_tokens": 67754718.0, "step": 1034 }, { "epoch": 0.13679619349722444, "grad_norm": 0.8870133209291085, "learning_rate": 9.963298492294388e-06, "loss": 0.65, "mean_token_accuracy": 0.7903186678886414, "num_tokens": 67820254.0, "step": 1035 }, { "epoch": 0.13692836373248746, "grad_norm": 0.9124377665904825, "learning_rate": 9.963033385194527e-06, "loss": 0.6422, "mean_token_accuracy": 0.7939659357070923, "num_tokens": 67885790.0, "step": 1036 }, { "epoch": 0.13706053396775048, "grad_norm": 0.8815330110595144, "learning_rate": 9.962767328008039e-06, "loss": 0.6778, "mean_token_accuracy": 0.7849774360656738, "num_tokens": 67951326.0, "step": 1037 }, { "epoch": 0.1371927042030135, "grad_norm": 0.8694792073974773, "learning_rate": 9.962500320791564e-06, "loss": 0.7314, "mean_token_accuracy": 0.7657032012939453, "num_tokens": 68016862.0, "step": 1038 }, { "epoch": 0.1373248744382765, "grad_norm": 0.8472605620903012, "learning_rate": 9.96223236360194e-06, "loss": 0.5867, "mean_token_accuracy": 0.812248170375824, "num_tokens": 68082398.0, "step": 1039 }, { "epoch": 0.13745704467353953, "grad_norm": 0.892615536850603, "learning_rate": 9.96196345649621e-06, "loss": 0.6687, "mean_token_accuracy": 0.7866713404655457, "num_tokens": 68147934.0, "step": 1040 }, { "epoch": 0.13758921490880255, "grad_norm": 0.9250037265609001, "learning_rate": 9.96169359953162e-06, "loss": 0.7265, "mean_token_accuracy": 0.7697625160217285, "num_tokens": 68213470.0, "step": 1041 }, { "epoch": 0.13772138514406557, "grad_norm": 0.8853640486800978, "learning_rate": 9.961422792765618e-06, "loss": 0.6322, "mean_token_accuracy": 0.7965754866600037, "num_tokens": 68279006.0, "step": 1042 }, { "epoch": 0.13785355537932859, "grad_norm": 0.9149721736260128, "learning_rate": 9.961151036255849e-06, "loss": 0.6447, "mean_token_accuracy": 0.7911274433135986, "num_tokens": 68344542.0, "step": 1043 }, { "epoch": 0.1379857256145916, "grad_norm": 0.7688083515422365, "learning_rate": 9.960878330060167e-06, "loss": 0.5718, "mean_token_accuracy": 0.8184897899627686, "num_tokens": 68410078.0, "step": 1044 }, { "epoch": 0.13811789584985462, "grad_norm": 0.8499400553425173, "learning_rate": 9.960604674236624e-06, "loss": 0.6557, "mean_token_accuracy": 0.7900897264480591, "num_tokens": 68475614.0, "step": 1045 }, { "epoch": 0.13825006608511764, "grad_norm": 0.962603756822811, "learning_rate": 9.960330068843477e-06, "loss": 0.6981, "mean_token_accuracy": 0.7744628190994263, "num_tokens": 68541150.0, "step": 1046 }, { "epoch": 0.13838223632038066, "grad_norm": 0.9098215114956856, "learning_rate": 9.960054513939183e-06, "loss": 0.6595, "mean_token_accuracy": 0.7869613170623779, "num_tokens": 68606686.0, "step": 1047 }, { "epoch": 0.13851440655564368, "grad_norm": 0.908042831892974, "learning_rate": 9.959778009582403e-06, "loss": 0.6831, "mean_token_accuracy": 0.779193639755249, "num_tokens": 68672222.0, "step": 1048 }, { "epoch": 0.1386465767909067, "grad_norm": 0.929749077630452, "learning_rate": 9.959500555831999e-06, "loss": 0.6744, "mean_token_accuracy": 0.7838786244392395, "num_tokens": 68737758.0, "step": 1049 }, { "epoch": 0.1387787470261697, "grad_norm": 0.8911745950645077, "learning_rate": 9.959222152747032e-06, "loss": 0.656, "mean_token_accuracy": 0.7905170321464539, "num_tokens": 68803294.0, "step": 1050 }, { "epoch": 0.13891091726143273, "grad_norm": 0.8556219335054143, "learning_rate": 9.958942800386772e-06, "loss": 0.6384, "mean_token_accuracy": 0.7995818853378296, "num_tokens": 68868830.0, "step": 1051 }, { "epoch": 0.13904308749669575, "grad_norm": 0.9721488583298523, "learning_rate": 9.958662498810687e-06, "loss": 0.7195, "mean_token_accuracy": 0.7736082077026367, "num_tokens": 68934366.0, "step": 1052 }, { "epoch": 0.13917525773195877, "grad_norm": 1.1517631349247677, "learning_rate": 9.958381248078443e-06, "loss": 0.767, "mean_token_accuracy": 0.7611860632896423, "num_tokens": 68999902.0, "step": 1053 }, { "epoch": 0.1393074279672218, "grad_norm": 0.9519639276651319, "learning_rate": 9.958099048249917e-06, "loss": 0.6818, "mean_token_accuracy": 0.7804297208786011, "num_tokens": 69065438.0, "step": 1054 }, { "epoch": 0.1394395982024848, "grad_norm": 0.9362198204494818, "learning_rate": 9.957815899385183e-06, "loss": 0.6882, "mean_token_accuracy": 0.7770571112632751, "num_tokens": 69130974.0, "step": 1055 }, { "epoch": 0.13957176843774782, "grad_norm": 0.9314685796412732, "learning_rate": 9.957531801544516e-06, "loss": 0.7108, "mean_token_accuracy": 0.779163122177124, "num_tokens": 69196510.0, "step": 1056 }, { "epoch": 0.13970393867301084, "grad_norm": 0.8687470557155744, "learning_rate": 9.957246754788396e-06, "loss": 0.6913, "mean_token_accuracy": 0.7811775207519531, "num_tokens": 69262046.0, "step": 1057 }, { "epoch": 0.13983610890827386, "grad_norm": 0.8788134273281856, "learning_rate": 9.956960759177503e-06, "loss": 0.6613, "mean_token_accuracy": 0.7894945740699768, "num_tokens": 69327582.0, "step": 1058 }, { "epoch": 0.13996827914353688, "grad_norm": 0.8862691241191796, "learning_rate": 9.956673814772715e-06, "loss": 0.6717, "mean_token_accuracy": 0.788227915763855, "num_tokens": 69393118.0, "step": 1059 }, { "epoch": 0.1401004493787999, "grad_norm": 0.8421897361364111, "learning_rate": 9.956385921635124e-06, "loss": 0.6494, "mean_token_accuracy": 0.7915242314338684, "num_tokens": 69458654.0, "step": 1060 }, { "epoch": 0.14023261961406291, "grad_norm": 0.9042812043388583, "learning_rate": 9.95609707982601e-06, "loss": 0.6274, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 69524190.0, "step": 1061 }, { "epoch": 0.14036478984932593, "grad_norm": 0.8377943476936827, "learning_rate": 9.955807289406866e-06, "loss": 0.6128, "mean_token_accuracy": 0.8047246932983398, "num_tokens": 69589726.0, "step": 1062 }, { "epoch": 0.14049696008458895, "grad_norm": 0.8740613395997775, "learning_rate": 9.955516550439382e-06, "loss": 0.6478, "mean_token_accuracy": 0.7916005253791809, "num_tokens": 69655262.0, "step": 1063 }, { "epoch": 0.14062913031985197, "grad_norm": 0.8578712182233826, "learning_rate": 9.955224862985447e-06, "loss": 0.6114, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 69720798.0, "step": 1064 }, { "epoch": 0.140761300555115, "grad_norm": 0.9344980299386291, "learning_rate": 9.95493222710716e-06, "loss": 0.6649, "mean_token_accuracy": 0.7897234559059143, "num_tokens": 69786334.0, "step": 1065 }, { "epoch": 0.140893470790378, "grad_norm": 1.0111411487789108, "learning_rate": 9.954638642866813e-06, "loss": 0.696, "mean_token_accuracy": 0.778140664100647, "num_tokens": 69851870.0, "step": 1066 }, { "epoch": 0.14102564102564102, "grad_norm": 0.9355115301639955, "learning_rate": 9.954344110326903e-06, "loss": 0.677, "mean_token_accuracy": 0.782932460308075, "num_tokens": 69917406.0, "step": 1067 }, { "epoch": 0.14115781126090404, "grad_norm": 0.9646068673774497, "learning_rate": 9.954048629550131e-06, "loss": 0.6802, "mean_token_accuracy": 0.7832682132720947, "num_tokens": 69982942.0, "step": 1068 }, { "epoch": 0.14128998149616706, "grad_norm": 0.894926872508572, "learning_rate": 9.953752200599403e-06, "loss": 0.6811, "mean_token_accuracy": 0.7812538146972656, "num_tokens": 70048478.0, "step": 1069 }, { "epoch": 0.14142215173143008, "grad_norm": 0.9682420386283926, "learning_rate": 9.953454823537817e-06, "loss": 0.6934, "mean_token_accuracy": 0.7777438759803772, "num_tokens": 70114014.0, "step": 1070 }, { "epoch": 0.1415543219666931, "grad_norm": 0.9578478292043625, "learning_rate": 9.953156498428682e-06, "loss": 0.6152, "mean_token_accuracy": 0.7996276617050171, "num_tokens": 70179550.0, "step": 1071 }, { "epoch": 0.14168649220195612, "grad_norm": 0.9069208508718571, "learning_rate": 9.952857225335502e-06, "loss": 0.6384, "mean_token_accuracy": 0.7930197715759277, "num_tokens": 70245086.0, "step": 1072 }, { "epoch": 0.14181866243721913, "grad_norm": 0.9241188288501931, "learning_rate": 9.952557004321988e-06, "loss": 0.6845, "mean_token_accuracy": 0.7820931673049927, "num_tokens": 70310622.0, "step": 1073 }, { "epoch": 0.14195083267248215, "grad_norm": 0.9595858671296851, "learning_rate": 9.952255835452053e-06, "loss": 0.6955, "mean_token_accuracy": 0.7802923917770386, "num_tokens": 70376158.0, "step": 1074 }, { "epoch": 0.14208300290774517, "grad_norm": 0.8462433844736211, "learning_rate": 9.951953718789805e-06, "loss": 0.644, "mean_token_accuracy": 0.7964839339256287, "num_tokens": 70441694.0, "step": 1075 }, { "epoch": 0.1422151731430082, "grad_norm": 0.9302675975624267, "learning_rate": 9.95165065439956e-06, "loss": 0.6508, "mean_token_accuracy": 0.7920430898666382, "num_tokens": 70507230.0, "step": 1076 }, { "epoch": 0.1423473433782712, "grad_norm": 1.1360993545209015, "learning_rate": 9.951346642345835e-06, "loss": 0.672, "mean_token_accuracy": 0.7861365675926208, "num_tokens": 70565593.0, "step": 1077 }, { "epoch": 0.14247951361353423, "grad_norm": 0.9294808024564196, "learning_rate": 9.951041682693347e-06, "loss": 0.6423, "mean_token_accuracy": 0.7987425327301025, "num_tokens": 70631129.0, "step": 1078 }, { "epoch": 0.14261168384879724, "grad_norm": 0.9028711966880869, "learning_rate": 9.950735775507017e-06, "loss": 0.6306, "mean_token_accuracy": 0.7974911332130432, "num_tokens": 70696665.0, "step": 1079 }, { "epoch": 0.14274385408406026, "grad_norm": 0.8554119907211457, "learning_rate": 9.950428920851964e-06, "loss": 0.6636, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 70762201.0, "step": 1080 }, { "epoch": 0.14287602431932328, "grad_norm": 1.0886202879117104, "learning_rate": 9.950121118793512e-06, "loss": 0.6988, "mean_token_accuracy": 0.7726467847824097, "num_tokens": 70827737.0, "step": 1081 }, { "epoch": 0.1430081945545863, "grad_norm": 0.9267026622972465, "learning_rate": 9.949812369397184e-06, "loss": 0.6507, "mean_token_accuracy": 0.7920888662338257, "num_tokens": 70893273.0, "step": 1082 }, { "epoch": 0.14314036478984932, "grad_norm": 0.876566409561952, "learning_rate": 9.94950267272871e-06, "loss": 0.6767, "mean_token_accuracy": 0.784794270992279, "num_tokens": 70958809.0, "step": 1083 }, { "epoch": 0.14327253502511234, "grad_norm": 0.9893632770939967, "learning_rate": 9.949192028854017e-06, "loss": 0.6826, "mean_token_accuracy": 0.7842143774032593, "num_tokens": 71024345.0, "step": 1084 }, { "epoch": 0.14340470526037535, "grad_norm": 0.9974199535010636, "learning_rate": 9.94888043783923e-06, "loss": 0.681, "mean_token_accuracy": 0.7841075658798218, "num_tokens": 71089881.0, "step": 1085 }, { "epoch": 0.14353687549563837, "grad_norm": 0.9744568566702576, "learning_rate": 9.948567899750687e-06, "loss": 0.657, "mean_token_accuracy": 0.7886704802513123, "num_tokens": 71155417.0, "step": 1086 }, { "epoch": 0.1436690457309014, "grad_norm": 0.8564215519787488, "learning_rate": 9.948254414654918e-06, "loss": 0.6092, "mean_token_accuracy": 0.8028476238250732, "num_tokens": 71220953.0, "step": 1087 }, { "epoch": 0.1438012159661644, "grad_norm": 0.9453253381400458, "learning_rate": 9.947939982618655e-06, "loss": 0.6827, "mean_token_accuracy": 0.7818947434425354, "num_tokens": 71286489.0, "step": 1088 }, { "epoch": 0.14393338620142743, "grad_norm": 0.9188220598560102, "learning_rate": 9.947624603708835e-06, "loss": 0.6478, "mean_token_accuracy": 0.7928061485290527, "num_tokens": 71352025.0, "step": 1089 }, { "epoch": 0.14406555643669045, "grad_norm": 0.9544505781383696, "learning_rate": 9.947308277992599e-06, "loss": 0.7005, "mean_token_accuracy": 0.7769502997398376, "num_tokens": 71417561.0, "step": 1090 }, { "epoch": 0.14419772667195346, "grad_norm": 0.9379415523908812, "learning_rate": 9.946991005537283e-06, "loss": 0.6697, "mean_token_accuracy": 0.7875869870185852, "num_tokens": 71483097.0, "step": 1091 }, { "epoch": 0.14432989690721648, "grad_norm": 0.9014399525751047, "learning_rate": 9.946672786410427e-06, "loss": 0.6603, "mean_token_accuracy": 0.7884873747825623, "num_tokens": 71548633.0, "step": 1092 }, { "epoch": 0.1444620671424795, "grad_norm": 0.887491491456773, "learning_rate": 9.946353620679775e-06, "loss": 0.6651, "mean_token_accuracy": 0.7907917499542236, "num_tokens": 71614169.0, "step": 1093 }, { "epoch": 0.14459423737774255, "grad_norm": 0.8414730325543806, "learning_rate": 9.946033508413273e-06, "loss": 0.6701, "mean_token_accuracy": 0.7877395749092102, "num_tokens": 71679705.0, "step": 1094 }, { "epoch": 0.14472640761300556, "grad_norm": 0.9836641641056272, "learning_rate": 9.94571244967906e-06, "loss": 0.6957, "mean_token_accuracy": 0.779071569442749, "num_tokens": 71745241.0, "step": 1095 }, { "epoch": 0.14485857784826858, "grad_norm": 1.0177965595221194, "learning_rate": 9.94539044454549e-06, "loss": 0.703, "mean_token_accuracy": 0.7775149345397949, "num_tokens": 71810777.0, "step": 1096 }, { "epoch": 0.1449907480835316, "grad_norm": 0.892041054382695, "learning_rate": 9.945067493081105e-06, "loss": 0.634, "mean_token_accuracy": 0.7974911332130432, "num_tokens": 71876313.0, "step": 1097 }, { "epoch": 0.14512291831879462, "grad_norm": 0.8596095748702742, "learning_rate": 9.944743595354657e-06, "loss": 0.6819, "mean_token_accuracy": 0.7832529544830322, "num_tokens": 71941849.0, "step": 1098 }, { "epoch": 0.14525508855405764, "grad_norm": 0.9536641891447016, "learning_rate": 9.944418751435096e-06, "loss": 0.7069, "mean_token_accuracy": 0.7727231383323669, "num_tokens": 72007385.0, "step": 1099 }, { "epoch": 0.14538725878932066, "grad_norm": 1.024072206529439, "learning_rate": 9.944092961391579e-06, "loss": 0.7428, "mean_token_accuracy": 0.7628189325332642, "num_tokens": 72072921.0, "step": 1100 }, { "epoch": 0.14551942902458367, "grad_norm": 0.9885288599980392, "learning_rate": 9.943766225293452e-06, "loss": 0.7432, "mean_token_accuracy": 0.7627121210098267, "num_tokens": 72138457.0, "step": 1101 }, { "epoch": 0.1456515992598467, "grad_norm": 0.9529569413786795, "learning_rate": 9.943438543210279e-06, "loss": 0.6961, "mean_token_accuracy": 0.7797277569770813, "num_tokens": 72203993.0, "step": 1102 }, { "epoch": 0.1457837694951097, "grad_norm": 1.0229010233428322, "learning_rate": 9.943109915211811e-06, "loss": 0.7333, "mean_token_accuracy": 0.7676718235015869, "num_tokens": 72269529.0, "step": 1103 }, { "epoch": 0.14591593973037273, "grad_norm": 0.8690515644350644, "learning_rate": 9.942780341368008e-06, "loss": 0.5911, "mean_token_accuracy": 0.8147357106208801, "num_tokens": 72335065.0, "step": 1104 }, { "epoch": 0.14604810996563575, "grad_norm": 0.8171505596428643, "learning_rate": 9.942449821749028e-06, "loss": 0.6454, "mean_token_accuracy": 0.7939659357070923, "num_tokens": 72400601.0, "step": 1105 }, { "epoch": 0.14618028020089877, "grad_norm": 0.9002863485756976, "learning_rate": 9.942118356425232e-06, "loss": 0.6749, "mean_token_accuracy": 0.784763753414154, "num_tokens": 72466137.0, "step": 1106 }, { "epoch": 0.14631245043616178, "grad_norm": 0.9757659146864498, "learning_rate": 9.941785945467184e-06, "loss": 0.6575, "mean_token_accuracy": 0.7904102206230164, "num_tokens": 72531673.0, "step": 1107 }, { "epoch": 0.1464446206714248, "grad_norm": 0.882514253211209, "learning_rate": 9.941452588945645e-06, "loss": 0.677, "mean_token_accuracy": 0.7855725884437561, "num_tokens": 72597209.0, "step": 1108 }, { "epoch": 0.14657679090668782, "grad_norm": 0.8404195887579491, "learning_rate": 9.941118286931582e-06, "loss": 0.6423, "mean_token_accuracy": 0.7938286066055298, "num_tokens": 72662745.0, "step": 1109 }, { "epoch": 0.14670896114195084, "grad_norm": 0.8678826075720316, "learning_rate": 9.940783039496157e-06, "loss": 0.6471, "mean_token_accuracy": 0.7946984767913818, "num_tokens": 72728281.0, "step": 1110 }, { "epoch": 0.14684113137721386, "grad_norm": 1.03492095625639, "learning_rate": 9.940446846710742e-06, "loss": 0.6902, "mean_token_accuracy": 0.7776675820350647, "num_tokens": 72793817.0, "step": 1111 }, { "epoch": 0.14697330161247688, "grad_norm": 0.9765971206295362, "learning_rate": 9.940109708646903e-06, "loss": 0.6369, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 72859353.0, "step": 1112 }, { "epoch": 0.1471054718477399, "grad_norm": 0.8450152159427284, "learning_rate": 9.93977162537641e-06, "loss": 0.6271, "mean_token_accuracy": 0.7991698384284973, "num_tokens": 72924889.0, "step": 1113 }, { "epoch": 0.1472376420830029, "grad_norm": 0.9163421324409412, "learning_rate": 9.939432596971233e-06, "loss": 0.6368, "mean_token_accuracy": 0.7956140637397766, "num_tokens": 72990425.0, "step": 1114 }, { "epoch": 0.14736981231826593, "grad_norm": 0.9328841581520145, "learning_rate": 9.939092623503545e-06, "loss": 0.7029, "mean_token_accuracy": 0.7782169580459595, "num_tokens": 73055961.0, "step": 1115 }, { "epoch": 0.14750198255352895, "grad_norm": 1.0164345425446133, "learning_rate": 9.938751705045719e-06, "loss": 0.6996, "mean_token_accuracy": 0.7747833132743835, "num_tokens": 73121497.0, "step": 1116 }, { "epoch": 0.14763415278879197, "grad_norm": 0.9410879293347721, "learning_rate": 9.93840984167033e-06, "loss": 0.7075, "mean_token_accuracy": 0.7752716541290283, "num_tokens": 73187033.0, "step": 1117 }, { "epoch": 0.14776632302405499, "grad_norm": 0.8738992535147295, "learning_rate": 9.938067033450151e-06, "loss": 0.6442, "mean_token_accuracy": 0.79343181848526, "num_tokens": 73252569.0, "step": 1118 }, { "epoch": 0.147898493259318, "grad_norm": 1.026179328164957, "learning_rate": 9.937723280458163e-06, "loss": 0.6501, "mean_token_accuracy": 0.7897692322731018, "num_tokens": 73318105.0, "step": 1119 }, { "epoch": 0.14803066349458102, "grad_norm": 0.8671622756479657, "learning_rate": 9.93737858276754e-06, "loss": 0.6777, "mean_token_accuracy": 0.7850689888000488, "num_tokens": 73383641.0, "step": 1120 }, { "epoch": 0.14816283372984404, "grad_norm": 0.9859404621275468, "learning_rate": 9.937032940451665e-06, "loss": 0.7012, "mean_token_accuracy": 0.7732419967651367, "num_tokens": 73449177.0, "step": 1121 }, { "epoch": 0.14829500396510706, "grad_norm": 0.9443533478969246, "learning_rate": 9.936686353584115e-06, "loss": 0.67, "mean_token_accuracy": 0.7849316596984863, "num_tokens": 73514713.0, "step": 1122 }, { "epoch": 0.14842717420037008, "grad_norm": 0.8488910261759884, "learning_rate": 9.936338822238669e-06, "loss": 0.6817, "mean_token_accuracy": 0.7821389436721802, "num_tokens": 73580249.0, "step": 1123 }, { "epoch": 0.1485593444356331, "grad_norm": 1.075368144847345, "learning_rate": 9.935990346489315e-06, "loss": 0.6829, "mean_token_accuracy": 0.7797430157661438, "num_tokens": 73645785.0, "step": 1124 }, { "epoch": 0.1486915146708961, "grad_norm": 0.9152910139642173, "learning_rate": 9.93564092641023e-06, "loss": 0.6714, "mean_token_accuracy": 0.7861982583999634, "num_tokens": 73711321.0, "step": 1125 }, { "epoch": 0.14882368490615913, "grad_norm": 0.8775265267873539, "learning_rate": 9.935290562075803e-06, "loss": 0.6259, "mean_token_accuracy": 0.7994139790534973, "num_tokens": 73776857.0, "step": 1126 }, { "epoch": 0.14895585514142215, "grad_norm": 1.0473585944196988, "learning_rate": 9.934939253560618e-06, "loss": 0.7172, "mean_token_accuracy": 0.7704492807388306, "num_tokens": 73842393.0, "step": 1127 }, { "epoch": 0.14908802537668517, "grad_norm": 1.0932646096042495, "learning_rate": 9.93458700093946e-06, "loss": 0.6736, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 73907929.0, "step": 1128 }, { "epoch": 0.1492201956119482, "grad_norm": 1.0471250377043435, "learning_rate": 9.934233804287317e-06, "loss": 0.7085, "mean_token_accuracy": 0.774737536907196, "num_tokens": 73973465.0, "step": 1129 }, { "epoch": 0.1493523658472112, "grad_norm": 0.9503237862489462, "learning_rate": 9.933879663679378e-06, "loss": 0.6919, "mean_token_accuracy": 0.778079628944397, "num_tokens": 74039001.0, "step": 1130 }, { "epoch": 0.14948453608247422, "grad_norm": 0.9284450141417617, "learning_rate": 9.933524579191029e-06, "loss": 0.7235, "mean_token_accuracy": 0.7681601643562317, "num_tokens": 74104537.0, "step": 1131 }, { "epoch": 0.14961670631773724, "grad_norm": 0.9195764956867251, "learning_rate": 9.933168550897865e-06, "loss": 0.6654, "mean_token_accuracy": 0.7896013855934143, "num_tokens": 74170073.0, "step": 1132 }, { "epoch": 0.14974887655300026, "grad_norm": 0.9265434447544788, "learning_rate": 9.932811578875673e-06, "loss": 0.6508, "mean_token_accuracy": 0.7902728319168091, "num_tokens": 74235609.0, "step": 1133 }, { "epoch": 0.14988104678826328, "grad_norm": 0.9716520964618579, "learning_rate": 9.932453663200446e-06, "loss": 0.759, "mean_token_accuracy": 0.7608503103256226, "num_tokens": 74301145.0, "step": 1134 }, { "epoch": 0.1500132170235263, "grad_norm": 0.9107597198686291, "learning_rate": 9.932094803948379e-06, "loss": 0.6224, "mean_token_accuracy": 0.7987120151519775, "num_tokens": 74366681.0, "step": 1135 }, { "epoch": 0.15014538725878931, "grad_norm": 0.8661413514356903, "learning_rate": 9.931735001195862e-06, "loss": 0.6647, "mean_token_accuracy": 0.7870070934295654, "num_tokens": 74432217.0, "step": 1136 }, { "epoch": 0.15027755749405233, "grad_norm": 0.8737401580557701, "learning_rate": 9.931374255019493e-06, "loss": 0.6074, "mean_token_accuracy": 0.8045110702514648, "num_tokens": 74497753.0, "step": 1137 }, { "epoch": 0.15040972772931535, "grad_norm": 1.2897952600455251, "learning_rate": 9.931012565496065e-06, "loss": 0.6978, "mean_token_accuracy": 0.7770571112632751, "num_tokens": 74563289.0, "step": 1138 }, { "epoch": 0.15054189796457837, "grad_norm": 0.9774119286026643, "learning_rate": 9.930649932702573e-06, "loss": 0.6488, "mean_token_accuracy": 0.7904102206230164, "num_tokens": 74628825.0, "step": 1139 }, { "epoch": 0.1506740681998414, "grad_norm": 1.063288191490499, "learning_rate": 9.930286356716217e-06, "loss": 0.6998, "mean_token_accuracy": 0.7717006206512451, "num_tokens": 74694361.0, "step": 1140 }, { "epoch": 0.1508062384351044, "grad_norm": 0.9619404741463552, "learning_rate": 9.929921837614392e-06, "loss": 0.7164, "mean_token_accuracy": 0.7723568677902222, "num_tokens": 74759897.0, "step": 1141 }, { "epoch": 0.15093840867036742, "grad_norm": 1.0050478743753513, "learning_rate": 9.9295563754747e-06, "loss": 0.6387, "mean_token_accuracy": 0.7974453568458557, "num_tokens": 74825433.0, "step": 1142 }, { "epoch": 0.15107057890563044, "grad_norm": 0.9365515582355503, "learning_rate": 9.929189970374935e-06, "loss": 0.6602, "mean_token_accuracy": 0.78819739818573, "num_tokens": 74890969.0, "step": 1143 }, { "epoch": 0.15120274914089346, "grad_norm": 0.9393418080789252, "learning_rate": 9.928822622393104e-06, "loss": 0.6762, "mean_token_accuracy": 0.7843669652938843, "num_tokens": 74956505.0, "step": 1144 }, { "epoch": 0.15133491937615648, "grad_norm": 0.9762670350296174, "learning_rate": 9.928454331607403e-06, "loss": 0.6987, "mean_token_accuracy": 0.775073230266571, "num_tokens": 75022041.0, "step": 1145 }, { "epoch": 0.1514670896114195, "grad_norm": 0.946611930293099, "learning_rate": 9.928085098096232e-06, "loss": 0.6515, "mean_token_accuracy": 0.7925161719322205, "num_tokens": 75087577.0, "step": 1146 }, { "epoch": 0.15159925984668252, "grad_norm": 1.1250556994894747, "learning_rate": 9.927714921938197e-06, "loss": 0.6801, "mean_token_accuracy": 0.7830393314361572, "num_tokens": 75153113.0, "step": 1147 }, { "epoch": 0.15173143008194553, "grad_norm": 0.9054581262689795, "learning_rate": 9.927343803212099e-06, "loss": 0.6008, "mean_token_accuracy": 0.8059760928153992, "num_tokens": 75218649.0, "step": 1148 }, { "epoch": 0.15186360031720855, "grad_norm": 1.033619247545382, "learning_rate": 9.92697174199694e-06, "loss": 0.6993, "mean_token_accuracy": 0.7759431004524231, "num_tokens": 75284185.0, "step": 1149 }, { "epoch": 0.15199577055247157, "grad_norm": 0.8847812361215461, "learning_rate": 9.926598738371927e-06, "loss": 0.69, "mean_token_accuracy": 0.7801703214645386, "num_tokens": 75349721.0, "step": 1150 }, { "epoch": 0.1521279407877346, "grad_norm": 0.987622181418002, "learning_rate": 9.926224792416465e-06, "loss": 0.7021, "mean_token_accuracy": 0.7757294774055481, "num_tokens": 75415257.0, "step": 1151 }, { "epoch": 0.1522601110229976, "grad_norm": 1.03148460017212, "learning_rate": 9.925849904210154e-06, "loss": 0.6811, "mean_token_accuracy": 0.7818642258644104, "num_tokens": 75480793.0, "step": 1152 }, { "epoch": 0.15239228125826065, "grad_norm": 0.9532651715463231, "learning_rate": 9.925474073832807e-06, "loss": 0.6629, "mean_token_accuracy": 0.788975715637207, "num_tokens": 75546329.0, "step": 1153 }, { "epoch": 0.15252445149352367, "grad_norm": 0.895857411971557, "learning_rate": 9.925097301364425e-06, "loss": 0.6541, "mean_token_accuracy": 0.789250373840332, "num_tokens": 75611865.0, "step": 1154 }, { "epoch": 0.1526566217287867, "grad_norm": 0.8795835598156908, "learning_rate": 9.924719586885216e-06, "loss": 0.6268, "mean_token_accuracy": 0.8010163307189941, "num_tokens": 75677401.0, "step": 1155 }, { "epoch": 0.1527887919640497, "grad_norm": 0.9788247130207582, "learning_rate": 9.924340930475591e-06, "loss": 0.6839, "mean_token_accuracy": 0.7808875441551208, "num_tokens": 75742937.0, "step": 1156 }, { "epoch": 0.15292096219931273, "grad_norm": 0.899389872614553, "learning_rate": 9.923961332216154e-06, "loss": 0.6635, "mean_token_accuracy": 0.7899981737136841, "num_tokens": 75808473.0, "step": 1157 }, { "epoch": 0.15305313243457574, "grad_norm": 0.8616069921870672, "learning_rate": 9.923580792187714e-06, "loss": 0.6311, "mean_token_accuracy": 0.8009095191955566, "num_tokens": 75874009.0, "step": 1158 }, { "epoch": 0.15318530266983876, "grad_norm": 1.0889486729331206, "learning_rate": 9.923199310471284e-06, "loss": 0.7136, "mean_token_accuracy": 0.7716395854949951, "num_tokens": 75939545.0, "step": 1159 }, { "epoch": 0.15331747290510178, "grad_norm": 0.898305466001605, "learning_rate": 9.922816887148068e-06, "loss": 0.6677, "mean_token_accuracy": 0.7885484099388123, "num_tokens": 76005081.0, "step": 1160 }, { "epoch": 0.1534496431403648, "grad_norm": 0.8368443441672713, "learning_rate": 9.92243352229948e-06, "loss": 0.6906, "mean_token_accuracy": 0.7819710373878479, "num_tokens": 76070617.0, "step": 1161 }, { "epoch": 0.15358181337562782, "grad_norm": 0.8557533758249913, "learning_rate": 9.922049216007128e-06, "loss": 0.6422, "mean_token_accuracy": 0.7960566282272339, "num_tokens": 76136153.0, "step": 1162 }, { "epoch": 0.15371398361089084, "grad_norm": 1.0285949479040217, "learning_rate": 9.921663968352825e-06, "loss": 0.7183, "mean_token_accuracy": 0.7716090679168701, "num_tokens": 76201689.0, "step": 1163 }, { "epoch": 0.15384615384615385, "grad_norm": 0.9140787859216385, "learning_rate": 9.92127777941858e-06, "loss": 0.646, "mean_token_accuracy": 0.7895861268043518, "num_tokens": 76267225.0, "step": 1164 }, { "epoch": 0.15397832408141687, "grad_norm": 0.9120412807153789, "learning_rate": 9.920890649286603e-06, "loss": 0.6374, "mean_token_accuracy": 0.7966518402099609, "num_tokens": 76332761.0, "step": 1165 }, { "epoch": 0.1541104943166799, "grad_norm": 0.966162616578369, "learning_rate": 9.920502578039312e-06, "loss": 0.6558, "mean_token_accuracy": 0.7887162566184998, "num_tokens": 76398297.0, "step": 1166 }, { "epoch": 0.1542426645519429, "grad_norm": 0.8410943956558941, "learning_rate": 9.920113565759313e-06, "loss": 0.5745, "mean_token_accuracy": 0.8186119198799133, "num_tokens": 76463833.0, "step": 1167 }, { "epoch": 0.15437483478720593, "grad_norm": 1.0633779453668328, "learning_rate": 9.919723612529423e-06, "loss": 0.6813, "mean_token_accuracy": 0.7820626497268677, "num_tokens": 76529369.0, "step": 1168 }, { "epoch": 0.15450700502246895, "grad_norm": 1.0308604268997452, "learning_rate": 9.919332718432652e-06, "loss": 0.6789, "mean_token_accuracy": 0.7790104746818542, "num_tokens": 76594905.0, "step": 1169 }, { "epoch": 0.15463917525773196, "grad_norm": 0.8693764579217933, "learning_rate": 9.918940883552216e-06, "loss": 0.6351, "mean_token_accuracy": 0.7956904172897339, "num_tokens": 76660441.0, "step": 1170 }, { "epoch": 0.15477134549299498, "grad_norm": 0.9690846698839317, "learning_rate": 9.918548107971525e-06, "loss": 0.7109, "mean_token_accuracy": 0.7728604674339294, "num_tokens": 76725977.0, "step": 1171 }, { "epoch": 0.154903515728258, "grad_norm": 0.9514476359059755, "learning_rate": 9.918154391774194e-06, "loss": 0.6714, "mean_token_accuracy": 0.7823678255081177, "num_tokens": 76791513.0, "step": 1172 }, { "epoch": 0.15503568596352102, "grad_norm": 0.8787515564126637, "learning_rate": 9.917759735044038e-06, "loss": 0.6046, "mean_token_accuracy": 0.8056708574295044, "num_tokens": 76857049.0, "step": 1173 }, { "epoch": 0.15516785619878404, "grad_norm": 0.9486633748824271, "learning_rate": 9.917364137865072e-06, "loss": 0.694, "mean_token_accuracy": 0.7806739211082458, "num_tokens": 76922585.0, "step": 1174 }, { "epoch": 0.15530002643404706, "grad_norm": 0.9045819019807799, "learning_rate": 9.916967600321506e-06, "loss": 0.6958, "mean_token_accuracy": 0.7804449796676636, "num_tokens": 76988121.0, "step": 1175 }, { "epoch": 0.15543219666931007, "grad_norm": 0.8842170142209329, "learning_rate": 9.91657012249776e-06, "loss": 0.6382, "mean_token_accuracy": 0.7945458292961121, "num_tokens": 77053657.0, "step": 1176 }, { "epoch": 0.1555643669045731, "grad_norm": 0.9060609406726621, "learning_rate": 9.916171704478445e-06, "loss": 0.6569, "mean_token_accuracy": 0.788823127746582, "num_tokens": 77119193.0, "step": 1177 }, { "epoch": 0.1556965371398361, "grad_norm": 0.9474565314583847, "learning_rate": 9.915772346348378e-06, "loss": 0.7216, "mean_token_accuracy": 0.7709833979606628, "num_tokens": 77184729.0, "step": 1178 }, { "epoch": 0.15582870737509913, "grad_norm": 0.9431042735008638, "learning_rate": 9.915372048192573e-06, "loss": 0.7101, "mean_token_accuracy": 0.7718074917793274, "num_tokens": 77250265.0, "step": 1179 }, { "epoch": 0.15596087761036215, "grad_norm": 0.8481946039158941, "learning_rate": 9.914970810096245e-06, "loss": 0.6159, "mean_token_accuracy": 0.8050146698951721, "num_tokens": 77315801.0, "step": 1180 }, { "epoch": 0.15609304784562517, "grad_norm": 0.9256076392534206, "learning_rate": 9.91456863214481e-06, "loss": 0.694, "mean_token_accuracy": 0.778171181678772, "num_tokens": 77381337.0, "step": 1181 }, { "epoch": 0.15622521808088818, "grad_norm": 0.9198135619918268, "learning_rate": 9.914165514423881e-06, "loss": 0.6821, "mean_token_accuracy": 0.7819710373878479, "num_tokens": 77446873.0, "step": 1182 }, { "epoch": 0.1563573883161512, "grad_norm": 0.8743815188758204, "learning_rate": 9.913761457019276e-06, "loss": 0.6464, "mean_token_accuracy": 0.7915700078010559, "num_tokens": 77512409.0, "step": 1183 }, { "epoch": 0.15648955855141422, "grad_norm": 0.8816638499066038, "learning_rate": 9.91335646001701e-06, "loss": 0.6785, "mean_token_accuracy": 0.7819405198097229, "num_tokens": 77577945.0, "step": 1184 }, { "epoch": 0.15662172878667724, "grad_norm": 1.0084161914729342, "learning_rate": 9.912950523503298e-06, "loss": 0.7162, "mean_token_accuracy": 0.7701135277748108, "num_tokens": 77643481.0, "step": 1185 }, { "epoch": 0.15675389902194026, "grad_norm": 0.9235780220735706, "learning_rate": 9.912543647564555e-06, "loss": 0.6421, "mean_token_accuracy": 0.7934470772743225, "num_tokens": 77709017.0, "step": 1186 }, { "epoch": 0.15688606925720328, "grad_norm": 0.8676808488248516, "learning_rate": 9.912135832287396e-06, "loss": 0.6806, "mean_token_accuracy": 0.7838481068611145, "num_tokens": 77774553.0, "step": 1187 }, { "epoch": 0.1570182394924663, "grad_norm": 0.9429710325562722, "learning_rate": 9.91172707775864e-06, "loss": 0.655, "mean_token_accuracy": 0.7891435623168945, "num_tokens": 77840089.0, "step": 1188 }, { "epoch": 0.1571504097277293, "grad_norm": 0.8704267682848033, "learning_rate": 9.911317384065297e-06, "loss": 0.6624, "mean_token_accuracy": 0.7931418418884277, "num_tokens": 77905625.0, "step": 1189 }, { "epoch": 0.15728257996299233, "grad_norm": 0.9136358102079282, "learning_rate": 9.910906751294585e-06, "loss": 0.6773, "mean_token_accuracy": 0.7821694612503052, "num_tokens": 77971161.0, "step": 1190 }, { "epoch": 0.15741475019825535, "grad_norm": 0.893623590648476, "learning_rate": 9.91049517953392e-06, "loss": 0.6628, "mean_token_accuracy": 0.7857862114906311, "num_tokens": 78036697.0, "step": 1191 }, { "epoch": 0.15754692043351837, "grad_norm": 0.8212530188017377, "learning_rate": 9.910082668870915e-06, "loss": 0.6033, "mean_token_accuracy": 0.8056250810623169, "num_tokens": 78102233.0, "step": 1192 }, { "epoch": 0.15767909066878139, "grad_norm": 0.9769688540562474, "learning_rate": 9.909669219393383e-06, "loss": 0.653, "mean_token_accuracy": 0.7928366661071777, "num_tokens": 78167769.0, "step": 1193 }, { "epoch": 0.1578112609040444, "grad_norm": 0.8642823273011028, "learning_rate": 9.909254831189344e-06, "loss": 0.6831, "mean_token_accuracy": 0.7817116379737854, "num_tokens": 78233305.0, "step": 1194 }, { "epoch": 0.15794343113930742, "grad_norm": 0.9935799202482417, "learning_rate": 9.90883950434701e-06, "loss": 0.6954, "mean_token_accuracy": 0.7779269814491272, "num_tokens": 78298841.0, "step": 1195 }, { "epoch": 0.15807560137457044, "grad_norm": 0.9545068757485078, "learning_rate": 9.908423238954796e-06, "loss": 0.6383, "mean_token_accuracy": 0.7945458292961121, "num_tokens": 78364377.0, "step": 1196 }, { "epoch": 0.15820777160983346, "grad_norm": 0.8946062641116954, "learning_rate": 9.908006035101313e-06, "loss": 0.6276, "mean_token_accuracy": 0.7983762621879578, "num_tokens": 78429913.0, "step": 1197 }, { "epoch": 0.15833994184509648, "grad_norm": 0.9120656456796986, "learning_rate": 9.90758789287538e-06, "loss": 0.617, "mean_token_accuracy": 0.8047739267349243, "num_tokens": 78494103.0, "step": 1198 }, { "epoch": 0.1584721120803595, "grad_norm": 0.9994974718951292, "learning_rate": 9.907168812366008e-06, "loss": 0.6292, "mean_token_accuracy": 0.7960413694381714, "num_tokens": 78559639.0, "step": 1199 }, { "epoch": 0.1586042823156225, "grad_norm": 1.1030041706797222, "learning_rate": 9.906748793662412e-06, "loss": 0.6913, "mean_token_accuracy": 0.7772097587585449, "num_tokens": 78625175.0, "step": 1200 }, { "epoch": 0.15873645255088553, "grad_norm": 0.9840862447824148, "learning_rate": 9.906327836854002e-06, "loss": 0.671, "mean_token_accuracy": 0.7847484946250916, "num_tokens": 78690711.0, "step": 1201 }, { "epoch": 0.15886862278614855, "grad_norm": 0.9286744209744534, "learning_rate": 9.905905942030394e-06, "loss": 0.6043, "mean_token_accuracy": 0.8042210936546326, "num_tokens": 78756247.0, "step": 1202 }, { "epoch": 0.15900079302141157, "grad_norm": 0.9676816185046337, "learning_rate": 9.9054831092814e-06, "loss": 0.6263, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 78821783.0, "step": 1203 }, { "epoch": 0.1591329632566746, "grad_norm": 0.9998497109909211, "learning_rate": 9.905059338697031e-06, "loss": 0.6939, "mean_token_accuracy": 0.7792699337005615, "num_tokens": 78887319.0, "step": 1204 }, { "epoch": 0.1592651334919376, "grad_norm": 1.1193031834118379, "learning_rate": 9.9046346303675e-06, "loss": 0.6871, "mean_token_accuracy": 0.7806891798973083, "num_tokens": 78952855.0, "step": 1205 }, { "epoch": 0.15939730372720062, "grad_norm": 0.9362426436458505, "learning_rate": 9.904208984383218e-06, "loss": 0.6415, "mean_token_accuracy": 0.7930045127868652, "num_tokens": 79018391.0, "step": 1206 }, { "epoch": 0.15952947396246364, "grad_norm": 0.9408356248486869, "learning_rate": 9.903782400834799e-06, "loss": 0.6025, "mean_token_accuracy": 0.8049230575561523, "num_tokens": 79083927.0, "step": 1207 }, { "epoch": 0.15966164419772666, "grad_norm": 0.9174871717190558, "learning_rate": 9.90335487981305e-06, "loss": 0.6426, "mean_token_accuracy": 0.7955988049507141, "num_tokens": 79149463.0, "step": 1208 }, { "epoch": 0.15979381443298968, "grad_norm": 1.1256232969742122, "learning_rate": 9.902926421408979e-06, "loss": 0.6483, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 79214999.0, "step": 1209 }, { "epoch": 0.1599259846682527, "grad_norm": 0.9364421010879062, "learning_rate": 9.902497025713803e-06, "loss": 0.6065, "mean_token_accuracy": 0.8080667853355408, "num_tokens": 79280535.0, "step": 1210 }, { "epoch": 0.16005815490351571, "grad_norm": 0.9455608865770978, "learning_rate": 9.902066692818929e-06, "loss": 0.6428, "mean_token_accuracy": 0.7956904172897339, "num_tokens": 79346071.0, "step": 1211 }, { "epoch": 0.16019032513877876, "grad_norm": 0.8191576689163846, "learning_rate": 9.901635422815961e-06, "loss": 0.6611, "mean_token_accuracy": 0.7897844910621643, "num_tokens": 79411607.0, "step": 1212 }, { "epoch": 0.16032249537404178, "grad_norm": 0.9720608633502831, "learning_rate": 9.901203215796711e-06, "loss": 0.6519, "mean_token_accuracy": 0.787922739982605, "num_tokens": 79477143.0, "step": 1213 }, { "epoch": 0.1604546656093048, "grad_norm": 0.8900243386984109, "learning_rate": 9.90077007185319e-06, "loss": 0.6474, "mean_token_accuracy": 0.7932028770446777, "num_tokens": 79542679.0, "step": 1214 }, { "epoch": 0.16058683584456782, "grad_norm": 0.9347260902616729, "learning_rate": 9.9003359910776e-06, "loss": 0.6593, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 79608215.0, "step": 1215 }, { "epoch": 0.16071900607983083, "grad_norm": 0.8711783208408314, "learning_rate": 9.899900973562351e-06, "loss": 0.6147, "mean_token_accuracy": 0.8028934001922607, "num_tokens": 79673751.0, "step": 1216 }, { "epoch": 0.16085117631509385, "grad_norm": 0.9648024794303179, "learning_rate": 9.899465019400048e-06, "loss": 0.6308, "mean_token_accuracy": 0.798071026802063, "num_tokens": 79739287.0, "step": 1217 }, { "epoch": 0.16098334655035687, "grad_norm": 0.9763297406209126, "learning_rate": 9.899028128683496e-06, "loss": 0.6255, "mean_token_accuracy": 0.7982847094535828, "num_tokens": 79804823.0, "step": 1218 }, { "epoch": 0.1611155167856199, "grad_norm": 0.8579891697349534, "learning_rate": 9.8985903015057e-06, "loss": 0.6522, "mean_token_accuracy": 0.7906848788261414, "num_tokens": 79870359.0, "step": 1219 }, { "epoch": 0.1612476870208829, "grad_norm": 0.9176485030936076, "learning_rate": 9.898151537959866e-06, "loss": 0.6517, "mean_token_accuracy": 0.7908833026885986, "num_tokens": 79935895.0, "step": 1220 }, { "epoch": 0.16137985725614593, "grad_norm": 0.9368885006627146, "learning_rate": 9.897711838139396e-06, "loss": 0.662, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 80001431.0, "step": 1221 }, { "epoch": 0.16151202749140894, "grad_norm": 0.9993551892374481, "learning_rate": 9.897271202137892e-06, "loss": 0.6643, "mean_token_accuracy": 0.7873581051826477, "num_tokens": 80066967.0, "step": 1222 }, { "epoch": 0.16164419772667196, "grad_norm": 0.8133803806447244, "learning_rate": 9.89682963004916e-06, "loss": 0.6438, "mean_token_accuracy": 0.7959040403366089, "num_tokens": 80132503.0, "step": 1223 }, { "epoch": 0.16177636796193498, "grad_norm": 0.8907991277509718, "learning_rate": 9.896387121967196e-06, "loss": 0.707, "mean_token_accuracy": 0.7744780778884888, "num_tokens": 80198039.0, "step": 1224 }, { "epoch": 0.161908538197198, "grad_norm": 0.9290537368066726, "learning_rate": 9.895943677986207e-06, "loss": 0.6985, "mean_token_accuracy": 0.7789341807365417, "num_tokens": 80263575.0, "step": 1225 }, { "epoch": 0.16204070843246102, "grad_norm": 0.839735689753056, "learning_rate": 9.895499298200588e-06, "loss": 0.6611, "mean_token_accuracy": 0.7907153964042664, "num_tokens": 80329111.0, "step": 1226 }, { "epoch": 0.16217287866772404, "grad_norm": 0.9506271433798448, "learning_rate": 9.895053982704938e-06, "loss": 0.7258, "mean_token_accuracy": 0.7698693871498108, "num_tokens": 80394647.0, "step": 1227 }, { "epoch": 0.16230504890298705, "grad_norm": 0.9558424543945652, "learning_rate": 9.894607731594058e-06, "loss": 0.6629, "mean_token_accuracy": 0.7882126569747925, "num_tokens": 80460183.0, "step": 1228 }, { "epoch": 0.16243721913825007, "grad_norm": 0.9397506140671511, "learning_rate": 9.894160544962946e-06, "loss": 0.6587, "mean_token_accuracy": 0.7902117967605591, "num_tokens": 80525719.0, "step": 1229 }, { "epoch": 0.1625693893735131, "grad_norm": 0.9055123233634069, "learning_rate": 9.893712422906797e-06, "loss": 0.646, "mean_token_accuracy": 0.7895556092262268, "num_tokens": 80591255.0, "step": 1230 }, { "epoch": 0.1627015596087761, "grad_norm": 0.8879108363965126, "learning_rate": 9.893263365521008e-06, "loss": 0.6809, "mean_token_accuracy": 0.7809638381004333, "num_tokens": 80656791.0, "step": 1231 }, { "epoch": 0.16283372984403913, "grad_norm": 0.9792412526000229, "learning_rate": 9.892813372901173e-06, "loss": 0.7307, "mean_token_accuracy": 0.7653064131736755, "num_tokens": 80722327.0, "step": 1232 }, { "epoch": 0.16296590007930215, "grad_norm": 0.87612075225771, "learning_rate": 9.892362445143086e-06, "loss": 0.6816, "mean_token_accuracy": 0.783558189868927, "num_tokens": 80787863.0, "step": 1233 }, { "epoch": 0.16309807031456516, "grad_norm": 0.893255338015196, "learning_rate": 9.891910582342739e-06, "loss": 0.6446, "mean_token_accuracy": 0.792500913143158, "num_tokens": 80853399.0, "step": 1234 }, { "epoch": 0.16323024054982818, "grad_norm": 0.8643788800073087, "learning_rate": 9.891457784596327e-06, "loss": 0.6527, "mean_token_accuracy": 0.7900744676589966, "num_tokens": 80918935.0, "step": 1235 }, { "epoch": 0.1633624107850912, "grad_norm": 0.9495119992089435, "learning_rate": 9.891004052000239e-06, "loss": 0.6869, "mean_token_accuracy": 0.7806281447410583, "num_tokens": 80984471.0, "step": 1236 }, { "epoch": 0.16349458102035422, "grad_norm": 0.8849969095518591, "learning_rate": 9.890549384651067e-06, "loss": 0.6214, "mean_token_accuracy": 0.7983152270317078, "num_tokens": 81050007.0, "step": 1237 }, { "epoch": 0.16362675125561724, "grad_norm": 0.9385893344106662, "learning_rate": 9.890093782645598e-06, "loss": 0.6627, "mean_token_accuracy": 0.7846264243125916, "num_tokens": 81115543.0, "step": 1238 }, { "epoch": 0.16375892149088025, "grad_norm": 0.9565969955380248, "learning_rate": 9.88963724608082e-06, "loss": 0.6721, "mean_token_accuracy": 0.783558189868927, "num_tokens": 81181079.0, "step": 1239 }, { "epoch": 0.16389109172614327, "grad_norm": 1.0093918455951816, "learning_rate": 9.889179775053925e-06, "loss": 0.654, "mean_token_accuracy": 0.7885636687278748, "num_tokens": 81246615.0, "step": 1240 }, { "epoch": 0.1640232619614063, "grad_norm": 0.9557635762195118, "learning_rate": 9.888721369662293e-06, "loss": 0.6358, "mean_token_accuracy": 0.7952478528022766, "num_tokens": 81312151.0, "step": 1241 }, { "epoch": 0.1641554321966693, "grad_norm": 0.9165506551092701, "learning_rate": 9.888262030003512e-06, "loss": 0.6441, "mean_token_accuracy": 0.7942101359367371, "num_tokens": 81377687.0, "step": 1242 }, { "epoch": 0.16428760243193233, "grad_norm": 1.0323300392190096, "learning_rate": 9.887801756175365e-06, "loss": 0.7344, "mean_token_accuracy": 0.7665120363235474, "num_tokens": 81443223.0, "step": 1243 }, { "epoch": 0.16441977266719535, "grad_norm": 0.9977001976903689, "learning_rate": 9.887340548275833e-06, "loss": 0.6954, "mean_token_accuracy": 0.7763246297836304, "num_tokens": 81508759.0, "step": 1244 }, { "epoch": 0.16455194290245836, "grad_norm": 1.1429945831891475, "learning_rate": 9.886878406403101e-06, "loss": 0.7449, "mean_token_accuracy": 0.7609266042709351, "num_tokens": 81574295.0, "step": 1245 }, { "epoch": 0.16468411313772138, "grad_norm": 0.9392456901563346, "learning_rate": 9.886415330655547e-06, "loss": 0.7212, "mean_token_accuracy": 0.7704340219497681, "num_tokens": 81639831.0, "step": 1246 }, { "epoch": 0.1648162833729844, "grad_norm": 0.9047333757801315, "learning_rate": 9.885951321131752e-06, "loss": 0.6963, "mean_token_accuracy": 0.7772555351257324, "num_tokens": 81705367.0, "step": 1247 }, { "epoch": 0.16494845360824742, "grad_norm": 0.9488332812076795, "learning_rate": 9.885486377930492e-06, "loss": 0.69, "mean_token_accuracy": 0.7774538993835449, "num_tokens": 81770903.0, "step": 1248 }, { "epoch": 0.16508062384351044, "grad_norm": 0.9190988383465787, "learning_rate": 9.885020501150746e-06, "loss": 0.6379, "mean_token_accuracy": 0.7970333099365234, "num_tokens": 81836439.0, "step": 1249 }, { "epoch": 0.16521279407877346, "grad_norm": 0.9519558869659647, "learning_rate": 9.884553690891687e-06, "loss": 0.6604, "mean_token_accuracy": 0.7863050699234009, "num_tokens": 81901975.0, "step": 1250 }, { "epoch": 0.16534496431403647, "grad_norm": 0.912744701865262, "learning_rate": 9.884085947252688e-06, "loss": 0.6304, "mean_token_accuracy": 0.7949578762054443, "num_tokens": 81967511.0, "step": 1251 }, { "epoch": 0.1654771345492995, "grad_norm": 1.0527587168903114, "learning_rate": 9.883617270333326e-06, "loss": 0.6895, "mean_token_accuracy": 0.7782169580459595, "num_tokens": 82033047.0, "step": 1252 }, { "epoch": 0.1656093047845625, "grad_norm": 0.9696671126027039, "learning_rate": 9.88314766023337e-06, "loss": 0.7464, "mean_token_accuracy": 0.758942723274231, "num_tokens": 82098583.0, "step": 1253 }, { "epoch": 0.16574147501982553, "grad_norm": 0.9232313565697643, "learning_rate": 9.88267711705279e-06, "loss": 0.6241, "mean_token_accuracy": 0.7990171909332275, "num_tokens": 82164119.0, "step": 1254 }, { "epoch": 0.16587364525508855, "grad_norm": 0.9802605924957764, "learning_rate": 9.882205640891757e-06, "loss": 0.664, "mean_token_accuracy": 0.78831946849823, "num_tokens": 82229655.0, "step": 1255 }, { "epoch": 0.16600581549035157, "grad_norm": 0.9053292451273841, "learning_rate": 9.881733231850637e-06, "loss": 0.6041, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 82295191.0, "step": 1256 }, { "epoch": 0.16613798572561458, "grad_norm": 1.0035373494472681, "learning_rate": 9.881259890029996e-06, "loss": 0.6561, "mean_token_accuracy": 0.7924551367759705, "num_tokens": 82360727.0, "step": 1257 }, { "epoch": 0.1662701559608776, "grad_norm": 1.0034888422736006, "learning_rate": 9.880785615530597e-06, "loss": 0.6659, "mean_token_accuracy": 0.7878158688545227, "num_tokens": 82426263.0, "step": 1258 }, { "epoch": 0.16640232619614062, "grad_norm": 0.9677517373876307, "learning_rate": 9.880310408453405e-06, "loss": 0.705, "mean_token_accuracy": 0.7737608551979065, "num_tokens": 82491799.0, "step": 1259 }, { "epoch": 0.16653449643140364, "grad_norm": 0.9947021537524772, "learning_rate": 9.879834268899582e-06, "loss": 0.6498, "mean_token_accuracy": 0.792561948299408, "num_tokens": 82557335.0, "step": 1260 }, { "epoch": 0.16666666666666666, "grad_norm": 0.8263185837974127, "learning_rate": 9.879357196970488e-06, "loss": 0.6187, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 82622871.0, "step": 1261 }, { "epoch": 0.16679883690192968, "grad_norm": 0.9471726747300354, "learning_rate": 9.878879192767679e-06, "loss": 0.6819, "mean_token_accuracy": 0.7807654738426208, "num_tokens": 82688407.0, "step": 1262 }, { "epoch": 0.1669310071371927, "grad_norm": 1.004256583995644, "learning_rate": 9.878400256392916e-06, "loss": 0.6553, "mean_token_accuracy": 0.7911732196807861, "num_tokens": 82753943.0, "step": 1263 }, { "epoch": 0.1670631773724557, "grad_norm": 1.0124224653495666, "learning_rate": 9.877920387948153e-06, "loss": 0.6941, "mean_token_accuracy": 0.7801092863082886, "num_tokens": 82819479.0, "step": 1264 }, { "epoch": 0.16719534760771873, "grad_norm": 0.9384625722186272, "learning_rate": 9.877439587535542e-06, "loss": 0.6828, "mean_token_accuracy": 0.7808417677879333, "num_tokens": 82885015.0, "step": 1265 }, { "epoch": 0.16732751784298175, "grad_norm": 1.3284726082886855, "learning_rate": 9.876957855257438e-06, "loss": 0.6336, "mean_token_accuracy": 0.7945000529289246, "num_tokens": 82950551.0, "step": 1266 }, { "epoch": 0.16745968807824477, "grad_norm": 1.1432618966741628, "learning_rate": 9.876475191216391e-06, "loss": 0.6807, "mean_token_accuracy": 0.7848095297813416, "num_tokens": 83016087.0, "step": 1267 }, { "epoch": 0.16759185831350779, "grad_norm": 0.9579713169020756, "learning_rate": 9.875991595515152e-06, "loss": 0.6548, "mean_token_accuracy": 0.7901965379714966, "num_tokens": 83081623.0, "step": 1268 }, { "epoch": 0.1677240285487708, "grad_norm": 0.8840167448007369, "learning_rate": 9.875507068256664e-06, "loss": 0.6408, "mean_token_accuracy": 0.7960718870162964, "num_tokens": 83147159.0, "step": 1269 }, { "epoch": 0.16785619878403382, "grad_norm": 0.9785101933911587, "learning_rate": 9.875021609544076e-06, "loss": 0.6572, "mean_token_accuracy": 0.7894030213356018, "num_tokens": 83212695.0, "step": 1270 }, { "epoch": 0.16798836901929687, "grad_norm": 0.9360547107971098, "learning_rate": 9.874535219480732e-06, "loss": 0.6691, "mean_token_accuracy": 0.7847179770469666, "num_tokens": 83278231.0, "step": 1271 }, { "epoch": 0.1681205392545599, "grad_norm": 0.8591775776925492, "learning_rate": 9.874047898170172e-06, "loss": 0.6601, "mean_token_accuracy": 0.7881516218185425, "num_tokens": 83343767.0, "step": 1272 }, { "epoch": 0.1682527094898229, "grad_norm": 0.951815654511452, "learning_rate": 9.873559645716139e-06, "loss": 0.6638, "mean_token_accuracy": 0.7853894233703613, "num_tokens": 83409303.0, "step": 1273 }, { "epoch": 0.16838487972508592, "grad_norm": 0.8478201321180603, "learning_rate": 9.87307046222257e-06, "loss": 0.709, "mean_token_accuracy": 0.7753174304962158, "num_tokens": 83474839.0, "step": 1274 }, { "epoch": 0.16851704996034894, "grad_norm": 0.9871590583051391, "learning_rate": 9.872580347793603e-06, "loss": 0.6565, "mean_token_accuracy": 0.7881211042404175, "num_tokens": 83540375.0, "step": 1275 }, { "epoch": 0.16864922019561196, "grad_norm": 1.371591920062193, "learning_rate": 9.872089302533573e-06, "loss": 0.689, "mean_token_accuracy": 0.7803076505661011, "num_tokens": 83605911.0, "step": 1276 }, { "epoch": 0.16878139043087498, "grad_norm": 0.9825047881604158, "learning_rate": 9.871597326547014e-06, "loss": 0.6687, "mean_token_accuracy": 0.7852063179016113, "num_tokens": 83671447.0, "step": 1277 }, { "epoch": 0.168913560666138, "grad_norm": 1.082303823529906, "learning_rate": 9.871104419938656e-06, "loss": 0.6825, "mean_token_accuracy": 0.7801550626754761, "num_tokens": 83736983.0, "step": 1278 }, { "epoch": 0.16904573090140101, "grad_norm": 0.8640464693059535, "learning_rate": 9.870610582813429e-06, "loss": 0.5873, "mean_token_accuracy": 0.8119887709617615, "num_tokens": 83802519.0, "step": 1279 }, { "epoch": 0.16917790113666403, "grad_norm": 0.9624843124434719, "learning_rate": 9.87011581527646e-06, "loss": 0.6843, "mean_token_accuracy": 0.7796362042427063, "num_tokens": 83868055.0, "step": 1280 }, { "epoch": 0.16931007137192705, "grad_norm": 0.9300884183272228, "learning_rate": 9.869620117433076e-06, "loss": 0.6663, "mean_token_accuracy": 0.783710777759552, "num_tokens": 83933591.0, "step": 1281 }, { "epoch": 0.16944224160719007, "grad_norm": 0.936221849243793, "learning_rate": 9.869123489388801e-06, "loss": 0.6742, "mean_token_accuracy": 0.784763753414154, "num_tokens": 83999127.0, "step": 1282 }, { "epoch": 0.1695744118424531, "grad_norm": 0.9591014688156937, "learning_rate": 9.868625931249353e-06, "loss": 0.6892, "mean_token_accuracy": 0.7782169580459595, "num_tokens": 84064663.0, "step": 1283 }, { "epoch": 0.1697065820777161, "grad_norm": 0.9529369920554657, "learning_rate": 9.868127443120658e-06, "loss": 0.6812, "mean_token_accuracy": 0.7847484946250916, "num_tokens": 84130199.0, "step": 1284 }, { "epoch": 0.16983875231297912, "grad_norm": 0.9350907906406681, "learning_rate": 9.867628025108828e-06, "loss": 0.6657, "mean_token_accuracy": 0.7861219644546509, "num_tokens": 84195735.0, "step": 1285 }, { "epoch": 0.16997092254824214, "grad_norm": 0.9287949961913327, "learning_rate": 9.867127677320182e-06, "loss": 0.6644, "mean_token_accuracy": 0.7864271998405457, "num_tokens": 84261271.0, "step": 1286 }, { "epoch": 0.17010309278350516, "grad_norm": 0.9635621034614099, "learning_rate": 9.866626399861233e-06, "loss": 0.6639, "mean_token_accuracy": 0.7845653891563416, "num_tokens": 84326807.0, "step": 1287 }, { "epoch": 0.17023526301876818, "grad_norm": 0.9434038535120434, "learning_rate": 9.866124192838691e-06, "loss": 0.6489, "mean_token_accuracy": 0.7943016886711121, "num_tokens": 84392343.0, "step": 1288 }, { "epoch": 0.1703674332540312, "grad_norm": 1.1781475131187245, "learning_rate": 9.865621056359467e-06, "loss": 0.6941, "mean_token_accuracy": 0.7801092863082886, "num_tokens": 84457879.0, "step": 1289 }, { "epoch": 0.17049960348929422, "grad_norm": 0.8833479302523939, "learning_rate": 9.865116990530666e-06, "loss": 0.612, "mean_token_accuracy": 0.8043736815452576, "num_tokens": 84523415.0, "step": 1290 }, { "epoch": 0.17063177372455723, "grad_norm": 0.9871268129658752, "learning_rate": 9.864611995459597e-06, "loss": 0.6422, "mean_token_accuracy": 0.7953699231147766, "num_tokens": 84588951.0, "step": 1291 }, { "epoch": 0.17076394395982025, "grad_norm": 1.0910890040786683, "learning_rate": 9.864106071253759e-06, "loss": 0.6945, "mean_token_accuracy": 0.7795446515083313, "num_tokens": 84654487.0, "step": 1292 }, { "epoch": 0.17089611419508327, "grad_norm": 0.9139742840554016, "learning_rate": 9.863599218020857e-06, "loss": 0.6328, "mean_token_accuracy": 0.7972012162208557, "num_tokens": 84720023.0, "step": 1293 }, { "epoch": 0.1710282844303463, "grad_norm": 0.922275218360911, "learning_rate": 9.863091435868785e-06, "loss": 0.6824, "mean_token_accuracy": 0.7812690734863281, "num_tokens": 84785559.0, "step": 1294 }, { "epoch": 0.1711604546656093, "grad_norm": 0.8887213710499219, "learning_rate": 9.86258272490564e-06, "loss": 0.6184, "mean_token_accuracy": 0.8017488718032837, "num_tokens": 84851095.0, "step": 1295 }, { "epoch": 0.17129262490087233, "grad_norm": 1.039124507481392, "learning_rate": 9.86207308523972e-06, "loss": 0.7356, "mean_token_accuracy": 0.7633988261222839, "num_tokens": 84916631.0, "step": 1296 }, { "epoch": 0.17142479513613534, "grad_norm": 0.8900941581491117, "learning_rate": 9.861562516979512e-06, "loss": 0.6431, "mean_token_accuracy": 0.7921041250228882, "num_tokens": 84982167.0, "step": 1297 }, { "epoch": 0.17155696537139836, "grad_norm": 0.8975754975217649, "learning_rate": 9.861051020233707e-06, "loss": 0.6413, "mean_token_accuracy": 0.7949120998382568, "num_tokens": 85047703.0, "step": 1298 }, { "epoch": 0.17168913560666138, "grad_norm": 1.0799519124563421, "learning_rate": 9.86053859511119e-06, "loss": 0.697, "mean_token_accuracy": 0.7729520201683044, "num_tokens": 85113239.0, "step": 1299 }, { "epoch": 0.1718213058419244, "grad_norm": 0.9641748201121061, "learning_rate": 9.86002524172105e-06, "loss": 0.6716, "mean_token_accuracy": 0.78789222240448, "num_tokens": 85178775.0, "step": 1300 }, { "epoch": 0.17195347607718742, "grad_norm": 0.8802971700761781, "learning_rate": 9.859510960172566e-06, "loss": 0.6319, "mean_token_accuracy": 0.7978421449661255, "num_tokens": 85244311.0, "step": 1301 }, { "epoch": 0.17208564631245044, "grad_norm": 0.8819053126711329, "learning_rate": 9.85899575057522e-06, "loss": 0.6862, "mean_token_accuracy": 0.7787815928459167, "num_tokens": 85309847.0, "step": 1302 }, { "epoch": 0.17221781654771345, "grad_norm": 0.9131809927097894, "learning_rate": 9.858479613038688e-06, "loss": 0.6142, "mean_token_accuracy": 0.8017946481704712, "num_tokens": 85375383.0, "step": 1303 }, { "epoch": 0.17234998678297647, "grad_norm": 1.0458234254847547, "learning_rate": 9.857962547672844e-06, "loss": 0.6661, "mean_token_accuracy": 0.783710777759552, "num_tokens": 85440919.0, "step": 1304 }, { "epoch": 0.1724821570182395, "grad_norm": 0.8982556181722628, "learning_rate": 9.85744455458776e-06, "loss": 0.6296, "mean_token_accuracy": 0.798818826675415, "num_tokens": 85506455.0, "step": 1305 }, { "epoch": 0.1726143272535025, "grad_norm": 1.0140451092015126, "learning_rate": 9.856925633893708e-06, "loss": 0.6362, "mean_token_accuracy": 0.7952478528022766, "num_tokens": 85571991.0, "step": 1306 }, { "epoch": 0.17274649748876553, "grad_norm": 0.9234142015921987, "learning_rate": 9.856405785701156e-06, "loss": 0.6506, "mean_token_accuracy": 0.7900592088699341, "num_tokens": 85637527.0, "step": 1307 }, { "epoch": 0.17287866772402855, "grad_norm": 0.9162896603344114, "learning_rate": 9.855885010120767e-06, "loss": 0.6635, "mean_token_accuracy": 0.7861982583999634, "num_tokens": 85703063.0, "step": 1308 }, { "epoch": 0.17301083795929156, "grad_norm": 0.9511791432064948, "learning_rate": 9.855363307263403e-06, "loss": 0.6441, "mean_token_accuracy": 0.7930960655212402, "num_tokens": 85768599.0, "step": 1309 }, { "epoch": 0.17314300819455458, "grad_norm": 0.9096258060742883, "learning_rate": 9.854840677240125e-06, "loss": 0.6246, "mean_token_accuracy": 0.7997497320175171, "num_tokens": 85834135.0, "step": 1310 }, { "epoch": 0.1732751784298176, "grad_norm": 0.8609116190299754, "learning_rate": 9.854317120162187e-06, "loss": 0.6489, "mean_token_accuracy": 0.7914326786994934, "num_tokens": 85899671.0, "step": 1311 }, { "epoch": 0.17340734866508062, "grad_norm": 0.9846639737432349, "learning_rate": 9.85379263614105e-06, "loss": 0.6394, "mean_token_accuracy": 0.7927908897399902, "num_tokens": 85965207.0, "step": 1312 }, { "epoch": 0.17353951890034364, "grad_norm": 0.9555170099784397, "learning_rate": 9.853267225288359e-06, "loss": 0.6667, "mean_token_accuracy": 0.7811470031738281, "num_tokens": 86030743.0, "step": 1313 }, { "epoch": 0.17367168913560665, "grad_norm": 1.0039293584716256, "learning_rate": 9.852740887715964e-06, "loss": 0.6566, "mean_token_accuracy": 0.7914631962776184, "num_tokens": 86096279.0, "step": 1314 }, { "epoch": 0.17380385937086967, "grad_norm": 0.9894639884676422, "learning_rate": 9.852213623535914e-06, "loss": 0.6291, "mean_token_accuracy": 0.7966823577880859, "num_tokens": 86161815.0, "step": 1315 }, { "epoch": 0.1739360296061327, "grad_norm": 0.9878855614082256, "learning_rate": 9.851685432860451e-06, "loss": 0.6733, "mean_token_accuracy": 0.7830851078033447, "num_tokens": 86227351.0, "step": 1316 }, { "epoch": 0.1740681998413957, "grad_norm": 1.0139887222774475, "learning_rate": 9.851156315802016e-06, "loss": 0.7212, "mean_token_accuracy": 0.7684348821640015, "num_tokens": 86292887.0, "step": 1317 }, { "epoch": 0.17420037007665873, "grad_norm": 1.0005285771997272, "learning_rate": 9.850626272473248e-06, "loss": 0.6766, "mean_token_accuracy": 0.7822915315628052, "num_tokens": 86358423.0, "step": 1318 }, { "epoch": 0.17433254031192175, "grad_norm": 0.9677574596065749, "learning_rate": 9.85009530298698e-06, "loss": 0.6507, "mean_token_accuracy": 0.7952173352241516, "num_tokens": 86423959.0, "step": 1319 }, { "epoch": 0.17446471054718476, "grad_norm": 0.8629414619061192, "learning_rate": 9.849563407456245e-06, "loss": 0.6145, "mean_token_accuracy": 0.7998565435409546, "num_tokens": 86489495.0, "step": 1320 }, { "epoch": 0.17459688078244778, "grad_norm": 1.0061627372421336, "learning_rate": 9.849030585994275e-06, "loss": 0.6865, "mean_token_accuracy": 0.779163122177124, "num_tokens": 86555031.0, "step": 1321 }, { "epoch": 0.1747290510177108, "grad_norm": 0.9334561053089653, "learning_rate": 9.848496838714492e-06, "loss": 0.594, "mean_token_accuracy": 0.8090434670448303, "num_tokens": 86620567.0, "step": 1322 }, { "epoch": 0.17486122125297382, "grad_norm": 0.9075108913837475, "learning_rate": 9.847962165730525e-06, "loss": 0.6667, "mean_token_accuracy": 0.7876175045967102, "num_tokens": 86686103.0, "step": 1323 }, { "epoch": 0.17499339148823684, "grad_norm": 0.8844030824087846, "learning_rate": 9.84742656715619e-06, "loss": 0.6228, "mean_token_accuracy": 0.8004211783409119, "num_tokens": 86751639.0, "step": 1324 }, { "epoch": 0.17512556172349986, "grad_norm": 0.8235672848502303, "learning_rate": 9.846890043105507e-06, "loss": 0.5907, "mean_token_accuracy": 0.8149645924568176, "num_tokens": 86817175.0, "step": 1325 }, { "epoch": 0.17525773195876287, "grad_norm": 0.9567712579503406, "learning_rate": 9.846352593692689e-06, "loss": 0.6598, "mean_token_accuracy": 0.7865034937858582, "num_tokens": 86882711.0, "step": 1326 }, { "epoch": 0.1753899021940259, "grad_norm": 0.8713705018400273, "learning_rate": 9.84581421903215e-06, "loss": 0.6418, "mean_token_accuracy": 0.7944847941398621, "num_tokens": 86948247.0, "step": 1327 }, { "epoch": 0.1755220724292889, "grad_norm": 0.9926102519186335, "learning_rate": 9.845274919238499e-06, "loss": 0.659, "mean_token_accuracy": 0.7855725884437561, "num_tokens": 87013783.0, "step": 1328 }, { "epoch": 0.17565424266455193, "grad_norm": 0.926419017593105, "learning_rate": 9.844734694426538e-06, "loss": 0.5962, "mean_token_accuracy": 0.8098064661026001, "num_tokens": 87079319.0, "step": 1329 }, { "epoch": 0.17578641289981498, "grad_norm": 0.9818115514546042, "learning_rate": 9.844193544711273e-06, "loss": 0.7328, "mean_token_accuracy": 0.7673361301422119, "num_tokens": 87144855.0, "step": 1330 }, { "epoch": 0.175918583135078, "grad_norm": 0.8835296486756568, "learning_rate": 9.843651470207906e-06, "loss": 0.6247, "mean_token_accuracy": 0.7975369095802307, "num_tokens": 87210391.0, "step": 1331 }, { "epoch": 0.176050753370341, "grad_norm": 1.0100116861413588, "learning_rate": 9.843108471031826e-06, "loss": 0.6598, "mean_token_accuracy": 0.7869002819061279, "num_tokens": 87275927.0, "step": 1332 }, { "epoch": 0.17618292360560403, "grad_norm": 0.9293445742379592, "learning_rate": 9.842564547298629e-06, "loss": 0.6704, "mean_token_accuracy": 0.7824746966362, "num_tokens": 87341463.0, "step": 1333 }, { "epoch": 0.17631509384086705, "grad_norm": 0.9240201011586546, "learning_rate": 9.84201969912411e-06, "loss": 0.6203, "mean_token_accuracy": 0.8010609149932861, "num_tokens": 87406697.0, "step": 1334 }, { "epoch": 0.17644726407613007, "grad_norm": 0.939073430054967, "learning_rate": 9.841473926624249e-06, "loss": 0.6284, "mean_token_accuracy": 0.7978421449661255, "num_tokens": 87472233.0, "step": 1335 }, { "epoch": 0.17657943431139309, "grad_norm": 0.9660688716129535, "learning_rate": 9.840927229915235e-06, "loss": 0.6383, "mean_token_accuracy": 0.7946068644523621, "num_tokens": 87537769.0, "step": 1336 }, { "epoch": 0.1767116045466561, "grad_norm": 0.8994283263409526, "learning_rate": 9.840379609113445e-06, "loss": 0.6466, "mean_token_accuracy": 0.793584406375885, "num_tokens": 87603305.0, "step": 1337 }, { "epoch": 0.17684377478191912, "grad_norm": 0.8804278724223576, "learning_rate": 9.839831064335454e-06, "loss": 0.6136, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 87668841.0, "step": 1338 }, { "epoch": 0.17697594501718214, "grad_norm": 0.9604722405682151, "learning_rate": 9.839281595698042e-06, "loss": 0.6661, "mean_token_accuracy": 0.7867018580436707, "num_tokens": 87734377.0, "step": 1339 }, { "epoch": 0.17710811525244516, "grad_norm": 0.8266092081051303, "learning_rate": 9.838731203318175e-06, "loss": 0.6071, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 87799913.0, "step": 1340 }, { "epoch": 0.17724028548770818, "grad_norm": 1.0629284230087956, "learning_rate": 9.838179887313019e-06, "loss": 0.7084, "mean_token_accuracy": 0.7723873853683472, "num_tokens": 87865449.0, "step": 1341 }, { "epoch": 0.1773724557229712, "grad_norm": 0.9466943501105813, "learning_rate": 9.837627647799942e-06, "loss": 0.655, "mean_token_accuracy": 0.7917684316635132, "num_tokens": 87930985.0, "step": 1342 }, { "epoch": 0.1775046259582342, "grad_norm": 0.8772425343796448, "learning_rate": 9.837074484896501e-06, "loss": 0.6532, "mean_token_accuracy": 0.7895250916481018, "num_tokens": 87996521.0, "step": 1343 }, { "epoch": 0.17763679619349723, "grad_norm": 0.9862203104513761, "learning_rate": 9.836520398720456e-06, "loss": 0.6797, "mean_token_accuracy": 0.7814980149269104, "num_tokens": 88062057.0, "step": 1344 }, { "epoch": 0.17776896642876025, "grad_norm": 0.9358378369182915, "learning_rate": 9.835965389389756e-06, "loss": 0.689, "mean_token_accuracy": 0.7783085107803345, "num_tokens": 88127593.0, "step": 1345 }, { "epoch": 0.17790113666402327, "grad_norm": 0.9674383192109677, "learning_rate": 9.835409457022553e-06, "loss": 0.6792, "mean_token_accuracy": 0.7816963791847229, "num_tokens": 88193129.0, "step": 1346 }, { "epoch": 0.1780333068992863, "grad_norm": 0.9929386940809185, "learning_rate": 9.834852601737193e-06, "loss": 0.6884, "mean_token_accuracy": 0.7787358164787292, "num_tokens": 88258665.0, "step": 1347 }, { "epoch": 0.1781654771345493, "grad_norm": 0.935370976154982, "learning_rate": 9.83429482365222e-06, "loss": 0.6914, "mean_token_accuracy": 0.7783695459365845, "num_tokens": 88324201.0, "step": 1348 }, { "epoch": 0.17829764736981232, "grad_norm": 0.903262308820978, "learning_rate": 9.833736122886373e-06, "loss": 0.6709, "mean_token_accuracy": 0.7843517065048218, "num_tokens": 88389737.0, "step": 1349 }, { "epoch": 0.17842981760507534, "grad_norm": 0.8276959660944334, "learning_rate": 9.833176499558586e-06, "loss": 0.6482, "mean_token_accuracy": 0.7897539734840393, "num_tokens": 88455273.0, "step": 1350 }, { "epoch": 0.17856198784033836, "grad_norm": 0.9252480094605522, "learning_rate": 9.832615953787993e-06, "loss": 0.6343, "mean_token_accuracy": 0.7980557680130005, "num_tokens": 88520809.0, "step": 1351 }, { "epoch": 0.17869415807560138, "grad_norm": 0.9085146314555204, "learning_rate": 9.832054485693923e-06, "loss": 0.6545, "mean_token_accuracy": 0.788823127746582, "num_tokens": 88586345.0, "step": 1352 }, { "epoch": 0.1788263283108644, "grad_norm": 0.8637588697148874, "learning_rate": 9.831492095395899e-06, "loss": 0.6075, "mean_token_accuracy": 0.8034886121749878, "num_tokens": 88651881.0, "step": 1353 }, { "epoch": 0.17895849854612741, "grad_norm": 0.8677726142284643, "learning_rate": 9.830928783013642e-06, "loss": 0.634, "mean_token_accuracy": 0.7945763468742371, "num_tokens": 88717417.0, "step": 1354 }, { "epoch": 0.17909066878139043, "grad_norm": 0.9395627962614365, "learning_rate": 9.830364548667074e-06, "loss": 0.6792, "mean_token_accuracy": 0.7808875441551208, "num_tokens": 88782953.0, "step": 1355 }, { "epoch": 0.17922283901665345, "grad_norm": 0.9128859705861864, "learning_rate": 9.829799392476302e-06, "loss": 0.6604, "mean_token_accuracy": 0.789250373840332, "num_tokens": 88848489.0, "step": 1356 }, { "epoch": 0.17935500925191647, "grad_norm": 1.1141714370868738, "learning_rate": 9.829233314561637e-06, "loss": 0.704, "mean_token_accuracy": 0.768450140953064, "num_tokens": 88914025.0, "step": 1357 }, { "epoch": 0.1794871794871795, "grad_norm": 0.9016242371452367, "learning_rate": 9.828666315043591e-06, "loss": 0.6522, "mean_token_accuracy": 0.7896013855934143, "num_tokens": 88979561.0, "step": 1358 }, { "epoch": 0.1796193497224425, "grad_norm": 0.9924985955520164, "learning_rate": 9.82809839404286e-06, "loss": 0.7145, "mean_token_accuracy": 0.7674429416656494, "num_tokens": 89045097.0, "step": 1359 }, { "epoch": 0.17975151995770552, "grad_norm": 0.886043743566956, "learning_rate": 9.827529551680348e-06, "loss": 0.6108, "mean_token_accuracy": 0.8060981631278992, "num_tokens": 89110633.0, "step": 1360 }, { "epoch": 0.17988369019296854, "grad_norm": 0.9877129444719637, "learning_rate": 9.826959788077146e-06, "loss": 0.711, "mean_token_accuracy": 0.7722042202949524, "num_tokens": 89176169.0, "step": 1361 }, { "epoch": 0.18001586042823156, "grad_norm": 0.9631870531024037, "learning_rate": 9.826389103354543e-06, "loss": 0.6229, "mean_token_accuracy": 0.797704815864563, "num_tokens": 89241705.0, "step": 1362 }, { "epoch": 0.18014803066349458, "grad_norm": 1.0207167576372949, "learning_rate": 9.82581749763403e-06, "loss": 0.5982, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 89307241.0, "step": 1363 }, { "epoch": 0.1802802008987576, "grad_norm": 0.9089226335590224, "learning_rate": 9.825244971037289e-06, "loss": 0.6034, "mean_token_accuracy": 0.8057013750076294, "num_tokens": 89372777.0, "step": 1364 }, { "epoch": 0.18041237113402062, "grad_norm": 0.9354499220621271, "learning_rate": 9.824671523686199e-06, "loss": 0.6623, "mean_token_accuracy": 0.7904407382011414, "num_tokens": 89438313.0, "step": 1365 }, { "epoch": 0.18054454136928363, "grad_norm": 0.9315649805561835, "learning_rate": 9.824097155702831e-06, "loss": 0.5738, "mean_token_accuracy": 0.8167501091957092, "num_tokens": 89503849.0, "step": 1366 }, { "epoch": 0.18067671160454665, "grad_norm": 0.9662723856108263, "learning_rate": 9.823521867209461e-06, "loss": 0.6464, "mean_token_accuracy": 0.7928519248962402, "num_tokens": 89569385.0, "step": 1367 }, { "epoch": 0.18080888183980967, "grad_norm": 0.8476969402232826, "learning_rate": 9.822945658328555e-06, "loss": 0.6017, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 89634921.0, "step": 1368 }, { "epoch": 0.1809410520750727, "grad_norm": 0.9778234159891745, "learning_rate": 9.822368529182773e-06, "loss": 0.7147, "mean_token_accuracy": 0.7724941968917847, "num_tokens": 89700457.0, "step": 1369 }, { "epoch": 0.1810732223103357, "grad_norm": 0.9866251900466837, "learning_rate": 9.821790479894979e-06, "loss": 0.6481, "mean_token_accuracy": 0.7896166443824768, "num_tokens": 89765993.0, "step": 1370 }, { "epoch": 0.18120539254559873, "grad_norm": 0.9036453383514018, "learning_rate": 9.821211510588222e-06, "loss": 0.6281, "mean_token_accuracy": 0.7995055317878723, "num_tokens": 89831529.0, "step": 1371 }, { "epoch": 0.18133756278086174, "grad_norm": 0.9188090528381188, "learning_rate": 9.820631621385754e-06, "loss": 0.6523, "mean_token_accuracy": 0.792684018611908, "num_tokens": 89897065.0, "step": 1372 }, { "epoch": 0.18146973301612476, "grad_norm": 0.9643278415685957, "learning_rate": 9.820050812411024e-06, "loss": 0.6866, "mean_token_accuracy": 0.7774691581726074, "num_tokens": 89962601.0, "step": 1373 }, { "epoch": 0.18160190325138778, "grad_norm": 0.9397324879627454, "learning_rate": 9.819469083787672e-06, "loss": 0.6849, "mean_token_accuracy": 0.7816811203956604, "num_tokens": 90028137.0, "step": 1374 }, { "epoch": 0.1817340734866508, "grad_norm": 0.9448417356721739, "learning_rate": 9.818886435639535e-06, "loss": 0.6922, "mean_token_accuracy": 0.7802466154098511, "num_tokens": 90093673.0, "step": 1375 }, { "epoch": 0.18186624372191382, "grad_norm": 0.9238469271686846, "learning_rate": 9.818302868090646e-06, "loss": 0.6525, "mean_token_accuracy": 0.788945198059082, "num_tokens": 90159209.0, "step": 1376 }, { "epoch": 0.18199841395717684, "grad_norm": 1.0095325435568423, "learning_rate": 9.81771838126524e-06, "loss": 0.7288, "mean_token_accuracy": 0.7677175998687744, "num_tokens": 90224745.0, "step": 1377 }, { "epoch": 0.18213058419243985, "grad_norm": 1.0430123683002919, "learning_rate": 9.817132975287736e-06, "loss": 0.7015, "mean_token_accuracy": 0.7729215025901794, "num_tokens": 90290281.0, "step": 1378 }, { "epoch": 0.18226275442770287, "grad_norm": 0.9964682479244527, "learning_rate": 9.816546650282756e-06, "loss": 0.6621, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 90355817.0, "step": 1379 }, { "epoch": 0.1823949246629659, "grad_norm": 0.9285749977314866, "learning_rate": 9.815959406375119e-06, "loss": 0.6587, "mean_token_accuracy": 0.7896013855934143, "num_tokens": 90421353.0, "step": 1380 }, { "epoch": 0.1825270948982289, "grad_norm": 0.9451589339238917, "learning_rate": 9.815371243689832e-06, "loss": 0.661, "mean_token_accuracy": 0.7846264243125916, "num_tokens": 90486889.0, "step": 1381 }, { "epoch": 0.18265926513349193, "grad_norm": 0.9123405242577993, "learning_rate": 9.81478216235211e-06, "loss": 0.5907, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 90552425.0, "step": 1382 }, { "epoch": 0.18279143536875495, "grad_norm": 0.8932188787876817, "learning_rate": 9.81419216248735e-06, "loss": 0.5979, "mean_token_accuracy": 0.8051825165748596, "num_tokens": 90617961.0, "step": 1383 }, { "epoch": 0.18292360560401796, "grad_norm": 1.005368045538355, "learning_rate": 9.813601244221152e-06, "loss": 0.6551, "mean_token_accuracy": 0.7885026335716248, "num_tokens": 90683497.0, "step": 1384 }, { "epoch": 0.18305577583928098, "grad_norm": 0.9815458475585331, "learning_rate": 9.813009407679311e-06, "loss": 0.6717, "mean_token_accuracy": 0.7814064025878906, "num_tokens": 90749033.0, "step": 1385 }, { "epoch": 0.183187946074544, "grad_norm": 0.8574053625912543, "learning_rate": 9.812416652987816e-06, "loss": 0.5817, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 90814569.0, "step": 1386 }, { "epoch": 0.18332011630980702, "grad_norm": 0.8660295876199258, "learning_rate": 9.811822980272851e-06, "loss": 0.6132, "mean_token_accuracy": 0.8019777536392212, "num_tokens": 90880105.0, "step": 1387 }, { "epoch": 0.18345228654507004, "grad_norm": 1.0010940061197475, "learning_rate": 9.8112283896608e-06, "loss": 0.6777, "mean_token_accuracy": 0.7817268967628479, "num_tokens": 90945641.0, "step": 1388 }, { "epoch": 0.18358445678033306, "grad_norm": 0.8647786128423647, "learning_rate": 9.810632881278237e-06, "loss": 0.6455, "mean_token_accuracy": 0.7896776795387268, "num_tokens": 91011177.0, "step": 1389 }, { "epoch": 0.1837166270155961, "grad_norm": 0.9561463072009646, "learning_rate": 9.810036455251933e-06, "loss": 0.6752, "mean_token_accuracy": 0.7834055423736572, "num_tokens": 91076713.0, "step": 1390 }, { "epoch": 0.18384879725085912, "grad_norm": 0.9533836444395889, "learning_rate": 9.809439111708853e-06, "loss": 0.6377, "mean_token_accuracy": 0.7943016886711121, "num_tokens": 91142249.0, "step": 1391 }, { "epoch": 0.18398096748612214, "grad_norm": 0.9809188259828288, "learning_rate": 9.808840850776163e-06, "loss": 0.6844, "mean_token_accuracy": 0.7774067521095276, "num_tokens": 91204249.0, "step": 1392 }, { "epoch": 0.18411313772138516, "grad_norm": 0.9066952078344019, "learning_rate": 9.808241672581215e-06, "loss": 0.6503, "mean_token_accuracy": 0.7908833026885986, "num_tokens": 91269785.0, "step": 1393 }, { "epoch": 0.18424530795664817, "grad_norm": 0.9499554013234547, "learning_rate": 9.807641577251566e-06, "loss": 0.6932, "mean_token_accuracy": 0.7779727578163147, "num_tokens": 91335321.0, "step": 1394 }, { "epoch": 0.1843774781919112, "grad_norm": 0.9418567508264616, "learning_rate": 9.807040564914964e-06, "loss": 0.701, "mean_token_accuracy": 0.7759736180305481, "num_tokens": 91400857.0, "step": 1395 }, { "epoch": 0.1845096484271742, "grad_norm": 0.8202124462074374, "learning_rate": 9.806438635699347e-06, "loss": 0.577, "mean_token_accuracy": 0.8122940063476562, "num_tokens": 91466393.0, "step": 1396 }, { "epoch": 0.18464181866243723, "grad_norm": 0.8579670628193289, "learning_rate": 9.80583578973286e-06, "loss": 0.6026, "mean_token_accuracy": 0.8045415878295898, "num_tokens": 91531929.0, "step": 1397 }, { "epoch": 0.18477398889770025, "grad_norm": 0.8461055629616895, "learning_rate": 9.805232027143831e-06, "loss": 0.5647, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 91597465.0, "step": 1398 }, { "epoch": 0.18490615913296327, "grad_norm": 0.8638048880354257, "learning_rate": 9.80462734806079e-06, "loss": 0.5976, "mean_token_accuracy": 0.8050757050514221, "num_tokens": 91663001.0, "step": 1399 }, { "epoch": 0.18503832936822628, "grad_norm": 0.8859410271571465, "learning_rate": 9.804021752612462e-06, "loss": 0.6822, "mean_token_accuracy": 0.7850384712219238, "num_tokens": 91728537.0, "step": 1400 }, { "epoch": 0.1851704996034893, "grad_norm": 0.928596216499485, "learning_rate": 9.803415240927762e-06, "loss": 0.6222, "mean_token_accuracy": 0.7990171909332275, "num_tokens": 91794073.0, "step": 1401 }, { "epoch": 0.18530266983875232, "grad_norm": 1.0444295160699733, "learning_rate": 9.802807813135808e-06, "loss": 0.6293, "mean_token_accuracy": 0.7967891693115234, "num_tokens": 91859609.0, "step": 1402 }, { "epoch": 0.18543484007401534, "grad_norm": 0.92500748031671, "learning_rate": 9.802199469365904e-06, "loss": 0.6746, "mean_token_accuracy": 0.7849010825157166, "num_tokens": 91925145.0, "step": 1403 }, { "epoch": 0.18556701030927836, "grad_norm": 0.913272031525666, "learning_rate": 9.801590209747559e-06, "loss": 0.6025, "mean_token_accuracy": 0.8071663975715637, "num_tokens": 91990681.0, "step": 1404 }, { "epoch": 0.18569918054454138, "grad_norm": 1.0776290291765847, "learning_rate": 9.80098003441047e-06, "loss": 0.6279, "mean_token_accuracy": 0.7946832180023193, "num_tokens": 92056217.0, "step": 1405 }, { "epoch": 0.1858313507798044, "grad_norm": 0.9408250953500615, "learning_rate": 9.800368943484527e-06, "loss": 0.65, "mean_token_accuracy": 0.7916310429573059, "num_tokens": 92121753.0, "step": 1406 }, { "epoch": 0.1859635210150674, "grad_norm": 0.9681150740296479, "learning_rate": 9.799756937099823e-06, "loss": 0.6264, "mean_token_accuracy": 0.7979031801223755, "num_tokens": 92187289.0, "step": 1407 }, { "epoch": 0.18609569125033043, "grad_norm": 0.9444163750366893, "learning_rate": 9.79914401538664e-06, "loss": 0.6276, "mean_token_accuracy": 0.7978116273880005, "num_tokens": 92252825.0, "step": 1408 }, { "epoch": 0.18622786148559345, "grad_norm": 0.9432299196638532, "learning_rate": 9.798530178475453e-06, "loss": 0.6274, "mean_token_accuracy": 0.7963008284568787, "num_tokens": 92318361.0, "step": 1409 }, { "epoch": 0.18636003172085647, "grad_norm": 0.8945422793706773, "learning_rate": 9.79791542649694e-06, "loss": 0.6207, "mean_token_accuracy": 0.803122341632843, "num_tokens": 92383897.0, "step": 1410 }, { "epoch": 0.18649220195611949, "grad_norm": 0.8675879435017366, "learning_rate": 9.797299759581967e-06, "loss": 0.6345, "mean_token_accuracy": 0.7941795587539673, "num_tokens": 92449433.0, "step": 1411 }, { "epoch": 0.1866243721913825, "grad_norm": 0.8964170617619291, "learning_rate": 9.796683177861595e-06, "loss": 0.672, "mean_token_accuracy": 0.7840617895126343, "num_tokens": 92514969.0, "step": 1412 }, { "epoch": 0.18675654242664552, "grad_norm": 0.9259256489425671, "learning_rate": 9.796065681467083e-06, "loss": 0.6741, "mean_token_accuracy": 0.7854810357093811, "num_tokens": 92580505.0, "step": 1413 }, { "epoch": 0.18688871266190854, "grad_norm": 0.925153924410743, "learning_rate": 9.795447270529883e-06, "loss": 0.6649, "mean_token_accuracy": 0.7862592935562134, "num_tokens": 92646041.0, "step": 1414 }, { "epoch": 0.18702088289717156, "grad_norm": 1.0706615747803976, "learning_rate": 9.794827945181642e-06, "loss": 0.7271, "mean_token_accuracy": 0.7663747072219849, "num_tokens": 92711577.0, "step": 1415 }, { "epoch": 0.18715305313243458, "grad_norm": 0.9028351513094618, "learning_rate": 9.7942077055542e-06, "loss": 0.6704, "mean_token_accuracy": 0.7845043540000916, "num_tokens": 92777113.0, "step": 1416 }, { "epoch": 0.1872852233676976, "grad_norm": 0.8959123905701855, "learning_rate": 9.793586551779595e-06, "loss": 0.6597, "mean_token_accuracy": 0.7885941863059998, "num_tokens": 92842649.0, "step": 1417 }, { "epoch": 0.1874173936029606, "grad_norm": 0.8180733587396213, "learning_rate": 9.792964483990058e-06, "loss": 0.6017, "mean_token_accuracy": 0.8088756203651428, "num_tokens": 92908185.0, "step": 1418 }, { "epoch": 0.18754956383822363, "grad_norm": 0.956409794929243, "learning_rate": 9.792341502318012e-06, "loss": 0.633, "mean_token_accuracy": 0.7939811944961548, "num_tokens": 92973721.0, "step": 1419 }, { "epoch": 0.18768173407348665, "grad_norm": 0.8473071094740312, "learning_rate": 9.791717606896076e-06, "loss": 0.6321, "mean_token_accuracy": 0.7959955930709839, "num_tokens": 93039257.0, "step": 1420 }, { "epoch": 0.18781390430874967, "grad_norm": 0.8899920129706828, "learning_rate": 9.791092797857069e-06, "loss": 0.6733, "mean_token_accuracy": 0.784855306148529, "num_tokens": 93104793.0, "step": 1421 }, { "epoch": 0.1879460745440127, "grad_norm": 0.9329375208394649, "learning_rate": 9.790467075333994e-06, "loss": 0.6827, "mean_token_accuracy": 0.7790868282318115, "num_tokens": 93170329.0, "step": 1422 }, { "epoch": 0.1880782447792757, "grad_norm": 0.9062085012552816, "learning_rate": 9.789840439460058e-06, "loss": 0.6817, "mean_token_accuracy": 0.7810554504394531, "num_tokens": 93235865.0, "step": 1423 }, { "epoch": 0.18821041501453872, "grad_norm": 0.9981082310059484, "learning_rate": 9.78921289036866e-06, "loss": 0.6932, "mean_token_accuracy": 0.7785984873771667, "num_tokens": 93301401.0, "step": 1424 }, { "epoch": 0.18834258524980174, "grad_norm": 0.9008275043170475, "learning_rate": 9.788584428193389e-06, "loss": 0.7008, "mean_token_accuracy": 0.7789494395256042, "num_tokens": 93366937.0, "step": 1425 }, { "epoch": 0.18847475548506476, "grad_norm": 1.0162367330982403, "learning_rate": 9.78795505306803e-06, "loss": 0.6928, "mean_token_accuracy": 0.7766298651695251, "num_tokens": 93432473.0, "step": 1426 }, { "epoch": 0.18860692572032778, "grad_norm": 0.9345319619774259, "learning_rate": 9.78732476512657e-06, "loss": 0.7389, "mean_token_accuracy": 0.7631699442863464, "num_tokens": 93498009.0, "step": 1427 }, { "epoch": 0.1887390959555908, "grad_norm": 0.9657768084713326, "learning_rate": 9.786693564503176e-06, "loss": 0.6795, "mean_token_accuracy": 0.783527672290802, "num_tokens": 93563545.0, "step": 1428 }, { "epoch": 0.18887126619085381, "grad_norm": 1.0319355866880708, "learning_rate": 9.786061451332227e-06, "loss": 0.6895, "mean_token_accuracy": 0.779346227645874, "num_tokens": 93629081.0, "step": 1429 }, { "epoch": 0.18900343642611683, "grad_norm": 0.9790024573740668, "learning_rate": 9.785428425748278e-06, "loss": 0.6551, "mean_token_accuracy": 0.792531430721283, "num_tokens": 93694617.0, "step": 1430 }, { "epoch": 0.18913560666137985, "grad_norm": 0.9572590257865498, "learning_rate": 9.78479448788609e-06, "loss": 0.682, "mean_token_accuracy": 0.7796056866645813, "num_tokens": 93760153.0, "step": 1431 }, { "epoch": 0.18926777689664287, "grad_norm": 0.9405175116054256, "learning_rate": 9.784159637880617e-06, "loss": 0.674, "mean_token_accuracy": 0.78290194272995, "num_tokens": 93825689.0, "step": 1432 }, { "epoch": 0.1893999471319059, "grad_norm": 1.1243236051605074, "learning_rate": 9.783523875867004e-06, "loss": 0.6525, "mean_token_accuracy": 0.7895523905754089, "num_tokens": 93885213.0, "step": 1433 }, { "epoch": 0.1895321173671689, "grad_norm": 0.9252553515303926, "learning_rate": 9.782887201980592e-06, "loss": 0.6531, "mean_token_accuracy": 0.7916768193244934, "num_tokens": 93950749.0, "step": 1434 }, { "epoch": 0.18966428760243192, "grad_norm": 1.0281400867670165, "learning_rate": 9.782249616356914e-06, "loss": 0.6685, "mean_token_accuracy": 0.7847790122032166, "num_tokens": 94016285.0, "step": 1435 }, { "epoch": 0.18979645783769494, "grad_norm": 1.0169624669921602, "learning_rate": 9.781611119131699e-06, "loss": 0.6867, "mean_token_accuracy": 0.7804449796676636, "num_tokens": 94081821.0, "step": 1436 }, { "epoch": 0.18992862807295796, "grad_norm": 0.9377679796642454, "learning_rate": 9.780971710440872e-06, "loss": 0.6743, "mean_token_accuracy": 0.7822915315628052, "num_tokens": 94147357.0, "step": 1437 }, { "epoch": 0.19006079830822098, "grad_norm": 0.9982180451384826, "learning_rate": 9.780331390420548e-06, "loss": 0.7017, "mean_token_accuracy": 0.7770571112632751, "num_tokens": 94212893.0, "step": 1438 }, { "epoch": 0.190192968543484, "grad_norm": 0.9513267097312461, "learning_rate": 9.779690159207035e-06, "loss": 0.6455, "mean_token_accuracy": 0.7913411259651184, "num_tokens": 94278429.0, "step": 1439 }, { "epoch": 0.19032513877874702, "grad_norm": 0.9957413048070929, "learning_rate": 9.779048016936844e-06, "loss": 0.6818, "mean_token_accuracy": 0.7815132737159729, "num_tokens": 94343965.0, "step": 1440 }, { "epoch": 0.19045730901401003, "grad_norm": 0.9923671861432982, "learning_rate": 9.77840496374667e-06, "loss": 0.6558, "mean_token_accuracy": 0.7883500456809998, "num_tokens": 94409501.0, "step": 1441 }, { "epoch": 0.19058947924927305, "grad_norm": 0.9045682851222839, "learning_rate": 9.777760999773403e-06, "loss": 0.671, "mean_token_accuracy": 0.7803534269332886, "num_tokens": 94475037.0, "step": 1442 }, { "epoch": 0.19072164948453607, "grad_norm": 1.0343631941993414, "learning_rate": 9.777116125154136e-06, "loss": 0.6347, "mean_token_accuracy": 0.7948815822601318, "num_tokens": 94540573.0, "step": 1443 }, { "epoch": 0.1908538197197991, "grad_norm": 0.957962316671755, "learning_rate": 9.776470340026144e-06, "loss": 0.6484, "mean_token_accuracy": 0.7911121845245361, "num_tokens": 94606109.0, "step": 1444 }, { "epoch": 0.1909859899550621, "grad_norm": 1.0152715528457255, "learning_rate": 9.775823644526902e-06, "loss": 0.6002, "mean_token_accuracy": 0.8058692216873169, "num_tokens": 94671645.0, "step": 1445 }, { "epoch": 0.19111816019032513, "grad_norm": 0.8667517250252373, "learning_rate": 9.77517603879408e-06, "loss": 0.6228, "mean_token_accuracy": 0.8000396490097046, "num_tokens": 94737181.0, "step": 1446 }, { "epoch": 0.19125033042558814, "grad_norm": 1.0149030709402402, "learning_rate": 9.774527522965536e-06, "loss": 0.7195, "mean_token_accuracy": 0.7693657875061035, "num_tokens": 94802717.0, "step": 1447 }, { "epoch": 0.19138250066085116, "grad_norm": 0.930011498242128, "learning_rate": 9.77387809717933e-06, "loss": 0.6193, "mean_token_accuracy": 0.8012147545814514, "num_tokens": 94868253.0, "step": 1448 }, { "epoch": 0.1915146708961142, "grad_norm": 0.9155311771380652, "learning_rate": 9.773227761573712e-06, "loss": 0.639, "mean_token_accuracy": 0.7950341701507568, "num_tokens": 94933789.0, "step": 1449 }, { "epoch": 0.19164684113137723, "grad_norm": 1.12231666603925, "learning_rate": 9.77257651628712e-06, "loss": 0.7146, "mean_token_accuracy": 0.769655704498291, "num_tokens": 94999325.0, "step": 1450 }, { "epoch": 0.19177901136664025, "grad_norm": 0.9235678874850407, "learning_rate": 9.77192436145819e-06, "loss": 0.6519, "mean_token_accuracy": 0.7913716435432434, "num_tokens": 95064861.0, "step": 1451 }, { "epoch": 0.19191118160190326, "grad_norm": 0.9274238571508051, "learning_rate": 9.771271297225758e-06, "loss": 0.6632, "mean_token_accuracy": 0.7841838598251343, "num_tokens": 95130397.0, "step": 1452 }, { "epoch": 0.19204335183716628, "grad_norm": 0.9222527596335646, "learning_rate": 9.770617323728844e-06, "loss": 0.6874, "mean_token_accuracy": 0.7791478633880615, "num_tokens": 95195933.0, "step": 1453 }, { "epoch": 0.1921755220724293, "grad_norm": 0.8933771994033721, "learning_rate": 9.769962441106665e-06, "loss": 0.656, "mean_token_accuracy": 0.7919973134994507, "num_tokens": 95261469.0, "step": 1454 }, { "epoch": 0.19230769230769232, "grad_norm": 0.870286838166974, "learning_rate": 9.769306649498634e-06, "loss": 0.6191, "mean_token_accuracy": 0.8012910485267639, "num_tokens": 95327005.0, "step": 1455 }, { "epoch": 0.19243986254295534, "grad_norm": 0.9498506643795248, "learning_rate": 9.768649949044351e-06, "loss": 0.6853, "mean_token_accuracy": 0.7783390283584595, "num_tokens": 95392541.0, "step": 1456 }, { "epoch": 0.19257203277821835, "grad_norm": 0.9385986074135884, "learning_rate": 9.76799233988362e-06, "loss": 0.6641, "mean_token_accuracy": 0.7870528697967529, "num_tokens": 95458077.0, "step": 1457 }, { "epoch": 0.19270420301348137, "grad_norm": 0.9625605020917336, "learning_rate": 9.767333822156429e-06, "loss": 0.6792, "mean_token_accuracy": 0.7820016145706177, "num_tokens": 95523613.0, "step": 1458 }, { "epoch": 0.1928363732487444, "grad_norm": 0.8743884025431465, "learning_rate": 9.766674396002963e-06, "loss": 0.6316, "mean_token_accuracy": 0.7946832180023193, "num_tokens": 95589149.0, "step": 1459 }, { "epoch": 0.1929685434840074, "grad_norm": 0.9438718586938065, "learning_rate": 9.766014061563597e-06, "loss": 0.6283, "mean_token_accuracy": 0.7975369095802307, "num_tokens": 95654685.0, "step": 1460 }, { "epoch": 0.19310071371927043, "grad_norm": 0.9739413113526605, "learning_rate": 9.765352818978908e-06, "loss": 0.6588, "mean_token_accuracy": 0.7876022458076477, "num_tokens": 95720221.0, "step": 1461 }, { "epoch": 0.19323288395453345, "grad_norm": 0.8948593279916657, "learning_rate": 9.764690668389658e-06, "loss": 0.6628, "mean_token_accuracy": 0.7845043540000916, "num_tokens": 95785757.0, "step": 1462 }, { "epoch": 0.19336505418979646, "grad_norm": 1.0409145018795645, "learning_rate": 9.764027609936803e-06, "loss": 0.6104, "mean_token_accuracy": 0.803427517414093, "num_tokens": 95851293.0, "step": 1463 }, { "epoch": 0.19349722442505948, "grad_norm": 0.9052705882355866, "learning_rate": 9.763363643761497e-06, "loss": 0.6455, "mean_token_accuracy": 0.7919210195541382, "num_tokens": 95916829.0, "step": 1464 }, { "epoch": 0.1936293946603225, "grad_norm": 0.9934580783435344, "learning_rate": 9.762698770005084e-06, "loss": 0.6332, "mean_token_accuracy": 0.7984830737113953, "num_tokens": 95982365.0, "step": 1465 }, { "epoch": 0.19376156489558552, "grad_norm": 0.9605828375258496, "learning_rate": 9.762032988809103e-06, "loss": 0.6523, "mean_token_accuracy": 0.788914680480957, "num_tokens": 96047901.0, "step": 1466 }, { "epoch": 0.19389373513084854, "grad_norm": 0.9998875544257554, "learning_rate": 9.761366300315282e-06, "loss": 0.641, "mean_token_accuracy": 0.7967433929443359, "num_tokens": 96113437.0, "step": 1467 }, { "epoch": 0.19402590536611156, "grad_norm": 0.887171314927199, "learning_rate": 9.760698704665546e-06, "loss": 0.6254, "mean_token_accuracy": 0.7974758744239807, "num_tokens": 96178973.0, "step": 1468 }, { "epoch": 0.19415807560137457, "grad_norm": 0.9622286747669133, "learning_rate": 9.760030202002015e-06, "loss": 0.6296, "mean_token_accuracy": 0.7957361936569214, "num_tokens": 96244509.0, "step": 1469 }, { "epoch": 0.1942902458366376, "grad_norm": 0.9765417543518619, "learning_rate": 9.759360792466996e-06, "loss": 0.5966, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 96310045.0, "step": 1470 }, { "epoch": 0.1944224160719006, "grad_norm": 0.8745935829873474, "learning_rate": 9.758690476202991e-06, "loss": 0.6327, "mean_token_accuracy": 0.7961024045944214, "num_tokens": 96375581.0, "step": 1471 }, { "epoch": 0.19455458630716363, "grad_norm": 1.0061737240753048, "learning_rate": 9.758019253352702e-06, "loss": 0.7123, "mean_token_accuracy": 0.7712122797966003, "num_tokens": 96441117.0, "step": 1472 }, { "epoch": 0.19468675654242665, "grad_norm": 1.0512275273574168, "learning_rate": 9.757347124059014e-06, "loss": 0.7149, "mean_token_accuracy": 0.768419623374939, "num_tokens": 96506653.0, "step": 1473 }, { "epoch": 0.19481892677768967, "grad_norm": 1.0203510757189855, "learning_rate": 9.75667408846501e-06, "loss": 0.6298, "mean_token_accuracy": 0.798757791519165, "num_tokens": 96572189.0, "step": 1474 }, { "epoch": 0.19495109701295268, "grad_norm": 0.8868296730595451, "learning_rate": 9.756000146713966e-06, "loss": 0.6152, "mean_token_accuracy": 0.803427517414093, "num_tokens": 96637725.0, "step": 1475 }, { "epoch": 0.1950832672482157, "grad_norm": 0.8820692737631309, "learning_rate": 9.755325298949348e-06, "loss": 0.6229, "mean_token_accuracy": 0.7983610033988953, "num_tokens": 96703261.0, "step": 1476 }, { "epoch": 0.19521543748347872, "grad_norm": 1.021352522589375, "learning_rate": 9.754649545314821e-06, "loss": 0.646, "mean_token_accuracy": 0.7919973134994507, "num_tokens": 96768797.0, "step": 1477 }, { "epoch": 0.19534760771874174, "grad_norm": 1.026871041989811, "learning_rate": 9.753972885954235e-06, "loss": 0.6574, "mean_token_accuracy": 0.7874954342842102, "num_tokens": 96834333.0, "step": 1478 }, { "epoch": 0.19547977795400476, "grad_norm": 1.0051416805902722, "learning_rate": 9.753295321011638e-06, "loss": 0.657, "mean_token_accuracy": 0.788945198059082, "num_tokens": 96899869.0, "step": 1479 }, { "epoch": 0.19561194818926778, "grad_norm": 0.8581285005428064, "learning_rate": 9.752616850631271e-06, "loss": 0.565, "mean_token_accuracy": 0.8171926736831665, "num_tokens": 96965405.0, "step": 1480 }, { "epoch": 0.1957441184245308, "grad_norm": 1.0156984659765331, "learning_rate": 9.751937474957564e-06, "loss": 0.6543, "mean_token_accuracy": 0.7869460582733154, "num_tokens": 97030941.0, "step": 1481 }, { "epoch": 0.1958762886597938, "grad_norm": 0.9700574783986211, "learning_rate": 9.751257194135142e-06, "loss": 0.6277, "mean_token_accuracy": 0.7984830737113953, "num_tokens": 97096477.0, "step": 1482 }, { "epoch": 0.19600845889505683, "grad_norm": 0.9189192249328523, "learning_rate": 9.750576008308824e-06, "loss": 0.6001, "mean_token_accuracy": 0.8040990233421326, "num_tokens": 97162013.0, "step": 1483 }, { "epoch": 0.19614062913031985, "grad_norm": 0.9476214108657696, "learning_rate": 9.749893917623618e-06, "loss": 0.6439, "mean_token_accuracy": 0.7928519248962402, "num_tokens": 97227549.0, "step": 1484 }, { "epoch": 0.19627279936558287, "grad_norm": 0.885609086563772, "learning_rate": 9.74921092222473e-06, "loss": 0.6234, "mean_token_accuracy": 0.79878830909729, "num_tokens": 97293085.0, "step": 1485 }, { "epoch": 0.19640496960084589, "grad_norm": 0.9736418619039101, "learning_rate": 9.748527022257553e-06, "loss": 0.6847, "mean_token_accuracy": 0.7769197821617126, "num_tokens": 97358621.0, "step": 1486 }, { "epoch": 0.1965371398361089, "grad_norm": 0.8036672758816947, "learning_rate": 9.747842217867677e-06, "loss": 0.5545, "mean_token_accuracy": 0.8217403292655945, "num_tokens": 97424157.0, "step": 1487 }, { "epoch": 0.19666931007137192, "grad_norm": 0.962928633175098, "learning_rate": 9.747156509200878e-06, "loss": 0.657, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 97489693.0, "step": 1488 }, { "epoch": 0.19680148030663494, "grad_norm": 0.8726593267199936, "learning_rate": 9.746469896403138e-06, "loss": 0.6418, "mean_token_accuracy": 0.7910664081573486, "num_tokens": 97555229.0, "step": 1489 }, { "epoch": 0.19693365054189796, "grad_norm": 0.9127511240555193, "learning_rate": 9.745782379620612e-06, "loss": 0.6366, "mean_token_accuracy": 0.7963923811912537, "num_tokens": 97620765.0, "step": 1490 }, { "epoch": 0.19706582077716098, "grad_norm": 0.9435798472505577, "learning_rate": 9.745093958999664e-06, "loss": 0.6506, "mean_token_accuracy": 0.7906696200370789, "num_tokens": 97686301.0, "step": 1491 }, { "epoch": 0.197197991012424, "grad_norm": 1.0095533791938227, "learning_rate": 9.744404634686845e-06, "loss": 0.7158, "mean_token_accuracy": 0.7681449055671692, "num_tokens": 97751837.0, "step": 1492 }, { "epoch": 0.197330161247687, "grad_norm": 0.9861693913982669, "learning_rate": 9.743714406828896e-06, "loss": 0.6032, "mean_token_accuracy": 0.8039769530296326, "num_tokens": 97817373.0, "step": 1493 }, { "epoch": 0.19746233148295003, "grad_norm": 0.9176210619332914, "learning_rate": 9.74302327557275e-06, "loss": 0.6882, "mean_token_accuracy": 0.7798956036567688, "num_tokens": 97882909.0, "step": 1494 }, { "epoch": 0.19759450171821305, "grad_norm": 1.0055800461064726, "learning_rate": 9.742331241065541e-06, "loss": 0.6864, "mean_token_accuracy": 0.7787968516349792, "num_tokens": 97948445.0, "step": 1495 }, { "epoch": 0.19772667195347607, "grad_norm": 0.9131081113103628, "learning_rate": 9.741638303454579e-06, "loss": 0.6644, "mean_token_accuracy": 0.7848095297813416, "num_tokens": 98013981.0, "step": 1496 }, { "epoch": 0.1978588421887391, "grad_norm": 1.169037105757483, "learning_rate": 9.740944462887384e-06, "loss": 0.7211, "mean_token_accuracy": 0.7683128118515015, "num_tokens": 98079517.0, "step": 1497 }, { "epoch": 0.1979910124240021, "grad_norm": 0.9465639480298889, "learning_rate": 9.740249719511657e-06, "loss": 0.5936, "mean_token_accuracy": 0.8073647618293762, "num_tokens": 98145053.0, "step": 1498 }, { "epoch": 0.19812318265926512, "grad_norm": 0.9168648251477732, "learning_rate": 9.739554073475295e-06, "loss": 0.6476, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 98210589.0, "step": 1499 }, { "epoch": 0.19825535289452814, "grad_norm": 0.9371436351732721, "learning_rate": 9.738857524926385e-06, "loss": 0.66, "mean_token_accuracy": 0.7870681285858154, "num_tokens": 98276125.0, "step": 1500 }, { "epoch": 0.19838752312979116, "grad_norm": 0.8842688723924053, "learning_rate": 9.73816007401321e-06, "loss": 0.6452, "mean_token_accuracy": 0.7922109365463257, "num_tokens": 98341661.0, "step": 1501 }, { "epoch": 0.19851969336505418, "grad_norm": 0.9554878331508121, "learning_rate": 9.73746172088424e-06, "loss": 0.6337, "mean_token_accuracy": 0.7956140637397766, "num_tokens": 98407197.0, "step": 1502 }, { "epoch": 0.1986518636003172, "grad_norm": 0.8903078972000071, "learning_rate": 9.736762465688142e-06, "loss": 0.6614, "mean_token_accuracy": 0.789097785949707, "num_tokens": 98472733.0, "step": 1503 }, { "epoch": 0.19878403383558021, "grad_norm": 0.9099386580086062, "learning_rate": 9.736062308573775e-06, "loss": 0.6606, "mean_token_accuracy": 0.789067268371582, "num_tokens": 98538269.0, "step": 1504 }, { "epoch": 0.19891620407084323, "grad_norm": 0.9641323448503957, "learning_rate": 9.735361249690183e-06, "loss": 0.6454, "mean_token_accuracy": 0.7942101359367371, "num_tokens": 98603805.0, "step": 1505 }, { "epoch": 0.19904837430610625, "grad_norm": 1.11596203149268, "learning_rate": 9.73465928918661e-06, "loss": 0.633, "mean_token_accuracy": 0.7958430051803589, "num_tokens": 98669341.0, "step": 1506 }, { "epoch": 0.19918054454136927, "grad_norm": 0.9962287984584, "learning_rate": 9.733956427212487e-06, "loss": 0.6035, "mean_token_accuracy": 0.8053045868873596, "num_tokens": 98734877.0, "step": 1507 }, { "epoch": 0.19931271477663232, "grad_norm": 0.8144639898115308, "learning_rate": 9.733252663917439e-06, "loss": 0.6141, "mean_token_accuracy": 0.8026034832000732, "num_tokens": 98800413.0, "step": 1508 }, { "epoch": 0.19944488501189533, "grad_norm": 0.9594569389052402, "learning_rate": 9.732547999451284e-06, "loss": 0.6567, "mean_token_accuracy": 0.7892045974731445, "num_tokens": 98865949.0, "step": 1509 }, { "epoch": 0.19957705524715835, "grad_norm": 1.168791493434508, "learning_rate": 9.731842433964027e-06, "loss": 0.6813, "mean_token_accuracy": 0.7813148498535156, "num_tokens": 98931485.0, "step": 1510 }, { "epoch": 0.19970922548242137, "grad_norm": 1.265082160138673, "learning_rate": 9.73113596760587e-06, "loss": 0.7122, "mean_token_accuracy": 0.7699914574623108, "num_tokens": 98997021.0, "step": 1511 }, { "epoch": 0.1998413957176844, "grad_norm": 1.0281255686015152, "learning_rate": 9.730428600527207e-06, "loss": 0.6759, "mean_token_accuracy": 0.7840923070907593, "num_tokens": 99062557.0, "step": 1512 }, { "epoch": 0.1999735659529474, "grad_norm": 0.955307789788948, "learning_rate": 9.72972033287862e-06, "loss": 0.6629, "mean_token_accuracy": 0.7895403504371643, "num_tokens": 99128093.0, "step": 1513 }, { "epoch": 0.20010573618821043, "grad_norm": 1.074268062276227, "learning_rate": 9.729011164810885e-06, "loss": 0.6497, "mean_token_accuracy": 0.7919973134994507, "num_tokens": 99193629.0, "step": 1514 }, { "epoch": 0.20023790642347344, "grad_norm": 1.045498457651409, "learning_rate": 9.728301096474966e-06, "loss": 0.6452, "mean_token_accuracy": 0.7941948771476746, "num_tokens": 99259165.0, "step": 1515 }, { "epoch": 0.20037007665873646, "grad_norm": 1.0300515560971952, "learning_rate": 9.727590128022026e-06, "loss": 0.657, "mean_token_accuracy": 0.7861982583999634, "num_tokens": 99324701.0, "step": 1516 }, { "epoch": 0.20050224689399948, "grad_norm": 0.9571270682809071, "learning_rate": 9.726878259603411e-06, "loss": 0.6786, "mean_token_accuracy": 0.7795599102973938, "num_tokens": 99390237.0, "step": 1517 }, { "epoch": 0.2006344171292625, "grad_norm": 0.9939146345733468, "learning_rate": 9.726165491370668e-06, "loss": 0.6427, "mean_token_accuracy": 0.7924246191978455, "num_tokens": 99455773.0, "step": 1518 }, { "epoch": 0.20076658736452552, "grad_norm": 1.090350878242779, "learning_rate": 9.725451823475527e-06, "loss": 0.6836, "mean_token_accuracy": 0.7802923917770386, "num_tokens": 99521309.0, "step": 1519 }, { "epoch": 0.20089875759978854, "grad_norm": 1.1481766562359819, "learning_rate": 9.724737256069915e-06, "loss": 0.6635, "mean_token_accuracy": 0.7855420708656311, "num_tokens": 99586845.0, "step": 1520 }, { "epoch": 0.20103092783505155, "grad_norm": 0.9579081989508901, "learning_rate": 9.724021789305945e-06, "loss": 0.6533, "mean_token_accuracy": 0.7912037372589111, "num_tokens": 99652381.0, "step": 1521 }, { "epoch": 0.20116309807031457, "grad_norm": 0.999316833166846, "learning_rate": 9.72330542333593e-06, "loss": 0.6519, "mean_token_accuracy": 0.7903034090995789, "num_tokens": 99717917.0, "step": 1522 }, { "epoch": 0.2012952683055776, "grad_norm": 0.9893011323224558, "learning_rate": 9.722588158312367e-06, "loss": 0.6223, "mean_token_accuracy": 0.7992613911628723, "num_tokens": 99783453.0, "step": 1523 }, { "epoch": 0.2014274385408406, "grad_norm": 0.9815657809046194, "learning_rate": 9.721869994387943e-06, "loss": 0.6157, "mean_token_accuracy": 0.8033207058906555, "num_tokens": 99848989.0, "step": 1524 }, { "epoch": 0.20155960877610363, "grad_norm": 0.9872337035433623, "learning_rate": 9.721150931715545e-06, "loss": 0.664, "mean_token_accuracy": 0.7849010825157166, "num_tokens": 99914525.0, "step": 1525 }, { "epoch": 0.20169177901136665, "grad_norm": 1.0244069233551267, "learning_rate": 9.720430970448246e-06, "loss": 0.6656, "mean_token_accuracy": 0.7883500456809998, "num_tokens": 99980061.0, "step": 1526 }, { "epoch": 0.20182394924662966, "grad_norm": 1.0138698274506799, "learning_rate": 9.719710110739307e-06, "loss": 0.7196, "mean_token_accuracy": 0.768572211265564, "num_tokens": 100045597.0, "step": 1527 }, { "epoch": 0.20195611948189268, "grad_norm": 0.9872821492356394, "learning_rate": 9.718988352742189e-06, "loss": 0.6868, "mean_token_accuracy": 0.7816048264503479, "num_tokens": 100111133.0, "step": 1528 }, { "epoch": 0.2020882897171557, "grad_norm": 1.0134441280767594, "learning_rate": 9.718265696610537e-06, "loss": 0.6744, "mean_token_accuracy": 0.7828409075737, "num_tokens": 100176669.0, "step": 1529 }, { "epoch": 0.20222045995241872, "grad_norm": 1.1749009673541193, "learning_rate": 9.717542142498187e-06, "loss": 0.7119, "mean_token_accuracy": 0.7723110914230347, "num_tokens": 100242205.0, "step": 1530 }, { "epoch": 0.20235263018768174, "grad_norm": 0.887081277605388, "learning_rate": 9.716817690559173e-06, "loss": 0.6329, "mean_token_accuracy": 0.7946678996086121, "num_tokens": 100307741.0, "step": 1531 }, { "epoch": 0.20248480042294476, "grad_norm": 0.8574826016809914, "learning_rate": 9.716092340947712e-06, "loss": 0.6798, "mean_token_accuracy": 0.7816811203956604, "num_tokens": 100373277.0, "step": 1532 }, { "epoch": 0.20261697065820777, "grad_norm": 1.024465564062011, "learning_rate": 9.715366093818218e-06, "loss": 0.5996, "mean_token_accuracy": 0.8048620223999023, "num_tokens": 100438813.0, "step": 1533 }, { "epoch": 0.2027491408934708, "grad_norm": 0.9155589302076796, "learning_rate": 9.714638949325295e-06, "loss": 0.6111, "mean_token_accuracy": 0.8044347167015076, "num_tokens": 100504349.0, "step": 1534 }, { "epoch": 0.2028813111287338, "grad_norm": 0.9812432212719909, "learning_rate": 9.713910907623732e-06, "loss": 0.6183, "mean_token_accuracy": 0.7980252504348755, "num_tokens": 100569885.0, "step": 1535 }, { "epoch": 0.20301348136399683, "grad_norm": 1.0335824461265497, "learning_rate": 9.713181968868518e-06, "loss": 0.7148, "mean_token_accuracy": 0.7662983536720276, "num_tokens": 100635421.0, "step": 1536 }, { "epoch": 0.20314565159925985, "grad_norm": 0.85805277669818, "learning_rate": 9.712452133214828e-06, "loss": 0.6501, "mean_token_accuracy": 0.792714536190033, "num_tokens": 100700957.0, "step": 1537 }, { "epoch": 0.20327782183452286, "grad_norm": 0.8896585196595803, "learning_rate": 9.711721400818028e-06, "loss": 0.6288, "mean_token_accuracy": 0.7973538041114807, "num_tokens": 100766493.0, "step": 1538 }, { "epoch": 0.20340999206978588, "grad_norm": 1.035017232490739, "learning_rate": 9.710989771833676e-06, "loss": 0.6898, "mean_token_accuracy": 0.7759125828742981, "num_tokens": 100832029.0, "step": 1539 }, { "epoch": 0.2035421623050489, "grad_norm": 0.9507170562350905, "learning_rate": 9.710257246417522e-06, "loss": 0.6412, "mean_token_accuracy": 0.7935081124305725, "num_tokens": 100897565.0, "step": 1540 }, { "epoch": 0.20367433254031192, "grad_norm": 1.0825260882107708, "learning_rate": 9.709523824725503e-06, "loss": 0.6325, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 100963101.0, "step": 1541 }, { "epoch": 0.20380650277557494, "grad_norm": 0.9080851698011513, "learning_rate": 9.708789506913751e-06, "loss": 0.6618, "mean_token_accuracy": 0.784580647945404, "num_tokens": 101028637.0, "step": 1542 }, { "epoch": 0.20393867301083796, "grad_norm": 0.9337132906956461, "learning_rate": 9.708054293138586e-06, "loss": 0.6508, "mean_token_accuracy": 0.7899066209793091, "num_tokens": 101094173.0, "step": 1543 }, { "epoch": 0.20407084324610097, "grad_norm": 1.0252753575144258, "learning_rate": 9.707318183556519e-06, "loss": 0.5972, "mean_token_accuracy": 0.8051672577857971, "num_tokens": 101159709.0, "step": 1544 }, { "epoch": 0.204203013481364, "grad_norm": 0.9161673252143762, "learning_rate": 9.706581178324253e-06, "loss": 0.6549, "mean_token_accuracy": 0.7898608446121216, "num_tokens": 101225245.0, "step": 1545 }, { "epoch": 0.204335183716627, "grad_norm": 0.9836941677133094, "learning_rate": 9.705843277598683e-06, "loss": 0.6256, "mean_token_accuracy": 0.7959040403366089, "num_tokens": 101290781.0, "step": 1546 }, { "epoch": 0.20446735395189003, "grad_norm": 0.9494269924536325, "learning_rate": 9.70510448153689e-06, "loss": 0.611, "mean_token_accuracy": 0.8018099069595337, "num_tokens": 101356317.0, "step": 1547 }, { "epoch": 0.20459952418715305, "grad_norm": 0.9518011086037416, "learning_rate": 9.704364790296147e-06, "loss": 0.685, "mean_token_accuracy": 0.778232216835022, "num_tokens": 101421853.0, "step": 1548 }, { "epoch": 0.20473169442241607, "grad_norm": 0.8844780639358497, "learning_rate": 9.703624204033925e-06, "loss": 0.6433, "mean_token_accuracy": 0.7929892539978027, "num_tokens": 101487389.0, "step": 1549 }, { "epoch": 0.20486386465767908, "grad_norm": 1.0408928346702688, "learning_rate": 9.70288272290787e-06, "loss": 0.6389, "mean_token_accuracy": 0.7937523126602173, "num_tokens": 101552925.0, "step": 1550 }, { "epoch": 0.2049960348929421, "grad_norm": 1.0382957922386056, "learning_rate": 9.702140347075839e-06, "loss": 0.6857, "mean_token_accuracy": 0.7816963791847229, "num_tokens": 101618461.0, "step": 1551 }, { "epoch": 0.20512820512820512, "grad_norm": 0.8832037934656131, "learning_rate": 9.70139707669586e-06, "loss": 0.6558, "mean_token_accuracy": 0.7912037372589111, "num_tokens": 101683997.0, "step": 1552 }, { "epoch": 0.20526037536346814, "grad_norm": 0.8435099666810807, "learning_rate": 9.700652911926163e-06, "loss": 0.6205, "mean_token_accuracy": 0.7990171909332275, "num_tokens": 101749533.0, "step": 1553 }, { "epoch": 0.20539254559873116, "grad_norm": 0.8843394725387055, "learning_rate": 9.699907852925162e-06, "loss": 0.6106, "mean_token_accuracy": 0.8021456599235535, "num_tokens": 101815069.0, "step": 1554 }, { "epoch": 0.20552471583399418, "grad_norm": 1.0520836606620807, "learning_rate": 9.699161899851467e-06, "loss": 0.669, "mean_token_accuracy": 0.784794270992279, "num_tokens": 101880605.0, "step": 1555 }, { "epoch": 0.2056568860692572, "grad_norm": 0.9074301750029589, "learning_rate": 9.698415052863878e-06, "loss": 0.6481, "mean_token_accuracy": 0.7906085848808289, "num_tokens": 101946141.0, "step": 1556 }, { "epoch": 0.2057890563045202, "grad_norm": 0.9310266025309851, "learning_rate": 9.69766731212138e-06, "loss": 0.6538, "mean_token_accuracy": 0.7869613170623779, "num_tokens": 102011677.0, "step": 1557 }, { "epoch": 0.20592122653978323, "grad_norm": 0.9193801615252842, "learning_rate": 9.696918677783153e-06, "loss": 0.6245, "mean_token_accuracy": 0.8009247779846191, "num_tokens": 102077213.0, "step": 1558 }, { "epoch": 0.20605339677504625, "grad_norm": 0.9018867712806348, "learning_rate": 9.696169150008563e-06, "loss": 0.6079, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 102142749.0, "step": 1559 }, { "epoch": 0.20618556701030927, "grad_norm": 1.0048869287333844, "learning_rate": 9.695418728957174e-06, "loss": 0.6824, "mean_token_accuracy": 0.7796209454536438, "num_tokens": 102208285.0, "step": 1560 }, { "epoch": 0.20631773724557229, "grad_norm": 0.9371625443134994, "learning_rate": 9.694667414788728e-06, "loss": 0.6688, "mean_token_accuracy": 0.7819710373878479, "num_tokens": 102273821.0, "step": 1561 }, { "epoch": 0.2064499074808353, "grad_norm": 0.8437241563382782, "learning_rate": 9.693915207663168e-06, "loss": 0.6443, "mean_token_accuracy": 0.7950799465179443, "num_tokens": 102339357.0, "step": 1562 }, { "epoch": 0.20658207771609832, "grad_norm": 1.009352942133089, "learning_rate": 9.693162107740624e-06, "loss": 0.6456, "mean_token_accuracy": 0.7889909744262695, "num_tokens": 102404893.0, "step": 1563 }, { "epoch": 0.20671424795136134, "grad_norm": 1.0235082337531658, "learning_rate": 9.692408115181415e-06, "loss": 0.6613, "mean_token_accuracy": 0.7853131294250488, "num_tokens": 102470429.0, "step": 1564 }, { "epoch": 0.20684641818662436, "grad_norm": 0.9958916018866385, "learning_rate": 9.691653230146049e-06, "loss": 0.6439, "mean_token_accuracy": 0.7909748554229736, "num_tokens": 102535965.0, "step": 1565 }, { "epoch": 0.20697858842188738, "grad_norm": 0.914413545903517, "learning_rate": 9.690897452795225e-06, "loss": 0.6553, "mean_token_accuracy": 0.7884568572044373, "num_tokens": 102601501.0, "step": 1566 }, { "epoch": 0.20711075865715042, "grad_norm": 1.1082745444736712, "learning_rate": 9.690140783289834e-06, "loss": 0.7226, "mean_token_accuracy": 0.7666035890579224, "num_tokens": 102667037.0, "step": 1567 }, { "epoch": 0.20724292889241344, "grad_norm": 0.9056305693739503, "learning_rate": 9.689383221790952e-06, "loss": 0.6223, "mean_token_accuracy": 0.7949883937835693, "num_tokens": 102732573.0, "step": 1568 }, { "epoch": 0.20737509912767646, "grad_norm": 0.9954590010607698, "learning_rate": 9.688624768459852e-06, "loss": 0.6627, "mean_token_accuracy": 0.784519612789154, "num_tokens": 102798109.0, "step": 1569 }, { "epoch": 0.20750726936293948, "grad_norm": 0.933465776303074, "learning_rate": 9.687865423457991e-06, "loss": 0.643, "mean_token_accuracy": 0.7920583486557007, "num_tokens": 102863645.0, "step": 1570 }, { "epoch": 0.2076394395982025, "grad_norm": 1.0376829770056086, "learning_rate": 9.687105186947017e-06, "loss": 0.7313, "mean_token_accuracy": 0.7674429416656494, "num_tokens": 102929181.0, "step": 1571 }, { "epoch": 0.20777160983346551, "grad_norm": 0.9330204233122568, "learning_rate": 9.686344059088769e-06, "loss": 0.6373, "mean_token_accuracy": 0.7918752431869507, "num_tokens": 102994717.0, "step": 1572 }, { "epoch": 0.20790378006872853, "grad_norm": 1.0162313337684123, "learning_rate": 9.685582040045277e-06, "loss": 0.6523, "mean_token_accuracy": 0.7868086695671082, "num_tokens": 103060253.0, "step": 1573 }, { "epoch": 0.20803595030399155, "grad_norm": 0.8339978139123361, "learning_rate": 9.684819129978754e-06, "loss": 0.5838, "mean_token_accuracy": 0.8108289837837219, "num_tokens": 103125789.0, "step": 1574 }, { "epoch": 0.20816812053925457, "grad_norm": 0.9234543165006864, "learning_rate": 9.684055329051613e-06, "loss": 0.6076, "mean_token_accuracy": 0.802191436290741, "num_tokens": 103191325.0, "step": 1575 }, { "epoch": 0.2083002907745176, "grad_norm": 0.8973299175804911, "learning_rate": 9.68329063742645e-06, "loss": 0.6171, "mean_token_accuracy": 0.797948956489563, "num_tokens": 103256861.0, "step": 1576 }, { "epoch": 0.2084324610097806, "grad_norm": 0.9037318922866201, "learning_rate": 9.68252505526605e-06, "loss": 0.6183, "mean_token_accuracy": 0.8019472360610962, "num_tokens": 103322397.0, "step": 1577 }, { "epoch": 0.20856463124504362, "grad_norm": 0.9822306050485301, "learning_rate": 9.68175858273339e-06, "loss": 0.6971, "mean_token_accuracy": 0.7793920040130615, "num_tokens": 103387933.0, "step": 1578 }, { "epoch": 0.20869680148030664, "grad_norm": 0.9672856577797303, "learning_rate": 9.680991219991637e-06, "loss": 0.6431, "mean_token_accuracy": 0.7891130447387695, "num_tokens": 103453469.0, "step": 1579 }, { "epoch": 0.20882897171556966, "grad_norm": 0.9537185477091711, "learning_rate": 9.680222967204147e-06, "loss": 0.6375, "mean_token_accuracy": 0.792500913143158, "num_tokens": 103519005.0, "step": 1580 }, { "epoch": 0.20896114195083268, "grad_norm": 0.9123528993983928, "learning_rate": 9.679453824534464e-06, "loss": 0.5941, "mean_token_accuracy": 0.8071358799934387, "num_tokens": 103584541.0, "step": 1581 }, { "epoch": 0.2090933121860957, "grad_norm": 0.9672526596176259, "learning_rate": 9.678683792146321e-06, "loss": 0.6613, "mean_token_accuracy": 0.7874954342842102, "num_tokens": 103650077.0, "step": 1582 }, { "epoch": 0.20922548242135872, "grad_norm": 0.9225914280745993, "learning_rate": 9.677912870203646e-06, "loss": 0.649, "mean_token_accuracy": 0.7910816669464111, "num_tokens": 103715613.0, "step": 1583 }, { "epoch": 0.20935765265662173, "grad_norm": 0.9431065917213796, "learning_rate": 9.677141058870548e-06, "loss": 0.6696, "mean_token_accuracy": 0.7836650013923645, "num_tokens": 103781149.0, "step": 1584 }, { "epoch": 0.20948982289188475, "grad_norm": 0.917533658327395, "learning_rate": 9.67636835831133e-06, "loss": 0.665, "mean_token_accuracy": 0.7840160131454468, "num_tokens": 103846685.0, "step": 1585 }, { "epoch": 0.20962199312714777, "grad_norm": 0.9301138677996366, "learning_rate": 9.675594768690487e-06, "loss": 0.6733, "mean_token_accuracy": 0.7834208011627197, "num_tokens": 103912221.0, "step": 1586 }, { "epoch": 0.2097541633624108, "grad_norm": 0.8669028902154545, "learning_rate": 9.674820290172695e-06, "loss": 0.6289, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 103977757.0, "step": 1587 }, { "epoch": 0.2098863335976738, "grad_norm": 0.9374584838647813, "learning_rate": 9.67404492292283e-06, "loss": 0.7093, "mean_token_accuracy": 0.775073230266571, "num_tokens": 104043293.0, "step": 1588 }, { "epoch": 0.21001850383293683, "grad_norm": 0.8830002769410339, "learning_rate": 9.673268667105944e-06, "loss": 0.6619, "mean_token_accuracy": 0.7856946587562561, "num_tokens": 104108829.0, "step": 1589 }, { "epoch": 0.21015067406819984, "grad_norm": 0.928915006551426, "learning_rate": 9.672491522887295e-06, "loss": 0.6582, "mean_token_accuracy": 0.7847790122032166, "num_tokens": 104174365.0, "step": 1590 }, { "epoch": 0.21028284430346286, "grad_norm": 0.850228759470226, "learning_rate": 9.671713490432315e-06, "loss": 0.6171, "mean_token_accuracy": 0.7994597554206848, "num_tokens": 104239901.0, "step": 1591 }, { "epoch": 0.21041501453872588, "grad_norm": 0.8962460308301236, "learning_rate": 9.670934569906632e-06, "loss": 0.6522, "mean_token_accuracy": 0.7922415137290955, "num_tokens": 104305437.0, "step": 1592 }, { "epoch": 0.2105471847739889, "grad_norm": 0.810120802803322, "learning_rate": 9.670154761476062e-06, "loss": 0.5981, "mean_token_accuracy": 0.8061286807060242, "num_tokens": 104370973.0, "step": 1593 }, { "epoch": 0.21067935500925192, "grad_norm": 0.8938639671174305, "learning_rate": 9.669374065306611e-06, "loss": 0.6656, "mean_token_accuracy": 0.7857251763343811, "num_tokens": 104436509.0, "step": 1594 }, { "epoch": 0.21081152524451494, "grad_norm": 0.9490271729139327, "learning_rate": 9.668592481564469e-06, "loss": 0.6839, "mean_token_accuracy": 0.7807196974754333, "num_tokens": 104502045.0, "step": 1595 }, { "epoch": 0.21094369547977795, "grad_norm": 0.890054197772487, "learning_rate": 9.667810010416023e-06, "loss": 0.585, "mean_token_accuracy": 0.8095775842666626, "num_tokens": 104567581.0, "step": 1596 }, { "epoch": 0.21107586571504097, "grad_norm": 0.8332538895514837, "learning_rate": 9.667026652027843e-06, "loss": 0.6336, "mean_token_accuracy": 0.7956598997116089, "num_tokens": 104633117.0, "step": 1597 }, { "epoch": 0.211208035950304, "grad_norm": 0.9523443046739365, "learning_rate": 9.666242406566691e-06, "loss": 0.6694, "mean_token_accuracy": 0.7841228246688843, "num_tokens": 104698653.0, "step": 1598 }, { "epoch": 0.211340206185567, "grad_norm": 0.9314480012640454, "learning_rate": 9.665457274199515e-06, "loss": 0.6473, "mean_token_accuracy": 0.7906391024589539, "num_tokens": 104764189.0, "step": 1599 }, { "epoch": 0.21147237642083003, "grad_norm": 0.9320717004715612, "learning_rate": 9.664671255093454e-06, "loss": 0.6756, "mean_token_accuracy": 0.7849926948547363, "num_tokens": 104829725.0, "step": 1600 }, { "epoch": 0.21160454665609305, "grad_norm": 0.9194674824241271, "learning_rate": 9.663884349415835e-06, "loss": 0.6395, "mean_token_accuracy": 0.7928061485290527, "num_tokens": 104895261.0, "step": 1601 }, { "epoch": 0.21173671689135606, "grad_norm": 0.893671965985691, "learning_rate": 9.663096557334172e-06, "loss": 0.6176, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 104960797.0, "step": 1602 }, { "epoch": 0.21186888712661908, "grad_norm": 0.8770623747694093, "learning_rate": 9.662307879016173e-06, "loss": 0.6352, "mean_token_accuracy": 0.7957209348678589, "num_tokens": 105026333.0, "step": 1603 }, { "epoch": 0.2120010573618821, "grad_norm": 0.8582516739579196, "learning_rate": 9.661518314629729e-06, "loss": 0.5945, "mean_token_accuracy": 0.8062354922294617, "num_tokens": 105091869.0, "step": 1604 }, { "epoch": 0.21213322759714512, "grad_norm": 1.0044531281153133, "learning_rate": 9.660727864342922e-06, "loss": 0.7155, "mean_token_accuracy": 0.769625186920166, "num_tokens": 105157405.0, "step": 1605 }, { "epoch": 0.21226539783240814, "grad_norm": 0.8511540963497923, "learning_rate": 9.659936528324021e-06, "loss": 0.5992, "mean_token_accuracy": 0.8054572343826294, "num_tokens": 105222941.0, "step": 1606 }, { "epoch": 0.21239756806767116, "grad_norm": 0.8235950800202272, "learning_rate": 9.659144306741487e-06, "loss": 0.5735, "mean_token_accuracy": 0.8148730397224426, "num_tokens": 105288477.0, "step": 1607 }, { "epoch": 0.21252973830293417, "grad_norm": 0.898535905580162, "learning_rate": 9.658351199763967e-06, "loss": 0.6965, "mean_token_accuracy": 0.775103747844696, "num_tokens": 105354013.0, "step": 1608 }, { "epoch": 0.2126619085381972, "grad_norm": 0.9576452101520214, "learning_rate": 9.657557207560299e-06, "loss": 0.6596, "mean_token_accuracy": 0.7872817516326904, "num_tokens": 105419549.0, "step": 1609 }, { "epoch": 0.2127940787734602, "grad_norm": 0.8731569544249325, "learning_rate": 9.656762330299501e-06, "loss": 0.6134, "mean_token_accuracy": 0.7984220385551453, "num_tokens": 105485085.0, "step": 1610 }, { "epoch": 0.21292624900872323, "grad_norm": 0.9352779609045179, "learning_rate": 9.655966568150792e-06, "loss": 0.6715, "mean_token_accuracy": 0.7832987308502197, "num_tokens": 105550621.0, "step": 1611 }, { "epoch": 0.21305841924398625, "grad_norm": 0.9283058375710963, "learning_rate": 9.65516992128357e-06, "loss": 0.6848, "mean_token_accuracy": 0.7809333205223083, "num_tokens": 105616157.0, "step": 1612 }, { "epoch": 0.21319058947924927, "grad_norm": 0.9635307011518228, "learning_rate": 9.654372389867426e-06, "loss": 0.6459, "mean_token_accuracy": 0.7910969257354736, "num_tokens": 105681693.0, "step": 1613 }, { "epoch": 0.21332275971451228, "grad_norm": 0.9214023783163029, "learning_rate": 9.65357397407214e-06, "loss": 0.6914, "mean_token_accuracy": 0.7803992033004761, "num_tokens": 105747229.0, "step": 1614 }, { "epoch": 0.2134549299497753, "grad_norm": 0.9583027271205059, "learning_rate": 9.652774674067672e-06, "loss": 0.6438, "mean_token_accuracy": 0.7939201593399048, "num_tokens": 105812765.0, "step": 1615 }, { "epoch": 0.21358710018503832, "grad_norm": 1.0202310197033955, "learning_rate": 9.65197449002418e-06, "loss": 0.6498, "mean_token_accuracy": 0.7910969257354736, "num_tokens": 105878301.0, "step": 1616 }, { "epoch": 0.21371927042030134, "grad_norm": 0.8991894413337899, "learning_rate": 9.651173422112006e-06, "loss": 0.6186, "mean_token_accuracy": 0.8018556833267212, "num_tokens": 105943837.0, "step": 1617 }, { "epoch": 0.21385144065556436, "grad_norm": 0.9868905177437365, "learning_rate": 9.65037147050168e-06, "loss": 0.6336, "mean_token_accuracy": 0.7955377697944641, "num_tokens": 106009373.0, "step": 1618 }, { "epoch": 0.21398361089082737, "grad_norm": 0.9418834599313504, "learning_rate": 9.649568635363924e-06, "loss": 0.6271, "mean_token_accuracy": 0.7979642152786255, "num_tokens": 106074909.0, "step": 1619 }, { "epoch": 0.2141157811260904, "grad_norm": 0.9261297692839241, "learning_rate": 9.648764916869641e-06, "loss": 0.6162, "mean_token_accuracy": 0.7991393208503723, "num_tokens": 106140445.0, "step": 1620 }, { "epoch": 0.2142479513613534, "grad_norm": 1.0377110576698765, "learning_rate": 9.647960315189927e-06, "loss": 0.6594, "mean_token_accuracy": 0.7874343991279602, "num_tokens": 106205981.0, "step": 1621 }, { "epoch": 0.21438012159661643, "grad_norm": 0.9209268052409788, "learning_rate": 9.647154830496064e-06, "loss": 0.6557, "mean_token_accuracy": 0.7892351150512695, "num_tokens": 106271517.0, "step": 1622 }, { "epoch": 0.21451229183187945, "grad_norm": 0.9301566295442876, "learning_rate": 9.646348462959526e-06, "loss": 0.6356, "mean_token_accuracy": 0.7927298545837402, "num_tokens": 106337053.0, "step": 1623 }, { "epoch": 0.21464446206714247, "grad_norm": 0.8697870490433027, "learning_rate": 9.64554121275197e-06, "loss": 0.6292, "mean_token_accuracy": 0.7981168627738953, "num_tokens": 106402589.0, "step": 1624 }, { "epoch": 0.21477663230240548, "grad_norm": 0.8828305669101331, "learning_rate": 9.64473308004524e-06, "loss": 0.5812, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 106468125.0, "step": 1625 }, { "epoch": 0.21490880253766853, "grad_norm": 1.0174872207088737, "learning_rate": 9.643924065011374e-06, "loss": 0.6352, "mean_token_accuracy": 0.7925466895103455, "num_tokens": 106533661.0, "step": 1626 }, { "epoch": 0.21504097277293155, "grad_norm": 0.9387633933874101, "learning_rate": 9.643114167822593e-06, "loss": 0.6253, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 106599197.0, "step": 1627 }, { "epoch": 0.21517314300819457, "grad_norm": 0.8729137392886279, "learning_rate": 9.642303388651309e-06, "loss": 0.5978, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 106664733.0, "step": 1628 }, { "epoch": 0.21530531324345759, "grad_norm": 0.9728426503468591, "learning_rate": 9.64149172767012e-06, "loss": 0.659, "mean_token_accuracy": 0.7866408228874207, "num_tokens": 106730269.0, "step": 1629 }, { "epoch": 0.2154374834787206, "grad_norm": 0.9329724624250868, "learning_rate": 9.640679185051808e-06, "loss": 0.6545, "mean_token_accuracy": 0.7894640564918518, "num_tokens": 106795805.0, "step": 1630 }, { "epoch": 0.21556965371398362, "grad_norm": 0.9217443665662405, "learning_rate": 9.63986576096935e-06, "loss": 0.6471, "mean_token_accuracy": 0.792714536190033, "num_tokens": 106861341.0, "step": 1631 }, { "epoch": 0.21570182394924664, "grad_norm": 0.9702911913826888, "learning_rate": 9.639051455595903e-06, "loss": 0.6442, "mean_token_accuracy": 0.7916310429573059, "num_tokens": 106926877.0, "step": 1632 }, { "epoch": 0.21583399418450966, "grad_norm": 0.9877690822694865, "learning_rate": 9.63823626910482e-06, "loss": 0.692, "mean_token_accuracy": 0.7744780778884888, "num_tokens": 106992413.0, "step": 1633 }, { "epoch": 0.21596616441977268, "grad_norm": 0.9015081242023226, "learning_rate": 9.637420201669636e-06, "loss": 0.601, "mean_token_accuracy": 0.8040379881858826, "num_tokens": 107057949.0, "step": 1634 }, { "epoch": 0.2160983346550357, "grad_norm": 0.9936309686541511, "learning_rate": 9.636603253464072e-06, "loss": 0.6823, "mean_token_accuracy": 0.7806891798973083, "num_tokens": 107123485.0, "step": 1635 }, { "epoch": 0.2162305048902987, "grad_norm": 0.9058100506272008, "learning_rate": 9.635785424662044e-06, "loss": 0.6147, "mean_token_accuracy": 0.8048467636108398, "num_tokens": 107189021.0, "step": 1636 }, { "epoch": 0.21636267512556173, "grad_norm": 1.081887885675095, "learning_rate": 9.634966715437646e-06, "loss": 0.6971, "mean_token_accuracy": 0.7743407487869263, "num_tokens": 107254557.0, "step": 1637 }, { "epoch": 0.21649484536082475, "grad_norm": 0.8826656298727678, "learning_rate": 9.634147125965168e-06, "loss": 0.6055, "mean_token_accuracy": 0.807456374168396, "num_tokens": 107320093.0, "step": 1638 }, { "epoch": 0.21662701559608777, "grad_norm": 0.9156912627230275, "learning_rate": 9.633326656419081e-06, "loss": 0.6846, "mean_token_accuracy": 0.7805823683738708, "num_tokens": 107385629.0, "step": 1639 }, { "epoch": 0.2167591858313508, "grad_norm": 0.8459386153345732, "learning_rate": 9.632505306974047e-06, "loss": 0.6545, "mean_token_accuracy": 0.792592465877533, "num_tokens": 107451165.0, "step": 1640 }, { "epoch": 0.2168913560666138, "grad_norm": 1.0760178560923754, "learning_rate": 9.631683077804912e-06, "loss": 0.7096, "mean_token_accuracy": 0.7698999047279358, "num_tokens": 107516701.0, "step": 1641 }, { "epoch": 0.21702352630187682, "grad_norm": 0.9123784373972544, "learning_rate": 9.630859969086715e-06, "loss": 0.6268, "mean_token_accuracy": 0.7966060042381287, "num_tokens": 107582237.0, "step": 1642 }, { "epoch": 0.21715569653713984, "grad_norm": 0.9394532037144382, "learning_rate": 9.630035980994676e-06, "loss": 0.6486, "mean_token_accuracy": 0.793401300907135, "num_tokens": 107647773.0, "step": 1643 }, { "epoch": 0.21728786677240286, "grad_norm": 0.9238138648859595, "learning_rate": 9.629211113704205e-06, "loss": 0.6999, "mean_token_accuracy": 0.7772402763366699, "num_tokens": 107713309.0, "step": 1644 }, { "epoch": 0.21742003700766588, "grad_norm": 0.9534112788085889, "learning_rate": 9.6283853673909e-06, "loss": 0.6498, "mean_token_accuracy": 0.7903797030448914, "num_tokens": 107778845.0, "step": 1645 }, { "epoch": 0.2175522072429289, "grad_norm": 0.9756141584447888, "learning_rate": 9.627558742230544e-06, "loss": 0.662, "mean_token_accuracy": 0.7853741645812988, "num_tokens": 107844381.0, "step": 1646 }, { "epoch": 0.21768437747819191, "grad_norm": 1.0519930538734898, "learning_rate": 9.62673123839911e-06, "loss": 0.7124, "mean_token_accuracy": 0.7685874700546265, "num_tokens": 107909917.0, "step": 1647 }, { "epoch": 0.21781654771345493, "grad_norm": 0.972511791994036, "learning_rate": 9.625902856072757e-06, "loss": 0.6729, "mean_token_accuracy": 0.7834513783454895, "num_tokens": 107975453.0, "step": 1648 }, { "epoch": 0.21794871794871795, "grad_norm": 0.9280795600691134, "learning_rate": 9.625073595427826e-06, "loss": 0.6688, "mean_token_accuracy": 0.783893883228302, "num_tokens": 108040989.0, "step": 1649 }, { "epoch": 0.21808088818398097, "grad_norm": 0.885390369896784, "learning_rate": 9.624243456640853e-06, "loss": 0.6032, "mean_token_accuracy": 0.8043889403343201, "num_tokens": 108106525.0, "step": 1650 }, { "epoch": 0.218213058419244, "grad_norm": 1.0252911054450393, "learning_rate": 9.623412439888556e-06, "loss": 0.6957, "mean_token_accuracy": 0.7784000635147095, "num_tokens": 108172061.0, "step": 1651 }, { "epoch": 0.218345228654507, "grad_norm": 1.006613900133797, "learning_rate": 9.62258054534784e-06, "loss": 0.6628, "mean_token_accuracy": 0.7855420708656311, "num_tokens": 108237597.0, "step": 1652 }, { "epoch": 0.21847739888977002, "grad_norm": 0.8610590617097599, "learning_rate": 9.6217477731958e-06, "loss": 0.6249, "mean_token_accuracy": 0.8003448843955994, "num_tokens": 108303133.0, "step": 1653 }, { "epoch": 0.21860956912503304, "grad_norm": 0.9236784177742748, "learning_rate": 9.620914123609715e-06, "loss": 0.6815, "mean_token_accuracy": 0.7807959914207458, "num_tokens": 108368669.0, "step": 1654 }, { "epoch": 0.21874173936029606, "grad_norm": 0.88883993414048, "learning_rate": 9.620079596767052e-06, "loss": 0.6352, "mean_token_accuracy": 0.793218195438385, "num_tokens": 108434205.0, "step": 1655 }, { "epoch": 0.21887390959555908, "grad_norm": 0.9181167812346851, "learning_rate": 9.619244192845463e-06, "loss": 0.6278, "mean_token_accuracy": 0.7969112396240234, "num_tokens": 108499741.0, "step": 1656 }, { "epoch": 0.2190060798308221, "grad_norm": 0.9808774512689974, "learning_rate": 9.618407912022789e-06, "loss": 0.6945, "mean_token_accuracy": 0.7746612429618835, "num_tokens": 108565277.0, "step": 1657 }, { "epoch": 0.21913825006608512, "grad_norm": 0.9117488604454976, "learning_rate": 9.617570754477056e-06, "loss": 0.6304, "mean_token_accuracy": 0.7930719256401062, "num_tokens": 108625475.0, "step": 1658 }, { "epoch": 0.21927042030134813, "grad_norm": 0.9473917202686751, "learning_rate": 9.61673272038648e-06, "loss": 0.6497, "mean_token_accuracy": 0.78813636302948, "num_tokens": 108691011.0, "step": 1659 }, { "epoch": 0.21940259053661115, "grad_norm": 1.0247585232242011, "learning_rate": 9.615893809929458e-06, "loss": 0.7088, "mean_token_accuracy": 0.7716548442840576, "num_tokens": 108756547.0, "step": 1660 }, { "epoch": 0.21953476077187417, "grad_norm": 0.9025305792653157, "learning_rate": 9.615054023284578e-06, "loss": 0.6581, "mean_token_accuracy": 0.7868544459342957, "num_tokens": 108822083.0, "step": 1661 }, { "epoch": 0.2196669310071372, "grad_norm": 0.8751807338868604, "learning_rate": 9.614213360630614e-06, "loss": 0.6232, "mean_token_accuracy": 0.7994292378425598, "num_tokens": 108887619.0, "step": 1662 }, { "epoch": 0.2197991012424002, "grad_norm": 0.8387761034632993, "learning_rate": 9.61337182214652e-06, "loss": 0.5831, "mean_token_accuracy": 0.8127670884132385, "num_tokens": 108953155.0, "step": 1663 }, { "epoch": 0.21993127147766323, "grad_norm": 0.8854687599751099, "learning_rate": 9.61252940801145e-06, "loss": 0.6074, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 109018691.0, "step": 1664 }, { "epoch": 0.22006344171292624, "grad_norm": 0.9018300193482071, "learning_rate": 9.61168611840473e-06, "loss": 0.6538, "mean_token_accuracy": 0.7867934107780457, "num_tokens": 109084227.0, "step": 1665 }, { "epoch": 0.22019561194818926, "grad_norm": 0.8270614717811327, "learning_rate": 9.610841953505883e-06, "loss": 0.5776, "mean_token_accuracy": 0.8128586411476135, "num_tokens": 109149763.0, "step": 1666 }, { "epoch": 0.22032778218345228, "grad_norm": 0.9103998431449722, "learning_rate": 9.609996913494612e-06, "loss": 0.6263, "mean_token_accuracy": 0.8014283776283264, "num_tokens": 109215299.0, "step": 1667 }, { "epoch": 0.2204599524187153, "grad_norm": 0.9680445058661334, "learning_rate": 9.609150998550808e-06, "loss": 0.6597, "mean_token_accuracy": 0.7858625054359436, "num_tokens": 109280835.0, "step": 1668 }, { "epoch": 0.22059212265397832, "grad_norm": 0.9647725382233888, "learning_rate": 9.60830420885455e-06, "loss": 0.6546, "mean_token_accuracy": 0.784855306148529, "num_tokens": 109346371.0, "step": 1669 }, { "epoch": 0.22072429288924134, "grad_norm": 0.8836218186469366, "learning_rate": 9.6074565445861e-06, "loss": 0.5617, "mean_token_accuracy": 0.8133164644241333, "num_tokens": 109411907.0, "step": 1670 }, { "epoch": 0.22085646312450435, "grad_norm": 0.866793120090719, "learning_rate": 9.60660800592591e-06, "loss": 0.6014, "mean_token_accuracy": 0.806433916091919, "num_tokens": 109477443.0, "step": 1671 }, { "epoch": 0.22098863335976737, "grad_norm": 1.039205698149019, "learning_rate": 9.605758593054615e-06, "loss": 0.7259, "mean_token_accuracy": 0.7659168839454651, "num_tokens": 109542979.0, "step": 1672 }, { "epoch": 0.2211208035950304, "grad_norm": 0.9112030689108493, "learning_rate": 9.604908306153037e-06, "loss": 0.6684, "mean_token_accuracy": 0.7828409075737, "num_tokens": 109608515.0, "step": 1673 }, { "epoch": 0.2212529738302934, "grad_norm": 0.8555709902287687, "learning_rate": 9.604057145402186e-06, "loss": 0.6618, "mean_token_accuracy": 0.7859236001968384, "num_tokens": 109674051.0, "step": 1674 }, { "epoch": 0.22138514406555643, "grad_norm": 0.8713294060756834, "learning_rate": 9.603205110983256e-06, "loss": 0.6286, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 109739587.0, "step": 1675 }, { "epoch": 0.22151731430081945, "grad_norm": 0.9187817633555885, "learning_rate": 9.602352203077625e-06, "loss": 0.6567, "mean_token_accuracy": 0.7903644442558289, "num_tokens": 109805123.0, "step": 1676 }, { "epoch": 0.22164948453608246, "grad_norm": 0.9615589479308542, "learning_rate": 9.601498421866862e-06, "loss": 0.6838, "mean_token_accuracy": 0.7792394161224365, "num_tokens": 109870659.0, "step": 1677 }, { "epoch": 0.22178165477134548, "grad_norm": 0.9279028358914845, "learning_rate": 9.600643767532717e-06, "loss": 0.6899, "mean_token_accuracy": 0.779071569442749, "num_tokens": 109936195.0, "step": 1678 }, { "epoch": 0.2219138250066085, "grad_norm": 0.8366181048400669, "learning_rate": 9.59978824025713e-06, "loss": 0.6424, "mean_token_accuracy": 0.7921193838119507, "num_tokens": 110001731.0, "step": 1679 }, { "epoch": 0.22204599524187152, "grad_norm": 0.8414835127573824, "learning_rate": 9.598931840222225e-06, "loss": 0.6488, "mean_token_accuracy": 0.793645441532135, "num_tokens": 110067267.0, "step": 1680 }, { "epoch": 0.22217816547713454, "grad_norm": 0.916298867138001, "learning_rate": 9.598074567610308e-06, "loss": 0.6464, "mean_token_accuracy": 0.7902117967605591, "num_tokens": 110132803.0, "step": 1681 }, { "epoch": 0.22231033571239756, "grad_norm": 0.900506820212438, "learning_rate": 9.59721642260388e-06, "loss": 0.5967, "mean_token_accuracy": 0.8108137249946594, "num_tokens": 110198339.0, "step": 1682 }, { "epoch": 0.22244250594766057, "grad_norm": 0.8717547357225575, "learning_rate": 9.596357405385617e-06, "loss": 0.5823, "mean_token_accuracy": 0.8114393949508667, "num_tokens": 110263875.0, "step": 1683 }, { "epoch": 0.2225746761829236, "grad_norm": 0.8384641875606359, "learning_rate": 9.59549751613839e-06, "loss": 0.6122, "mean_token_accuracy": 0.8030765652656555, "num_tokens": 110329411.0, "step": 1684 }, { "epoch": 0.22270684641818664, "grad_norm": 0.9233702219865482, "learning_rate": 9.594636755045249e-06, "loss": 0.5888, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 110394947.0, "step": 1685 }, { "epoch": 0.22283901665344966, "grad_norm": 1.0895071927082993, "learning_rate": 9.593775122289431e-06, "loss": 0.691, "mean_token_accuracy": 0.7756684422492981, "num_tokens": 110460483.0, "step": 1686 }, { "epoch": 0.22297118688871267, "grad_norm": 1.0262019120212662, "learning_rate": 9.592912618054363e-06, "loss": 0.6994, "mean_token_accuracy": 0.7711970210075378, "num_tokens": 110526019.0, "step": 1687 }, { "epoch": 0.2231033571239757, "grad_norm": 0.9047082193697615, "learning_rate": 9.592049242523654e-06, "loss": 0.6217, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 110591555.0, "step": 1688 }, { "epoch": 0.2232355273592387, "grad_norm": 1.0185608569059437, "learning_rate": 9.591184995881095e-06, "loss": 0.7062, "mean_token_accuracy": 0.7723263502120972, "num_tokens": 110657091.0, "step": 1689 }, { "epoch": 0.22336769759450173, "grad_norm": 0.835285637422747, "learning_rate": 9.59031987831067e-06, "loss": 0.557, "mean_token_accuracy": 0.8191002607345581, "num_tokens": 110722627.0, "step": 1690 }, { "epoch": 0.22349986782976475, "grad_norm": 1.0497076968452759, "learning_rate": 9.589453889996539e-06, "loss": 0.7421, "mean_token_accuracy": 0.7618269920349121, "num_tokens": 110788163.0, "step": 1691 }, { "epoch": 0.22363203806502777, "grad_norm": 0.9101438081513669, "learning_rate": 9.588587031123061e-06, "loss": 0.6456, "mean_token_accuracy": 0.793462336063385, "num_tokens": 110853699.0, "step": 1692 }, { "epoch": 0.22376420830029078, "grad_norm": 0.8880244375074078, "learning_rate": 9.587719301874765e-06, "loss": 0.6109, "mean_token_accuracy": 0.8052740693092346, "num_tokens": 110919235.0, "step": 1693 }, { "epoch": 0.2238963785355538, "grad_norm": 0.9612777670776533, "learning_rate": 9.586850702436376e-06, "loss": 0.67, "mean_token_accuracy": 0.7824593782424927, "num_tokens": 110984771.0, "step": 1694 }, { "epoch": 0.22402854877081682, "grad_norm": 0.9153666091689532, "learning_rate": 9.585981232992798e-06, "loss": 0.6242, "mean_token_accuracy": 0.7990477085113525, "num_tokens": 111050307.0, "step": 1695 }, { "epoch": 0.22416071900607984, "grad_norm": 0.9602110925718562, "learning_rate": 9.585110893729126e-06, "loss": 0.6663, "mean_token_accuracy": 0.7849469184875488, "num_tokens": 111115843.0, "step": 1696 }, { "epoch": 0.22429288924134286, "grad_norm": 0.8067597239168758, "learning_rate": 9.584239684830636e-06, "loss": 0.5755, "mean_token_accuracy": 0.8176962733268738, "num_tokens": 111181379.0, "step": 1697 }, { "epoch": 0.22442505947660588, "grad_norm": 0.8663762255253344, "learning_rate": 9.58336760648279e-06, "loss": 0.6096, "mean_token_accuracy": 0.8014436364173889, "num_tokens": 111246915.0, "step": 1698 }, { "epoch": 0.2245572297118689, "grad_norm": 0.8694591956206948, "learning_rate": 9.582494658871234e-06, "loss": 0.5961, "mean_token_accuracy": 0.8061286807060242, "num_tokens": 111312451.0, "step": 1699 }, { "epoch": 0.2246893999471319, "grad_norm": 0.9189852273440423, "learning_rate": 9.581620842181803e-06, "loss": 0.6691, "mean_token_accuracy": 0.7867476344108582, "num_tokens": 111377987.0, "step": 1700 }, { "epoch": 0.22482157018239493, "grad_norm": 0.9724533160954082, "learning_rate": 9.58074615660051e-06, "loss": 0.6618, "mean_token_accuracy": 0.7874038815498352, "num_tokens": 111443523.0, "step": 1701 }, { "epoch": 0.22495374041765795, "grad_norm": 0.8922929371862186, "learning_rate": 9.579870602313564e-06, "loss": 0.698, "mean_token_accuracy": 0.7764924764633179, "num_tokens": 111509059.0, "step": 1702 }, { "epoch": 0.22508591065292097, "grad_norm": 0.82478283336254, "learning_rate": 9.578994179507347e-06, "loss": 0.6115, "mean_token_accuracy": 0.8027865886688232, "num_tokens": 111574595.0, "step": 1703 }, { "epoch": 0.22521808088818399, "grad_norm": 1.0116739986464884, "learning_rate": 9.57811688836843e-06, "loss": 0.7007, "mean_token_accuracy": 0.7716243267059326, "num_tokens": 111640131.0, "step": 1704 }, { "epoch": 0.225350251123447, "grad_norm": 0.9845841326270807, "learning_rate": 9.577238729083575e-06, "loss": 0.678, "mean_token_accuracy": 0.7807654738426208, "num_tokens": 111705667.0, "step": 1705 }, { "epoch": 0.22548242135871002, "grad_norm": 0.9959533426713644, "learning_rate": 9.576359701839721e-06, "loss": 0.6675, "mean_token_accuracy": 0.7847179770469666, "num_tokens": 111771203.0, "step": 1706 }, { "epoch": 0.22561459159397304, "grad_norm": 0.8719639919464789, "learning_rate": 9.575479806823994e-06, "loss": 0.6525, "mean_token_accuracy": 0.7895098328590393, "num_tokens": 111836739.0, "step": 1707 }, { "epoch": 0.22574676182923606, "grad_norm": 0.8736829819915821, "learning_rate": 9.574599044223706e-06, "loss": 0.6245, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 111902275.0, "step": 1708 }, { "epoch": 0.22587893206449908, "grad_norm": 0.9786606559539217, "learning_rate": 9.573717414226353e-06, "loss": 0.6194, "mean_token_accuracy": 0.7999175786972046, "num_tokens": 111967811.0, "step": 1709 }, { "epoch": 0.2260111022997621, "grad_norm": 0.9862321840136307, "learning_rate": 9.572834917019616e-06, "loss": 0.6325, "mean_token_accuracy": 0.7952478528022766, "num_tokens": 112033347.0, "step": 1710 }, { "epoch": 0.2261432725350251, "grad_norm": 0.9142193257261422, "learning_rate": 9.57195155279136e-06, "loss": 0.5976, "mean_token_accuracy": 0.8067543506622314, "num_tokens": 112098883.0, "step": 1711 }, { "epoch": 0.22627544277028813, "grad_norm": 0.9096385892881067, "learning_rate": 9.571067321729634e-06, "loss": 0.5952, "mean_token_accuracy": 0.8091045022010803, "num_tokens": 112164419.0, "step": 1712 }, { "epoch": 0.22640761300555115, "grad_norm": 0.9096203323607365, "learning_rate": 9.570182224022674e-06, "loss": 0.6425, "mean_token_accuracy": 0.7913869023323059, "num_tokens": 112229955.0, "step": 1713 }, { "epoch": 0.22653978324081417, "grad_norm": 0.9385981547916755, "learning_rate": 9.569296259858896e-06, "loss": 0.6317, "mean_token_accuracy": 0.7965602278709412, "num_tokens": 112295491.0, "step": 1714 }, { "epoch": 0.2266719534760772, "grad_norm": 0.9254115177651856, "learning_rate": 9.568409429426906e-06, "loss": 0.6234, "mean_token_accuracy": 0.7998870611190796, "num_tokens": 112361027.0, "step": 1715 }, { "epoch": 0.2268041237113402, "grad_norm": 0.9402637377117012, "learning_rate": 9.567521732915492e-06, "loss": 0.62, "mean_token_accuracy": 0.7950341701507568, "num_tokens": 112426563.0, "step": 1716 }, { "epoch": 0.22693629394660322, "grad_norm": 0.9392198782914645, "learning_rate": 9.566633170513622e-06, "loss": 0.6549, "mean_token_accuracy": 0.7878464460372925, "num_tokens": 112492099.0, "step": 1717 }, { "epoch": 0.22706846418186624, "grad_norm": 0.909681051332698, "learning_rate": 9.565743742410456e-06, "loss": 0.6229, "mean_token_accuracy": 0.7966976165771484, "num_tokens": 112557635.0, "step": 1718 }, { "epoch": 0.22720063441712926, "grad_norm": 0.8912270467166583, "learning_rate": 9.564853448795337e-06, "loss": 0.6368, "mean_token_accuracy": 0.7912342548370361, "num_tokens": 112623171.0, "step": 1719 }, { "epoch": 0.22733280465239228, "grad_norm": 0.9136759000882673, "learning_rate": 9.563962289857783e-06, "loss": 0.6435, "mean_token_accuracy": 0.792409360408783, "num_tokens": 112688707.0, "step": 1720 }, { "epoch": 0.2274649748876553, "grad_norm": 1.0331778382108843, "learning_rate": 9.56307026578751e-06, "loss": 0.6688, "mean_token_accuracy": 0.7796819806098938, "num_tokens": 112754243.0, "step": 1721 }, { "epoch": 0.22759714512291832, "grad_norm": 1.1338237171716021, "learning_rate": 9.562177376774406e-06, "loss": 0.7308, "mean_token_accuracy": 0.7678244709968567, "num_tokens": 112819779.0, "step": 1722 }, { "epoch": 0.22772931535818133, "grad_norm": 0.9072299642738041, "learning_rate": 9.561283623008551e-06, "loss": 0.6162, "mean_token_accuracy": 0.7984525561332703, "num_tokens": 112885315.0, "step": 1723 }, { "epoch": 0.22786148559344435, "grad_norm": 0.8986072889771849, "learning_rate": 9.560389004680207e-06, "loss": 0.6415, "mean_token_accuracy": 0.7929434776306152, "num_tokens": 112950851.0, "step": 1724 }, { "epoch": 0.22799365582870737, "grad_norm": 1.0011480400730555, "learning_rate": 9.559493521979818e-06, "loss": 0.6835, "mean_token_accuracy": 0.7778506875038147, "num_tokens": 113016387.0, "step": 1725 }, { "epoch": 0.2281258260639704, "grad_norm": 0.9365786918325993, "learning_rate": 9.558597175098014e-06, "loss": 0.6414, "mean_token_accuracy": 0.7918752431869507, "num_tokens": 113081923.0, "step": 1726 }, { "epoch": 0.2282579962992334, "grad_norm": 1.060058090632705, "learning_rate": 9.55769996422561e-06, "loss": 0.6638, "mean_token_accuracy": 0.7876785397529602, "num_tokens": 113147459.0, "step": 1727 }, { "epoch": 0.22839016653449642, "grad_norm": 0.891698701152852, "learning_rate": 9.5568018895536e-06, "loss": 0.6268, "mean_token_accuracy": 0.7931113243103027, "num_tokens": 113212995.0, "step": 1728 }, { "epoch": 0.22852233676975944, "grad_norm": 0.957708918375653, "learning_rate": 9.555902951273169e-06, "loss": 0.6395, "mean_token_accuracy": 0.7944847941398621, "num_tokens": 113278531.0, "step": 1729 }, { "epoch": 0.22865450700502246, "grad_norm": 1.0534184442776535, "learning_rate": 9.555003149575682e-06, "loss": 0.6365, "mean_token_accuracy": 0.7972622513771057, "num_tokens": 113344067.0, "step": 1730 }, { "epoch": 0.22878667724028548, "grad_norm": 0.8981785122100139, "learning_rate": 9.554102484652685e-06, "loss": 0.54, "mean_token_accuracy": 0.8249908685684204, "num_tokens": 113409603.0, "step": 1731 }, { "epoch": 0.2289188474755485, "grad_norm": 1.0246093787940425, "learning_rate": 9.553200956695915e-06, "loss": 0.6577, "mean_token_accuracy": 0.7870833873748779, "num_tokens": 113475139.0, "step": 1732 }, { "epoch": 0.22905101771081152, "grad_norm": 1.0632267057365188, "learning_rate": 9.552298565897284e-06, "loss": 0.6777, "mean_token_accuracy": 0.7772402763366699, "num_tokens": 113540675.0, "step": 1733 }, { "epoch": 0.22918318794607453, "grad_norm": 0.9585412502045005, "learning_rate": 9.551395312448895e-06, "loss": 0.6706, "mean_token_accuracy": 0.7838175892829895, "num_tokens": 113606211.0, "step": 1734 }, { "epoch": 0.22931535818133755, "grad_norm": 0.9730006364300338, "learning_rate": 9.55049119654303e-06, "loss": 0.6754, "mean_token_accuracy": 0.7806586623191833, "num_tokens": 113671747.0, "step": 1735 }, { "epoch": 0.22944752841660057, "grad_norm": 0.9978412384817414, "learning_rate": 9.549586218372159e-06, "loss": 0.6125, "mean_token_accuracy": 0.8012300133705139, "num_tokens": 113737283.0, "step": 1736 }, { "epoch": 0.2295796986518636, "grad_norm": 0.9554238428051832, "learning_rate": 9.548680378128929e-06, "loss": 0.6184, "mean_token_accuracy": 0.8003143668174744, "num_tokens": 113802819.0, "step": 1737 }, { "epoch": 0.2297118688871266, "grad_norm": 0.9403971956032114, "learning_rate": 9.547773676006178e-06, "loss": 0.6255, "mean_token_accuracy": 0.7970943450927734, "num_tokens": 113868355.0, "step": 1738 }, { "epoch": 0.22984403912238963, "grad_norm": 1.0336375318544322, "learning_rate": 9.546866112196924e-06, "loss": 0.6381, "mean_token_accuracy": 0.7941948771476746, "num_tokens": 113933891.0, "step": 1739 }, { "epoch": 0.22997620935765264, "grad_norm": 0.9938138720583297, "learning_rate": 9.545957686894367e-06, "loss": 0.579, "mean_token_accuracy": 0.8098217844963074, "num_tokens": 113999427.0, "step": 1740 }, { "epoch": 0.23010837959291566, "grad_norm": 0.931469436562264, "learning_rate": 9.545048400291891e-06, "loss": 0.6417, "mean_token_accuracy": 0.7918294668197632, "num_tokens": 114064963.0, "step": 1741 }, { "epoch": 0.23024054982817868, "grad_norm": 0.9350555450310755, "learning_rate": 9.544138252583066e-06, "loss": 0.6391, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 114130499.0, "step": 1742 }, { "epoch": 0.2303727200634417, "grad_norm": 0.981459648304761, "learning_rate": 9.543227243961641e-06, "loss": 0.6172, "mean_token_accuracy": 0.7991545796394348, "num_tokens": 114196035.0, "step": 1743 }, { "epoch": 0.23050489029870475, "grad_norm": 1.0111489822885797, "learning_rate": 9.542315374621554e-06, "loss": 0.6972, "mean_token_accuracy": 0.7760041356086731, "num_tokens": 114261571.0, "step": 1744 }, { "epoch": 0.23063706053396776, "grad_norm": 0.9405139083656008, "learning_rate": 9.541402644756921e-06, "loss": 0.6579, "mean_token_accuracy": 0.7870376110076904, "num_tokens": 114327107.0, "step": 1745 }, { "epoch": 0.23076923076923078, "grad_norm": 0.9626351867898938, "learning_rate": 9.540489054562042e-06, "loss": 0.6725, "mean_token_accuracy": 0.7827951312065125, "num_tokens": 114392643.0, "step": 1746 }, { "epoch": 0.2309014010044938, "grad_norm": 0.8869523993873011, "learning_rate": 9.539574604231403e-06, "loss": 0.6121, "mean_token_accuracy": 0.8017488718032837, "num_tokens": 114458179.0, "step": 1747 }, { "epoch": 0.23103357123975682, "grad_norm": 0.9513566172830842, "learning_rate": 9.538659293959673e-06, "loss": 0.6437, "mean_token_accuracy": 0.7916615605354309, "num_tokens": 114523715.0, "step": 1748 }, { "epoch": 0.23116574147501984, "grad_norm": 0.8986070829871733, "learning_rate": 9.537743123941697e-06, "loss": 0.6491, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 114589251.0, "step": 1749 }, { "epoch": 0.23129791171028286, "grad_norm": 0.9109411462342658, "learning_rate": 9.536826094372514e-06, "loss": 0.6293, "mean_token_accuracy": 0.797796368598938, "num_tokens": 114654787.0, "step": 1750 }, { "epoch": 0.23143008194554587, "grad_norm": 0.9630063618408642, "learning_rate": 9.535908205447338e-06, "loss": 0.6581, "mean_token_accuracy": 0.7878158688545227, "num_tokens": 114720323.0, "step": 1751 }, { "epoch": 0.2315622521808089, "grad_norm": 0.8835217598389769, "learning_rate": 9.53498945736157e-06, "loss": 0.6467, "mean_token_accuracy": 0.7910511493682861, "num_tokens": 114785859.0, "step": 1752 }, { "epoch": 0.2316944224160719, "grad_norm": 0.9967870132547957, "learning_rate": 9.53406985031079e-06, "loss": 0.6846, "mean_token_accuracy": 0.7790563106536865, "num_tokens": 114851395.0, "step": 1753 }, { "epoch": 0.23182659265133493, "grad_norm": 0.851077914583138, "learning_rate": 9.533149384490765e-06, "loss": 0.6171, "mean_token_accuracy": 0.7995971441268921, "num_tokens": 114916931.0, "step": 1754 }, { "epoch": 0.23195876288659795, "grad_norm": 0.8892569080202939, "learning_rate": 9.53222806009744e-06, "loss": 0.6255, "mean_token_accuracy": 0.7978116273880005, "num_tokens": 114982467.0, "step": 1755 }, { "epoch": 0.23209093312186097, "grad_norm": 0.8729834215561599, "learning_rate": 9.531305877326949e-06, "loss": 0.6261, "mean_token_accuracy": 0.7971554398536682, "num_tokens": 115048003.0, "step": 1756 }, { "epoch": 0.23222310335712398, "grad_norm": 1.0111507112971583, "learning_rate": 9.530382836375604e-06, "loss": 0.691, "mean_token_accuracy": 0.779407262802124, "num_tokens": 115113539.0, "step": 1757 }, { "epoch": 0.232355273592387, "grad_norm": 0.8977026022179888, "learning_rate": 9.529458937439902e-06, "loss": 0.6432, "mean_token_accuracy": 0.7913258671760559, "num_tokens": 115179075.0, "step": 1758 }, { "epoch": 0.23248744382765002, "grad_norm": 0.9450806605925338, "learning_rate": 9.528534180716522e-06, "loss": 0.6713, "mean_token_accuracy": 0.7846874594688416, "num_tokens": 115244611.0, "step": 1759 }, { "epoch": 0.23261961406291304, "grad_norm": 0.9831689089117459, "learning_rate": 9.527608566402323e-06, "loss": 0.6786, "mean_token_accuracy": 0.7801245450973511, "num_tokens": 115310147.0, "step": 1760 }, { "epoch": 0.23275178429817606, "grad_norm": 0.983828503894242, "learning_rate": 9.526682094694352e-06, "loss": 0.6317, "mean_token_accuracy": 0.7967433929443359, "num_tokens": 115375683.0, "step": 1761 }, { "epoch": 0.23288395453343907, "grad_norm": 0.9770409658490196, "learning_rate": 9.525754765789835e-06, "loss": 0.6821, "mean_token_accuracy": 0.7804144620895386, "num_tokens": 115441219.0, "step": 1762 }, { "epoch": 0.2330161247687021, "grad_norm": 0.8824297821426986, "learning_rate": 9.524826579886178e-06, "loss": 0.6244, "mean_token_accuracy": 0.8009705543518066, "num_tokens": 115506755.0, "step": 1763 }, { "epoch": 0.2331482950039651, "grad_norm": 1.0409518338091426, "learning_rate": 9.523897537180976e-06, "loss": 0.6809, "mean_token_accuracy": 0.7775607109069824, "num_tokens": 115572291.0, "step": 1764 }, { "epoch": 0.23328046523922813, "grad_norm": 0.9788798471411949, "learning_rate": 9.522967637872e-06, "loss": 0.6686, "mean_token_accuracy": 0.7831461429595947, "num_tokens": 115637827.0, "step": 1765 }, { "epoch": 0.23341263547449115, "grad_norm": 0.9548203844559258, "learning_rate": 9.522036882157208e-06, "loss": 0.6544, "mean_token_accuracy": 0.7869918346405029, "num_tokens": 115703363.0, "step": 1766 }, { "epoch": 0.23354480570975417, "grad_norm": 0.8863347324429928, "learning_rate": 9.521105270234737e-06, "loss": 0.6034, "mean_token_accuracy": 0.8042058348655701, "num_tokens": 115768899.0, "step": 1767 }, { "epoch": 0.23367697594501718, "grad_norm": 0.9507668733887814, "learning_rate": 9.520172802302908e-06, "loss": 0.6283, "mean_token_accuracy": 0.79897141456604, "num_tokens": 115834435.0, "step": 1768 }, { "epoch": 0.2338091461802802, "grad_norm": 1.0705024871930293, "learning_rate": 9.519239478560227e-06, "loss": 0.6796, "mean_token_accuracy": 0.7788273692131042, "num_tokens": 115899971.0, "step": 1769 }, { "epoch": 0.23394131641554322, "grad_norm": 0.9944554362641693, "learning_rate": 9.518305299205375e-06, "loss": 0.683, "mean_token_accuracy": 0.7794225215911865, "num_tokens": 115965507.0, "step": 1770 }, { "epoch": 0.23407348665080624, "grad_norm": 0.9235674166191595, "learning_rate": 9.517370264437223e-06, "loss": 0.6585, "mean_token_accuracy": 0.7872512340545654, "num_tokens": 116031043.0, "step": 1771 }, { "epoch": 0.23420565688606926, "grad_norm": 0.9737735662797847, "learning_rate": 9.516434374454816e-06, "loss": 0.6226, "mean_token_accuracy": 0.7998718023300171, "num_tokens": 116096579.0, "step": 1772 }, { "epoch": 0.23433782712133228, "grad_norm": 1.0171293051243362, "learning_rate": 9.515497629457387e-06, "loss": 0.6553, "mean_token_accuracy": 0.7877853512763977, "num_tokens": 116162115.0, "step": 1773 }, { "epoch": 0.2344699973565953, "grad_norm": 0.9263620542348565, "learning_rate": 9.514560029644353e-06, "loss": 0.5983, "mean_token_accuracy": 0.8038853406906128, "num_tokens": 116227651.0, "step": 1774 }, { "epoch": 0.2346021675918583, "grad_norm": 0.8616439807116294, "learning_rate": 9.513621575215306e-06, "loss": 0.6057, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 116293187.0, "step": 1775 }, { "epoch": 0.23473433782712133, "grad_norm": 0.8387625738177749, "learning_rate": 9.512682266370024e-06, "loss": 0.6317, "mean_token_accuracy": 0.7961787581443787, "num_tokens": 116358723.0, "step": 1776 }, { "epoch": 0.23486650806238435, "grad_norm": 0.9121710649952514, "learning_rate": 9.511742103308464e-06, "loss": 0.6516, "mean_token_accuracy": 0.792653501033783, "num_tokens": 116424259.0, "step": 1777 }, { "epoch": 0.23499867829764737, "grad_norm": 1.098113907421151, "learning_rate": 9.510801086230773e-06, "loss": 0.6973, "mean_token_accuracy": 0.773806631565094, "num_tokens": 116489795.0, "step": 1778 }, { "epoch": 0.23513084853291039, "grad_norm": 0.9369657744193574, "learning_rate": 9.509859215337268e-06, "loss": 0.6519, "mean_token_accuracy": 0.7888994216918945, "num_tokens": 116555331.0, "step": 1779 }, { "epoch": 0.2352630187681734, "grad_norm": 0.8279706332255649, "learning_rate": 9.508916490828456e-06, "loss": 0.6381, "mean_token_accuracy": 0.79337078332901, "num_tokens": 116620867.0, "step": 1780 }, { "epoch": 0.23539518900343642, "grad_norm": 0.920734419382891, "learning_rate": 9.507972912905024e-06, "loss": 0.6332, "mean_token_accuracy": 0.7935996651649475, "num_tokens": 116686403.0, "step": 1781 }, { "epoch": 0.23552735923869944, "grad_norm": 0.9760075926271272, "learning_rate": 9.507028481767838e-06, "loss": 0.6136, "mean_token_accuracy": 0.8035954236984253, "num_tokens": 116751939.0, "step": 1782 }, { "epoch": 0.23565952947396246, "grad_norm": 1.0235316872072395, "learning_rate": 9.50608319761795e-06, "loss": 0.6624, "mean_token_accuracy": 0.7838481068611145, "num_tokens": 116817475.0, "step": 1783 }, { "epoch": 0.23579169970922548, "grad_norm": 0.9674695420497033, "learning_rate": 9.505137060656591e-06, "loss": 0.7, "mean_token_accuracy": 0.7760346531867981, "num_tokens": 116883011.0, "step": 1784 }, { "epoch": 0.2359238699444885, "grad_norm": 0.8323816180532012, "learning_rate": 9.504190071085172e-06, "loss": 0.5986, "mean_token_accuracy": 0.808753490447998, "num_tokens": 116948547.0, "step": 1785 }, { "epoch": 0.23605604017975151, "grad_norm": 0.9231304500785604, "learning_rate": 9.503242229105288e-06, "loss": 0.6452, "mean_token_accuracy": 0.7897998094558716, "num_tokens": 117014083.0, "step": 1786 }, { "epoch": 0.23618821041501453, "grad_norm": 0.9675886368130625, "learning_rate": 9.502293534918716e-06, "loss": 0.6032, "mean_token_accuracy": 0.8073342442512512, "num_tokens": 117079619.0, "step": 1787 }, { "epoch": 0.23632038065027755, "grad_norm": 1.0402133207841135, "learning_rate": 9.501343988727411e-06, "loss": 0.6307, "mean_token_accuracy": 0.7952783703804016, "num_tokens": 117145155.0, "step": 1788 }, { "epoch": 0.23645255088554057, "grad_norm": 1.0962742017463678, "learning_rate": 9.500393590733514e-06, "loss": 0.6916, "mean_token_accuracy": 0.7780032753944397, "num_tokens": 117210691.0, "step": 1789 }, { "epoch": 0.2365847211208036, "grad_norm": 0.8703580572066396, "learning_rate": 9.499442341139344e-06, "loss": 0.6393, "mean_token_accuracy": 0.7939049005508423, "num_tokens": 117276227.0, "step": 1790 }, { "epoch": 0.2367168913560666, "grad_norm": 0.9780959367431494, "learning_rate": 9.498490240147401e-06, "loss": 0.6472, "mean_token_accuracy": 0.788853645324707, "num_tokens": 117341763.0, "step": 1791 }, { "epoch": 0.23684906159132962, "grad_norm": 1.0994172164699159, "learning_rate": 9.497537287960368e-06, "loss": 0.6267, "mean_token_accuracy": 0.7967739105224609, "num_tokens": 117407299.0, "step": 1792 }, { "epoch": 0.23698123182659264, "grad_norm": 1.0867119973717485, "learning_rate": 9.49658348478111e-06, "loss": 0.6219, "mean_token_accuracy": 0.7988340854644775, "num_tokens": 117472835.0, "step": 1793 }, { "epoch": 0.23711340206185566, "grad_norm": 0.957614507556028, "learning_rate": 9.49562883081267e-06, "loss": 0.6209, "mean_token_accuracy": 0.798818826675415, "num_tokens": 117538371.0, "step": 1794 }, { "epoch": 0.23724557229711868, "grad_norm": 0.9891367804497504, "learning_rate": 9.494673326258274e-06, "loss": 0.6808, "mean_token_accuracy": 0.7777438759803772, "num_tokens": 117603907.0, "step": 1795 }, { "epoch": 0.2373777425323817, "grad_norm": 0.8726047723478764, "learning_rate": 9.493716971321331e-06, "loss": 0.5887, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 117669443.0, "step": 1796 }, { "epoch": 0.23750991276764472, "grad_norm": 0.9617327007019884, "learning_rate": 9.492759766205427e-06, "loss": 0.6121, "mean_token_accuracy": 0.8005127310752869, "num_tokens": 117734979.0, "step": 1797 }, { "epoch": 0.23764208300290773, "grad_norm": 0.971036495779359, "learning_rate": 9.491801711114333e-06, "loss": 0.6056, "mean_token_accuracy": 0.8048467636108398, "num_tokens": 117800515.0, "step": 1798 }, { "epoch": 0.23777425323817075, "grad_norm": 0.9977557351854482, "learning_rate": 9.490842806251996e-06, "loss": 0.6359, "mean_token_accuracy": 0.7945000529289246, "num_tokens": 117866051.0, "step": 1799 }, { "epoch": 0.23790642347343377, "grad_norm": 0.9401287382733089, "learning_rate": 9.489883051822549e-06, "loss": 0.6226, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 117931587.0, "step": 1800 }, { "epoch": 0.2380385937086968, "grad_norm": 0.9394301712772216, "learning_rate": 9.488922448030302e-06, "loss": 0.6427, "mean_token_accuracy": 0.7895403504371643, "num_tokens": 117997123.0, "step": 1801 }, { "epoch": 0.2381707639439598, "grad_norm": 0.9423909029139864, "learning_rate": 9.48796099507975e-06, "loss": 0.5849, "mean_token_accuracy": 0.8124160766601562, "num_tokens": 118062659.0, "step": 1802 }, { "epoch": 0.23830293417922285, "grad_norm": 0.9750122341906503, "learning_rate": 9.486998693175564e-06, "loss": 0.6747, "mean_token_accuracy": 0.7842754125595093, "num_tokens": 118128195.0, "step": 1803 }, { "epoch": 0.23843510441448587, "grad_norm": 0.9095549215045416, "learning_rate": 9.486035542522598e-06, "loss": 0.6234, "mean_token_accuracy": 0.7973538041114807, "num_tokens": 118193731.0, "step": 1804 }, { "epoch": 0.2385672746497489, "grad_norm": 0.9294210544296613, "learning_rate": 9.485071543325888e-06, "loss": 0.609, "mean_token_accuracy": 0.8026645183563232, "num_tokens": 118259267.0, "step": 1805 }, { "epoch": 0.2386994448850119, "grad_norm": 0.9487351980817182, "learning_rate": 9.484106695790648e-06, "loss": 0.6597, "mean_token_accuracy": 0.7853589057922363, "num_tokens": 118324803.0, "step": 1806 }, { "epoch": 0.23883161512027493, "grad_norm": 0.9021619543170522, "learning_rate": 9.483141000122277e-06, "loss": 0.5997, "mean_token_accuracy": 0.8072732090950012, "num_tokens": 118390339.0, "step": 1807 }, { "epoch": 0.23896378535553794, "grad_norm": 1.024143567327278, "learning_rate": 9.482174456526348e-06, "loss": 0.703, "mean_token_accuracy": 0.7695794105529785, "num_tokens": 118455875.0, "step": 1808 }, { "epoch": 0.23909595559080096, "grad_norm": 0.8781357498060012, "learning_rate": 9.481207065208617e-06, "loss": 0.6167, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 118521411.0, "step": 1809 }, { "epoch": 0.23922812582606398, "grad_norm": 0.8948355471998354, "learning_rate": 9.480238826375025e-06, "loss": 0.6046, "mean_token_accuracy": 0.802282989025116, "num_tokens": 118586947.0, "step": 1810 }, { "epoch": 0.239360296061327, "grad_norm": 0.8823524276361052, "learning_rate": 9.47926974023169e-06, "loss": 0.6047, "mean_token_accuracy": 0.8035801649093628, "num_tokens": 118652483.0, "step": 1811 }, { "epoch": 0.23949246629659002, "grad_norm": 0.9188034435322265, "learning_rate": 9.478299806984907e-06, "loss": 0.6298, "mean_token_accuracy": 0.7955225110054016, "num_tokens": 118718019.0, "step": 1812 }, { "epoch": 0.23962463653185304, "grad_norm": 0.8626260851409737, "learning_rate": 9.477329026841154e-06, "loss": 0.5994, "mean_token_accuracy": 0.8055487871170044, "num_tokens": 118783555.0, "step": 1813 }, { "epoch": 0.23975680676711605, "grad_norm": 0.9336602398375997, "learning_rate": 9.476357400007095e-06, "loss": 0.6969, "mean_token_accuracy": 0.7726010084152222, "num_tokens": 118849091.0, "step": 1814 }, { "epoch": 0.23988897700237907, "grad_norm": 0.8878695624217039, "learning_rate": 9.475384926689564e-06, "loss": 0.6275, "mean_token_accuracy": 0.79343181848526, "num_tokens": 118914627.0, "step": 1815 }, { "epoch": 0.2400211472376421, "grad_norm": 0.9577532185957872, "learning_rate": 9.474411607095583e-06, "loss": 0.6088, "mean_token_accuracy": 0.8008332252502441, "num_tokens": 118980163.0, "step": 1816 }, { "epoch": 0.2401533174729051, "grad_norm": 0.8592423961092747, "learning_rate": 9.47343744143235e-06, "loss": 0.584, "mean_token_accuracy": 0.807395339012146, "num_tokens": 119045699.0, "step": 1817 }, { "epoch": 0.24028548770816813, "grad_norm": 0.8404818599905308, "learning_rate": 9.472462429907245e-06, "loss": 0.54, "mean_token_accuracy": 0.8274325728416443, "num_tokens": 119111235.0, "step": 1818 }, { "epoch": 0.24041765794343115, "grad_norm": 0.9065235335038258, "learning_rate": 9.471486572727828e-06, "loss": 0.6249, "mean_token_accuracy": 0.7983915209770203, "num_tokens": 119176771.0, "step": 1819 }, { "epoch": 0.24054982817869416, "grad_norm": 1.1007150204659122, "learning_rate": 9.470509870101839e-06, "loss": 0.6703, "mean_token_accuracy": 0.7847790122032166, "num_tokens": 119242307.0, "step": 1820 }, { "epoch": 0.24068199841395718, "grad_norm": 0.8766266248908138, "learning_rate": 9.469532322237197e-06, "loss": 0.5772, "mean_token_accuracy": 0.8139879107475281, "num_tokens": 119307843.0, "step": 1821 }, { "epoch": 0.2408141686492202, "grad_norm": 0.9324861370697489, "learning_rate": 9.468553929342001e-06, "loss": 0.6373, "mean_token_accuracy": 0.7930808067321777, "num_tokens": 119373379.0, "step": 1822 }, { "epoch": 0.24094633888448322, "grad_norm": 0.9830177634309687, "learning_rate": 9.46757469162453e-06, "loss": 0.6743, "mean_token_accuracy": 0.7789189219474792, "num_tokens": 119438915.0, "step": 1823 }, { "epoch": 0.24107850911974624, "grad_norm": 1.094645677116268, "learning_rate": 9.466594609293244e-06, "loss": 0.6488, "mean_token_accuracy": 0.7869918346405029, "num_tokens": 119504451.0, "step": 1824 }, { "epoch": 0.24121067935500926, "grad_norm": 1.011658555017231, "learning_rate": 9.465613682556784e-06, "loss": 0.7131, "mean_token_accuracy": 0.7725857496261597, "num_tokens": 119569987.0, "step": 1825 }, { "epoch": 0.24134284959027227, "grad_norm": 0.971048383159339, "learning_rate": 9.464631911623965e-06, "loss": 0.6739, "mean_token_accuracy": 0.7810554504394531, "num_tokens": 119635523.0, "step": 1826 }, { "epoch": 0.2414750198255353, "grad_norm": 0.9436663923492297, "learning_rate": 9.463649296703789e-06, "loss": 0.6377, "mean_token_accuracy": 0.7950189113616943, "num_tokens": 119701059.0, "step": 1827 }, { "epoch": 0.2416071900607983, "grad_norm": 1.093044159303958, "learning_rate": 9.462665838005432e-06, "loss": 0.6552, "mean_token_accuracy": 0.7864577174186707, "num_tokens": 119766595.0, "step": 1828 }, { "epoch": 0.24173936029606133, "grad_norm": 1.1981361093465523, "learning_rate": 9.461681535738255e-06, "loss": 0.6927, "mean_token_accuracy": 0.7739439606666565, "num_tokens": 119832131.0, "step": 1829 }, { "epoch": 0.24187153053132435, "grad_norm": 0.9446165338006295, "learning_rate": 9.460696390111791e-06, "loss": 0.6048, "mean_token_accuracy": 0.8040685057640076, "num_tokens": 119897667.0, "step": 1830 }, { "epoch": 0.24200370076658737, "grad_norm": 0.9265321231223395, "learning_rate": 9.459710401335758e-06, "loss": 0.6443, "mean_token_accuracy": 0.7905170321464539, "num_tokens": 119963203.0, "step": 1831 }, { "epoch": 0.24213587100185038, "grad_norm": 0.8932170737005273, "learning_rate": 9.458723569620054e-06, "loss": 0.6303, "mean_token_accuracy": 0.8003296256065369, "num_tokens": 120028739.0, "step": 1832 }, { "epoch": 0.2422680412371134, "grad_norm": 1.0513436710108381, "learning_rate": 9.457735895174756e-06, "loss": 0.6366, "mean_token_accuracy": 0.7952936291694641, "num_tokens": 120094275.0, "step": 1833 }, { "epoch": 0.24240021147237642, "grad_norm": 0.9797281509955714, "learning_rate": 9.456747378210117e-06, "loss": 0.6636, "mean_token_accuracy": 0.7845959067344666, "num_tokens": 120159811.0, "step": 1834 }, { "epoch": 0.24253238170763944, "grad_norm": 0.8359022697160189, "learning_rate": 9.45575801893657e-06, "loss": 0.5729, "mean_token_accuracy": 0.8140184283256531, "num_tokens": 120225347.0, "step": 1835 }, { "epoch": 0.24266455194290246, "grad_norm": 0.8608301953947987, "learning_rate": 9.454767817564731e-06, "loss": 0.591, "mean_token_accuracy": 0.808356761932373, "num_tokens": 120290883.0, "step": 1836 }, { "epoch": 0.24279672217816547, "grad_norm": 1.076431190705244, "learning_rate": 9.453776774305396e-06, "loss": 0.6456, "mean_token_accuracy": 0.7886399626731873, "num_tokens": 120356419.0, "step": 1837 }, { "epoch": 0.2429288924134285, "grad_norm": 0.9738480731187309, "learning_rate": 9.452784889369533e-06, "loss": 0.6395, "mean_token_accuracy": 0.7907764911651611, "num_tokens": 120421955.0, "step": 1838 }, { "epoch": 0.2430610626486915, "grad_norm": 0.9395254009627747, "learning_rate": 9.451792162968295e-06, "loss": 0.6212, "mean_token_accuracy": 0.7984220385551453, "num_tokens": 120487491.0, "step": 1839 }, { "epoch": 0.24319323288395453, "grad_norm": 0.9921967085677467, "learning_rate": 9.450798595313013e-06, "loss": 0.6233, "mean_token_accuracy": 0.7952478528022766, "num_tokens": 120553027.0, "step": 1840 }, { "epoch": 0.24332540311921755, "grad_norm": 0.9258901360103242, "learning_rate": 9.449804186615199e-06, "loss": 0.6713, "mean_token_accuracy": 0.7813148498535156, "num_tokens": 120618563.0, "step": 1841 }, { "epoch": 0.24345757335448057, "grad_norm": 0.9585442848944425, "learning_rate": 9.448808937086536e-06, "loss": 0.6948, "mean_token_accuracy": 0.7757294774055481, "num_tokens": 120684099.0, "step": 1842 }, { "epoch": 0.24358974358974358, "grad_norm": 0.9465315593402656, "learning_rate": 9.447812846938897e-06, "loss": 0.6464, "mean_token_accuracy": 0.792470395565033, "num_tokens": 120749635.0, "step": 1843 }, { "epoch": 0.2437219138250066, "grad_norm": 0.8854996653914763, "learning_rate": 9.446815916384329e-06, "loss": 0.6129, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 120815171.0, "step": 1844 }, { "epoch": 0.24385408406026962, "grad_norm": 1.0128035654184662, "learning_rate": 9.445818145635053e-06, "loss": 0.6465, "mean_token_accuracy": 0.7912189960479736, "num_tokens": 120880707.0, "step": 1845 }, { "epoch": 0.24398625429553264, "grad_norm": 0.8694186613047896, "learning_rate": 9.44481953490348e-06, "loss": 0.5779, "mean_token_accuracy": 0.8144915103912354, "num_tokens": 120946243.0, "step": 1846 }, { "epoch": 0.24411842453079566, "grad_norm": 0.919548192643193, "learning_rate": 9.443820084402187e-06, "loss": 0.6354, "mean_token_accuracy": 0.7929129600524902, "num_tokens": 121011779.0, "step": 1847 }, { "epoch": 0.24425059476605868, "grad_norm": 0.8360918455768028, "learning_rate": 9.442819794343941e-06, "loss": 0.6124, "mean_token_accuracy": 0.8018404245376587, "num_tokens": 121077315.0, "step": 1848 }, { "epoch": 0.2443827650013217, "grad_norm": 0.9519054963977452, "learning_rate": 9.44181866494168e-06, "loss": 0.6931, "mean_token_accuracy": 0.7741270661354065, "num_tokens": 121142851.0, "step": 1849 }, { "epoch": 0.2445149352365847, "grad_norm": 0.9686707920353679, "learning_rate": 9.440816696408524e-06, "loss": 0.6394, "mean_token_accuracy": 0.7939964532852173, "num_tokens": 121208387.0, "step": 1850 }, { "epoch": 0.24464710547184773, "grad_norm": 1.051825683139164, "learning_rate": 9.439813888957773e-06, "loss": 0.6893, "mean_token_accuracy": 0.7786290049552917, "num_tokens": 121273923.0, "step": 1851 }, { "epoch": 0.24477927570711075, "grad_norm": 0.8439170402332249, "learning_rate": 9.438810242802902e-06, "loss": 0.6242, "mean_token_accuracy": 0.7973690629005432, "num_tokens": 121339459.0, "step": 1852 }, { "epoch": 0.24491144594237377, "grad_norm": 0.8617445610914055, "learning_rate": 9.437805758157565e-06, "loss": 0.6014, "mean_token_accuracy": 0.8042668700218201, "num_tokens": 121404995.0, "step": 1853 }, { "epoch": 0.24504361617763679, "grad_norm": 0.8833147700251998, "learning_rate": 9.436800435235599e-06, "loss": 0.6313, "mean_token_accuracy": 0.7943779826164246, "num_tokens": 121470531.0, "step": 1854 }, { "epoch": 0.2451757864128998, "grad_norm": 0.8786957668019648, "learning_rate": 9.435794274251014e-06, "loss": 0.6313, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 121536067.0, "step": 1855 }, { "epoch": 0.24530795664816282, "grad_norm": 0.8817475212809229, "learning_rate": 9.434787275418002e-06, "loss": 0.5722, "mean_token_accuracy": 0.8139421343803406, "num_tokens": 121601603.0, "step": 1856 }, { "epoch": 0.24544012688342584, "grad_norm": 0.8109231932231885, "learning_rate": 9.433779438950932e-06, "loss": 0.5709, "mean_token_accuracy": 0.813179075717926, "num_tokens": 121667139.0, "step": 1857 }, { "epoch": 0.24557229711868886, "grad_norm": 0.8881966242482727, "learning_rate": 9.432770765064351e-06, "loss": 0.6337, "mean_token_accuracy": 0.798879861831665, "num_tokens": 121732675.0, "step": 1858 }, { "epoch": 0.24570446735395188, "grad_norm": 0.9364255348415131, "learning_rate": 9.431761253972983e-06, "loss": 0.6818, "mean_token_accuracy": 0.7800176739692688, "num_tokens": 121798211.0, "step": 1859 }, { "epoch": 0.2458366375892149, "grad_norm": 0.8994408609791411, "learning_rate": 9.430750905891735e-06, "loss": 0.599, "mean_token_accuracy": 0.8047857284545898, "num_tokens": 121863747.0, "step": 1860 }, { "epoch": 0.24596880782447791, "grad_norm": 0.9673483922028961, "learning_rate": 9.429739721035687e-06, "loss": 0.6225, "mean_token_accuracy": 0.7986814975738525, "num_tokens": 121929283.0, "step": 1861 }, { "epoch": 0.24610097805974096, "grad_norm": 0.8895851503203144, "learning_rate": 9.428727699620099e-06, "loss": 0.626, "mean_token_accuracy": 0.797918438911438, "num_tokens": 121994819.0, "step": 1862 }, { "epoch": 0.24623314829500398, "grad_norm": 0.8779562880777441, "learning_rate": 9.42771484186041e-06, "loss": 0.6267, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 122060355.0, "step": 1863 }, { "epoch": 0.246365318530267, "grad_norm": 0.8912655252859336, "learning_rate": 9.426701147972234e-06, "loss": 0.6496, "mean_token_accuracy": 0.7900134325027466, "num_tokens": 122125891.0, "step": 1864 }, { "epoch": 0.24649748876553002, "grad_norm": 0.9684826648320622, "learning_rate": 9.42568661817137e-06, "loss": 0.6512, "mean_token_accuracy": 0.7910358905792236, "num_tokens": 122191427.0, "step": 1865 }, { "epoch": 0.24662965900079303, "grad_norm": 0.9722461890949902, "learning_rate": 9.424671252673788e-06, "loss": 0.6066, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 122256963.0, "step": 1866 }, { "epoch": 0.24676182923605605, "grad_norm": 0.9484152078624973, "learning_rate": 9.423655051695635e-06, "loss": 0.6239, "mean_token_accuracy": 0.8000549674034119, "num_tokens": 122322499.0, "step": 1867 }, { "epoch": 0.24689399947131907, "grad_norm": 1.0138753259672488, "learning_rate": 9.422638015453243e-06, "loss": 0.6846, "mean_token_accuracy": 0.7761415243148804, "num_tokens": 122388035.0, "step": 1868 }, { "epoch": 0.2470261697065821, "grad_norm": 1.0922476428814787, "learning_rate": 9.421620144163113e-06, "loss": 0.592, "mean_token_accuracy": 0.806555986404419, "num_tokens": 122453571.0, "step": 1869 }, { "epoch": 0.2471583399418451, "grad_norm": 0.869666189802078, "learning_rate": 9.420601438041935e-06, "loss": 0.6045, "mean_token_accuracy": 0.8039616942405701, "num_tokens": 122519107.0, "step": 1870 }, { "epoch": 0.24729051017710812, "grad_norm": 0.8958144936076364, "learning_rate": 9.419581897306567e-06, "loss": 0.5725, "mean_token_accuracy": 0.8129807114601135, "num_tokens": 122584643.0, "step": 1871 }, { "epoch": 0.24742268041237114, "grad_norm": 1.0849115641458815, "learning_rate": 9.418561522174048e-06, "loss": 0.6846, "mean_token_accuracy": 0.7796514630317688, "num_tokens": 122650179.0, "step": 1872 }, { "epoch": 0.24755485064763416, "grad_norm": 0.9296809516710642, "learning_rate": 9.417540312861596e-06, "loss": 0.6485, "mean_token_accuracy": 0.7885484099388123, "num_tokens": 122715715.0, "step": 1873 }, { "epoch": 0.24768702088289718, "grad_norm": 0.9028084289480859, "learning_rate": 9.4165182695866e-06, "loss": 0.5946, "mean_token_accuracy": 0.8075021505355835, "num_tokens": 122781251.0, "step": 1874 }, { "epoch": 0.2478191911181602, "grad_norm": 0.8803806843174133, "learning_rate": 9.415495392566638e-06, "loss": 0.6332, "mean_token_accuracy": 0.7950189113616943, "num_tokens": 122846787.0, "step": 1875 }, { "epoch": 0.24795136135342322, "grad_norm": 0.9109903154283326, "learning_rate": 9.414471682019455e-06, "loss": 0.6074, "mean_token_accuracy": 0.8002685904502869, "num_tokens": 122912323.0, "step": 1876 }, { "epoch": 0.24808353158868623, "grad_norm": 0.9215063273364997, "learning_rate": 9.413447138162982e-06, "loss": 0.655, "mean_token_accuracy": 0.7855115532875061, "num_tokens": 122977859.0, "step": 1877 }, { "epoch": 0.24821570182394925, "grad_norm": 0.8025902578179923, "learning_rate": 9.412421761215319e-06, "loss": 0.556, "mean_token_accuracy": 0.8196954131126404, "num_tokens": 123043395.0, "step": 1878 }, { "epoch": 0.24834787205921227, "grad_norm": 0.8680901907886, "learning_rate": 9.411395551394746e-06, "loss": 0.5927, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 123108931.0, "step": 1879 }, { "epoch": 0.2484800422944753, "grad_norm": 0.9604845706839323, "learning_rate": 9.410368508919726e-06, "loss": 0.6324, "mean_token_accuracy": 0.7947442531585693, "num_tokens": 123174467.0, "step": 1880 }, { "epoch": 0.2486122125297383, "grad_norm": 0.9894532912585291, "learning_rate": 9.409340634008893e-06, "loss": 0.6259, "mean_token_accuracy": 0.7993834614753723, "num_tokens": 123240003.0, "step": 1881 }, { "epoch": 0.24874438276500133, "grad_norm": 0.8821388000739481, "learning_rate": 9.40831192688106e-06, "loss": 0.6225, "mean_token_accuracy": 0.7995055317878723, "num_tokens": 123305539.0, "step": 1882 }, { "epoch": 0.24887655300026434, "grad_norm": 1.0285495851627264, "learning_rate": 9.407282387755218e-06, "loss": 0.6664, "mean_token_accuracy": 0.7840007543563843, "num_tokens": 123371075.0, "step": 1883 }, { "epoch": 0.24900872323552736, "grad_norm": 0.9390022299351458, "learning_rate": 9.406252016850535e-06, "loss": 0.6479, "mean_token_accuracy": 0.78825843334198, "num_tokens": 123436611.0, "step": 1884 }, { "epoch": 0.24914089347079038, "grad_norm": 0.8222026724124181, "learning_rate": 9.405220814386352e-06, "loss": 0.6172, "mean_token_accuracy": 0.7993834614753723, "num_tokens": 123502147.0, "step": 1885 }, { "epoch": 0.2492730637060534, "grad_norm": 0.9080389189956668, "learning_rate": 9.404188780582194e-06, "loss": 0.6615, "mean_token_accuracy": 0.7878311276435852, "num_tokens": 123567683.0, "step": 1886 }, { "epoch": 0.24940523394131642, "grad_norm": 0.9567316937655581, "learning_rate": 9.403155915657756e-06, "loss": 0.6518, "mean_token_accuracy": 0.789128303527832, "num_tokens": 123633219.0, "step": 1887 }, { "epoch": 0.24953740417657944, "grad_norm": 1.265417694786738, "learning_rate": 9.402122219832918e-06, "loss": 0.709, "mean_token_accuracy": 0.7735471725463867, "num_tokens": 123698755.0, "step": 1888 }, { "epoch": 0.24966957441184245, "grad_norm": 0.9225576738704834, "learning_rate": 9.40108769332773e-06, "loss": 0.6499, "mean_token_accuracy": 0.7873886227607727, "num_tokens": 123764291.0, "step": 1889 }, { "epoch": 0.24980174464710547, "grad_norm": 0.9683318213447094, "learning_rate": 9.40005233636242e-06, "loss": 0.6362, "mean_token_accuracy": 0.7947747707366943, "num_tokens": 123829827.0, "step": 1890 }, { "epoch": 0.2499339148823685, "grad_norm": 0.9109430925121125, "learning_rate": 9.399016149157396e-06, "loss": 0.6724, "mean_token_accuracy": 0.7827951312065125, "num_tokens": 123895363.0, "step": 1891 }, { "epoch": 0.2500660851176315, "grad_norm": 0.9896005381235387, "learning_rate": 9.397979131933237e-06, "loss": 0.6305, "mean_token_accuracy": 0.7958582639694214, "num_tokens": 123960899.0, "step": 1892 }, { "epoch": 0.2501982553528945, "grad_norm": 0.9326907880298554, "learning_rate": 9.396941284910706e-06, "loss": 0.6302, "mean_token_accuracy": 0.7949120998382568, "num_tokens": 124026435.0, "step": 1893 }, { "epoch": 0.25033042558815755, "grad_norm": 0.8580925415764915, "learning_rate": 9.395902608310738e-06, "loss": 0.6214, "mean_token_accuracy": 0.7991240620613098, "num_tokens": 124091971.0, "step": 1894 }, { "epoch": 0.25046259582342056, "grad_norm": 1.002672729145666, "learning_rate": 9.394863102354446e-06, "loss": 0.6884, "mean_token_accuracy": 0.7766451239585876, "num_tokens": 124157507.0, "step": 1895 }, { "epoch": 0.2505947660586836, "grad_norm": 0.8829138464557914, "learning_rate": 9.393822767263116e-06, "loss": 0.5631, "mean_token_accuracy": 0.8178946375846863, "num_tokens": 124223043.0, "step": 1896 }, { "epoch": 0.2507269362939466, "grad_norm": 0.898031599973139, "learning_rate": 9.39278160325822e-06, "loss": 0.5819, "mean_token_accuracy": 0.8103559017181396, "num_tokens": 124288579.0, "step": 1897 }, { "epoch": 0.2508591065292096, "grad_norm": 0.8670335842682684, "learning_rate": 9.391739610561393e-06, "loss": 0.6189, "mean_token_accuracy": 0.7994902729988098, "num_tokens": 124354115.0, "step": 1898 }, { "epoch": 0.25099127676447264, "grad_norm": 0.8747583071833034, "learning_rate": 9.390696789394456e-06, "loss": 0.6329, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 124419651.0, "step": 1899 }, { "epoch": 0.25112344699973566, "grad_norm": 0.9182519830993956, "learning_rate": 9.389653139979407e-06, "loss": 0.62, "mean_token_accuracy": 0.802374541759491, "num_tokens": 124485187.0, "step": 1900 }, { "epoch": 0.2512556172349987, "grad_norm": 0.7959109772649561, "learning_rate": 9.388608662538413e-06, "loss": 0.5944, "mean_token_accuracy": 0.8096691370010376, "num_tokens": 124550723.0, "step": 1901 }, { "epoch": 0.2513877874702617, "grad_norm": 0.9001432851425325, "learning_rate": 9.387563357293822e-06, "loss": 0.6714, "mean_token_accuracy": 0.7821999788284302, "num_tokens": 124616259.0, "step": 1902 }, { "epoch": 0.2515199577055247, "grad_norm": 1.0218151681463752, "learning_rate": 9.386517224468162e-06, "loss": 0.6567, "mean_token_accuracy": 0.78795325756073, "num_tokens": 124681795.0, "step": 1903 }, { "epoch": 0.25165212794078773, "grad_norm": 1.0025895504828934, "learning_rate": 9.385470264284127e-06, "loss": 0.6431, "mean_token_accuracy": 0.7888689041137695, "num_tokens": 124747331.0, "step": 1904 }, { "epoch": 0.25178429817605075, "grad_norm": 0.9536443007971245, "learning_rate": 9.384422476964596e-06, "loss": 0.6435, "mean_token_accuracy": 0.7952325940132141, "num_tokens": 124812867.0, "step": 1905 }, { "epoch": 0.25191646841131377, "grad_norm": 0.895142577116058, "learning_rate": 9.383373862732622e-06, "loss": 0.6301, "mean_token_accuracy": 0.7932944893836975, "num_tokens": 124878403.0, "step": 1906 }, { "epoch": 0.2520486386465768, "grad_norm": 0.8927248149963283, "learning_rate": 9.382324421811429e-06, "loss": 0.6464, "mean_token_accuracy": 0.7903339266777039, "num_tokens": 124943939.0, "step": 1907 }, { "epoch": 0.2521808088818398, "grad_norm": 1.0312193925388382, "learning_rate": 9.381274154424425e-06, "loss": 0.6264, "mean_token_accuracy": 0.7958277463912964, "num_tokens": 125009475.0, "step": 1908 }, { "epoch": 0.2523129791171028, "grad_norm": 0.9478767947160446, "learning_rate": 9.380223060795188e-06, "loss": 0.6225, "mean_token_accuracy": 0.7989256381988525, "num_tokens": 125075011.0, "step": 1909 }, { "epoch": 0.25244514935236584, "grad_norm": 0.9017555594436641, "learning_rate": 9.379171141147475e-06, "loss": 0.6201, "mean_token_accuracy": 0.799062967300415, "num_tokens": 125140547.0, "step": 1910 }, { "epoch": 0.25257731958762886, "grad_norm": 0.8587726247173779, "learning_rate": 9.378118395705218e-06, "loss": 0.6328, "mean_token_accuracy": 0.7940880060195923, "num_tokens": 125206083.0, "step": 1911 }, { "epoch": 0.2527094898228919, "grad_norm": 0.9449879130545829, "learning_rate": 9.377064824692525e-06, "loss": 0.6556, "mean_token_accuracy": 0.7883653044700623, "num_tokens": 125271619.0, "step": 1912 }, { "epoch": 0.2528416600581549, "grad_norm": 0.9330165230760934, "learning_rate": 9.376010428333676e-06, "loss": 0.6028, "mean_token_accuracy": 0.803122341632843, "num_tokens": 125337155.0, "step": 1913 }, { "epoch": 0.2529738302934179, "grad_norm": 1.077869105256551, "learning_rate": 9.374955206853131e-06, "loss": 0.6466, "mean_token_accuracy": 0.7883653044700623, "num_tokens": 125402691.0, "step": 1914 }, { "epoch": 0.25310600052868093, "grad_norm": 0.9669899621004887, "learning_rate": 9.373899160475527e-06, "loss": 0.6463, "mean_token_accuracy": 0.7947289943695068, "num_tokens": 125468227.0, "step": 1915 }, { "epoch": 0.25323817076394395, "grad_norm": 0.8676014790433758, "learning_rate": 9.372842289425674e-06, "loss": 0.6121, "mean_token_accuracy": 0.803030788898468, "num_tokens": 125533763.0, "step": 1916 }, { "epoch": 0.25337034099920697, "grad_norm": 0.9530647598012428, "learning_rate": 9.371784593928557e-06, "loss": 0.6724, "mean_token_accuracy": 0.7836955189704895, "num_tokens": 125599299.0, "step": 1917 }, { "epoch": 0.25350251123447, "grad_norm": 0.9131996295859557, "learning_rate": 9.370726074209333e-06, "loss": 0.5992, "mean_token_accuracy": 0.8087992668151855, "num_tokens": 125664835.0, "step": 1918 }, { "epoch": 0.253634681469733, "grad_norm": 0.8695229446908014, "learning_rate": 9.369666730493348e-06, "loss": 0.5485, "mean_token_accuracy": 0.8234189748764038, "num_tokens": 125730371.0, "step": 1919 }, { "epoch": 0.253766851704996, "grad_norm": 0.9209069654955194, "learning_rate": 9.368606563006107e-06, "loss": 0.5843, "mean_token_accuracy": 0.8091502785682678, "num_tokens": 125795907.0, "step": 1920 }, { "epoch": 0.25389902194025904, "grad_norm": 0.9489068352803224, "learning_rate": 9.367545571973298e-06, "loss": 0.6369, "mean_token_accuracy": 0.7940422296524048, "num_tokens": 125861443.0, "step": 1921 }, { "epoch": 0.25403119217552206, "grad_norm": 0.9135831250302456, "learning_rate": 9.366483757620787e-06, "loss": 0.6135, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 125926979.0, "step": 1922 }, { "epoch": 0.2541633624107851, "grad_norm": 0.8692861399511864, "learning_rate": 9.36542112017461e-06, "loss": 0.6036, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 125992515.0, "step": 1923 }, { "epoch": 0.2542955326460481, "grad_norm": 0.986639728282799, "learning_rate": 9.36435765986098e-06, "loss": 0.6478, "mean_token_accuracy": 0.7898455858230591, "num_tokens": 126058051.0, "step": 1924 }, { "epoch": 0.2544277028813111, "grad_norm": 1.0042463582479513, "learning_rate": 9.363293376906287e-06, "loss": 0.6909, "mean_token_accuracy": 0.7776370644569397, "num_tokens": 126123587.0, "step": 1925 }, { "epoch": 0.25455987311657413, "grad_norm": 0.9719932345121652, "learning_rate": 9.362228271537092e-06, "loss": 0.6419, "mean_token_accuracy": 0.7915852665901184, "num_tokens": 126189123.0, "step": 1926 }, { "epoch": 0.25469204335183715, "grad_norm": 0.9548230311675161, "learning_rate": 9.361162343980136e-06, "loss": 0.6325, "mean_token_accuracy": 0.7947900295257568, "num_tokens": 126254659.0, "step": 1927 }, { "epoch": 0.25482421358710017, "grad_norm": 0.8880913941987343, "learning_rate": 9.360095594462335e-06, "loss": 0.656, "mean_token_accuracy": 0.788105845451355, "num_tokens": 126320195.0, "step": 1928 }, { "epoch": 0.2549563838223632, "grad_norm": 0.9362563007593189, "learning_rate": 9.359028023210772e-06, "loss": 0.6619, "mean_token_accuracy": 0.7852826118469238, "num_tokens": 126385731.0, "step": 1929 }, { "epoch": 0.2550885540576262, "grad_norm": 0.9771472549387902, "learning_rate": 9.357959630452715e-06, "loss": 0.6122, "mean_token_accuracy": 0.8016725778579712, "num_tokens": 126451267.0, "step": 1930 }, { "epoch": 0.2552207242928892, "grad_norm": 1.0281526042328841, "learning_rate": 9.356890416415601e-06, "loss": 0.6287, "mean_token_accuracy": 0.7954919934272766, "num_tokens": 126516803.0, "step": 1931 }, { "epoch": 0.25535289452815224, "grad_norm": 0.9689502421750376, "learning_rate": 9.355820381327042e-06, "loss": 0.6912, "mean_token_accuracy": 0.7752411365509033, "num_tokens": 126582339.0, "step": 1932 }, { "epoch": 0.25548506476341526, "grad_norm": 0.8569725818230675, "learning_rate": 9.35474952541483e-06, "loss": 0.5918, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 126647875.0, "step": 1933 }, { "epoch": 0.2556172349986783, "grad_norm": 1.0317783682240733, "learning_rate": 9.353677848906923e-06, "loss": 0.5853, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 126713411.0, "step": 1934 }, { "epoch": 0.2557494052339413, "grad_norm": 1.0159933214150967, "learning_rate": 9.352605352031461e-06, "loss": 0.6006, "mean_token_accuracy": 0.8052740693092346, "num_tokens": 126778947.0, "step": 1935 }, { "epoch": 0.2558815754692043, "grad_norm": 0.9947913626653268, "learning_rate": 9.351532035016758e-06, "loss": 0.6554, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 126844483.0, "step": 1936 }, { "epoch": 0.25601374570446733, "grad_norm": 0.9252258090899483, "learning_rate": 9.3504578980913e-06, "loss": 0.6314, "mean_token_accuracy": 0.7963923811912537, "num_tokens": 126910019.0, "step": 1937 }, { "epoch": 0.25614591593973035, "grad_norm": 0.9284015045465931, "learning_rate": 9.349382941483743e-06, "loss": 0.6542, "mean_token_accuracy": 0.7853894233703613, "num_tokens": 126975555.0, "step": 1938 }, { "epoch": 0.25627808617499337, "grad_norm": 0.9239840410363293, "learning_rate": 9.34830716542293e-06, "loss": 0.6353, "mean_token_accuracy": 0.792500913143158, "num_tokens": 127041091.0, "step": 1939 }, { "epoch": 0.2564102564102564, "grad_norm": 0.944581044049613, "learning_rate": 9.347230570137868e-06, "loss": 0.633, "mean_token_accuracy": 0.7932028770446777, "num_tokens": 127106627.0, "step": 1940 }, { "epoch": 0.2565424266455194, "grad_norm": 0.9046602199286512, "learning_rate": 9.346153155857742e-06, "loss": 0.6163, "mean_token_accuracy": 0.8020235896110535, "num_tokens": 127172163.0, "step": 1941 }, { "epoch": 0.2566745968807824, "grad_norm": 0.8914984298161447, "learning_rate": 9.345074922811912e-06, "loss": 0.5757, "mean_token_accuracy": 0.8155750036239624, "num_tokens": 127237699.0, "step": 1942 }, { "epoch": 0.25680676711604544, "grad_norm": 0.8558838566823536, "learning_rate": 9.343995871229909e-06, "loss": 0.6207, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 127303235.0, "step": 1943 }, { "epoch": 0.25693893735130846, "grad_norm": 0.9161928842941214, "learning_rate": 9.34291600134144e-06, "loss": 0.5815, "mean_token_accuracy": 0.8107526302337646, "num_tokens": 127368771.0, "step": 1944 }, { "epoch": 0.2570711075865715, "grad_norm": 0.9478351110872871, "learning_rate": 9.341835313376392e-06, "loss": 0.6765, "mean_token_accuracy": 0.7828561663627625, "num_tokens": 127434307.0, "step": 1945 }, { "epoch": 0.2572032778218345, "grad_norm": 0.9063043822795291, "learning_rate": 9.340753807564817e-06, "loss": 0.6054, "mean_token_accuracy": 0.806403398513794, "num_tokens": 127499843.0, "step": 1946 }, { "epoch": 0.2573354480570975, "grad_norm": 0.8605475316097865, "learning_rate": 9.339671484136943e-06, "loss": 0.5685, "mean_token_accuracy": 0.8149493336677551, "num_tokens": 127565379.0, "step": 1947 }, { "epoch": 0.25746761829236053, "grad_norm": 0.8666103258221681, "learning_rate": 9.338588343323178e-06, "loss": 0.61, "mean_token_accuracy": 0.8025729656219482, "num_tokens": 127630915.0, "step": 1948 }, { "epoch": 0.25759978852762355, "grad_norm": 0.9622761107691434, "learning_rate": 9.3375043853541e-06, "loss": 0.6444, "mean_token_accuracy": 0.789158821105957, "num_tokens": 127696451.0, "step": 1949 }, { "epoch": 0.25773195876288657, "grad_norm": 0.9606905016780597, "learning_rate": 9.336419610460458e-06, "loss": 0.6562, "mean_token_accuracy": 0.7859846353530884, "num_tokens": 127761987.0, "step": 1950 }, { "epoch": 0.25786412899814964, "grad_norm": 0.9014035185839344, "learning_rate": 9.33533401887318e-06, "loss": 0.609, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 127827523.0, "step": 1951 }, { "epoch": 0.25799629923341266, "grad_norm": 0.9965171138540403, "learning_rate": 9.334247610823364e-06, "loss": 0.6251, "mean_token_accuracy": 0.7965449690818787, "num_tokens": 127893059.0, "step": 1952 }, { "epoch": 0.2581284694686757, "grad_norm": 0.990936172350865, "learning_rate": 9.333160386542289e-06, "loss": 0.6623, "mean_token_accuracy": 0.7862287759780884, "num_tokens": 127958595.0, "step": 1953 }, { "epoch": 0.2582606397039387, "grad_norm": 0.8927811908497439, "learning_rate": 9.332072346261395e-06, "loss": 0.6084, "mean_token_accuracy": 0.8004974722862244, "num_tokens": 128024131.0, "step": 1954 }, { "epoch": 0.2583928099392017, "grad_norm": 0.8728901704657233, "learning_rate": 9.330983490212308e-06, "loss": 0.6445, "mean_token_accuracy": 0.7920736074447632, "num_tokens": 128089667.0, "step": 1955 }, { "epoch": 0.25852498017446474, "grad_norm": 0.8532734864840654, "learning_rate": 9.32989381862682e-06, "loss": 0.5861, "mean_token_accuracy": 0.8086771965026855, "num_tokens": 128155203.0, "step": 1956 }, { "epoch": 0.25865715040972775, "grad_norm": 0.9673771189817012, "learning_rate": 9.3288033317369e-06, "loss": 0.6595, "mean_token_accuracy": 0.7870528697967529, "num_tokens": 128220739.0, "step": 1957 }, { "epoch": 0.2587893206449908, "grad_norm": 1.006717031393974, "learning_rate": 9.327712029774692e-06, "loss": 0.6361, "mean_token_accuracy": 0.792287290096283, "num_tokens": 128286275.0, "step": 1958 }, { "epoch": 0.2589214908802538, "grad_norm": 0.8520237306924824, "learning_rate": 9.326619912972509e-06, "loss": 0.6019, "mean_token_accuracy": 0.8066627979278564, "num_tokens": 128351811.0, "step": 1959 }, { "epoch": 0.2590536611155168, "grad_norm": 0.8133174422623788, "learning_rate": 9.32552698156284e-06, "loss": 0.5832, "mean_token_accuracy": 0.8112257122993469, "num_tokens": 128417347.0, "step": 1960 }, { "epoch": 0.2591858313507798, "grad_norm": 0.9001422978098214, "learning_rate": 9.324433235778348e-06, "loss": 0.6475, "mean_token_accuracy": 0.7874191403388977, "num_tokens": 128482883.0, "step": 1961 }, { "epoch": 0.25931800158604285, "grad_norm": 0.9093498055789148, "learning_rate": 9.32333867585187e-06, "loss": 0.6214, "mean_token_accuracy": 0.7956598997116089, "num_tokens": 128548419.0, "step": 1962 }, { "epoch": 0.25945017182130586, "grad_norm": 0.951344909010195, "learning_rate": 9.322243302016412e-06, "loss": 0.6514, "mean_token_accuracy": 0.7864577174186707, "num_tokens": 128613955.0, "step": 1963 }, { "epoch": 0.2595823420565689, "grad_norm": 0.9843881191061121, "learning_rate": 9.32114711450516e-06, "loss": 0.7037, "mean_token_accuracy": 0.7721889615058899, "num_tokens": 128679491.0, "step": 1964 }, { "epoch": 0.2597145122918319, "grad_norm": 0.902195689389118, "learning_rate": 9.320050113551464e-06, "loss": 0.6319, "mean_token_accuracy": 0.7955835461616516, "num_tokens": 128745027.0, "step": 1965 }, { "epoch": 0.2598466825270949, "grad_norm": 1.0945998055221586, "learning_rate": 9.318952299388857e-06, "loss": 0.679, "mean_token_accuracy": 0.7786442637443542, "num_tokens": 128810563.0, "step": 1966 }, { "epoch": 0.25997885276235794, "grad_norm": 0.8416144383827343, "learning_rate": 9.317853672251038e-06, "loss": 0.5782, "mean_token_accuracy": 0.8081430792808533, "num_tokens": 128876099.0, "step": 1967 }, { "epoch": 0.26011102299762096, "grad_norm": 0.9905264463874381, "learning_rate": 9.316754232371886e-06, "loss": 0.6596, "mean_token_accuracy": 0.783863365650177, "num_tokens": 128941635.0, "step": 1968 }, { "epoch": 0.260243193232884, "grad_norm": 0.8977125845485203, "learning_rate": 9.315653979985444e-06, "loss": 0.6303, "mean_token_accuracy": 0.7926992774009705, "num_tokens": 129007171.0, "step": 1969 }, { "epoch": 0.260375363468147, "grad_norm": 0.8689691530567921, "learning_rate": 9.314552915325935e-06, "loss": 0.5791, "mean_token_accuracy": 0.8108137249946594, "num_tokens": 129072707.0, "step": 1970 }, { "epoch": 0.26050753370341, "grad_norm": 0.9672470934114653, "learning_rate": 9.313451038627753e-06, "loss": 0.6141, "mean_token_accuracy": 0.8010469079017639, "num_tokens": 129138243.0, "step": 1971 }, { "epoch": 0.26063970393867303, "grad_norm": 1.1333660196598698, "learning_rate": 9.312348350125463e-06, "loss": 0.6704, "mean_token_accuracy": 0.7793920040130615, "num_tokens": 129203779.0, "step": 1972 }, { "epoch": 0.26077187417393605, "grad_norm": 0.9472020022423586, "learning_rate": 9.311244850053808e-06, "loss": 0.5937, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 129269315.0, "step": 1973 }, { "epoch": 0.26090404440919907, "grad_norm": 0.9230769307582414, "learning_rate": 9.310140538647694e-06, "loss": 0.658, "mean_token_accuracy": 0.7892045974731445, "num_tokens": 129334851.0, "step": 1974 }, { "epoch": 0.2610362146444621, "grad_norm": 0.8914346853362635, "learning_rate": 9.30903541614221e-06, "loss": 0.6103, "mean_token_accuracy": 0.802160918712616, "num_tokens": 129400387.0, "step": 1975 }, { "epoch": 0.2611683848797251, "grad_norm": 1.0289720596255028, "learning_rate": 9.307929482772617e-06, "loss": 0.6522, "mean_token_accuracy": 0.78801429271698, "num_tokens": 129465923.0, "step": 1976 }, { "epoch": 0.2613005551149881, "grad_norm": 0.9109540483628203, "learning_rate": 9.306822738774338e-06, "loss": 0.5973, "mean_token_accuracy": 0.8037785291671753, "num_tokens": 129531459.0, "step": 1977 }, { "epoch": 0.26143272535025114, "grad_norm": 0.9623092993518011, "learning_rate": 9.305715184382978e-06, "loss": 0.6285, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 129596995.0, "step": 1978 }, { "epoch": 0.26156489558551416, "grad_norm": 0.8926592245682847, "learning_rate": 9.304606819834313e-06, "loss": 0.6475, "mean_token_accuracy": 0.7931571006774902, "num_tokens": 129662531.0, "step": 1979 }, { "epoch": 0.2616970658207772, "grad_norm": 0.8934024279835936, "learning_rate": 9.303497645364291e-06, "loss": 0.6231, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 129728067.0, "step": 1980 }, { "epoch": 0.2618292360560402, "grad_norm": 0.9508622324431005, "learning_rate": 9.30238766120903e-06, "loss": 0.6609, "mean_token_accuracy": 0.7864119410514832, "num_tokens": 129793603.0, "step": 1981 }, { "epoch": 0.2619614062913032, "grad_norm": 1.006344272181762, "learning_rate": 9.301276867604826e-06, "loss": 0.6761, "mean_token_accuracy": 0.7823830842971802, "num_tokens": 129859139.0, "step": 1982 }, { "epoch": 0.26209357652656623, "grad_norm": 0.9864023802397275, "learning_rate": 9.300165264788141e-06, "loss": 0.6521, "mean_token_accuracy": 0.7914937138557434, "num_tokens": 129924675.0, "step": 1983 }, { "epoch": 0.26222574676182925, "grad_norm": 0.910773758785726, "learning_rate": 9.299052852995614e-06, "loss": 0.6652, "mean_token_accuracy": 0.7851300239562988, "num_tokens": 129990211.0, "step": 1984 }, { "epoch": 0.26235791699709227, "grad_norm": 0.9475398605285934, "learning_rate": 9.297939632464053e-06, "loss": 0.6515, "mean_token_accuracy": 0.7876327633857727, "num_tokens": 130055747.0, "step": 1985 }, { "epoch": 0.2624900872323553, "grad_norm": 0.8936272300521232, "learning_rate": 9.296825603430438e-06, "loss": 0.5721, "mean_token_accuracy": 0.8117445707321167, "num_tokens": 130121283.0, "step": 1986 }, { "epoch": 0.2626222574676183, "grad_norm": 1.0220037262475277, "learning_rate": 9.295710766131925e-06, "loss": 0.6423, "mean_token_accuracy": 0.7907001376152039, "num_tokens": 130186819.0, "step": 1987 }, { "epoch": 0.2627544277028813, "grad_norm": 0.9167540229392905, "learning_rate": 9.29459512080584e-06, "loss": 0.6275, "mean_token_accuracy": 0.79891037940979, "num_tokens": 130252355.0, "step": 1988 }, { "epoch": 0.26288659793814434, "grad_norm": 0.9576579675882241, "learning_rate": 9.293478667689675e-06, "loss": 0.6259, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 130317891.0, "step": 1989 }, { "epoch": 0.26301876817340736, "grad_norm": 1.0140462256392249, "learning_rate": 9.292361407021108e-06, "loss": 0.6472, "mean_token_accuracy": 0.7899523973464966, "num_tokens": 130383427.0, "step": 1990 }, { "epoch": 0.2631509384086704, "grad_norm": 1.011970289002674, "learning_rate": 9.291243339037974e-06, "loss": 0.6439, "mean_token_accuracy": 0.7920125722885132, "num_tokens": 130448963.0, "step": 1991 }, { "epoch": 0.2632831086439334, "grad_norm": 0.9916175189213574, "learning_rate": 9.290124463978289e-06, "loss": 0.6164, "mean_token_accuracy": 0.7975979447364807, "num_tokens": 130514499.0, "step": 1992 }, { "epoch": 0.2634152788791964, "grad_norm": 0.8769900199175689, "learning_rate": 9.28900478208024e-06, "loss": 0.6465, "mean_token_accuracy": 0.7870070934295654, "num_tokens": 130580035.0, "step": 1993 }, { "epoch": 0.26354744911445943, "grad_norm": 0.8698454754896124, "learning_rate": 9.28788429358218e-06, "loss": 0.6572, "mean_token_accuracy": 0.7865340113639832, "num_tokens": 130645571.0, "step": 1994 }, { "epoch": 0.26367961934972245, "grad_norm": 0.9983251848306638, "learning_rate": 9.286762998722636e-06, "loss": 0.6247, "mean_token_accuracy": 0.7971249222755432, "num_tokens": 130711107.0, "step": 1995 }, { "epoch": 0.26381178958498547, "grad_norm": 0.8501816821938127, "learning_rate": 9.285640897740316e-06, "loss": 0.6023, "mean_token_accuracy": 0.8056555986404419, "num_tokens": 130776643.0, "step": 1996 }, { "epoch": 0.2639439598202485, "grad_norm": 0.9326037612127177, "learning_rate": 9.284517990874086e-06, "loss": 0.6054, "mean_token_accuracy": 0.8025577068328857, "num_tokens": 130842179.0, "step": 1997 }, { "epoch": 0.2640761300555115, "grad_norm": 0.8619419764121355, "learning_rate": 9.28339427836299e-06, "loss": 0.664, "mean_token_accuracy": 0.7849010825157166, "num_tokens": 130907715.0, "step": 1998 }, { "epoch": 0.2642083002907745, "grad_norm": 0.9855429625619249, "learning_rate": 9.282269760446244e-06, "loss": 0.6215, "mean_token_accuracy": 0.7962245345115662, "num_tokens": 130973251.0, "step": 1999 }, { "epoch": 0.26434047052603754, "grad_norm": 0.9086477411709121, "learning_rate": 9.281144437363235e-06, "loss": 0.6328, "mean_token_accuracy": 0.7942101359367371, "num_tokens": 131038787.0, "step": 2000 }, { "epoch": 0.26447264076130056, "grad_norm": 0.9727373243057662, "learning_rate": 9.280018309353519e-06, "loss": 0.6692, "mean_token_accuracy": 0.7803076505661011, "num_tokens": 131104323.0, "step": 2001 }, { "epoch": 0.2646048109965636, "grad_norm": 0.8580551389918405, "learning_rate": 9.278891376656826e-06, "loss": 0.574, "mean_token_accuracy": 0.8155292272567749, "num_tokens": 131169859.0, "step": 2002 }, { "epoch": 0.2647369812318266, "grad_norm": 0.8974483687321598, "learning_rate": 9.277763639513058e-06, "loss": 0.5704, "mean_token_accuracy": 0.812873899936676, "num_tokens": 131235395.0, "step": 2003 }, { "epoch": 0.2648691514670896, "grad_norm": 1.0730307089569828, "learning_rate": 9.276635098162283e-06, "loss": 0.6793, "mean_token_accuracy": 0.7818337082862854, "num_tokens": 131300931.0, "step": 2004 }, { "epoch": 0.26500132170235263, "grad_norm": 0.9382138076384446, "learning_rate": 9.275505752844747e-06, "loss": 0.6686, "mean_token_accuracy": 0.7843517065048218, "num_tokens": 131366467.0, "step": 2005 }, { "epoch": 0.26513349193761565, "grad_norm": 0.958477115918525, "learning_rate": 9.274375603800864e-06, "loss": 0.6483, "mean_token_accuracy": 0.787861704826355, "num_tokens": 131432003.0, "step": 2006 }, { "epoch": 0.26526566217287867, "grad_norm": 0.9942005961521566, "learning_rate": 9.273244651271217e-06, "loss": 0.6574, "mean_token_accuracy": 0.7860761880874634, "num_tokens": 131497539.0, "step": 2007 }, { "epoch": 0.2653978324081417, "grad_norm": 1.0004310075895206, "learning_rate": 9.272112895496564e-06, "loss": 0.6511, "mean_token_accuracy": 0.7873581051826477, "num_tokens": 131563075.0, "step": 2008 }, { "epoch": 0.2655300026434047, "grad_norm": 0.9571689302050029, "learning_rate": 9.270980336717831e-06, "loss": 0.6707, "mean_token_accuracy": 0.7829782962799072, "num_tokens": 131628611.0, "step": 2009 }, { "epoch": 0.2656621728786677, "grad_norm": 0.9692660967439264, "learning_rate": 9.269846975176116e-06, "loss": 0.6854, "mean_token_accuracy": 0.7786290049552917, "num_tokens": 131694147.0, "step": 2010 }, { "epoch": 0.26579434311393074, "grad_norm": 0.9007509371821422, "learning_rate": 9.268712811112692e-06, "loss": 0.6129, "mean_token_accuracy": 0.8009858131408691, "num_tokens": 131759683.0, "step": 2011 }, { "epoch": 0.26592651334919376, "grad_norm": 0.9483126275535705, "learning_rate": 9.267577844768994e-06, "loss": 0.662, "mean_token_accuracy": 0.7835429310798645, "num_tokens": 131825219.0, "step": 2012 }, { "epoch": 0.2660586835844568, "grad_norm": 1.020927252489909, "learning_rate": 9.266442076386632e-06, "loss": 0.6422, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 131890755.0, "step": 2013 }, { "epoch": 0.2661908538197198, "grad_norm": 0.9110978066034525, "learning_rate": 9.265305506207392e-06, "loss": 0.6011, "mean_token_accuracy": 0.8051977753639221, "num_tokens": 131956291.0, "step": 2014 }, { "epoch": 0.2663230240549828, "grad_norm": 0.8986714756604866, "learning_rate": 9.264168134473222e-06, "loss": 0.6119, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 132021827.0, "step": 2015 }, { "epoch": 0.26645519429024583, "grad_norm": 0.9513089383787798, "learning_rate": 9.263029961426246e-06, "loss": 0.6607, "mean_token_accuracy": 0.7838481068611145, "num_tokens": 132087363.0, "step": 2016 }, { "epoch": 0.26658736452550885, "grad_norm": 0.8825958820151437, "learning_rate": 9.26189098730876e-06, "loss": 0.6403, "mean_token_accuracy": 0.792378842830658, "num_tokens": 132152899.0, "step": 2017 }, { "epoch": 0.26671953476077187, "grad_norm": 0.8763386373632599, "learning_rate": 9.260751212363224e-06, "loss": 0.63, "mean_token_accuracy": 0.7955682873725891, "num_tokens": 132218435.0, "step": 2018 }, { "epoch": 0.2668517049960349, "grad_norm": 0.9573298576349816, "learning_rate": 9.259610636832275e-06, "loss": 0.6602, "mean_token_accuracy": 0.783497154712677, "num_tokens": 132283971.0, "step": 2019 }, { "epoch": 0.2669838752312979, "grad_norm": 0.9467728204820139, "learning_rate": 9.258469260958714e-06, "loss": 0.6159, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 132349507.0, "step": 2020 }, { "epoch": 0.2671160454665609, "grad_norm": 0.9442248862025644, "learning_rate": 9.25732708498552e-06, "loss": 0.6295, "mean_token_accuracy": 0.7963160872459412, "num_tokens": 132415043.0, "step": 2021 }, { "epoch": 0.26724821570182394, "grad_norm": 0.8602970994376822, "learning_rate": 9.256184109155837e-06, "loss": 0.5755, "mean_token_accuracy": 0.8114241361618042, "num_tokens": 132480579.0, "step": 2022 }, { "epoch": 0.26738038593708696, "grad_norm": 0.9253657296584566, "learning_rate": 9.255040333712982e-06, "loss": 0.6045, "mean_token_accuracy": 0.8045873641967773, "num_tokens": 132546115.0, "step": 2023 }, { "epoch": 0.26751255617235, "grad_norm": 0.9004344306863008, "learning_rate": 9.253895758900437e-06, "loss": 0.5922, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 132611651.0, "step": 2024 }, { "epoch": 0.267644726407613, "grad_norm": 0.944229027823422, "learning_rate": 9.252750384961862e-06, "loss": 0.6408, "mean_token_accuracy": 0.7897539734840393, "num_tokens": 132677187.0, "step": 2025 }, { "epoch": 0.267776896642876, "grad_norm": 0.9374527240708559, "learning_rate": 9.251604212141083e-06, "loss": 0.619, "mean_token_accuracy": 0.8002228140830994, "num_tokens": 132742723.0, "step": 2026 }, { "epoch": 0.26790906687813903, "grad_norm": 0.9509940780520482, "learning_rate": 9.250457240682092e-06, "loss": 0.6396, "mean_token_accuracy": 0.7915242314338684, "num_tokens": 132808259.0, "step": 2027 }, { "epoch": 0.26804123711340205, "grad_norm": 0.9536575229393336, "learning_rate": 9.24930947082906e-06, "loss": 0.644, "mean_token_accuracy": 0.7919362783432007, "num_tokens": 132873795.0, "step": 2028 }, { "epoch": 0.26817340734866507, "grad_norm": 0.853673489919148, "learning_rate": 9.248160902826321e-06, "loss": 0.5906, "mean_token_accuracy": 0.8114851713180542, "num_tokens": 132939331.0, "step": 2029 }, { "epoch": 0.2683055775839281, "grad_norm": 0.9011508094031913, "learning_rate": 9.247011536918383e-06, "loss": 0.6167, "mean_token_accuracy": 0.8021761775016785, "num_tokens": 133004867.0, "step": 2030 }, { "epoch": 0.2684377478191911, "grad_norm": 0.8578004381527786, "learning_rate": 9.24586137334992e-06, "loss": 0.5989, "mean_token_accuracy": 0.8071663975715637, "num_tokens": 133070403.0, "step": 2031 }, { "epoch": 0.2685699180544541, "grad_norm": 0.9852604596388704, "learning_rate": 9.244710412365778e-06, "loss": 0.637, "mean_token_accuracy": 0.7928824424743652, "num_tokens": 133135939.0, "step": 2032 }, { "epoch": 0.26870208828971714, "grad_norm": 0.8368121456589511, "learning_rate": 9.243558654210973e-06, "loss": 0.6269, "mean_token_accuracy": 0.7975674271583557, "num_tokens": 133201475.0, "step": 2033 }, { "epoch": 0.26883425852498016, "grad_norm": 0.9729672698125699, "learning_rate": 9.242406099130691e-06, "loss": 0.6511, "mean_token_accuracy": 0.7871444225311279, "num_tokens": 133267011.0, "step": 2034 }, { "epoch": 0.2689664287602432, "grad_norm": 0.9571863473957036, "learning_rate": 9.241252747370288e-06, "loss": 0.6323, "mean_token_accuracy": 0.7971706986427307, "num_tokens": 133332547.0, "step": 2035 }, { "epoch": 0.2690985989955062, "grad_norm": 0.911255702578991, "learning_rate": 9.240098599175284e-06, "loss": 0.5683, "mean_token_accuracy": 0.812873899936676, "num_tokens": 133398083.0, "step": 2036 }, { "epoch": 0.2692307692307692, "grad_norm": 0.8312248395015077, "learning_rate": 9.238943654791378e-06, "loss": 0.546, "mean_token_accuracy": 0.8215113878250122, "num_tokens": 133463619.0, "step": 2037 }, { "epoch": 0.26936293946603224, "grad_norm": 0.9888101287443096, "learning_rate": 9.237787914464431e-06, "loss": 0.6899, "mean_token_accuracy": 0.7752411365509033, "num_tokens": 133529155.0, "step": 2038 }, { "epoch": 0.26949510970129525, "grad_norm": 0.8708937827634208, "learning_rate": 9.236631378440476e-06, "loss": 0.6649, "mean_token_accuracy": 0.7868544459342957, "num_tokens": 133594691.0, "step": 2039 }, { "epoch": 0.2696272799365583, "grad_norm": 0.8683594303309803, "learning_rate": 9.235474046965716e-06, "loss": 0.5867, "mean_token_accuracy": 0.8094555139541626, "num_tokens": 133660227.0, "step": 2040 }, { "epoch": 0.2697594501718213, "grad_norm": 1.0011184094364383, "learning_rate": 9.234315920286522e-06, "loss": 0.6794, "mean_token_accuracy": 0.7812076807022095, "num_tokens": 133725219.0, "step": 2041 }, { "epoch": 0.2698916204070843, "grad_norm": 0.8542215015718062, "learning_rate": 9.233156998649435e-06, "loss": 0.5829, "mean_token_accuracy": 0.808356761932373, "num_tokens": 133790755.0, "step": 2042 }, { "epoch": 0.2700237906423473, "grad_norm": 0.8456385659536794, "learning_rate": 9.231997282301168e-06, "loss": 0.6509, "mean_token_accuracy": 0.78795325756073, "num_tokens": 133856291.0, "step": 2043 }, { "epoch": 0.27015596087761035, "grad_norm": 0.8528216110632207, "learning_rate": 9.230836771488596e-06, "loss": 0.6203, "mean_token_accuracy": 0.8014436364173889, "num_tokens": 133921827.0, "step": 2044 }, { "epoch": 0.27028813111287336, "grad_norm": 0.9057463675079385, "learning_rate": 9.22967546645877e-06, "loss": 0.651, "mean_token_accuracy": 0.7887315154075623, "num_tokens": 133987363.0, "step": 2045 }, { "epoch": 0.2704203013481364, "grad_norm": 0.9499030864012954, "learning_rate": 9.228513367458907e-06, "loss": 0.6973, "mean_token_accuracy": 0.7746306657791138, "num_tokens": 134052899.0, "step": 2046 }, { "epoch": 0.2705524715833994, "grad_norm": 0.8492059761092982, "learning_rate": 9.227350474736394e-06, "loss": 0.5925, "mean_token_accuracy": 0.8080820441246033, "num_tokens": 134118435.0, "step": 2047 }, { "epoch": 0.2706846418186624, "grad_norm": 0.9202230860830348, "learning_rate": 9.226186788538786e-06, "loss": 0.6253, "mean_token_accuracy": 0.7944542765617371, "num_tokens": 134183971.0, "step": 2048 }, { "epoch": 0.27081681205392544, "grad_norm": 0.9224651132163894, "learning_rate": 9.22502230911381e-06, "loss": 0.6384, "mean_token_accuracy": 0.7914021611213684, "num_tokens": 134249507.0, "step": 2049 }, { "epoch": 0.27094898228918846, "grad_norm": 0.9453341493041125, "learning_rate": 9.223857036709355e-06, "loss": 0.6416, "mean_token_accuracy": 0.7887162566184998, "num_tokens": 134315043.0, "step": 2050 }, { "epoch": 0.2710811525244515, "grad_norm": 1.0552538943367065, "learning_rate": 9.222690971573486e-06, "loss": 0.6487, "mean_token_accuracy": 0.7899523973464966, "num_tokens": 134380579.0, "step": 2051 }, { "epoch": 0.2712133227597145, "grad_norm": 0.8743489571339375, "learning_rate": 9.221524113954434e-06, "loss": 0.629, "mean_token_accuracy": 0.7964686751365662, "num_tokens": 134446115.0, "step": 2052 }, { "epoch": 0.2713454929949775, "grad_norm": 0.8864706077068325, "learning_rate": 9.220356464100597e-06, "loss": 0.5795, "mean_token_accuracy": 0.812904417514801, "num_tokens": 134511651.0, "step": 2053 }, { "epoch": 0.27147766323024053, "grad_norm": 0.980771766224713, "learning_rate": 9.219188022260545e-06, "loss": 0.6553, "mean_token_accuracy": 0.7849010825157166, "num_tokens": 134577187.0, "step": 2054 }, { "epoch": 0.27160983346550355, "grad_norm": 0.9568417289088551, "learning_rate": 9.218018788683012e-06, "loss": 0.6171, "mean_token_accuracy": 0.7968349456787109, "num_tokens": 134642723.0, "step": 2055 }, { "epoch": 0.27174200370076657, "grad_norm": 0.8472809786723513, "learning_rate": 9.216848763616908e-06, "loss": 0.5969, "mean_token_accuracy": 0.8072732090950012, "num_tokens": 134708259.0, "step": 2056 }, { "epoch": 0.2718741739360296, "grad_norm": 0.8463958558842979, "learning_rate": 9.215677947311305e-06, "loss": 0.5596, "mean_token_accuracy": 0.8200463652610779, "num_tokens": 134773795.0, "step": 2057 }, { "epoch": 0.2720063441712926, "grad_norm": 1.0040506240205824, "learning_rate": 9.214506340015443e-06, "loss": 0.6446, "mean_token_accuracy": 0.792348325252533, "num_tokens": 134839331.0, "step": 2058 }, { "epoch": 0.2721385144065556, "grad_norm": 0.9221129722199896, "learning_rate": 9.213333941978734e-06, "loss": 0.6432, "mean_token_accuracy": 0.788823127746582, "num_tokens": 134904867.0, "step": 2059 }, { "epoch": 0.27227068464181864, "grad_norm": 0.8675064509034546, "learning_rate": 9.21216075345076e-06, "loss": 0.639, "mean_token_accuracy": 0.7932639718055725, "num_tokens": 134970403.0, "step": 2060 }, { "epoch": 0.27240285487708166, "grad_norm": 0.9132132776678961, "learning_rate": 9.210986774681264e-06, "loss": 0.6376, "mean_token_accuracy": 0.793401300907135, "num_tokens": 135035939.0, "step": 2061 }, { "epoch": 0.2725350251123447, "grad_norm": 0.8701167648024186, "learning_rate": 9.209812005920163e-06, "loss": 0.6226, "mean_token_accuracy": 0.7976895570755005, "num_tokens": 135101475.0, "step": 2062 }, { "epoch": 0.2726671953476077, "grad_norm": 0.9460436263604217, "learning_rate": 9.208636447417543e-06, "loss": 0.6825, "mean_token_accuracy": 0.7799413800239563, "num_tokens": 135167011.0, "step": 2063 }, { "epoch": 0.2727993655828707, "grad_norm": 1.1139448695784306, "learning_rate": 9.207460099423653e-06, "loss": 0.703, "mean_token_accuracy": 0.7728757262229919, "num_tokens": 135232547.0, "step": 2064 }, { "epoch": 0.27293153581813373, "grad_norm": 0.8737320150891918, "learning_rate": 9.206282962188916e-06, "loss": 0.6612, "mean_token_accuracy": 0.7858167290687561, "num_tokens": 135298083.0, "step": 2065 }, { "epoch": 0.27306370605339675, "grad_norm": 0.885045793534967, "learning_rate": 9.205105035963914e-06, "loss": 0.6303, "mean_token_accuracy": 0.7955225110054016, "num_tokens": 135363619.0, "step": 2066 }, { "epoch": 0.27319587628865977, "grad_norm": 1.014711918527345, "learning_rate": 9.203926320999411e-06, "loss": 0.6476, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 135429155.0, "step": 2067 }, { "epoch": 0.2733280465239228, "grad_norm": 0.9112517648203219, "learning_rate": 9.202746817546326e-06, "loss": 0.6208, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 135494691.0, "step": 2068 }, { "epoch": 0.27346021675918586, "grad_norm": 0.8740669206083822, "learning_rate": 9.201566525855749e-06, "loss": 0.5995, "mean_token_accuracy": 0.8037174940109253, "num_tokens": 135560227.0, "step": 2069 }, { "epoch": 0.2735923869944489, "grad_norm": 0.9411109635662787, "learning_rate": 9.200385446178943e-06, "loss": 0.6116, "mean_token_accuracy": 0.8018709421157837, "num_tokens": 135625763.0, "step": 2070 }, { "epoch": 0.2737245572297119, "grad_norm": 1.0225254668825812, "learning_rate": 9.199203578767335e-06, "loss": 0.6637, "mean_token_accuracy": 0.7845959067344666, "num_tokens": 135691299.0, "step": 2071 }, { "epoch": 0.2738567274649749, "grad_norm": 0.9102602530530413, "learning_rate": 9.198020923872517e-06, "loss": 0.6436, "mean_token_accuracy": 0.7905322909355164, "num_tokens": 135756835.0, "step": 2072 }, { "epoch": 0.27398889770023793, "grad_norm": 0.8623650075471355, "learning_rate": 9.196837481746254e-06, "loss": 0.6143, "mean_token_accuracy": 0.8019319772720337, "num_tokens": 135822371.0, "step": 2073 }, { "epoch": 0.27412106793550095, "grad_norm": 0.8963275523043815, "learning_rate": 9.195653252640476e-06, "loss": 0.6874, "mean_token_accuracy": 0.7800635099411011, "num_tokens": 135887907.0, "step": 2074 }, { "epoch": 0.27425323817076397, "grad_norm": 0.9503660860123728, "learning_rate": 9.19446823680728e-06, "loss": 0.6987, "mean_token_accuracy": 0.774737536907196, "num_tokens": 135953443.0, "step": 2075 }, { "epoch": 0.274385408406027, "grad_norm": 0.9376964456240039, "learning_rate": 9.193282434498933e-06, "loss": 0.5817, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 136018979.0, "step": 2076 }, { "epoch": 0.27451757864129, "grad_norm": 0.885588908523124, "learning_rate": 9.192095845967864e-06, "loss": 0.5877, "mean_token_accuracy": 0.8106000423431396, "num_tokens": 136084515.0, "step": 2077 }, { "epoch": 0.274649748876553, "grad_norm": 0.7860216869244849, "learning_rate": 9.190908471466674e-06, "loss": 0.5481, "mean_token_accuracy": 0.8197106719017029, "num_tokens": 136150051.0, "step": 2078 }, { "epoch": 0.27478191911181604, "grad_norm": 0.9758708410160224, "learning_rate": 9.189720311248132e-06, "loss": 0.6622, "mean_token_accuracy": 0.782932460308075, "num_tokens": 136215587.0, "step": 2079 }, { "epoch": 0.27491408934707906, "grad_norm": 1.0508389992379679, "learning_rate": 9.188531365565171e-06, "loss": 0.6705, "mean_token_accuracy": 0.7817268967628479, "num_tokens": 136281123.0, "step": 2080 }, { "epoch": 0.2750462595823421, "grad_norm": 0.9751724447367962, "learning_rate": 9.187341634670896e-06, "loss": 0.6776, "mean_token_accuracy": 0.7797277569770813, "num_tokens": 136346659.0, "step": 2081 }, { "epoch": 0.2751784298176051, "grad_norm": 0.8718911282814809, "learning_rate": 9.186151118818571e-06, "loss": 0.5808, "mean_token_accuracy": 0.8114241361618042, "num_tokens": 136412195.0, "step": 2082 }, { "epoch": 0.2753106000528681, "grad_norm": 0.9253287541424678, "learning_rate": 9.184959818261632e-06, "loss": 0.6388, "mean_token_accuracy": 0.7929587364196777, "num_tokens": 136477731.0, "step": 2083 }, { "epoch": 0.27544277028813113, "grad_norm": 0.9693536053359141, "learning_rate": 9.183767733253688e-06, "loss": 0.6191, "mean_token_accuracy": 0.8003906607627869, "num_tokens": 136543267.0, "step": 2084 }, { "epoch": 0.27557494052339415, "grad_norm": 0.9310605567915836, "learning_rate": 9.182574864048504e-06, "loss": 0.5683, "mean_token_accuracy": 0.8144152164459229, "num_tokens": 136608803.0, "step": 2085 }, { "epoch": 0.27570711075865717, "grad_norm": 0.9364292172680679, "learning_rate": 9.181381210900016e-06, "loss": 0.6839, "mean_token_accuracy": 0.7764924764633179, "num_tokens": 136674339.0, "step": 2086 }, { "epoch": 0.2758392809939202, "grad_norm": 1.1427888302490217, "learning_rate": 9.180186774062332e-06, "loss": 0.6394, "mean_token_accuracy": 0.7906085848808289, "num_tokens": 136739875.0, "step": 2087 }, { "epoch": 0.2759714512291832, "grad_norm": 0.9116778884638216, "learning_rate": 9.178991553789721e-06, "loss": 0.5975, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 136805411.0, "step": 2088 }, { "epoch": 0.2761036214644462, "grad_norm": 0.8880208533373062, "learning_rate": 9.177795550336618e-06, "loss": 0.6321, "mean_token_accuracy": 0.799001932144165, "num_tokens": 136870947.0, "step": 2089 }, { "epoch": 0.27623579169970924, "grad_norm": 0.9665429519513113, "learning_rate": 9.176598763957629e-06, "loss": 0.6375, "mean_token_accuracy": 0.7926077246665955, "num_tokens": 136936483.0, "step": 2090 }, { "epoch": 0.27636796193497226, "grad_norm": 0.9698027754198711, "learning_rate": 9.175401194907527e-06, "loss": 0.6125, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 137002019.0, "step": 2091 }, { "epoch": 0.2765001321702353, "grad_norm": 0.9128547516905146, "learning_rate": 9.174202843441246e-06, "loss": 0.6272, "mean_token_accuracy": 0.7961329221725464, "num_tokens": 137067555.0, "step": 2092 }, { "epoch": 0.2766323024054983, "grad_norm": 0.9551600581910106, "learning_rate": 9.173003709813893e-06, "loss": 0.6327, "mean_token_accuracy": 0.792561948299408, "num_tokens": 137133091.0, "step": 2093 }, { "epoch": 0.2767644726407613, "grad_norm": 0.8906900889382183, "learning_rate": 9.171803794280736e-06, "loss": 0.6127, "mean_token_accuracy": 0.8006501197814941, "num_tokens": 137198627.0, "step": 2094 }, { "epoch": 0.27689664287602433, "grad_norm": 0.9653609268271502, "learning_rate": 9.170603097097213e-06, "loss": 0.6481, "mean_token_accuracy": 0.7905322909355164, "num_tokens": 137264163.0, "step": 2095 }, { "epoch": 0.27702881311128735, "grad_norm": 0.876728077159865, "learning_rate": 9.169401618518927e-06, "loss": 0.6212, "mean_token_accuracy": 0.7988646030426025, "num_tokens": 137329699.0, "step": 2096 }, { "epoch": 0.27716098334655037, "grad_norm": 0.9420004363015568, "learning_rate": 9.168199358801649e-06, "loss": 0.6292, "mean_token_accuracy": 0.7940880060195923, "num_tokens": 137395235.0, "step": 2097 }, { "epoch": 0.2772931535818134, "grad_norm": 0.9173422717828118, "learning_rate": 9.166996318201313e-06, "loss": 0.632, "mean_token_accuracy": 0.7908680438995361, "num_tokens": 137460771.0, "step": 2098 }, { "epoch": 0.2774253238170764, "grad_norm": 0.931761586530365, "learning_rate": 9.165792496974021e-06, "loss": 0.6101, "mean_token_accuracy": 0.8001617789268494, "num_tokens": 137526307.0, "step": 2099 }, { "epoch": 0.2775574940523394, "grad_norm": 0.964479862088126, "learning_rate": 9.164587895376045e-06, "loss": 0.652, "mean_token_accuracy": 0.7855268120765686, "num_tokens": 137591843.0, "step": 2100 }, { "epoch": 0.27768966428760244, "grad_norm": 0.968047658555628, "learning_rate": 9.163382513663817e-06, "loss": 0.615, "mean_token_accuracy": 0.798818826675415, "num_tokens": 137657379.0, "step": 2101 }, { "epoch": 0.27782183452286546, "grad_norm": 0.926207201374567, "learning_rate": 9.162176352093937e-06, "loss": 0.6151, "mean_token_accuracy": 0.8017030954360962, "num_tokens": 137722915.0, "step": 2102 }, { "epoch": 0.2779540047581285, "grad_norm": 0.9566363561445175, "learning_rate": 9.160969410923173e-06, "loss": 0.6601, "mean_token_accuracy": 0.7851300239562988, "num_tokens": 137788451.0, "step": 2103 }, { "epoch": 0.2780861749933915, "grad_norm": 0.8418754305889412, "learning_rate": 9.159761690408456e-06, "loss": 0.6237, "mean_token_accuracy": 0.7992766499519348, "num_tokens": 137853987.0, "step": 2104 }, { "epoch": 0.2782183452286545, "grad_norm": 0.9281951429078775, "learning_rate": 9.158553190806884e-06, "loss": 0.6021, "mean_token_accuracy": 0.8011232018470764, "num_tokens": 137919523.0, "step": 2105 }, { "epoch": 0.27835051546391754, "grad_norm": 1.0483434122953017, "learning_rate": 9.157343912375726e-06, "loss": 0.6662, "mean_token_accuracy": 0.7818642258644104, "num_tokens": 137985059.0, "step": 2106 }, { "epoch": 0.27848268569918055, "grad_norm": 0.947414147472181, "learning_rate": 9.156133855372407e-06, "loss": 0.6256, "mean_token_accuracy": 0.7957056760787964, "num_tokens": 138050595.0, "step": 2107 }, { "epoch": 0.2786148559344436, "grad_norm": 0.9200642495175755, "learning_rate": 9.154923020054525e-06, "loss": 0.6165, "mean_token_accuracy": 0.7995360493659973, "num_tokens": 138116131.0, "step": 2108 }, { "epoch": 0.2787470261697066, "grad_norm": 0.8291666366633876, "learning_rate": 9.153711406679842e-06, "loss": 0.6669, "mean_token_accuracy": 0.7870070934295654, "num_tokens": 138181667.0, "step": 2109 }, { "epoch": 0.2788791964049696, "grad_norm": 0.9161246675392296, "learning_rate": 9.152499015506285e-06, "loss": 0.5857, "mean_token_accuracy": 0.8088756203651428, "num_tokens": 138247203.0, "step": 2110 }, { "epoch": 0.2790113666402326, "grad_norm": 0.9122504293291069, "learning_rate": 9.151285846791945e-06, "loss": 0.6434, "mean_token_accuracy": 0.7894182801246643, "num_tokens": 138312739.0, "step": 2111 }, { "epoch": 0.27914353687549565, "grad_norm": 0.9523651996260055, "learning_rate": 9.150071900795082e-06, "loss": 0.6795, "mean_token_accuracy": 0.7787968516349792, "num_tokens": 138378275.0, "step": 2112 }, { "epoch": 0.27927570711075866, "grad_norm": 0.812349295379888, "learning_rate": 9.14885717777412e-06, "loss": 0.5683, "mean_token_accuracy": 0.812782347202301, "num_tokens": 138443811.0, "step": 2113 }, { "epoch": 0.2794078773460217, "grad_norm": 0.9044466381317885, "learning_rate": 9.147641677987645e-06, "loss": 0.6151, "mean_token_accuracy": 0.8017336130142212, "num_tokens": 138509347.0, "step": 2114 }, { "epoch": 0.2795400475812847, "grad_norm": 0.907369865387713, "learning_rate": 9.146425401694416e-06, "loss": 0.585, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 138574883.0, "step": 2115 }, { "epoch": 0.2796722178165477, "grad_norm": 0.926504032586058, "learning_rate": 9.145208349153351e-06, "loss": 0.6278, "mean_token_accuracy": 0.7956445813179016, "num_tokens": 138640419.0, "step": 2116 }, { "epoch": 0.27980438805181074, "grad_norm": 0.8660933235113314, "learning_rate": 9.143990520623534e-06, "loss": 0.6031, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 138705955.0, "step": 2117 }, { "epoch": 0.27993655828707376, "grad_norm": 0.9294183538067444, "learning_rate": 9.142771916364218e-06, "loss": 0.6356, "mean_token_accuracy": 0.793340265750885, "num_tokens": 138771491.0, "step": 2118 }, { "epoch": 0.2800687285223368, "grad_norm": 0.8414955246722519, "learning_rate": 9.141552536634816e-06, "loss": 0.6116, "mean_token_accuracy": 0.7999023199081421, "num_tokens": 138837027.0, "step": 2119 }, { "epoch": 0.2802008987575998, "grad_norm": 0.8307975015141212, "learning_rate": 9.140332381694907e-06, "loss": 0.5988, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 138902563.0, "step": 2120 }, { "epoch": 0.2803330689928628, "grad_norm": 0.8436374717558158, "learning_rate": 9.13911145180424e-06, "loss": 0.614, "mean_token_accuracy": 0.8016420602798462, "num_tokens": 138968099.0, "step": 2121 }, { "epoch": 0.28046523922812583, "grad_norm": 0.8872520438338045, "learning_rate": 9.137889747222726e-06, "loss": 0.5973, "mean_token_accuracy": 0.8066627979278564, "num_tokens": 139033635.0, "step": 2122 }, { "epoch": 0.28059740946338885, "grad_norm": 0.8580132375351381, "learning_rate": 9.136667268210436e-06, "loss": 0.6171, "mean_token_accuracy": 0.7996124029159546, "num_tokens": 139099171.0, "step": 2123 }, { "epoch": 0.28072957969865187, "grad_norm": 0.8290399574630088, "learning_rate": 9.135444015027612e-06, "loss": 0.6005, "mean_token_accuracy": 0.8048315048217773, "num_tokens": 139164707.0, "step": 2124 }, { "epoch": 0.2808617499339149, "grad_norm": 1.0093356465456491, "learning_rate": 9.134219987934662e-06, "loss": 0.6721, "mean_token_accuracy": 0.7816506028175354, "num_tokens": 139230243.0, "step": 2125 }, { "epoch": 0.2809939201691779, "grad_norm": 0.8984753015881387, "learning_rate": 9.132995187192154e-06, "loss": 0.5961, "mean_token_accuracy": 0.806861162185669, "num_tokens": 139295779.0, "step": 2126 }, { "epoch": 0.2811260904044409, "grad_norm": 0.928792833985375, "learning_rate": 9.131769613060819e-06, "loss": 0.6744, "mean_token_accuracy": 0.7805671095848083, "num_tokens": 139361315.0, "step": 2127 }, { "epoch": 0.28125826063970394, "grad_norm": 0.8500136343952563, "learning_rate": 9.130543265801562e-06, "loss": 0.6036, "mean_token_accuracy": 0.8030155301094055, "num_tokens": 139426851.0, "step": 2128 }, { "epoch": 0.28139043087496696, "grad_norm": 0.9068251125957514, "learning_rate": 9.129316145675443e-06, "loss": 0.6206, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 139492387.0, "step": 2129 }, { "epoch": 0.28152260111023, "grad_norm": 0.9700947789428213, "learning_rate": 9.128088252943691e-06, "loss": 0.6655, "mean_token_accuracy": 0.784458577632904, "num_tokens": 139557923.0, "step": 2130 }, { "epoch": 0.281654771345493, "grad_norm": 0.9068489302129902, "learning_rate": 9.1268595878677e-06, "loss": 0.6289, "mean_token_accuracy": 0.7947747707366943, "num_tokens": 139623459.0, "step": 2131 }, { "epoch": 0.281786941580756, "grad_norm": 0.9051861493117492, "learning_rate": 9.125630150709027e-06, "loss": 0.6196, "mean_token_accuracy": 0.7959498167037964, "num_tokens": 139688995.0, "step": 2132 }, { "epoch": 0.28191911181601903, "grad_norm": 0.9665755343560879, "learning_rate": 9.124399941729392e-06, "loss": 0.6805, "mean_token_accuracy": 0.7772707939147949, "num_tokens": 139754531.0, "step": 2133 }, { "epoch": 0.28205128205128205, "grad_norm": 0.9179947256901094, "learning_rate": 9.12316896119068e-06, "loss": 0.5904, "mean_token_accuracy": 0.8110120892524719, "num_tokens": 139820067.0, "step": 2134 }, { "epoch": 0.28218345228654507, "grad_norm": 0.8426598169676379, "learning_rate": 9.121937209354943e-06, "loss": 0.6373, "mean_token_accuracy": 0.7904559969902039, "num_tokens": 139885603.0, "step": 2135 }, { "epoch": 0.2823156225218081, "grad_norm": 0.8319220080012821, "learning_rate": 9.120704686484394e-06, "loss": 0.5943, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 139951139.0, "step": 2136 }, { "epoch": 0.2824477927570711, "grad_norm": 0.8881983345784793, "learning_rate": 9.119471392841413e-06, "loss": 0.6472, "mean_token_accuracy": 0.7876022458076477, "num_tokens": 140016675.0, "step": 2137 }, { "epoch": 0.2825799629923341, "grad_norm": 0.879265549880044, "learning_rate": 9.118237328688541e-06, "loss": 0.6213, "mean_token_accuracy": 0.7964991927146912, "num_tokens": 140082211.0, "step": 2138 }, { "epoch": 0.28271213322759714, "grad_norm": 0.8928634793493556, "learning_rate": 9.117002494288487e-06, "loss": 0.5805, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 140147747.0, "step": 2139 }, { "epoch": 0.28284430346286016, "grad_norm": 0.8477691456447004, "learning_rate": 9.11576688990412e-06, "loss": 0.6185, "mean_token_accuracy": 0.7984678149223328, "num_tokens": 140213283.0, "step": 2140 }, { "epoch": 0.2829764736981232, "grad_norm": 0.9521372547368884, "learning_rate": 9.11453051579847e-06, "loss": 0.6732, "mean_token_accuracy": 0.7805823683738708, "num_tokens": 140278819.0, "step": 2141 }, { "epoch": 0.2831086439333862, "grad_norm": 0.8719143660501445, "learning_rate": 9.113293372234741e-06, "loss": 0.6181, "mean_token_accuracy": 0.7992156147956848, "num_tokens": 140344355.0, "step": 2142 }, { "epoch": 0.2832408141686492, "grad_norm": 0.875504536911158, "learning_rate": 9.112055459476294e-06, "loss": 0.6047, "mean_token_accuracy": 0.8004059195518494, "num_tokens": 140409891.0, "step": 2143 }, { "epoch": 0.28337298440391223, "grad_norm": 0.892269693497411, "learning_rate": 9.110816777786653e-06, "loss": 0.6231, "mean_token_accuracy": 0.7957514524459839, "num_tokens": 140475427.0, "step": 2144 }, { "epoch": 0.28350515463917525, "grad_norm": 0.8191913435554, "learning_rate": 9.109577327429509e-06, "loss": 0.6095, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 140540963.0, "step": 2145 }, { "epoch": 0.28363732487443827, "grad_norm": 0.9030918728161837, "learning_rate": 9.108337108668715e-06, "loss": 0.6188, "mean_token_accuracy": 0.7960108518600464, "num_tokens": 140606499.0, "step": 2146 }, { "epoch": 0.2837694951097013, "grad_norm": 0.8760118501275307, "learning_rate": 9.107096121768285e-06, "loss": 0.6228, "mean_token_accuracy": 0.7959955930709839, "num_tokens": 140672035.0, "step": 2147 }, { "epoch": 0.2839016653449643, "grad_norm": 0.9594055917831532, "learning_rate": 9.105854366992404e-06, "loss": 0.6331, "mean_token_accuracy": 0.7951715588569641, "num_tokens": 140737571.0, "step": 2148 }, { "epoch": 0.2840338355802273, "grad_norm": 0.893534922482763, "learning_rate": 9.104611844605413e-06, "loss": 0.5978, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 140803107.0, "step": 2149 }, { "epoch": 0.28416600581549034, "grad_norm": 0.9591901974400653, "learning_rate": 9.103368554871816e-06, "loss": 0.651, "mean_token_accuracy": 0.7861219644546509, "num_tokens": 140868643.0, "step": 2150 }, { "epoch": 0.28429817605075336, "grad_norm": 1.0002760568071154, "learning_rate": 9.102124498056289e-06, "loss": 0.6904, "mean_token_accuracy": 0.7758362889289856, "num_tokens": 140934179.0, "step": 2151 }, { "epoch": 0.2844303462860164, "grad_norm": 0.8688617934235381, "learning_rate": 9.100879674423664e-06, "loss": 0.6269, "mean_token_accuracy": 0.7951715588569641, "num_tokens": 140999715.0, "step": 2152 }, { "epoch": 0.2845625165212794, "grad_norm": 0.8279023199966381, "learning_rate": 9.099634084238936e-06, "loss": 0.5632, "mean_token_accuracy": 0.8167348504066467, "num_tokens": 141065251.0, "step": 2153 }, { "epoch": 0.2846946867565424, "grad_norm": 0.986932872246236, "learning_rate": 9.098387727767268e-06, "loss": 0.6547, "mean_token_accuracy": 0.7890825271606445, "num_tokens": 141130787.0, "step": 2154 }, { "epoch": 0.28482685699180543, "grad_norm": 0.8671211903039557, "learning_rate": 9.09714060527398e-06, "loss": 0.6386, "mean_token_accuracy": 0.7935996651649475, "num_tokens": 141196323.0, "step": 2155 }, { "epoch": 0.28495902722706845, "grad_norm": 0.8968790235382021, "learning_rate": 9.09589271702456e-06, "loss": 0.6151, "mean_token_accuracy": 0.7983304858207703, "num_tokens": 141261859.0, "step": 2156 }, { "epoch": 0.28509119746233147, "grad_norm": 0.8876179499115773, "learning_rate": 9.094644063284659e-06, "loss": 0.629, "mean_token_accuracy": 0.7953241467475891, "num_tokens": 141327395.0, "step": 2157 }, { "epoch": 0.2852233676975945, "grad_norm": 0.9750610557584023, "learning_rate": 9.093394644320086e-06, "loss": 0.6609, "mean_token_accuracy": 0.783863365650177, "num_tokens": 141392931.0, "step": 2158 }, { "epoch": 0.2853555379328575, "grad_norm": 0.9590408026020236, "learning_rate": 9.092144460396818e-06, "loss": 0.6583, "mean_token_accuracy": 0.7842296361923218, "num_tokens": 141458467.0, "step": 2159 }, { "epoch": 0.2854877081681205, "grad_norm": 0.9920597249892767, "learning_rate": 9.090893511780994e-06, "loss": 0.6454, "mean_token_accuracy": 0.7904102206230164, "num_tokens": 141524003.0, "step": 2160 }, { "epoch": 0.28561987840338354, "grad_norm": 0.9384690558088382, "learning_rate": 9.089641798738914e-06, "loss": 0.6598, "mean_token_accuracy": 0.7853436470031738, "num_tokens": 141589539.0, "step": 2161 }, { "epoch": 0.28575204863864656, "grad_norm": 0.9712954030523906, "learning_rate": 9.08838932153704e-06, "loss": 0.6455, "mean_token_accuracy": 0.7884110808372498, "num_tokens": 141655075.0, "step": 2162 }, { "epoch": 0.2858842188739096, "grad_norm": 0.9841879699725222, "learning_rate": 9.087136080441999e-06, "loss": 0.6872, "mean_token_accuracy": 0.7778812050819397, "num_tokens": 141720611.0, "step": 2163 }, { "epoch": 0.2860163891091726, "grad_norm": 0.9425693619881355, "learning_rate": 9.085882075720583e-06, "loss": 0.5973, "mean_token_accuracy": 0.8045415878295898, "num_tokens": 141786147.0, "step": 2164 }, { "epoch": 0.2861485593444356, "grad_norm": 0.9859627939453438, "learning_rate": 9.084627307639739e-06, "loss": 0.6271, "mean_token_accuracy": 0.7965602278709412, "num_tokens": 141851683.0, "step": 2165 }, { "epoch": 0.28628072957969863, "grad_norm": 0.9106666653317991, "learning_rate": 9.083371776466584e-06, "loss": 0.6204, "mean_token_accuracy": 0.7980252504348755, "num_tokens": 141917219.0, "step": 2166 }, { "epoch": 0.28641289981496165, "grad_norm": 0.9861728365082755, "learning_rate": 9.082115482468391e-06, "loss": 0.6148, "mean_token_accuracy": 0.8012757897377014, "num_tokens": 141982755.0, "step": 2167 }, { "epoch": 0.28654507005022467, "grad_norm": 0.9394854985514932, "learning_rate": 9.080858425912604e-06, "loss": 0.6364, "mean_token_accuracy": 0.7940269708633423, "num_tokens": 142048291.0, "step": 2168 }, { "epoch": 0.2866772402854877, "grad_norm": 0.9179721083898756, "learning_rate": 9.07960060706682e-06, "loss": 0.6309, "mean_token_accuracy": 0.7972164750099182, "num_tokens": 142113827.0, "step": 2169 }, { "epoch": 0.2868094105207507, "grad_norm": 0.9231084772351387, "learning_rate": 9.078342026198803e-06, "loss": 0.5938, "mean_token_accuracy": 0.8066322803497314, "num_tokens": 142179363.0, "step": 2170 }, { "epoch": 0.2869415807560137, "grad_norm": 1.0085481035366408, "learning_rate": 9.077082683576479e-06, "loss": 0.6006, "mean_token_accuracy": 0.8055487871170044, "num_tokens": 142244899.0, "step": 2171 }, { "epoch": 0.28707375099127674, "grad_norm": 0.9183091879546013, "learning_rate": 9.075822579467937e-06, "loss": 0.5947, "mean_token_accuracy": 0.8043126463890076, "num_tokens": 142310435.0, "step": 2172 }, { "epoch": 0.28720592122653976, "grad_norm": 0.9282550694380084, "learning_rate": 9.074561714141424e-06, "loss": 0.6477, "mean_token_accuracy": 0.789097785949707, "num_tokens": 142375971.0, "step": 2173 }, { "epoch": 0.2873380914618028, "grad_norm": 0.9830020079625179, "learning_rate": 9.073300087865355e-06, "loss": 0.607, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 142441507.0, "step": 2174 }, { "epoch": 0.2874702616970658, "grad_norm": 0.8971501240490851, "learning_rate": 9.0720377009083e-06, "loss": 0.6119, "mean_token_accuracy": 0.8002685904502869, "num_tokens": 142507043.0, "step": 2175 }, { "epoch": 0.2876024319323288, "grad_norm": 0.9571824367972285, "learning_rate": 9.070774553539e-06, "loss": 0.6492, "mean_token_accuracy": 0.788853645324707, "num_tokens": 142572579.0, "step": 2176 }, { "epoch": 0.28773460216759184, "grad_norm": 0.9745280665515139, "learning_rate": 9.069510646026348e-06, "loss": 0.6429, "mean_token_accuracy": 0.7928214073181152, "num_tokens": 142638115.0, "step": 2177 }, { "epoch": 0.28786677240285485, "grad_norm": 0.859497319201066, "learning_rate": 9.068245978639406e-06, "loss": 0.6096, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 142703651.0, "step": 2178 }, { "epoch": 0.28799894263811787, "grad_norm": 0.9119296561756863, "learning_rate": 9.066980551647395e-06, "loss": 0.5926, "mean_token_accuracy": 0.8079141974449158, "num_tokens": 142769187.0, "step": 2179 }, { "epoch": 0.2881311128733809, "grad_norm": 1.0663892382161382, "learning_rate": 9.065714365319698e-06, "loss": 0.7139, "mean_token_accuracy": 0.7674734592437744, "num_tokens": 142834723.0, "step": 2180 }, { "epoch": 0.2882632831086439, "grad_norm": 0.9606482740497815, "learning_rate": 9.064447419925858e-06, "loss": 0.706, "mean_token_accuracy": 0.7724941968917847, "num_tokens": 142900259.0, "step": 2181 }, { "epoch": 0.2883954533439069, "grad_norm": 0.9615735376331483, "learning_rate": 9.063179715735586e-06, "loss": 0.6599, "mean_token_accuracy": 0.7831308841705322, "num_tokens": 142965795.0, "step": 2182 }, { "epoch": 0.28852762357916995, "grad_norm": 1.0097155567158933, "learning_rate": 9.061911253018742e-06, "loss": 0.6961, "mean_token_accuracy": 0.7736998200416565, "num_tokens": 143031331.0, "step": 2183 }, { "epoch": 0.28865979381443296, "grad_norm": 0.9519045200908987, "learning_rate": 9.060642032045361e-06, "loss": 0.6452, "mean_token_accuracy": 0.7921956777572632, "num_tokens": 143096867.0, "step": 2184 }, { "epoch": 0.288791964049696, "grad_norm": 0.8949013984099442, "learning_rate": 9.059372053085635e-06, "loss": 0.611, "mean_token_accuracy": 0.8010621666908264, "num_tokens": 143162403.0, "step": 2185 }, { "epoch": 0.288924134284959, "grad_norm": 0.8519853948192043, "learning_rate": 9.058101316409911e-06, "loss": 0.6153, "mean_token_accuracy": 0.7995055317878723, "num_tokens": 143227939.0, "step": 2186 }, { "epoch": 0.289056304520222, "grad_norm": 0.8489169665915418, "learning_rate": 9.056829822288705e-06, "loss": 0.5822, "mean_token_accuracy": 0.8113936185836792, "num_tokens": 143293475.0, "step": 2187 }, { "epoch": 0.2891884747554851, "grad_norm": 0.842929640517054, "learning_rate": 9.05555757099269e-06, "loss": 0.6239, "mean_token_accuracy": 0.7969570159912109, "num_tokens": 143359011.0, "step": 2188 }, { "epoch": 0.2893206449907481, "grad_norm": 0.9171088251588744, "learning_rate": 9.054284562792704e-06, "loss": 0.6039, "mean_token_accuracy": 0.8028781414031982, "num_tokens": 143424547.0, "step": 2189 }, { "epoch": 0.28945281522601113, "grad_norm": 0.9836071993677646, "learning_rate": 9.053010797959743e-06, "loss": 0.614, "mean_token_accuracy": 0.7967128753662109, "num_tokens": 143490083.0, "step": 2190 }, { "epoch": 0.28958498546127415, "grad_norm": 0.9055646510651447, "learning_rate": 9.051736276764963e-06, "loss": 0.6164, "mean_token_accuracy": 0.7995818853378296, "num_tokens": 143555619.0, "step": 2191 }, { "epoch": 0.28971715569653717, "grad_norm": 0.9685905887270542, "learning_rate": 9.050460999479685e-06, "loss": 0.6547, "mean_token_accuracy": 0.7864424586296082, "num_tokens": 143621155.0, "step": 2192 }, { "epoch": 0.2898493259318002, "grad_norm": 0.8945345955802729, "learning_rate": 9.04918496637539e-06, "loss": 0.6698, "mean_token_accuracy": 0.7816200852394104, "num_tokens": 143686691.0, "step": 2193 }, { "epoch": 0.2899814961670632, "grad_norm": 0.8433370821582323, "learning_rate": 9.047908177723715e-06, "loss": 0.627, "mean_token_accuracy": 0.7968654632568359, "num_tokens": 143752227.0, "step": 2194 }, { "epoch": 0.2901136664023262, "grad_norm": 0.9415098560343549, "learning_rate": 9.046630633796465e-06, "loss": 0.6015, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 143817763.0, "step": 2195 }, { "epoch": 0.29024583663758924, "grad_norm": 1.0293101626133758, "learning_rate": 9.045352334865602e-06, "loss": 0.6072, "mean_token_accuracy": 0.8047094345092773, "num_tokens": 143883299.0, "step": 2196 }, { "epoch": 0.29037800687285226, "grad_norm": 0.8756434721433147, "learning_rate": 9.044073281203248e-06, "loss": 0.5981, "mean_token_accuracy": 0.8050299286842346, "num_tokens": 143948835.0, "step": 2197 }, { "epoch": 0.2905101771081153, "grad_norm": 0.8070253366813488, "learning_rate": 9.042793473081687e-06, "loss": 0.5858, "mean_token_accuracy": 0.8126907348632812, "num_tokens": 144014371.0, "step": 2198 }, { "epoch": 0.2906423473433783, "grad_norm": 0.9955201883736418, "learning_rate": 9.041512910773365e-06, "loss": 0.6202, "mean_token_accuracy": 0.7949120998382568, "num_tokens": 144079907.0, "step": 2199 }, { "epoch": 0.2907745175786413, "grad_norm": 1.0120949182171766, "learning_rate": 9.040231594550886e-06, "loss": 0.5954, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 144145443.0, "step": 2200 }, { "epoch": 0.29090668781390433, "grad_norm": 1.245788538154328, "learning_rate": 9.038949524687014e-06, "loss": 0.6883, "mean_token_accuracy": 0.7744170427322388, "num_tokens": 144210979.0, "step": 2201 }, { "epoch": 0.29103885804916735, "grad_norm": 0.9811301656731158, "learning_rate": 9.037666701454676e-06, "loss": 0.6374, "mean_token_accuracy": 0.7897539734840393, "num_tokens": 144276515.0, "step": 2202 }, { "epoch": 0.29117102828443037, "grad_norm": 0.875480200219596, "learning_rate": 9.036383125126958e-06, "loss": 0.6301, "mean_token_accuracy": 0.7948510646820068, "num_tokens": 144342051.0, "step": 2203 }, { "epoch": 0.2913031985196934, "grad_norm": 0.9100600999867663, "learning_rate": 9.035098795977108e-06, "loss": 0.6681, "mean_token_accuracy": 0.7807654738426208, "num_tokens": 144407587.0, "step": 2204 }, { "epoch": 0.2914353687549564, "grad_norm": 0.8777544658238806, "learning_rate": 9.033813714278533e-06, "loss": 0.6029, "mean_token_accuracy": 0.807456374168396, "num_tokens": 144473123.0, "step": 2205 }, { "epoch": 0.2915675389902194, "grad_norm": 0.9294035836986703, "learning_rate": 9.032527880304797e-06, "loss": 0.6426, "mean_token_accuracy": 0.7899371385574341, "num_tokens": 144538659.0, "step": 2206 }, { "epoch": 0.29169970922548244, "grad_norm": 0.9491976300994068, "learning_rate": 9.03124129432963e-06, "loss": 0.6351, "mean_token_accuracy": 0.7920278310775757, "num_tokens": 144604195.0, "step": 2207 }, { "epoch": 0.29183187946074546, "grad_norm": 0.846594875562213, "learning_rate": 9.02995395662692e-06, "loss": 0.5856, "mean_token_accuracy": 0.8070901036262512, "num_tokens": 144669731.0, "step": 2208 }, { "epoch": 0.2919640496960085, "grad_norm": 0.8107449337898157, "learning_rate": 9.02866586747071e-06, "loss": 0.5705, "mean_token_accuracy": 0.8105237483978271, "num_tokens": 144735267.0, "step": 2209 }, { "epoch": 0.2920962199312715, "grad_norm": 0.892991045452015, "learning_rate": 9.027377027135212e-06, "loss": 0.6267, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 144800803.0, "step": 2210 }, { "epoch": 0.2922283901665345, "grad_norm": 0.9649537982809895, "learning_rate": 9.026087435894789e-06, "loss": 0.6555, "mean_token_accuracy": 0.7895098328590393, "num_tokens": 144866339.0, "step": 2211 }, { "epoch": 0.29236056040179753, "grad_norm": 0.9350628452041935, "learning_rate": 9.024797094023971e-06, "loss": 0.6532, "mean_token_accuracy": 0.7907001376152039, "num_tokens": 144931875.0, "step": 2212 }, { "epoch": 0.29249273063706055, "grad_norm": 0.9125815886021972, "learning_rate": 9.023506001797445e-06, "loss": 0.6101, "mean_token_accuracy": 0.8010469079017639, "num_tokens": 144997411.0, "step": 2213 }, { "epoch": 0.29262490087232357, "grad_norm": 0.8654405586804437, "learning_rate": 9.022214159490058e-06, "loss": 0.6243, "mean_token_accuracy": 0.7966518402099609, "num_tokens": 145062947.0, "step": 2214 }, { "epoch": 0.2927570711075866, "grad_norm": 0.9335341312891728, "learning_rate": 9.020921567376811e-06, "loss": 0.6424, "mean_token_accuracy": 0.7912189960479736, "num_tokens": 145128483.0, "step": 2215 }, { "epoch": 0.2928892413428496, "grad_norm": 0.9650986695481241, "learning_rate": 9.019628225732879e-06, "loss": 0.6247, "mean_token_accuracy": 0.7953088879585266, "num_tokens": 145194019.0, "step": 2216 }, { "epoch": 0.2930214115781126, "grad_norm": 0.8874418477118781, "learning_rate": 9.018334134833579e-06, "loss": 0.6425, "mean_token_accuracy": 0.7898913621902466, "num_tokens": 145259555.0, "step": 2217 }, { "epoch": 0.29315358181337564, "grad_norm": 0.9562561514659528, "learning_rate": 9.017039294954401e-06, "loss": 0.6277, "mean_token_accuracy": 0.7966670989990234, "num_tokens": 145325091.0, "step": 2218 }, { "epoch": 0.29328575204863866, "grad_norm": 0.8906996364539601, "learning_rate": 9.015743706370985e-06, "loss": 0.588, "mean_token_accuracy": 0.8103559017181396, "num_tokens": 145390627.0, "step": 2219 }, { "epoch": 0.2934179222839017, "grad_norm": 0.9625473921159745, "learning_rate": 9.014447369359138e-06, "loss": 0.6756, "mean_token_accuracy": 0.7813606262207031, "num_tokens": 145456163.0, "step": 2220 }, { "epoch": 0.2935500925191647, "grad_norm": 0.9971933663883187, "learning_rate": 9.013150284194824e-06, "loss": 0.6209, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 145521699.0, "step": 2221 }, { "epoch": 0.2936822627544277, "grad_norm": 0.9228615515196158, "learning_rate": 9.011852451154164e-06, "loss": 0.633, "mean_token_accuracy": 0.7941337823867798, "num_tokens": 145587235.0, "step": 2222 }, { "epoch": 0.29381443298969073, "grad_norm": 0.8606461263096068, "learning_rate": 9.010553870513437e-06, "loss": 0.5589, "mean_token_accuracy": 0.816521167755127, "num_tokens": 145652771.0, "step": 2223 }, { "epoch": 0.29394660322495375, "grad_norm": 0.9958382696359359, "learning_rate": 9.009254542549088e-06, "loss": 0.6478, "mean_token_accuracy": 0.7879990339279175, "num_tokens": 145718307.0, "step": 2224 }, { "epoch": 0.29407877346021677, "grad_norm": 0.9104333529483869, "learning_rate": 9.007954467537714e-06, "loss": 0.6142, "mean_token_accuracy": 0.8008484840393066, "num_tokens": 145783843.0, "step": 2225 }, { "epoch": 0.2942109436954798, "grad_norm": 0.8961724906660297, "learning_rate": 9.006653645756075e-06, "loss": 0.6163, "mean_token_accuracy": 0.7982541918754578, "num_tokens": 145849379.0, "step": 2226 }, { "epoch": 0.2943431139307428, "grad_norm": 0.9948765681547227, "learning_rate": 9.00535207748109e-06, "loss": 0.6725, "mean_token_accuracy": 0.7814369201660156, "num_tokens": 145914915.0, "step": 2227 }, { "epoch": 0.2944752841660058, "grad_norm": 0.9135068989850572, "learning_rate": 9.004049762989831e-06, "loss": 0.5818, "mean_token_accuracy": 0.8125076293945312, "num_tokens": 145980451.0, "step": 2228 }, { "epoch": 0.29460745440126884, "grad_norm": 0.9471397243128142, "learning_rate": 9.00274670255954e-06, "loss": 0.6432, "mean_token_accuracy": 0.7897387146949768, "num_tokens": 146045987.0, "step": 2229 }, { "epoch": 0.29473962463653186, "grad_norm": 1.0138121262905466, "learning_rate": 9.001442896467606e-06, "loss": 0.6855, "mean_token_accuracy": 0.7778812050819397, "num_tokens": 146111523.0, "step": 2230 }, { "epoch": 0.2948717948717949, "grad_norm": 0.9607842905566549, "learning_rate": 9.000138344991584e-06, "loss": 0.6864, "mean_token_accuracy": 0.7762178182601929, "num_tokens": 146177059.0, "step": 2231 }, { "epoch": 0.2950039651070579, "grad_norm": 0.9393360251450048, "learning_rate": 8.998833048409188e-06, "loss": 0.5856, "mean_token_accuracy": 0.8105847835540771, "num_tokens": 146242595.0, "step": 2232 }, { "epoch": 0.2951361353423209, "grad_norm": 1.0506070451764518, "learning_rate": 8.997527006998283e-06, "loss": 0.6557, "mean_token_accuracy": 0.788945198059082, "num_tokens": 146308131.0, "step": 2233 }, { "epoch": 0.29526830557758393, "grad_norm": 0.9164621335758351, "learning_rate": 8.996220221036903e-06, "loss": 0.6536, "mean_token_accuracy": 0.788975715637207, "num_tokens": 146373667.0, "step": 2234 }, { "epoch": 0.29540047581284695, "grad_norm": 0.9593689237895872, "learning_rate": 8.994912690803236e-06, "loss": 0.6457, "mean_token_accuracy": 0.7904559969902039, "num_tokens": 146439203.0, "step": 2235 }, { "epoch": 0.29553264604810997, "grad_norm": 0.9104682370648493, "learning_rate": 8.993604416575624e-06, "loss": 0.6352, "mean_token_accuracy": 0.7926077246665955, "num_tokens": 146504739.0, "step": 2236 }, { "epoch": 0.295664816283373, "grad_norm": 0.9654941431220915, "learning_rate": 8.992295398632572e-06, "loss": 0.6116, "mean_token_accuracy": 0.8005127310752869, "num_tokens": 146570275.0, "step": 2237 }, { "epoch": 0.295796986518636, "grad_norm": 0.9314046409574621, "learning_rate": 8.990985637252746e-06, "loss": 0.6609, "mean_token_accuracy": 0.7860456705093384, "num_tokens": 146635811.0, "step": 2238 }, { "epoch": 0.295929156753899, "grad_norm": 0.8998727791503571, "learning_rate": 8.989675132714962e-06, "loss": 0.6826, "mean_token_accuracy": 0.7748138308525085, "num_tokens": 146701347.0, "step": 2239 }, { "epoch": 0.29606132698916204, "grad_norm": 0.9871164522077248, "learning_rate": 8.988363885298202e-06, "loss": 0.7194, "mean_token_accuracy": 0.7677786350250244, "num_tokens": 146766883.0, "step": 2240 }, { "epoch": 0.29619349722442506, "grad_norm": 0.8756981518088149, "learning_rate": 8.987051895281602e-06, "loss": 0.6048, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 146832419.0, "step": 2241 }, { "epoch": 0.2963256674596881, "grad_norm": 0.9052983639158521, "learning_rate": 8.985739162944458e-06, "loss": 0.5762, "mean_token_accuracy": 0.8102948069572449, "num_tokens": 146897955.0, "step": 2242 }, { "epoch": 0.2964578376949511, "grad_norm": 0.994405461349212, "learning_rate": 8.984425688566224e-06, "loss": 0.6526, "mean_token_accuracy": 0.7846264243125916, "num_tokens": 146963491.0, "step": 2243 }, { "epoch": 0.2965900079302141, "grad_norm": 0.9219028670708382, "learning_rate": 8.98311147242651e-06, "loss": 0.6261, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 147029027.0, "step": 2244 }, { "epoch": 0.29672217816547714, "grad_norm": 0.8690884792059623, "learning_rate": 8.981796514805087e-06, "loss": 0.592, "mean_token_accuracy": 0.8079752326011658, "num_tokens": 147094563.0, "step": 2245 }, { "epoch": 0.29685434840074015, "grad_norm": 0.9019140514511353, "learning_rate": 8.980480815981878e-06, "loss": 0.6542, "mean_token_accuracy": 0.7872664928436279, "num_tokens": 147160099.0, "step": 2246 }, { "epoch": 0.29698651863600317, "grad_norm": 0.8955039071211136, "learning_rate": 8.979164376236973e-06, "loss": 0.6333, "mean_token_accuracy": 0.7967739105224609, "num_tokens": 147225635.0, "step": 2247 }, { "epoch": 0.2971186888712662, "grad_norm": 0.9796200876152187, "learning_rate": 8.97784719585061e-06, "loss": 0.6506, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 147291171.0, "step": 2248 }, { "epoch": 0.2972508591065292, "grad_norm": 0.8923175844909537, "learning_rate": 8.976529275103195e-06, "loss": 0.62, "mean_token_accuracy": 0.7988951206207275, "num_tokens": 147356707.0, "step": 2249 }, { "epoch": 0.2973830293417922, "grad_norm": 0.8764533727489483, "learning_rate": 8.975210614275277e-06, "loss": 0.6184, "mean_token_accuracy": 0.799001932144165, "num_tokens": 147422243.0, "step": 2250 }, { "epoch": 0.29751519957705524, "grad_norm": 0.9154310927771917, "learning_rate": 8.97389121364758e-06, "loss": 0.6329, "mean_token_accuracy": 0.7947747707366943, "num_tokens": 147487779.0, "step": 2251 }, { "epoch": 0.29764736981231826, "grad_norm": 0.9572134538817247, "learning_rate": 8.97257107350097e-06, "loss": 0.664, "mean_token_accuracy": 0.783680260181427, "num_tokens": 147553315.0, "step": 2252 }, { "epoch": 0.2977795400475813, "grad_norm": 0.8726238264790459, "learning_rate": 8.971250194116483e-06, "loss": 0.5794, "mean_token_accuracy": 0.8112714886665344, "num_tokens": 147618851.0, "step": 2253 }, { "epoch": 0.2979117102828443, "grad_norm": 0.8665489376689989, "learning_rate": 8.969928575775305e-06, "loss": 0.5947, "mean_token_accuracy": 0.8047857284545898, "num_tokens": 147684387.0, "step": 2254 }, { "epoch": 0.2980438805181073, "grad_norm": 0.7902455451784721, "learning_rate": 8.968606218758778e-06, "loss": 0.5733, "mean_token_accuracy": 0.8143389225006104, "num_tokens": 147749923.0, "step": 2255 }, { "epoch": 0.29817605075337034, "grad_norm": 0.9777708378179922, "learning_rate": 8.967283123348407e-06, "loss": 0.6877, "mean_token_accuracy": 0.7798650860786438, "num_tokens": 147815459.0, "step": 2256 }, { "epoch": 0.29830822098863335, "grad_norm": 0.9526743466823063, "learning_rate": 8.965959289825851e-06, "loss": 0.638, "mean_token_accuracy": 0.7934775948524475, "num_tokens": 147880995.0, "step": 2257 }, { "epoch": 0.2984403912238964, "grad_norm": 1.0255789935341049, "learning_rate": 8.964634718472924e-06, "loss": 0.6864, "mean_token_accuracy": 0.7766146063804626, "num_tokens": 147946531.0, "step": 2258 }, { "epoch": 0.2985725614591594, "grad_norm": 1.0387360782634174, "learning_rate": 8.963309409571605e-06, "loss": 0.6329, "mean_token_accuracy": 0.7921499013900757, "num_tokens": 148012067.0, "step": 2259 }, { "epoch": 0.2987047316944224, "grad_norm": 0.9808966506769792, "learning_rate": 8.96198336340402e-06, "loss": 0.6434, "mean_token_accuracy": 0.7930197715759277, "num_tokens": 148077603.0, "step": 2260 }, { "epoch": 0.29883690192968543, "grad_norm": 0.9104561636043859, "learning_rate": 8.960656580252458e-06, "loss": 0.5909, "mean_token_accuracy": 0.8066627979278564, "num_tokens": 148143139.0, "step": 2261 }, { "epoch": 0.29896907216494845, "grad_norm": 0.9044063783197026, "learning_rate": 8.959329060399365e-06, "loss": 0.6485, "mean_token_accuracy": 0.7905780673027039, "num_tokens": 148208675.0, "step": 2262 }, { "epoch": 0.29910124240021146, "grad_norm": 0.8801004206899724, "learning_rate": 8.958000804127338e-06, "loss": 0.6434, "mean_token_accuracy": 0.7922261953353882, "num_tokens": 148274211.0, "step": 2263 }, { "epoch": 0.2992334126354745, "grad_norm": 0.898166853885653, "learning_rate": 8.956671811719138e-06, "loss": 0.5862, "mean_token_accuracy": 0.8087992668151855, "num_tokens": 148339747.0, "step": 2264 }, { "epoch": 0.2993655828707375, "grad_norm": 0.8883578191864738, "learning_rate": 8.95534208345768e-06, "loss": 0.6471, "mean_token_accuracy": 0.7885789275169373, "num_tokens": 148405283.0, "step": 2265 }, { "epoch": 0.2994977531060005, "grad_norm": 0.9832147581755142, "learning_rate": 8.954011619626037e-06, "loss": 0.6614, "mean_token_accuracy": 0.7852673530578613, "num_tokens": 148470819.0, "step": 2266 }, { "epoch": 0.29962992334126354, "grad_norm": 0.9794672920026233, "learning_rate": 8.952680420507432e-06, "loss": 0.6002, "mean_token_accuracy": 0.802313506603241, "num_tokens": 148536355.0, "step": 2267 }, { "epoch": 0.29976209357652656, "grad_norm": 0.8637463468327917, "learning_rate": 8.951348486385251e-06, "loss": 0.5699, "mean_token_accuracy": 0.8144304752349854, "num_tokens": 148601891.0, "step": 2268 }, { "epoch": 0.2998942638117896, "grad_norm": 0.902504238846881, "learning_rate": 8.95001581754304e-06, "loss": 0.6113, "mean_token_accuracy": 0.8017793893814087, "num_tokens": 148667427.0, "step": 2269 }, { "epoch": 0.3000264340470526, "grad_norm": 0.878919587467285, "learning_rate": 8.94868241426449e-06, "loss": 0.625, "mean_token_accuracy": 0.7971401810646057, "num_tokens": 148732963.0, "step": 2270 }, { "epoch": 0.3001586042823156, "grad_norm": 0.9237709929478668, "learning_rate": 8.947348276833457e-06, "loss": 0.6414, "mean_token_accuracy": 0.7903797030448914, "num_tokens": 148798499.0, "step": 2271 }, { "epoch": 0.30029077451757863, "grad_norm": 0.8853390619963056, "learning_rate": 8.946013405533953e-06, "loss": 0.6113, "mean_token_accuracy": 0.8009944558143616, "num_tokens": 148862862.0, "step": 2272 }, { "epoch": 0.30042294475284165, "grad_norm": 0.8752419808358332, "learning_rate": 8.944677800650141e-06, "loss": 0.62, "mean_token_accuracy": 0.7991393208503723, "num_tokens": 148928398.0, "step": 2273 }, { "epoch": 0.30055511498810467, "grad_norm": 0.8831421372809154, "learning_rate": 8.943341462466347e-06, "loss": 0.6434, "mean_token_accuracy": 0.7920125722885132, "num_tokens": 148993934.0, "step": 2274 }, { "epoch": 0.3006872852233677, "grad_norm": 0.9148877116928811, "learning_rate": 8.942004391267045e-06, "loss": 0.6127, "mean_token_accuracy": 0.8018709421157837, "num_tokens": 149059470.0, "step": 2275 }, { "epoch": 0.3008194554586307, "grad_norm": 0.9363270588432501, "learning_rate": 8.940666587336873e-06, "loss": 0.5917, "mean_token_accuracy": 0.808478832244873, "num_tokens": 149125006.0, "step": 2276 }, { "epoch": 0.3009516256938937, "grad_norm": 1.0300078828099368, "learning_rate": 8.939328050960623e-06, "loss": 0.6758, "mean_token_accuracy": 0.7769045233726501, "num_tokens": 149190542.0, "step": 2277 }, { "epoch": 0.30108379592915674, "grad_norm": 0.9687364447019143, "learning_rate": 8.937988782423236e-06, "loss": 0.6303, "mean_token_accuracy": 0.7958582639694214, "num_tokens": 149256078.0, "step": 2278 }, { "epoch": 0.30121596616441976, "grad_norm": 0.8544211763172235, "learning_rate": 8.936648782009817e-06, "loss": 0.6058, "mean_token_accuracy": 0.8026034832000732, "num_tokens": 149321614.0, "step": 2279 }, { "epoch": 0.3013481363996828, "grad_norm": 0.8808142151189784, "learning_rate": 8.935308050005626e-06, "loss": 0.6306, "mean_token_accuracy": 0.7948358058929443, "num_tokens": 149387150.0, "step": 2280 }, { "epoch": 0.3014803066349458, "grad_norm": 0.8957030642075092, "learning_rate": 8.933966586696073e-06, "loss": 0.61, "mean_token_accuracy": 0.8025119304656982, "num_tokens": 149452686.0, "step": 2281 }, { "epoch": 0.3016124768702088, "grad_norm": 0.9875200332375679, "learning_rate": 8.932624392366732e-06, "loss": 0.651, "mean_token_accuracy": 0.7878464460372925, "num_tokens": 149518222.0, "step": 2282 }, { "epoch": 0.30174464710547183, "grad_norm": 0.8909338390775581, "learning_rate": 8.931281467303325e-06, "loss": 0.6133, "mean_token_accuracy": 0.7997649908065796, "num_tokens": 149583758.0, "step": 2283 }, { "epoch": 0.30187681734073485, "grad_norm": 0.9612504485597686, "learning_rate": 8.929937811791735e-06, "loss": 0.6321, "mean_token_accuracy": 0.7920888662338257, "num_tokens": 149649294.0, "step": 2284 }, { "epoch": 0.30200898757599787, "grad_norm": 0.8974192719272926, "learning_rate": 8.928593426117995e-06, "loss": 0.5982, "mean_token_accuracy": 0.8060371279716492, "num_tokens": 149714830.0, "step": 2285 }, { "epoch": 0.3021411578112609, "grad_norm": 0.8947208477091212, "learning_rate": 8.927248310568299e-06, "loss": 0.6428, "mean_token_accuracy": 0.7940422296524048, "num_tokens": 149780366.0, "step": 2286 }, { "epoch": 0.3022733280465239, "grad_norm": 0.8821240004462615, "learning_rate": 8.925902465428995e-06, "loss": 0.5883, "mean_token_accuracy": 0.8101879954338074, "num_tokens": 149845902.0, "step": 2287 }, { "epoch": 0.3024054982817869, "grad_norm": 0.9190223815900984, "learning_rate": 8.92455589098658e-06, "loss": 0.5823, "mean_token_accuracy": 0.8102185130119324, "num_tokens": 149911438.0, "step": 2288 }, { "epoch": 0.30253766851704994, "grad_norm": 1.0407329994889618, "learning_rate": 8.923208587527719e-06, "loss": 0.6692, "mean_token_accuracy": 0.7820473909378052, "num_tokens": 149976974.0, "step": 2289 }, { "epoch": 0.30266983875231296, "grad_norm": 0.9182211748715366, "learning_rate": 8.92186055533922e-06, "loss": 0.6019, "mean_token_accuracy": 0.8009705543518066, "num_tokens": 150042510.0, "step": 2290 }, { "epoch": 0.302802008987576, "grad_norm": 0.8485206157178055, "learning_rate": 8.920511794708052e-06, "loss": 0.5668, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 150108046.0, "step": 2291 }, { "epoch": 0.302934179222839, "grad_norm": 0.9030549664302973, "learning_rate": 8.919162305921338e-06, "loss": 0.6268, "mean_token_accuracy": 0.7992003560066223, "num_tokens": 150173582.0, "step": 2292 }, { "epoch": 0.303066349458102, "grad_norm": 0.9256612706476463, "learning_rate": 8.917812089266355e-06, "loss": 0.6116, "mean_token_accuracy": 0.7993376851081848, "num_tokens": 150239118.0, "step": 2293 }, { "epoch": 0.30319851969336503, "grad_norm": 0.9567557430447252, "learning_rate": 8.91646114503054e-06, "loss": 0.6277, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 150304654.0, "step": 2294 }, { "epoch": 0.30333068992862805, "grad_norm": 0.9996718572679602, "learning_rate": 8.915109473501475e-06, "loss": 0.7039, "mean_token_accuracy": 0.7728452086448669, "num_tokens": 150370190.0, "step": 2295 }, { "epoch": 0.30346286016389107, "grad_norm": 0.820178171463861, "learning_rate": 8.913757074966905e-06, "loss": 0.5896, "mean_token_accuracy": 0.807822585105896, "num_tokens": 150435726.0, "step": 2296 }, { "epoch": 0.3035950303991541, "grad_norm": 0.9141216889713629, "learning_rate": 8.91240394971473e-06, "loss": 0.6574, "mean_token_accuracy": 0.7869155406951904, "num_tokens": 150501262.0, "step": 2297 }, { "epoch": 0.3037272006344171, "grad_norm": 0.941639520942208, "learning_rate": 8.911050098032998e-06, "loss": 0.6282, "mean_token_accuracy": 0.7972012162208557, "num_tokens": 150566798.0, "step": 2298 }, { "epoch": 0.3038593708696801, "grad_norm": 0.9238971809775608, "learning_rate": 8.90969552020992e-06, "loss": 0.6517, "mean_token_accuracy": 0.7900134325027466, "num_tokens": 150632334.0, "step": 2299 }, { "epoch": 0.30399154110494314, "grad_norm": 0.9822798798837176, "learning_rate": 8.908340216533853e-06, "loss": 0.701, "mean_token_accuracy": 0.7745543718338013, "num_tokens": 150697870.0, "step": 2300 }, { "epoch": 0.30412371134020616, "grad_norm": 1.0091152259335516, "learning_rate": 8.906984187293316e-06, "loss": 0.6545, "mean_token_accuracy": 0.7852673530578613, "num_tokens": 150763406.0, "step": 2301 }, { "epoch": 0.3042558815754692, "grad_norm": 0.926199823351275, "learning_rate": 8.905627432776977e-06, "loss": 0.593, "mean_token_accuracy": 0.8060829043388367, "num_tokens": 150828942.0, "step": 2302 }, { "epoch": 0.3043880518107322, "grad_norm": 0.9942810150381893, "learning_rate": 8.904269953273665e-06, "loss": 0.5804, "mean_token_accuracy": 0.8147204518318176, "num_tokens": 150894478.0, "step": 2303 }, { "epoch": 0.3045202220459952, "grad_norm": 0.9174181417603734, "learning_rate": 8.902911749072353e-06, "loss": 0.5863, "mean_token_accuracy": 0.8069375157356262, "num_tokens": 150960014.0, "step": 2304 }, { "epoch": 0.30465239228125823, "grad_norm": 0.8408068811168339, "learning_rate": 8.901552820462178e-06, "loss": 0.593, "mean_token_accuracy": 0.8095928430557251, "num_tokens": 151025550.0, "step": 2305 }, { "epoch": 0.3047845625165213, "grad_norm": 0.8666558598478961, "learning_rate": 8.900193167732428e-06, "loss": 0.5774, "mean_token_accuracy": 0.8098217844963074, "num_tokens": 151091086.0, "step": 2306 }, { "epoch": 0.3049167327517843, "grad_norm": 0.8319985213905827, "learning_rate": 8.898832791172542e-06, "loss": 0.5872, "mean_token_accuracy": 0.8063423037528992, "num_tokens": 151156622.0, "step": 2307 }, { "epoch": 0.30504890298704734, "grad_norm": 0.800677810650676, "learning_rate": 8.897471691072118e-06, "loss": 0.5902, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 151222158.0, "step": 2308 }, { "epoch": 0.30518107322231036, "grad_norm": 0.9214466579370829, "learning_rate": 8.896109867720904e-06, "loss": 0.65, "mean_token_accuracy": 0.7863203287124634, "num_tokens": 151287694.0, "step": 2309 }, { "epoch": 0.3053132434575734, "grad_norm": 0.9583290630287203, "learning_rate": 8.894747321408805e-06, "loss": 0.6363, "mean_token_accuracy": 0.7941032648086548, "num_tokens": 151353230.0, "step": 2310 }, { "epoch": 0.3054454136928364, "grad_norm": 0.9512588379069223, "learning_rate": 8.89338405242588e-06, "loss": 0.6387, "mean_token_accuracy": 0.7903644442558289, "num_tokens": 151418766.0, "step": 2311 }, { "epoch": 0.3055775839280994, "grad_norm": 0.9196600159687395, "learning_rate": 8.892020061062338e-06, "loss": 0.6662, "mean_token_accuracy": 0.7829477190971375, "num_tokens": 151484302.0, "step": 2312 }, { "epoch": 0.30570975416336243, "grad_norm": 0.877172887253987, "learning_rate": 8.890655347608545e-06, "loss": 0.6106, "mean_token_accuracy": 0.8022066950798035, "num_tokens": 151549838.0, "step": 2313 }, { "epoch": 0.30584192439862545, "grad_norm": 0.8811557191914634, "learning_rate": 8.889289912355023e-06, "loss": 0.5621, "mean_token_accuracy": 0.8150561451911926, "num_tokens": 151615374.0, "step": 2314 }, { "epoch": 0.30597409463388847, "grad_norm": 0.8505527432234398, "learning_rate": 8.887923755592439e-06, "loss": 0.6042, "mean_token_accuracy": 0.8025729656219482, "num_tokens": 151680910.0, "step": 2315 }, { "epoch": 0.3061062648691515, "grad_norm": 0.8796927791743655, "learning_rate": 8.886556877611623e-06, "loss": 0.6018, "mean_token_accuracy": 0.8041753172874451, "num_tokens": 151746446.0, "step": 2316 }, { "epoch": 0.3062384351044145, "grad_norm": 0.7974328579028492, "learning_rate": 8.885189278703553e-06, "loss": 0.5552, "mean_token_accuracy": 0.8186119198799133, "num_tokens": 151811982.0, "step": 2317 }, { "epoch": 0.3063706053396775, "grad_norm": 0.9247334651787743, "learning_rate": 8.883820959159365e-06, "loss": 0.6364, "mean_token_accuracy": 0.7889909744262695, "num_tokens": 151877518.0, "step": 2318 }, { "epoch": 0.30650277557494054, "grad_norm": 0.8440498415365213, "learning_rate": 8.882451919270342e-06, "loss": 0.5656, "mean_token_accuracy": 0.8176352381706238, "num_tokens": 151943054.0, "step": 2319 }, { "epoch": 0.30663494581020356, "grad_norm": 0.8804169678585024, "learning_rate": 8.881082159327927e-06, "loss": 0.6445, "mean_token_accuracy": 0.7907917499542236, "num_tokens": 152008590.0, "step": 2320 }, { "epoch": 0.3067671160454666, "grad_norm": 0.8633762384913958, "learning_rate": 8.879711679623711e-06, "loss": 0.5715, "mean_token_accuracy": 0.8112257122993469, "num_tokens": 152074126.0, "step": 2321 }, { "epoch": 0.3068992862807296, "grad_norm": 0.8910635721868618, "learning_rate": 8.878340480449445e-06, "loss": 0.6148, "mean_token_accuracy": 0.8013215661048889, "num_tokens": 152139662.0, "step": 2322 }, { "epoch": 0.3070314565159926, "grad_norm": 0.9618887147112565, "learning_rate": 8.876968562097023e-06, "loss": 0.6654, "mean_token_accuracy": 0.7823983430862427, "num_tokens": 152205198.0, "step": 2323 }, { "epoch": 0.30716362675125564, "grad_norm": 0.9182609778328507, "learning_rate": 8.8755959248585e-06, "loss": 0.6572, "mean_token_accuracy": 0.7860914468765259, "num_tokens": 152270734.0, "step": 2324 }, { "epoch": 0.30729579698651865, "grad_norm": 0.8761002216613027, "learning_rate": 8.874222569026082e-06, "loss": 0.6087, "mean_token_accuracy": 0.8009400367736816, "num_tokens": 152336270.0, "step": 2325 }, { "epoch": 0.3074279672217817, "grad_norm": 0.8479180640428452, "learning_rate": 8.872848494892127e-06, "loss": 0.5962, "mean_token_accuracy": 0.8057318925857544, "num_tokens": 152401806.0, "step": 2326 }, { "epoch": 0.3075601374570447, "grad_norm": 0.9547513817912355, "learning_rate": 8.87147370274915e-06, "loss": 0.6205, "mean_token_accuracy": 0.7982084155082703, "num_tokens": 152467342.0, "step": 2327 }, { "epoch": 0.3076923076923077, "grad_norm": 1.009357768618026, "learning_rate": 8.870098192889812e-06, "loss": 0.6011, "mean_token_accuracy": 0.8023592829704285, "num_tokens": 152532878.0, "step": 2328 }, { "epoch": 0.3078244779275707, "grad_norm": 0.9192722045044818, "learning_rate": 8.86872196560693e-06, "loss": 0.5772, "mean_token_accuracy": 0.8112868070602417, "num_tokens": 152598414.0, "step": 2329 }, { "epoch": 0.30795664816283375, "grad_norm": 0.8593317422614496, "learning_rate": 8.867345021193475e-06, "loss": 0.6238, "mean_token_accuracy": 0.7975826859474182, "num_tokens": 152663950.0, "step": 2330 }, { "epoch": 0.30808881839809676, "grad_norm": 0.9580441183767402, "learning_rate": 8.865967359942573e-06, "loss": 0.6744, "mean_token_accuracy": 0.7790563106536865, "num_tokens": 152729486.0, "step": 2331 }, { "epoch": 0.3082209886333598, "grad_norm": 0.9268541483535062, "learning_rate": 8.864588982147495e-06, "loss": 0.672, "mean_token_accuracy": 0.7787052989006042, "num_tokens": 152795022.0, "step": 2332 }, { "epoch": 0.3083531588686228, "grad_norm": 0.8526021929364375, "learning_rate": 8.86320988810167e-06, "loss": 0.6399, "mean_token_accuracy": 0.7917225956916809, "num_tokens": 152860558.0, "step": 2333 }, { "epoch": 0.3084853291038858, "grad_norm": 0.970656227507614, "learning_rate": 8.861830078098678e-06, "loss": 0.6703, "mean_token_accuracy": 0.7828256487846375, "num_tokens": 152926094.0, "step": 2334 }, { "epoch": 0.30861749933914884, "grad_norm": 0.8407506841856429, "learning_rate": 8.860449552432252e-06, "loss": 0.6123, "mean_token_accuracy": 0.7971249222755432, "num_tokens": 152991630.0, "step": 2335 }, { "epoch": 0.30874966957441186, "grad_norm": 0.9284733596033186, "learning_rate": 8.85906831139628e-06, "loss": 0.6634, "mean_token_accuracy": 0.7862297296524048, "num_tokens": 153056357.0, "step": 2336 }, { "epoch": 0.3088818398096749, "grad_norm": 0.9494564743856722, "learning_rate": 8.857686355284793e-06, "loss": 0.6944, "mean_token_accuracy": 0.7757752537727356, "num_tokens": 153121893.0, "step": 2337 }, { "epoch": 0.3090140100449379, "grad_norm": 0.8488956869408912, "learning_rate": 8.856303684391988e-06, "loss": 0.6224, "mean_token_accuracy": 0.7949273586273193, "num_tokens": 153187429.0, "step": 2338 }, { "epoch": 0.3091461802802009, "grad_norm": 0.8571793289031248, "learning_rate": 8.8549202990122e-06, "loss": 0.5865, "mean_token_accuracy": 0.8101574778556824, "num_tokens": 153252965.0, "step": 2339 }, { "epoch": 0.30927835051546393, "grad_norm": 1.0455868728034923, "learning_rate": 8.85353619943993e-06, "loss": 0.6216, "mean_token_accuracy": 0.7969722747802734, "num_tokens": 153318501.0, "step": 2340 }, { "epoch": 0.30941052075072695, "grad_norm": 0.9581100655290463, "learning_rate": 8.852151385969816e-06, "loss": 0.6556, "mean_token_accuracy": 0.7866408228874207, "num_tokens": 153384037.0, "step": 2341 }, { "epoch": 0.30954269098598997, "grad_norm": 0.9574057253578101, "learning_rate": 8.850765858896661e-06, "loss": 0.6334, "mean_token_accuracy": 0.7946832180023193, "num_tokens": 153449573.0, "step": 2342 }, { "epoch": 0.309674861221253, "grad_norm": 0.9460673813713146, "learning_rate": 8.849379618515414e-06, "loss": 0.6597, "mean_token_accuracy": 0.7847179770469666, "num_tokens": 153515109.0, "step": 2343 }, { "epoch": 0.309807031456516, "grad_norm": 0.8291027064938943, "learning_rate": 8.847992665121177e-06, "loss": 0.6689, "mean_token_accuracy": 0.7809638381004333, "num_tokens": 153580645.0, "step": 2344 }, { "epoch": 0.309939201691779, "grad_norm": 0.9303781251053317, "learning_rate": 8.846604999009201e-06, "loss": 0.6592, "mean_token_accuracy": 0.7875106930732727, "num_tokens": 153646181.0, "step": 2345 }, { "epoch": 0.31007137192704204, "grad_norm": 0.8746430327781565, "learning_rate": 8.845216620474895e-06, "loss": 0.6305, "mean_token_accuracy": 0.7942406535148621, "num_tokens": 153711717.0, "step": 2346 }, { "epoch": 0.31020354216230506, "grad_norm": 0.9071427386459227, "learning_rate": 8.843827529813812e-06, "loss": 0.6613, "mean_token_accuracy": 0.7867781519889832, "num_tokens": 153777253.0, "step": 2347 }, { "epoch": 0.3103357123975681, "grad_norm": 0.901859585629064, "learning_rate": 8.842437727321662e-06, "loss": 0.6337, "mean_token_accuracy": 0.7928519248962402, "num_tokens": 153842789.0, "step": 2348 }, { "epoch": 0.3104678826328311, "grad_norm": 0.9651554127058362, "learning_rate": 8.841047213294306e-06, "loss": 0.6083, "mean_token_accuracy": 0.8012452721595764, "num_tokens": 153908325.0, "step": 2349 }, { "epoch": 0.3106000528680941, "grad_norm": 0.8962631845635136, "learning_rate": 8.839655988027752e-06, "loss": 0.575, "mean_token_accuracy": 0.8106763362884521, "num_tokens": 153973861.0, "step": 2350 }, { "epoch": 0.31073222310335713, "grad_norm": 0.8657333316666335, "learning_rate": 8.838264051818167e-06, "loss": 0.6296, "mean_token_accuracy": 0.7963313460350037, "num_tokens": 154039397.0, "step": 2351 }, { "epoch": 0.31086439333862015, "grad_norm": 0.8485189721674056, "learning_rate": 8.836871404961864e-06, "loss": 0.5887, "mean_token_accuracy": 0.8096691370010376, "num_tokens": 154104933.0, "step": 2352 }, { "epoch": 0.31099656357388317, "grad_norm": 0.9184782998053846, "learning_rate": 8.835478047755306e-06, "loss": 0.6353, "mean_token_accuracy": 0.7940422296524048, "num_tokens": 154170469.0, "step": 2353 }, { "epoch": 0.3111287338091462, "grad_norm": 0.8952155808715548, "learning_rate": 8.834083980495112e-06, "loss": 0.5986, "mean_token_accuracy": 0.8051977753639221, "num_tokens": 154236005.0, "step": 2354 }, { "epoch": 0.3112609040444092, "grad_norm": 0.9177458282671133, "learning_rate": 8.832689203478049e-06, "loss": 0.6946, "mean_token_accuracy": 0.7750884890556335, "num_tokens": 154301541.0, "step": 2355 }, { "epoch": 0.3113930742796722, "grad_norm": 1.2759820663864139, "learning_rate": 8.831293717001037e-06, "loss": 0.6679, "mean_token_accuracy": 0.78290194272995, "num_tokens": 154367077.0, "step": 2356 }, { "epoch": 0.31152524451493524, "grad_norm": 0.8884243239793926, "learning_rate": 8.829897521361144e-06, "loss": 0.5957, "mean_token_accuracy": 0.8058539628982544, "num_tokens": 154432613.0, "step": 2357 }, { "epoch": 0.31165741475019826, "grad_norm": 0.9399629352900049, "learning_rate": 8.828500616855591e-06, "loss": 0.6531, "mean_token_accuracy": 0.7887772917747498, "num_tokens": 154498149.0, "step": 2358 }, { "epoch": 0.3117895849854613, "grad_norm": 0.8937627366783282, "learning_rate": 8.827103003781754e-06, "loss": 0.6315, "mean_token_accuracy": 0.7915089726448059, "num_tokens": 154563685.0, "step": 2359 }, { "epoch": 0.3119217552207243, "grad_norm": 0.7612210302458428, "learning_rate": 8.825704682437151e-06, "loss": 0.5364, "mean_token_accuracy": 0.8247161507606506, "num_tokens": 154629221.0, "step": 2360 }, { "epoch": 0.3120539254559873, "grad_norm": 0.9237806932150346, "learning_rate": 8.824305653119457e-06, "loss": 0.6707, "mean_token_accuracy": 0.7829172015190125, "num_tokens": 154694757.0, "step": 2361 }, { "epoch": 0.31218609569125033, "grad_norm": 0.9481176749350988, "learning_rate": 8.822905916126496e-06, "loss": 0.6281, "mean_token_accuracy": 0.7946068644523621, "num_tokens": 154760293.0, "step": 2362 }, { "epoch": 0.31231826592651335, "grad_norm": 0.860107107147992, "learning_rate": 8.82150547175624e-06, "loss": 0.6275, "mean_token_accuracy": 0.7950494289398193, "num_tokens": 154825829.0, "step": 2363 }, { "epoch": 0.31245043616177637, "grad_norm": 0.9229779847990148, "learning_rate": 8.82010432030682e-06, "loss": 0.6508, "mean_token_accuracy": 0.7874191403388977, "num_tokens": 154891365.0, "step": 2364 }, { "epoch": 0.3125826063970394, "grad_norm": 0.8474869030324989, "learning_rate": 8.818702462076508e-06, "loss": 0.5973, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 154956901.0, "step": 2365 }, { "epoch": 0.3127147766323024, "grad_norm": 0.8545190841272248, "learning_rate": 8.817299897363731e-06, "loss": 0.6373, "mean_token_accuracy": 0.7914631962776184, "num_tokens": 155022437.0, "step": 2366 }, { "epoch": 0.3128469468675654, "grad_norm": 0.9005387913205672, "learning_rate": 8.815896626467068e-06, "loss": 0.6594, "mean_token_accuracy": 0.7856336236000061, "num_tokens": 155087973.0, "step": 2367 }, { "epoch": 0.31297911710282844, "grad_norm": 0.8887641689959317, "learning_rate": 8.814492649685243e-06, "loss": 0.6574, "mean_token_accuracy": 0.7832529544830322, "num_tokens": 155153509.0, "step": 2368 }, { "epoch": 0.31311128733809146, "grad_norm": 0.8267407175111777, "learning_rate": 8.813087967317133e-06, "loss": 0.6007, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 155219045.0, "step": 2369 }, { "epoch": 0.3132434575733545, "grad_norm": 0.910824737571424, "learning_rate": 8.81168257966177e-06, "loss": 0.651, "mean_token_accuracy": 0.7866255640983582, "num_tokens": 155284581.0, "step": 2370 }, { "epoch": 0.3133756278086175, "grad_norm": 0.8983170397798581, "learning_rate": 8.810276487018326e-06, "loss": 0.63, "mean_token_accuracy": 0.79324871301651, "num_tokens": 155350117.0, "step": 2371 }, { "epoch": 0.3135077980438805, "grad_norm": 0.871349081584937, "learning_rate": 8.808869689686133e-06, "loss": 0.6264, "mean_token_accuracy": 0.7947289943695068, "num_tokens": 155415653.0, "step": 2372 }, { "epoch": 0.31363996827914353, "grad_norm": 0.8948260495686086, "learning_rate": 8.807462187964666e-06, "loss": 0.6177, "mean_token_accuracy": 0.7988951206207275, "num_tokens": 155481189.0, "step": 2373 }, { "epoch": 0.31377213851440655, "grad_norm": 0.8520339334883554, "learning_rate": 8.806053982153556e-06, "loss": 0.6132, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 155546725.0, "step": 2374 }, { "epoch": 0.31390430874966957, "grad_norm": 0.8535408013113766, "learning_rate": 8.804645072552578e-06, "loss": 0.6286, "mean_token_accuracy": 0.79343181848526, "num_tokens": 155612261.0, "step": 2375 }, { "epoch": 0.3140364789849326, "grad_norm": 0.8771245047919561, "learning_rate": 8.803235459461658e-06, "loss": 0.5975, "mean_token_accuracy": 0.8062965273857117, "num_tokens": 155677797.0, "step": 2376 }, { "epoch": 0.3141686492201956, "grad_norm": 0.8790355176140782, "learning_rate": 8.801825143180878e-06, "loss": 0.5685, "mean_token_accuracy": 0.8141404986381531, "num_tokens": 155743333.0, "step": 2377 }, { "epoch": 0.3143008194554586, "grad_norm": 0.9068023419973125, "learning_rate": 8.800414124010461e-06, "loss": 0.6065, "mean_token_accuracy": 0.8035801649093628, "num_tokens": 155808869.0, "step": 2378 }, { "epoch": 0.31443298969072164, "grad_norm": 0.8061610337159947, "learning_rate": 8.799002402250785e-06, "loss": 0.606, "mean_token_accuracy": 0.802435576915741, "num_tokens": 155874405.0, "step": 2379 }, { "epoch": 0.31456515992598466, "grad_norm": 0.9096438546209022, "learning_rate": 8.797589978202376e-06, "loss": 0.6718, "mean_token_accuracy": 0.7808112502098083, "num_tokens": 155939941.0, "step": 2380 }, { "epoch": 0.3146973301612477, "grad_norm": 1.0553506264722528, "learning_rate": 8.79617685216591e-06, "loss": 0.6587, "mean_token_accuracy": 0.7863966822624207, "num_tokens": 156005477.0, "step": 2381 }, { "epoch": 0.3148295003965107, "grad_norm": 0.9349668328531844, "learning_rate": 8.794763024442208e-06, "loss": 0.6515, "mean_token_accuracy": 0.788227915763855, "num_tokens": 156071013.0, "step": 2382 }, { "epoch": 0.3149616706317737, "grad_norm": 0.9376000689893981, "learning_rate": 8.79334849533225e-06, "loss": 0.6665, "mean_token_accuracy": 0.7834360599517822, "num_tokens": 156136549.0, "step": 2383 }, { "epoch": 0.31509384086703673, "grad_norm": 0.8175411049248257, "learning_rate": 8.791933265137159e-06, "loss": 0.5796, "mean_token_accuracy": 0.8096233606338501, "num_tokens": 156202085.0, "step": 2384 }, { "epoch": 0.31522601110229975, "grad_norm": 0.8781561573164343, "learning_rate": 8.790517334158204e-06, "loss": 0.5835, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 156267621.0, "step": 2385 }, { "epoch": 0.31535818133756277, "grad_norm": 0.8581746306901659, "learning_rate": 8.78910070269681e-06, "loss": 0.601, "mean_token_accuracy": 0.8050604462623596, "num_tokens": 156333157.0, "step": 2386 }, { "epoch": 0.3154903515728258, "grad_norm": 0.8457604357462093, "learning_rate": 8.78768337105455e-06, "loss": 0.5988, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 156398693.0, "step": 2387 }, { "epoch": 0.3156225218080888, "grad_norm": 0.9136824802676242, "learning_rate": 8.786265339533142e-06, "loss": 0.6954, "mean_token_accuracy": 0.773684561252594, "num_tokens": 156464229.0, "step": 2388 }, { "epoch": 0.3157546920433518, "grad_norm": 0.9045947288632208, "learning_rate": 8.784846608434452e-06, "loss": 0.6095, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 156529765.0, "step": 2389 }, { "epoch": 0.31588686227861484, "grad_norm": 0.829486462052672, "learning_rate": 8.783427178060503e-06, "loss": 0.602, "mean_token_accuracy": 0.8028171062469482, "num_tokens": 156595301.0, "step": 2390 }, { "epoch": 0.31601903251387786, "grad_norm": 0.8809265304439264, "learning_rate": 8.782007048713462e-06, "loss": 0.6156, "mean_token_accuracy": 0.7982847094535828, "num_tokens": 156660837.0, "step": 2391 }, { "epoch": 0.3161512027491409, "grad_norm": 0.9154313875204084, "learning_rate": 8.780586220695644e-06, "loss": 0.6047, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 156726373.0, "step": 2392 }, { "epoch": 0.3162833729844039, "grad_norm": 0.8747855612649462, "learning_rate": 8.779164694309512e-06, "loss": 0.6247, "mean_token_accuracy": 0.7957056760787964, "num_tokens": 156791909.0, "step": 2393 }, { "epoch": 0.3164155432196669, "grad_norm": 0.8755986498461772, "learning_rate": 8.777742469857682e-06, "loss": 0.5796, "mean_token_accuracy": 0.8124618530273438, "num_tokens": 156857445.0, "step": 2394 }, { "epoch": 0.31654771345492994, "grad_norm": 0.9251187728156477, "learning_rate": 8.776319547642914e-06, "loss": 0.6309, "mean_token_accuracy": 0.7957972288131714, "num_tokens": 156922981.0, "step": 2395 }, { "epoch": 0.31667988369019295, "grad_norm": 1.0065501342807333, "learning_rate": 8.774895927968119e-06, "loss": 0.6276, "mean_token_accuracy": 0.7975521683692932, "num_tokens": 156988517.0, "step": 2396 }, { "epoch": 0.31681205392545597, "grad_norm": 0.8815043162037656, "learning_rate": 8.773471611136356e-06, "loss": 0.6477, "mean_token_accuracy": 0.7868391871452332, "num_tokens": 157054053.0, "step": 2397 }, { "epoch": 0.316944224160719, "grad_norm": 0.9362605629511517, "learning_rate": 8.772046597450831e-06, "loss": 0.654, "mean_token_accuracy": 0.7858777642250061, "num_tokens": 157119589.0, "step": 2398 }, { "epoch": 0.317076394395982, "grad_norm": 0.8715043598124985, "learning_rate": 8.770620887214903e-06, "loss": 0.617, "mean_token_accuracy": 0.7985899448394775, "num_tokens": 157185125.0, "step": 2399 }, { "epoch": 0.317208564631245, "grad_norm": 0.8847335351627498, "learning_rate": 8.769194480732073e-06, "loss": 0.6081, "mean_token_accuracy": 0.8028323650360107, "num_tokens": 157250661.0, "step": 2400 }, { "epoch": 0.31734073486650805, "grad_norm": 0.9041303730019081, "learning_rate": 8.76776737830599e-06, "loss": 0.5909, "mean_token_accuracy": 0.8061897158622742, "num_tokens": 157316197.0, "step": 2401 }, { "epoch": 0.31747290510177106, "grad_norm": 0.8579776535363052, "learning_rate": 8.766339580240462e-06, "loss": 0.5908, "mean_token_accuracy": 0.808417797088623, "num_tokens": 157381733.0, "step": 2402 }, { "epoch": 0.3176050753370341, "grad_norm": 0.8584772647370953, "learning_rate": 8.764911086839435e-06, "loss": 0.6022, "mean_token_accuracy": 0.8025119304656982, "num_tokens": 157447269.0, "step": 2403 }, { "epoch": 0.3177372455722971, "grad_norm": 0.9644817036661937, "learning_rate": 8.763481898407003e-06, "loss": 0.6454, "mean_token_accuracy": 0.7899066209793091, "num_tokens": 157512805.0, "step": 2404 }, { "epoch": 0.3178694158075601, "grad_norm": 0.8777891875064359, "learning_rate": 8.762052015247409e-06, "loss": 0.6251, "mean_token_accuracy": 0.7974758744239807, "num_tokens": 157578341.0, "step": 2405 }, { "epoch": 0.31800158604282314, "grad_norm": 0.9244177729406278, "learning_rate": 8.76062143766505e-06, "loss": 0.6065, "mean_token_accuracy": 0.8028934001922607, "num_tokens": 157643877.0, "step": 2406 }, { "epoch": 0.31813375627808615, "grad_norm": 0.9622046955067595, "learning_rate": 8.759190165964464e-06, "loss": 0.6828, "mean_token_accuracy": 0.7766451239585876, "num_tokens": 157709413.0, "step": 2407 }, { "epoch": 0.3182659265133492, "grad_norm": 0.8738529457195275, "learning_rate": 8.757758200450336e-06, "loss": 0.5822, "mean_token_accuracy": 0.8087687492370605, "num_tokens": 157774949.0, "step": 2408 }, { "epoch": 0.3183980967486122, "grad_norm": 0.9539434668000792, "learning_rate": 8.756325541427507e-06, "loss": 0.5883, "mean_token_accuracy": 0.806800127029419, "num_tokens": 157840485.0, "step": 2409 }, { "epoch": 0.3185302669838752, "grad_norm": 0.9768225179707614, "learning_rate": 8.754892189200955e-06, "loss": 0.6646, "mean_token_accuracy": 0.7837870717048645, "num_tokens": 157906021.0, "step": 2410 }, { "epoch": 0.31866243721913823, "grad_norm": 0.9151740209205181, "learning_rate": 8.753458144075815e-06, "loss": 0.6067, "mean_token_accuracy": 0.8040837645530701, "num_tokens": 157971557.0, "step": 2411 }, { "epoch": 0.31879460745440125, "grad_norm": 1.0184397723088412, "learning_rate": 8.752023406357365e-06, "loss": 0.6774, "mean_token_accuracy": 0.7833750247955322, "num_tokens": 158037093.0, "step": 2412 }, { "epoch": 0.31892677768966426, "grad_norm": 0.9008963484001334, "learning_rate": 8.750587976351028e-06, "loss": 0.5983, "mean_token_accuracy": 0.8049230575561523, "num_tokens": 158102629.0, "step": 2413 }, { "epoch": 0.3190589479249273, "grad_norm": 0.8589996359194224, "learning_rate": 8.749151854362376e-06, "loss": 0.5739, "mean_token_accuracy": 0.8143084049224854, "num_tokens": 158168165.0, "step": 2414 }, { "epoch": 0.3191911181601903, "grad_norm": 0.9000585953538921, "learning_rate": 8.747715040697131e-06, "loss": 0.6191, "mean_token_accuracy": 0.798071026802063, "num_tokens": 158233701.0, "step": 2415 }, { "epoch": 0.3193232883954533, "grad_norm": 0.9811262920255028, "learning_rate": 8.746277535661166e-06, "loss": 0.5706, "mean_token_accuracy": 0.8146288394927979, "num_tokens": 158299237.0, "step": 2416 }, { "epoch": 0.31945545863071634, "grad_norm": 0.8734756286028237, "learning_rate": 8.744839339560487e-06, "loss": 0.5784, "mean_token_accuracy": 0.8161702156066895, "num_tokens": 158364773.0, "step": 2417 }, { "epoch": 0.31958762886597936, "grad_norm": 0.978584476375097, "learning_rate": 8.743400452701261e-06, "loss": 0.6237, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 158430309.0, "step": 2418 }, { "epoch": 0.3197197991012424, "grad_norm": 0.8991936023951239, "learning_rate": 8.741960875389797e-06, "loss": 0.6134, "mean_token_accuracy": 0.7996886968612671, "num_tokens": 158495845.0, "step": 2419 }, { "epoch": 0.3198519693365054, "grad_norm": 0.9108808951851047, "learning_rate": 8.740520607932548e-06, "loss": 0.5806, "mean_token_accuracy": 0.8110120892524719, "num_tokens": 158561381.0, "step": 2420 }, { "epoch": 0.3199841395717684, "grad_norm": 0.8576376914896167, "learning_rate": 8.73907965063612e-06, "loss": 0.6126, "mean_token_accuracy": 0.7974911332130432, "num_tokens": 158626917.0, "step": 2421 }, { "epoch": 0.32011630980703143, "grad_norm": 0.905962294797804, "learning_rate": 8.737638003807262e-06, "loss": 0.6458, "mean_token_accuracy": 0.788975715637207, "num_tokens": 158692453.0, "step": 2422 }, { "epoch": 0.32024848004229445, "grad_norm": 0.9440678215552074, "learning_rate": 8.73619566775287e-06, "loss": 0.6359, "mean_token_accuracy": 0.7930502891540527, "num_tokens": 158757989.0, "step": 2423 }, { "epoch": 0.3203806502775575, "grad_norm": 0.918572916369652, "learning_rate": 8.734752642779985e-06, "loss": 0.626, "mean_token_accuracy": 0.797826886177063, "num_tokens": 158823525.0, "step": 2424 }, { "epoch": 0.32051282051282054, "grad_norm": 0.900964691435152, "learning_rate": 8.733308929195802e-06, "loss": 0.6466, "mean_token_accuracy": 0.7872207164764404, "num_tokens": 158889061.0, "step": 2425 }, { "epoch": 0.32064499074808356, "grad_norm": 0.8515654640999879, "learning_rate": 8.731864527307652e-06, "loss": 0.6081, "mean_token_accuracy": 0.8020846247673035, "num_tokens": 158954597.0, "step": 2426 }, { "epoch": 0.3207771609833466, "grad_norm": 0.9113865380545768, "learning_rate": 8.730419437423022e-06, "loss": 0.637, "mean_token_accuracy": 0.7911121845245361, "num_tokens": 159020133.0, "step": 2427 }, { "epoch": 0.3209093312186096, "grad_norm": 0.8784520508810029, "learning_rate": 8.728973659849542e-06, "loss": 0.5947, "mean_token_accuracy": 0.8058539628982544, "num_tokens": 159085669.0, "step": 2428 }, { "epoch": 0.3210415014538726, "grad_norm": 0.9456794027305049, "learning_rate": 8.727527194894984e-06, "loss": 0.6056, "mean_token_accuracy": 0.8008484840393066, "num_tokens": 159151205.0, "step": 2429 }, { "epoch": 0.32117367168913563, "grad_norm": 0.9036597773616318, "learning_rate": 8.726080042867272e-06, "loss": 0.5423, "mean_token_accuracy": 0.8222439289093018, "num_tokens": 159216741.0, "step": 2430 }, { "epoch": 0.32130584192439865, "grad_norm": 0.9293037241256877, "learning_rate": 8.724632204074474e-06, "loss": 0.6298, "mean_token_accuracy": 0.792256772518158, "num_tokens": 159282277.0, "step": 2431 }, { "epoch": 0.32143801215966167, "grad_norm": 0.8893095874488872, "learning_rate": 8.723183678824806e-06, "loss": 0.6595, "mean_token_accuracy": 0.7835429310798645, "num_tokens": 159347813.0, "step": 2432 }, { "epoch": 0.3215701823949247, "grad_norm": 0.9447572379766249, "learning_rate": 8.721734467426627e-06, "loss": 0.6316, "mean_token_accuracy": 0.7931876182556152, "num_tokens": 159413349.0, "step": 2433 }, { "epoch": 0.3217023526301877, "grad_norm": 0.9496660809144506, "learning_rate": 8.720284570188446e-06, "loss": 0.6097, "mean_token_accuracy": 0.8034886121749878, "num_tokens": 159478885.0, "step": 2434 }, { "epoch": 0.3218345228654507, "grad_norm": 0.956663039325495, "learning_rate": 8.718833987418915e-06, "loss": 0.668, "mean_token_accuracy": 0.7831461429595947, "num_tokens": 159544421.0, "step": 2435 }, { "epoch": 0.32196669310071374, "grad_norm": 0.8387695992263109, "learning_rate": 8.717382719426832e-06, "loss": 0.5719, "mean_token_accuracy": 0.8140794634819031, "num_tokens": 159609957.0, "step": 2436 }, { "epoch": 0.32209886333597676, "grad_norm": 0.9643963274836685, "learning_rate": 8.715930766521142e-06, "loss": 0.652, "mean_token_accuracy": 0.7909748554229736, "num_tokens": 159675493.0, "step": 2437 }, { "epoch": 0.3222310335712398, "grad_norm": 0.8785464584867426, "learning_rate": 8.714478129010935e-06, "loss": 0.654, "mean_token_accuracy": 0.784458577632904, "num_tokens": 159741029.0, "step": 2438 }, { "epoch": 0.3223632038065028, "grad_norm": 0.8786824515594955, "learning_rate": 8.713024807205448e-06, "loss": 0.6285, "mean_token_accuracy": 0.7939811944961548, "num_tokens": 159806565.0, "step": 2439 }, { "epoch": 0.3224953740417658, "grad_norm": 0.8737553025330725, "learning_rate": 8.711570801414064e-06, "loss": 0.5895, "mean_token_accuracy": 0.807395339012146, "num_tokens": 159872101.0, "step": 2440 }, { "epoch": 0.32262754427702883, "grad_norm": 1.094820694308025, "learning_rate": 8.710116111946309e-06, "loss": 0.6489, "mean_token_accuracy": 0.7864119410514832, "num_tokens": 159937637.0, "step": 2441 }, { "epoch": 0.32275971451229185, "grad_norm": 0.8515060112111363, "learning_rate": 8.708660739111857e-06, "loss": 0.6549, "mean_token_accuracy": 0.7871139049530029, "num_tokens": 160003173.0, "step": 2442 }, { "epoch": 0.32289188474755487, "grad_norm": 0.8691019567183977, "learning_rate": 8.707204683220522e-06, "loss": 0.611, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 160068709.0, "step": 2443 }, { "epoch": 0.3230240549828179, "grad_norm": 0.9115815918596598, "learning_rate": 8.705747944582273e-06, "loss": 0.6897, "mean_token_accuracy": 0.7762178182601929, "num_tokens": 160134245.0, "step": 2444 }, { "epoch": 0.3231562252180809, "grad_norm": 1.0561625818802622, "learning_rate": 8.704290523507219e-06, "loss": 0.7133, "mean_token_accuracy": 0.7663288712501526, "num_tokens": 160199781.0, "step": 2445 }, { "epoch": 0.3232883954533439, "grad_norm": 1.0251997119391083, "learning_rate": 8.702832420305614e-06, "loss": 0.6443, "mean_token_accuracy": 0.7899218797683716, "num_tokens": 160265317.0, "step": 2446 }, { "epoch": 0.32342056568860694, "grad_norm": 0.9757768619223063, "learning_rate": 8.701373635287855e-06, "loss": 0.582, "mean_token_accuracy": 0.8093639612197876, "num_tokens": 160330853.0, "step": 2447 }, { "epoch": 0.32355273592386996, "grad_norm": 0.848427757949133, "learning_rate": 8.69991416876449e-06, "loss": 0.6012, "mean_token_accuracy": 0.8039922118186951, "num_tokens": 160396389.0, "step": 2448 }, { "epoch": 0.323684906159133, "grad_norm": 0.9553710695838499, "learning_rate": 8.698454021046208e-06, "loss": 0.6482, "mean_token_accuracy": 0.78795325756073, "num_tokens": 160461925.0, "step": 2449 }, { "epoch": 0.323817076394396, "grad_norm": 1.088233123651795, "learning_rate": 8.696993192443843e-06, "loss": 0.64, "mean_token_accuracy": 0.7919973134994507, "num_tokens": 160527461.0, "step": 2450 }, { "epoch": 0.323949246629659, "grad_norm": 1.0766053460761353, "learning_rate": 8.695531683268376e-06, "loss": 0.609, "mean_token_accuracy": 0.8006043434143066, "num_tokens": 160592997.0, "step": 2451 }, { "epoch": 0.32408141686492203, "grad_norm": 0.9110114013830227, "learning_rate": 8.694069493830932e-06, "loss": 0.6258, "mean_token_accuracy": 0.7955530285835266, "num_tokens": 160658533.0, "step": 2452 }, { "epoch": 0.32421358710018505, "grad_norm": 0.9281361452352263, "learning_rate": 8.69260662444278e-06, "loss": 0.6212, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 160724069.0, "step": 2453 }, { "epoch": 0.32434575733544807, "grad_norm": 0.9152547501625761, "learning_rate": 8.691143075415334e-06, "loss": 0.5574, "mean_token_accuracy": 0.8187034726142883, "num_tokens": 160789605.0, "step": 2454 }, { "epoch": 0.3244779275707111, "grad_norm": 1.2569909960115986, "learning_rate": 8.689678847060155e-06, "loss": 0.6124, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 160855141.0, "step": 2455 }, { "epoch": 0.3246100978059741, "grad_norm": 0.8657246240190803, "learning_rate": 8.688213939688943e-06, "loss": 0.5187, "mean_token_accuracy": 0.8285160660743713, "num_tokens": 160920677.0, "step": 2456 }, { "epoch": 0.3247422680412371, "grad_norm": 0.9613112016742487, "learning_rate": 8.686748353613549e-06, "loss": 0.6111, "mean_token_accuracy": 0.8006501197814941, "num_tokens": 160986213.0, "step": 2457 }, { "epoch": 0.32487443827650014, "grad_norm": 0.9008236188114568, "learning_rate": 8.685282089145966e-06, "loss": 0.6214, "mean_token_accuracy": 0.7974453568458557, "num_tokens": 161051749.0, "step": 2458 }, { "epoch": 0.32500660851176316, "grad_norm": 0.9536134243831126, "learning_rate": 8.68381514659833e-06, "loss": 0.6255, "mean_token_accuracy": 0.798696756362915, "num_tokens": 161117285.0, "step": 2459 }, { "epoch": 0.3251387787470262, "grad_norm": 0.9395834457941873, "learning_rate": 8.682347526282922e-06, "loss": 0.6303, "mean_token_accuracy": 0.7961024045944214, "num_tokens": 161182821.0, "step": 2460 }, { "epoch": 0.3252709489822892, "grad_norm": 0.92869135703498, "learning_rate": 8.680879228512168e-06, "loss": 0.592, "mean_token_accuracy": 0.8053961396217346, "num_tokens": 161248357.0, "step": 2461 }, { "epoch": 0.3254031192175522, "grad_norm": 0.9210451293609164, "learning_rate": 8.679410253598638e-06, "loss": 0.6426, "mean_token_accuracy": 0.7923635840415955, "num_tokens": 161313893.0, "step": 2462 }, { "epoch": 0.32553528945281524, "grad_norm": 0.9032864346685816, "learning_rate": 8.677940601855047e-06, "loss": 0.6437, "mean_token_accuracy": 0.787983775138855, "num_tokens": 161379429.0, "step": 2463 }, { "epoch": 0.32566745968807825, "grad_norm": 0.9424515838918125, "learning_rate": 8.676470273594253e-06, "loss": 0.6674, "mean_token_accuracy": 0.7789646983146667, "num_tokens": 161444965.0, "step": 2464 }, { "epoch": 0.32579962992334127, "grad_norm": 0.8661446640745116, "learning_rate": 8.674999269129257e-06, "loss": 0.6221, "mean_token_accuracy": 0.7945763468742371, "num_tokens": 161510501.0, "step": 2465 }, { "epoch": 0.3259318001586043, "grad_norm": 0.8535855134796053, "learning_rate": 8.673527588773204e-06, "loss": 0.5749, "mean_token_accuracy": 0.8132401704788208, "num_tokens": 161576037.0, "step": 2466 }, { "epoch": 0.3260639703938673, "grad_norm": 0.9892252395925057, "learning_rate": 8.67205523283939e-06, "loss": 0.6306, "mean_token_accuracy": 0.7957819700241089, "num_tokens": 161641573.0, "step": 2467 }, { "epoch": 0.3261961406291303, "grad_norm": 0.8437042245719542, "learning_rate": 8.670582201641242e-06, "loss": 0.6325, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 161707109.0, "step": 2468 }, { "epoch": 0.32632831086439335, "grad_norm": 0.8433617308498834, "learning_rate": 8.66910849549234e-06, "loss": 0.5528, "mean_token_accuracy": 0.8191612958908081, "num_tokens": 161772645.0, "step": 2469 }, { "epoch": 0.32646048109965636, "grad_norm": 0.850740253039004, "learning_rate": 8.667634114706407e-06, "loss": 0.5662, "mean_token_accuracy": 0.8153766393661499, "num_tokens": 161838181.0, "step": 2470 }, { "epoch": 0.3265926513349194, "grad_norm": 0.9054288266126265, "learning_rate": 8.666159059597306e-06, "loss": 0.6361, "mean_token_accuracy": 0.7929739952087402, "num_tokens": 161903717.0, "step": 2471 }, { "epoch": 0.3267248215701824, "grad_norm": 0.9589750892229404, "learning_rate": 8.664683330479046e-06, "loss": 0.6088, "mean_token_accuracy": 0.7976895570755005, "num_tokens": 161969253.0, "step": 2472 }, { "epoch": 0.3268569918054454, "grad_norm": 0.9708999089042306, "learning_rate": 8.663206927665778e-06, "loss": 0.5781, "mean_token_accuracy": 0.812065064907074, "num_tokens": 162034789.0, "step": 2473 }, { "epoch": 0.32698916204070844, "grad_norm": 0.8671810057436768, "learning_rate": 8.661729851471796e-06, "loss": 0.6297, "mean_token_accuracy": 0.7942253947257996, "num_tokens": 162100325.0, "step": 2474 }, { "epoch": 0.32712133227597145, "grad_norm": 0.8963652262190156, "learning_rate": 8.660252102211546e-06, "loss": 0.5923, "mean_token_accuracy": 0.8073037266731262, "num_tokens": 162165861.0, "step": 2475 }, { "epoch": 0.3272535025112345, "grad_norm": 0.9414053616766523, "learning_rate": 8.6587736801996e-06, "loss": 0.6009, "mean_token_accuracy": 0.8050604462623596, "num_tokens": 162231397.0, "step": 2476 }, { "epoch": 0.3273856727464975, "grad_norm": 0.8515464338277571, "learning_rate": 8.65729458575069e-06, "loss": 0.5342, "mean_token_accuracy": 0.8298131823539734, "num_tokens": 162296933.0, "step": 2477 }, { "epoch": 0.3275178429817605, "grad_norm": 0.9781000367274026, "learning_rate": 8.655814819179683e-06, "loss": 0.6518, "mean_token_accuracy": 0.782627284526825, "num_tokens": 162362469.0, "step": 2478 }, { "epoch": 0.32765001321702353, "grad_norm": 1.0956978667833508, "learning_rate": 8.654334380801587e-06, "loss": 0.6562, "mean_token_accuracy": 0.7865034937858582, "num_tokens": 162428005.0, "step": 2479 }, { "epoch": 0.32778218345228655, "grad_norm": 0.9255717445887992, "learning_rate": 8.652853270931562e-06, "loss": 0.5946, "mean_token_accuracy": 0.806617021560669, "num_tokens": 162493541.0, "step": 2480 }, { "epoch": 0.32791435368754956, "grad_norm": 0.9770383033532589, "learning_rate": 8.651371489884902e-06, "loss": 0.6027, "mean_token_accuracy": 0.8048772811889648, "num_tokens": 162559077.0, "step": 2481 }, { "epoch": 0.3280465239228126, "grad_norm": 0.9977776550668475, "learning_rate": 8.649889037977046e-06, "loss": 0.6451, "mean_token_accuracy": 0.7880295515060425, "num_tokens": 162624613.0, "step": 2482 }, { "epoch": 0.3281786941580756, "grad_norm": 0.9350421853387343, "learning_rate": 8.64840591552358e-06, "loss": 0.6091, "mean_token_accuracy": 0.7993071675300598, "num_tokens": 162690149.0, "step": 2483 }, { "epoch": 0.3283108643933386, "grad_norm": 0.8951413305538918, "learning_rate": 8.646922122840228e-06, "loss": 0.5728, "mean_token_accuracy": 0.8094249963760376, "num_tokens": 162755685.0, "step": 2484 }, { "epoch": 0.32844303462860164, "grad_norm": 0.980269659607189, "learning_rate": 8.64543766024286e-06, "loss": 0.6223, "mean_token_accuracy": 0.7965144515037537, "num_tokens": 162821221.0, "step": 2485 }, { "epoch": 0.32857520486386466, "grad_norm": 0.9247754882151237, "learning_rate": 8.643952528047486e-06, "loss": 0.6189, "mean_token_accuracy": 0.7996124029159546, "num_tokens": 162886757.0, "step": 2486 }, { "epoch": 0.3287073750991277, "grad_norm": 0.926412181979683, "learning_rate": 8.642466726570259e-06, "loss": 0.583, "mean_token_accuracy": 0.8102337718009949, "num_tokens": 162952293.0, "step": 2487 }, { "epoch": 0.3288395453343907, "grad_norm": 1.0300365741516075, "learning_rate": 8.640980256127475e-06, "loss": 0.666, "mean_token_accuracy": 0.782505214214325, "num_tokens": 163017829.0, "step": 2488 }, { "epoch": 0.3289717155696537, "grad_norm": 0.9909674252845144, "learning_rate": 8.639493117035576e-06, "loss": 0.5987, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 163083365.0, "step": 2489 }, { "epoch": 0.32910388580491673, "grad_norm": 0.8857034512054988, "learning_rate": 8.638005309611137e-06, "loss": 0.6115, "mean_token_accuracy": 0.7992003560066223, "num_tokens": 163148901.0, "step": 2490 }, { "epoch": 0.32923605604017975, "grad_norm": 0.9090812682849421, "learning_rate": 8.636516834170889e-06, "loss": 0.6172, "mean_token_accuracy": 0.7965297102928162, "num_tokens": 163214437.0, "step": 2491 }, { "epoch": 0.32936822627544277, "grad_norm": 1.0693613053762672, "learning_rate": 8.63502769103169e-06, "loss": 0.6608, "mean_token_accuracy": 0.7847179770469666, "num_tokens": 163279973.0, "step": 2492 }, { "epoch": 0.3295003965107058, "grad_norm": 1.0121654179971742, "learning_rate": 8.63353788051055e-06, "loss": 0.6254, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 163345509.0, "step": 2493 }, { "epoch": 0.3296325667459688, "grad_norm": 0.9029346879264794, "learning_rate": 8.632047402924618e-06, "loss": 0.619, "mean_token_accuracy": 0.7978421449661255, "num_tokens": 163411045.0, "step": 2494 }, { "epoch": 0.3297647369812318, "grad_norm": 0.8225680504042098, "learning_rate": 8.630556258591188e-06, "loss": 0.6024, "mean_token_accuracy": 0.8024508357048035, "num_tokens": 163476581.0, "step": 2495 }, { "epoch": 0.32989690721649484, "grad_norm": 0.8159127389561031, "learning_rate": 8.629064447827691e-06, "loss": 0.5669, "mean_token_accuracy": 0.8158649802207947, "num_tokens": 163542117.0, "step": 2496 }, { "epoch": 0.33002907745175786, "grad_norm": 0.9714093283370153, "learning_rate": 8.627571970951703e-06, "loss": 0.5844, "mean_token_accuracy": 0.8070138096809387, "num_tokens": 163607653.0, "step": 2497 }, { "epoch": 0.3301612476870209, "grad_norm": 0.9234775698268015, "learning_rate": 8.626078828280942e-06, "loss": 0.6461, "mean_token_accuracy": 0.7895861268043518, "num_tokens": 163673189.0, "step": 2498 }, { "epoch": 0.3302934179222839, "grad_norm": 0.9659099815741066, "learning_rate": 8.624585020133266e-06, "loss": 0.6525, "mean_token_accuracy": 0.784855306148529, "num_tokens": 163738725.0, "step": 2499 }, { "epoch": 0.3304255881575469, "grad_norm": 0.9181327999690032, "learning_rate": 8.623090546826676e-06, "loss": 0.6206, "mean_token_accuracy": 0.7962245345115662, "num_tokens": 163804261.0, "step": 2500 }, { "epoch": 0.33055775839280993, "grad_norm": 0.8615395619337111, "learning_rate": 8.621595408679315e-06, "loss": 0.646, "mean_token_accuracy": 0.7901507616043091, "num_tokens": 65536.0, "step": 2501 }, { "epoch": 0.33068992862807295, "grad_norm": 0.8361202755543437, "learning_rate": 8.620099606009464e-06, "loss": 0.5898, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 131072.0, "step": 2502 }, { "epoch": 0.33082209886333597, "grad_norm": 0.9753098298978115, "learning_rate": 8.61860313913555e-06, "loss": 0.6661, "mean_token_accuracy": 0.7840160131454468, "num_tokens": 196608.0, "step": 2503 }, { "epoch": 0.330954269098599, "grad_norm": 0.864610922938665, "learning_rate": 8.61710600837614e-06, "loss": 0.5828, "mean_token_accuracy": 0.8098523020744324, "num_tokens": 262144.0, "step": 2504 }, { "epoch": 0.331086439333862, "grad_norm": 0.8715608986944299, "learning_rate": 8.615608214049942e-06, "loss": 0.597, "mean_token_accuracy": 0.806403398513794, "num_tokens": 327680.0, "step": 2505 }, { "epoch": 0.331218609569125, "grad_norm": 0.9138407680191035, "learning_rate": 8.614109756475805e-06, "loss": 0.64, "mean_token_accuracy": 0.7937523126602173, "num_tokens": 393216.0, "step": 2506 }, { "epoch": 0.33135077980438804, "grad_norm": 0.8973688941027385, "learning_rate": 8.612610635972717e-06, "loss": 0.6276, "mean_token_accuracy": 0.7970333099365234, "num_tokens": 458752.0, "step": 2507 }, { "epoch": 0.33148295003965106, "grad_norm": 0.8697632410479386, "learning_rate": 8.611110852859811e-06, "loss": 0.5849, "mean_token_accuracy": 0.808631420135498, "num_tokens": 524288.0, "step": 2508 }, { "epoch": 0.3316151202749141, "grad_norm": 0.942270062307416, "learning_rate": 8.609610407456362e-06, "loss": 0.6359, "mean_token_accuracy": 0.7903491854667664, "num_tokens": 589824.0, "step": 2509 }, { "epoch": 0.3317472905101771, "grad_norm": 0.9234924710445781, "learning_rate": 8.60810930008178e-06, "loss": 0.6343, "mean_token_accuracy": 0.7918599843978882, "num_tokens": 655360.0, "step": 2510 }, { "epoch": 0.3318794607454401, "grad_norm": 0.8066238695861249, "learning_rate": 8.606607531055618e-06, "loss": 0.5415, "mean_token_accuracy": 0.8214350938796997, "num_tokens": 720896.0, "step": 2511 }, { "epoch": 0.33201163098070313, "grad_norm": 0.9281750879795779, "learning_rate": 8.605105100697578e-06, "loss": 0.6695, "mean_token_accuracy": 0.7813148498535156, "num_tokens": 786432.0, "step": 2512 }, { "epoch": 0.33214380121596615, "grad_norm": 0.9018210140640728, "learning_rate": 8.60360200932749e-06, "loss": 0.6298, "mean_token_accuracy": 0.7935081124305725, "num_tokens": 851968.0, "step": 2513 }, { "epoch": 0.33227597145122917, "grad_norm": 0.905160848827546, "learning_rate": 8.602098257265334e-06, "loss": 0.5861, "mean_token_accuracy": 0.8075631856918335, "num_tokens": 917504.0, "step": 2514 }, { "epoch": 0.3324081416864922, "grad_norm": 0.9771154053462509, "learning_rate": 8.600593844831225e-06, "loss": 0.655, "mean_token_accuracy": 0.7874801754951477, "num_tokens": 983040.0, "step": 2515 }, { "epoch": 0.3325403119217552, "grad_norm": 0.9036032578943896, "learning_rate": 8.599088772345422e-06, "loss": 0.6125, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 1048576.0, "step": 2516 }, { "epoch": 0.3326724821570182, "grad_norm": 0.9635874735480556, "learning_rate": 8.597583040128324e-06, "loss": 0.6799, "mean_token_accuracy": 0.7798345685005188, "num_tokens": 1114112.0, "step": 2517 }, { "epoch": 0.33280465239228124, "grad_norm": 0.8376600439261577, "learning_rate": 8.596076648500469e-06, "loss": 0.625, "mean_token_accuracy": 0.7949120998382568, "num_tokens": 1179648.0, "step": 2518 }, { "epoch": 0.33293682262754426, "grad_norm": 0.9146350699020118, "learning_rate": 8.594569597782538e-06, "loss": 0.5631, "mean_token_accuracy": 0.818413496017456, "num_tokens": 1245184.0, "step": 2519 }, { "epoch": 0.3330689928628073, "grad_norm": 0.8979884355346542, "learning_rate": 8.59306188829535e-06, "loss": 0.6354, "mean_token_accuracy": 0.7893877625465393, "num_tokens": 1310720.0, "step": 2520 }, { "epoch": 0.3332011630980703, "grad_norm": 0.9641074490164697, "learning_rate": 8.591553520359864e-06, "loss": 0.6847, "mean_token_accuracy": 0.7772860527038574, "num_tokens": 1376256.0, "step": 2521 }, { "epoch": 0.3333333333333333, "grad_norm": 0.8087351958384632, "learning_rate": 8.59004449429718e-06, "loss": 0.5788, "mean_token_accuracy": 0.8113478422164917, "num_tokens": 1441792.0, "step": 2522 }, { "epoch": 0.33346550356859633, "grad_norm": 0.8734812062448818, "learning_rate": 8.58853481042854e-06, "loss": 0.5593, "mean_token_accuracy": 0.8170705437660217, "num_tokens": 1507328.0, "step": 2523 }, { "epoch": 0.33359767380385935, "grad_norm": 0.8874764425810446, "learning_rate": 8.58702446907532e-06, "loss": 0.6564, "mean_token_accuracy": 0.7862898111343384, "num_tokens": 1572864.0, "step": 2524 }, { "epoch": 0.33372984403912237, "grad_norm": 0.8323640912003499, "learning_rate": 8.585513470559045e-06, "loss": 0.5602, "mean_token_accuracy": 0.8150561451911926, "num_tokens": 1638400.0, "step": 2525 }, { "epoch": 0.3338620142743854, "grad_norm": 0.8617487264293487, "learning_rate": 8.584001815201373e-06, "loss": 0.5723, "mean_token_accuracy": 0.8139116168022156, "num_tokens": 1703936.0, "step": 2526 }, { "epoch": 0.3339941845096484, "grad_norm": 0.9981588971943244, "learning_rate": 8.582489503324104e-06, "loss": 0.6037, "mean_token_accuracy": 0.8019014596939087, "num_tokens": 1769472.0, "step": 2527 }, { "epoch": 0.3341263547449114, "grad_norm": 0.8938183444686, "learning_rate": 8.580976535249177e-06, "loss": 0.5958, "mean_token_accuracy": 0.8038548231124878, "num_tokens": 1835008.0, "step": 2528 }, { "epoch": 0.33425852498017444, "grad_norm": 0.8786098376009605, "learning_rate": 8.579462911298671e-06, "loss": 0.6341, "mean_token_accuracy": 0.7947900295257568, "num_tokens": 1900544.0, "step": 2529 }, { "epoch": 0.33439069521543746, "grad_norm": 0.9514764716621615, "learning_rate": 8.577948631794809e-06, "loss": 0.6178, "mean_token_accuracy": 0.7948968410491943, "num_tokens": 1966080.0, "step": 2530 }, { "epoch": 0.3345228654507005, "grad_norm": 0.9649872643913437, "learning_rate": 8.576433697059943e-06, "loss": 0.6389, "mean_token_accuracy": 0.7917378544807434, "num_tokens": 2031616.0, "step": 2531 }, { "epoch": 0.3346550356859635, "grad_norm": 0.8746794807700539, "learning_rate": 8.574918107416574e-06, "loss": 0.6119, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 2097152.0, "step": 2532 }, { "epoch": 0.3347872059212265, "grad_norm": 0.8749038353805336, "learning_rate": 8.573401863187342e-06, "loss": 0.5729, "mean_token_accuracy": 0.8136674165725708, "num_tokens": 2162688.0, "step": 2533 }, { "epoch": 0.33491937615648953, "grad_norm": 0.8987688466746694, "learning_rate": 8.57188496469502e-06, "loss": 0.6163, "mean_token_accuracy": 0.8001617789268494, "num_tokens": 2228224.0, "step": 2534 }, { "epoch": 0.33505154639175255, "grad_norm": 0.8943684388164892, "learning_rate": 8.570367412262524e-06, "loss": 0.6222, "mean_token_accuracy": 0.7955072522163391, "num_tokens": 2293760.0, "step": 2535 }, { "epoch": 0.33518371662701557, "grad_norm": 0.9111568602328286, "learning_rate": 8.568849206212914e-06, "loss": 0.6414, "mean_token_accuracy": 0.7886857390403748, "num_tokens": 2359296.0, "step": 2536 }, { "epoch": 0.3353158868622786, "grad_norm": 0.8748495934892646, "learning_rate": 8.567330346869378e-06, "loss": 0.5663, "mean_token_accuracy": 0.8168721795082092, "num_tokens": 2424832.0, "step": 2537 }, { "epoch": 0.3354480570975416, "grad_norm": 0.9631984986870546, "learning_rate": 8.565810834555253e-06, "loss": 0.6678, "mean_token_accuracy": 0.7813148498535156, "num_tokens": 2490368.0, "step": 2538 }, { "epoch": 0.3355802273328046, "grad_norm": 0.8760919124219432, "learning_rate": 8.564290669594012e-06, "loss": 0.5629, "mean_token_accuracy": 0.8152698278427124, "num_tokens": 2555904.0, "step": 2539 }, { "epoch": 0.33571239756806764, "grad_norm": 0.9431365359598192, "learning_rate": 8.562769852309263e-06, "loss": 0.6544, "mean_token_accuracy": 0.7851452827453613, "num_tokens": 2621440.0, "step": 2540 }, { "epoch": 0.33584456780333066, "grad_norm": 0.7884253797422751, "learning_rate": 8.56124838302476e-06, "loss": 0.5121, "mean_token_accuracy": 0.8334910273551941, "num_tokens": 2686976.0, "step": 2541 }, { "epoch": 0.33597673803859374, "grad_norm": 0.9252110020612568, "learning_rate": 8.55972626206439e-06, "loss": 0.6259, "mean_token_accuracy": 0.7992766499519348, "num_tokens": 2752512.0, "step": 2542 }, { "epoch": 0.33610890827385675, "grad_norm": 0.842647977487357, "learning_rate": 8.558203489752182e-06, "loss": 0.5163, "mean_token_accuracy": 0.8351696729660034, "num_tokens": 2818048.0, "step": 2543 }, { "epoch": 0.3362410785091198, "grad_norm": 1.0994118788415392, "learning_rate": 8.556680066412302e-06, "loss": 0.6871, "mean_token_accuracy": 0.779529333114624, "num_tokens": 2883584.0, "step": 2544 }, { "epoch": 0.3363732487443828, "grad_norm": 0.8659120364439395, "learning_rate": 8.555155992369053e-06, "loss": 0.5759, "mean_token_accuracy": 0.8112409710884094, "num_tokens": 2949120.0, "step": 2545 }, { "epoch": 0.3365054189796458, "grad_norm": 0.8332172860516502, "learning_rate": 8.553631267946884e-06, "loss": 0.5616, "mean_token_accuracy": 0.8176657557487488, "num_tokens": 3014656.0, "step": 2546 }, { "epoch": 0.33663758921490883, "grad_norm": 0.9791537720753495, "learning_rate": 8.55210589347037e-06, "loss": 0.6623, "mean_token_accuracy": 0.7858777642250061, "num_tokens": 3080192.0, "step": 2547 }, { "epoch": 0.33676975945017185, "grad_norm": 0.9066973286821759, "learning_rate": 8.550579869264233e-06, "loss": 0.5807, "mean_token_accuracy": 0.8093487024307251, "num_tokens": 3145728.0, "step": 2548 }, { "epoch": 0.33690192968543486, "grad_norm": 0.837226100227639, "learning_rate": 8.549053195653337e-06, "loss": 0.5232, "mean_token_accuracy": 0.827081561088562, "num_tokens": 3211264.0, "step": 2549 }, { "epoch": 0.3370340999206979, "grad_norm": 0.9519742698685549, "learning_rate": 8.547525872962675e-06, "loss": 0.6569, "mean_token_accuracy": 0.7869460582733154, "num_tokens": 3276800.0, "step": 2550 }, { "epoch": 0.3371662701559609, "grad_norm": 0.8890820769852875, "learning_rate": 8.54599790151738e-06, "loss": 0.6249, "mean_token_accuracy": 0.7960413694381714, "num_tokens": 3342336.0, "step": 2551 }, { "epoch": 0.3372984403912239, "grad_norm": 0.8435523226450744, "learning_rate": 8.54446928164273e-06, "loss": 0.6048, "mean_token_accuracy": 0.802405059337616, "num_tokens": 3407872.0, "step": 2552 }, { "epoch": 0.33743061062648694, "grad_norm": 0.8946791249867131, "learning_rate": 8.542940013664132e-06, "loss": 0.5801, "mean_token_accuracy": 0.8084330558776855, "num_tokens": 3473408.0, "step": 2553 }, { "epoch": 0.33756278086174996, "grad_norm": 0.9049299846532431, "learning_rate": 8.541410097907139e-06, "loss": 0.6051, "mean_token_accuracy": 0.7995971441268921, "num_tokens": 3538944.0, "step": 2554 }, { "epoch": 0.337694951097013, "grad_norm": 0.9801434809885438, "learning_rate": 8.539879534697436e-06, "loss": 0.6116, "mean_token_accuracy": 0.79878830909729, "num_tokens": 3604480.0, "step": 2555 }, { "epoch": 0.337827121332276, "grad_norm": 0.8438828477932089, "learning_rate": 8.538348324360846e-06, "loss": 0.5439, "mean_token_accuracy": 0.821633517742157, "num_tokens": 3670016.0, "step": 2556 }, { "epoch": 0.337959291567539, "grad_norm": 0.8680684318458061, "learning_rate": 8.536816467223336e-06, "loss": 0.5762, "mean_token_accuracy": 0.8122787475585938, "num_tokens": 3735552.0, "step": 2557 }, { "epoch": 0.33809146180280203, "grad_norm": 0.9400367596996524, "learning_rate": 8.535283963611003e-06, "loss": 0.5908, "mean_token_accuracy": 0.8098523020744324, "num_tokens": 3801088.0, "step": 2558 }, { "epoch": 0.33822363203806505, "grad_norm": 0.919680285121244, "learning_rate": 8.533750813850087e-06, "loss": 0.65, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 3866624.0, "step": 2559 }, { "epoch": 0.33835580227332807, "grad_norm": 0.9453790677787148, "learning_rate": 8.532217018266963e-06, "loss": 0.6088, "mean_token_accuracy": 0.7999175786972046, "num_tokens": 3932160.0, "step": 2560 }, { "epoch": 0.3384879725085911, "grad_norm": 0.8759131974990597, "learning_rate": 8.530682577188144e-06, "loss": 0.6122, "mean_token_accuracy": 0.7985441088676453, "num_tokens": 3997696.0, "step": 2561 }, { "epoch": 0.3386201427438541, "grad_norm": 0.9050247128733986, "learning_rate": 8.52914749094028e-06, "loss": 0.628, "mean_token_accuracy": 0.7945153117179871, "num_tokens": 4063232.0, "step": 2562 }, { "epoch": 0.3387523129791171, "grad_norm": 0.9844801766343543, "learning_rate": 8.527611759850158e-06, "loss": 0.6145, "mean_token_accuracy": 0.8007721900939941, "num_tokens": 4128768.0, "step": 2563 }, { "epoch": 0.33888448321438014, "grad_norm": 0.8581068809904097, "learning_rate": 8.526075384244707e-06, "loss": 0.5894, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 4194304.0, "step": 2564 }, { "epoch": 0.33901665344964316, "grad_norm": 0.8763895919617987, "learning_rate": 8.524538364450984e-06, "loss": 0.5773, "mean_token_accuracy": 0.8125381469726562, "num_tokens": 4259840.0, "step": 2565 }, { "epoch": 0.3391488236849062, "grad_norm": 0.9248688057629052, "learning_rate": 8.523000700796193e-06, "loss": 0.6567, "mean_token_accuracy": 0.7843974828720093, "num_tokens": 4325376.0, "step": 2566 }, { "epoch": 0.3392809939201692, "grad_norm": 0.9239042893418359, "learning_rate": 8.52146239360767e-06, "loss": 0.5902, "mean_token_accuracy": 0.8049688935279846, "num_tokens": 4390912.0, "step": 2567 }, { "epoch": 0.3394131641554322, "grad_norm": 0.945357268270459, "learning_rate": 8.519923443212882e-06, "loss": 0.6425, "mean_token_accuracy": 0.7900744676589966, "num_tokens": 4456448.0, "step": 2568 }, { "epoch": 0.33954533439069523, "grad_norm": 0.9331970375665739, "learning_rate": 8.518383849939447e-06, "loss": 0.6261, "mean_token_accuracy": 0.7948205471038818, "num_tokens": 4521984.0, "step": 2569 }, { "epoch": 0.33967750462595825, "grad_norm": 0.8679908362617621, "learning_rate": 8.516843614115111e-06, "loss": 0.6138, "mean_token_accuracy": 0.79884934425354, "num_tokens": 4587520.0, "step": 2570 }, { "epoch": 0.33980967486122127, "grad_norm": 0.9812539350795672, "learning_rate": 8.515302736067755e-06, "loss": 0.6363, "mean_token_accuracy": 0.7916463017463684, "num_tokens": 4653056.0, "step": 2571 }, { "epoch": 0.3399418450964843, "grad_norm": 0.9121191170857811, "learning_rate": 8.513761216125403e-06, "loss": 0.5986, "mean_token_accuracy": 0.8035343885421753, "num_tokens": 4718592.0, "step": 2572 }, { "epoch": 0.3400740153317473, "grad_norm": 1.0808678715634603, "learning_rate": 8.51221905461621e-06, "loss": 0.6186, "mean_token_accuracy": 0.797857403755188, "num_tokens": 4784128.0, "step": 2573 }, { "epoch": 0.3402061855670103, "grad_norm": 1.073150148116403, "learning_rate": 8.51067625186847e-06, "loss": 0.6188, "mean_token_accuracy": 0.8007874488830566, "num_tokens": 4849664.0, "step": 2574 }, { "epoch": 0.34033835580227334, "grad_norm": 0.9200140503978027, "learning_rate": 8.509132808210616e-06, "loss": 0.6109, "mean_token_accuracy": 0.8012300133705139, "num_tokens": 4915200.0, "step": 2575 }, { "epoch": 0.34047052603753636, "grad_norm": 0.8490112091153091, "learning_rate": 8.507588723971213e-06, "loss": 0.5848, "mean_token_accuracy": 0.8087687492370605, "num_tokens": 4980736.0, "step": 2576 }, { "epoch": 0.3406026962727994, "grad_norm": 0.8865278716369352, "learning_rate": 8.506043999478963e-06, "loss": 0.5645, "mean_token_accuracy": 0.8143084049224854, "num_tokens": 5046272.0, "step": 2577 }, { "epoch": 0.3407348665080624, "grad_norm": 0.9511436109757307, "learning_rate": 8.504498635062707e-06, "loss": 0.6802, "mean_token_accuracy": 0.7770876288414001, "num_tokens": 5111808.0, "step": 2578 }, { "epoch": 0.3408670367433254, "grad_norm": 0.8918167692387188, "learning_rate": 8.502952631051425e-06, "loss": 0.5933, "mean_token_accuracy": 0.806525468826294, "num_tokens": 5177344.0, "step": 2579 }, { "epoch": 0.34099920697858843, "grad_norm": 0.9390629603274221, "learning_rate": 8.501405987774222e-06, "loss": 0.5859, "mean_token_accuracy": 0.8063880801200867, "num_tokens": 5242880.0, "step": 2580 }, { "epoch": 0.34113137721385145, "grad_norm": 0.9028995027140456, "learning_rate": 8.49985870556035e-06, "loss": 0.5623, "mean_token_accuracy": 0.8157429099082947, "num_tokens": 5308416.0, "step": 2581 }, { "epoch": 0.34126354744911447, "grad_norm": 0.8223076375578332, "learning_rate": 8.498310784739192e-06, "loss": 0.6052, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 5373952.0, "step": 2582 }, { "epoch": 0.3413957176843775, "grad_norm": 1.0021514295411542, "learning_rate": 8.496762225640267e-06, "loss": 0.6191, "mean_token_accuracy": 0.7985441088676453, "num_tokens": 5439488.0, "step": 2583 }, { "epoch": 0.3415278879196405, "grad_norm": 0.9484837172717233, "learning_rate": 8.495213028593232e-06, "loss": 0.6309, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 5505024.0, "step": 2584 }, { "epoch": 0.3416600581549035, "grad_norm": 0.9087395899757372, "learning_rate": 8.493663193927878e-06, "loss": 0.5681, "mean_token_accuracy": 0.8148577809333801, "num_tokens": 5570560.0, "step": 2585 }, { "epoch": 0.34179222839016654, "grad_norm": 0.8693947364104772, "learning_rate": 8.492112721974136e-06, "loss": 0.5348, "mean_token_accuracy": 0.82615065574646, "num_tokens": 5636096.0, "step": 2586 }, { "epoch": 0.34192439862542956, "grad_norm": 0.8994116053313308, "learning_rate": 8.490561613062064e-06, "loss": 0.5852, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 5701632.0, "step": 2587 }, { "epoch": 0.3420565688606926, "grad_norm": 0.8715868389507313, "learning_rate": 8.489009867521865e-06, "loss": 0.6133, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 5767168.0, "step": 2588 }, { "epoch": 0.3421887390959556, "grad_norm": 0.8921604716543382, "learning_rate": 8.487457485683871e-06, "loss": 0.6307, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 5832704.0, "step": 2589 }, { "epoch": 0.3423209093312186, "grad_norm": 0.9023547585270546, "learning_rate": 8.485904467878549e-06, "loss": 0.6318, "mean_token_accuracy": 0.7916615605354309, "num_tokens": 5898240.0, "step": 2590 }, { "epoch": 0.34245307956648163, "grad_norm": 0.9259969030439005, "learning_rate": 8.48435081443651e-06, "loss": 0.6255, "mean_token_accuracy": 0.7949731349945068, "num_tokens": 5963776.0, "step": 2591 }, { "epoch": 0.34258524980174465, "grad_norm": 0.8869874546691676, "learning_rate": 8.48279652568849e-06, "loss": 0.6085, "mean_token_accuracy": 0.7998107671737671, "num_tokens": 6029312.0, "step": 2592 }, { "epoch": 0.34271742003700767, "grad_norm": 0.876232385992557, "learning_rate": 8.481241601965367e-06, "loss": 0.623, "mean_token_accuracy": 0.7992156147956848, "num_tokens": 6094848.0, "step": 2593 }, { "epoch": 0.3428495902722707, "grad_norm": 0.9253752569508531, "learning_rate": 8.47968604359815e-06, "loss": 0.6195, "mean_token_accuracy": 0.797826886177063, "num_tokens": 6160384.0, "step": 2594 }, { "epoch": 0.3429817605075337, "grad_norm": 0.8679588611065133, "learning_rate": 8.478129850917986e-06, "loss": 0.6564, "mean_token_accuracy": 0.7841686010360718, "num_tokens": 6225920.0, "step": 2595 }, { "epoch": 0.3431139307427967, "grad_norm": 0.8801550003118228, "learning_rate": 8.476573024256156e-06, "loss": 0.6107, "mean_token_accuracy": 0.7981473803520203, "num_tokens": 6291456.0, "step": 2596 }, { "epoch": 0.34324610097805974, "grad_norm": 0.898979549875201, "learning_rate": 8.475015563944074e-06, "loss": 0.6199, "mean_token_accuracy": 0.7964686751365662, "num_tokens": 6356992.0, "step": 2597 }, { "epoch": 0.34337827121332276, "grad_norm": 0.9098761028440198, "learning_rate": 8.473457470313294e-06, "loss": 0.5803, "mean_token_accuracy": 0.8093334436416626, "num_tokens": 6422528.0, "step": 2598 }, { "epoch": 0.3435104414485858, "grad_norm": 0.8778103162161616, "learning_rate": 8.471898743695499e-06, "loss": 0.609, "mean_token_accuracy": 0.8017336130142212, "num_tokens": 6488064.0, "step": 2599 }, { "epoch": 0.3436426116838488, "grad_norm": 0.9395499629916776, "learning_rate": 8.47033938442251e-06, "loss": 0.6117, "mean_token_accuracy": 0.7983610033988953, "num_tokens": 6553600.0, "step": 2600 }, { "epoch": 0.3437747819191118, "grad_norm": 0.9527378360984219, "learning_rate": 8.468779392826283e-06, "loss": 0.6423, "mean_token_accuracy": 0.7892351150512695, "num_tokens": 6619136.0, "step": 2601 }, { "epoch": 0.34390695215437483, "grad_norm": 0.8655789860988728, "learning_rate": 8.467218769238907e-06, "loss": 0.6005, "mean_token_accuracy": 0.8028628826141357, "num_tokens": 6684672.0, "step": 2602 }, { "epoch": 0.34403912238963785, "grad_norm": 0.8898249116967958, "learning_rate": 8.465657513992604e-06, "loss": 0.5598, "mean_token_accuracy": 0.8191918134689331, "num_tokens": 6750208.0, "step": 2603 }, { "epoch": 0.34417129262490087, "grad_norm": 0.9691506278501141, "learning_rate": 8.464095627419737e-06, "loss": 0.6388, "mean_token_accuracy": 0.7916463017463684, "num_tokens": 6815744.0, "step": 2604 }, { "epoch": 0.3443034628601639, "grad_norm": 0.9532931177865578, "learning_rate": 8.462533109852797e-06, "loss": 0.6693, "mean_token_accuracy": 0.7811164855957031, "num_tokens": 6881280.0, "step": 2605 }, { "epoch": 0.3444356330954269, "grad_norm": 0.8804862889491312, "learning_rate": 8.46096996162441e-06, "loss": 0.6398, "mean_token_accuracy": 0.7903491854667664, "num_tokens": 6946816.0, "step": 2606 }, { "epoch": 0.3445678033306899, "grad_norm": 0.8597445105107968, "learning_rate": 8.45940618306734e-06, "loss": 0.6433, "mean_token_accuracy": 0.789189338684082, "num_tokens": 7012352.0, "step": 2607 }, { "epoch": 0.34469997356595294, "grad_norm": 0.8550717765690822, "learning_rate": 8.457841774514482e-06, "loss": 0.6036, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 7077888.0, "step": 2608 }, { "epoch": 0.34483214380121596, "grad_norm": 0.8249283312186265, "learning_rate": 8.456276736298861e-06, "loss": 0.6024, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 7143424.0, "step": 2609 }, { "epoch": 0.344964314036479, "grad_norm": 0.8469367780500178, "learning_rate": 8.45471106875365e-06, "loss": 0.613, "mean_token_accuracy": 0.7991393208503723, "num_tokens": 7208960.0, "step": 2610 }, { "epoch": 0.345096484271742, "grad_norm": 0.9503187288115993, "learning_rate": 8.453144772212143e-06, "loss": 0.6321, "mean_token_accuracy": 0.7960108518600464, "num_tokens": 7274496.0, "step": 2611 }, { "epoch": 0.345228654507005, "grad_norm": 0.8193712651569655, "learning_rate": 8.451577847007768e-06, "loss": 0.5788, "mean_token_accuracy": 0.8125839233398438, "num_tokens": 7340032.0, "step": 2612 }, { "epoch": 0.34536082474226804, "grad_norm": 0.9493837545188833, "learning_rate": 8.450010293474097e-06, "loss": 0.6668, "mean_token_accuracy": 0.7842448949813843, "num_tokens": 7405568.0, "step": 2613 }, { "epoch": 0.34549299497753105, "grad_norm": 0.9767074856182781, "learning_rate": 8.448442111944823e-06, "loss": 0.6451, "mean_token_accuracy": 0.7897539734840393, "num_tokens": 7471104.0, "step": 2614 }, { "epoch": 0.34562516521279407, "grad_norm": 0.8956163852289335, "learning_rate": 8.446873302753783e-06, "loss": 0.6055, "mean_token_accuracy": 0.803274929523468, "num_tokens": 7536640.0, "step": 2615 }, { "epoch": 0.3457573354480571, "grad_norm": 0.8710934169733913, "learning_rate": 8.445303866234944e-06, "loss": 0.5929, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 7602176.0, "step": 2616 }, { "epoch": 0.3458895056833201, "grad_norm": 0.988924113847472, "learning_rate": 8.443733802722406e-06, "loss": 0.6624, "mean_token_accuracy": 0.7872664928436279, "num_tokens": 7667712.0, "step": 2617 }, { "epoch": 0.3460216759185831, "grad_norm": 0.871252795212821, "learning_rate": 8.442163112550401e-06, "loss": 0.672, "mean_token_accuracy": 0.7816353440284729, "num_tokens": 7733248.0, "step": 2618 }, { "epoch": 0.34615384615384615, "grad_norm": 0.8530485316534514, "learning_rate": 8.440591796053299e-06, "loss": 0.5893, "mean_token_accuracy": 0.8052740693092346, "num_tokens": 7798784.0, "step": 2619 }, { "epoch": 0.34628601638910916, "grad_norm": 0.9148438520230256, "learning_rate": 8.439019853565598e-06, "loss": 0.6734, "mean_token_accuracy": 0.7801855802536011, "num_tokens": 7864320.0, "step": 2620 }, { "epoch": 0.3464181866243722, "grad_norm": 0.8757258597380174, "learning_rate": 8.437447285421931e-06, "loss": 0.6427, "mean_token_accuracy": 0.7894793152809143, "num_tokens": 7929856.0, "step": 2621 }, { "epoch": 0.3465503568596352, "grad_norm": 0.8934561763508148, "learning_rate": 8.435874091957067e-06, "loss": 0.6081, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 7995392.0, "step": 2622 }, { "epoch": 0.3466825270948982, "grad_norm": 1.0480098668627473, "learning_rate": 8.434300273505907e-06, "loss": 0.6552, "mean_token_accuracy": 0.7823373079299927, "num_tokens": 8060928.0, "step": 2623 }, { "epoch": 0.34681469733016124, "grad_norm": 0.8595023485632383, "learning_rate": 8.432725830403481e-06, "loss": 0.619, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 8126464.0, "step": 2624 }, { "epoch": 0.34694686756542426, "grad_norm": 0.7907233201130566, "learning_rate": 8.431150762984958e-06, "loss": 0.5543, "mean_token_accuracy": 0.817268967628479, "num_tokens": 8192000.0, "step": 2625 }, { "epoch": 0.3470790378006873, "grad_norm": 0.9035914050355538, "learning_rate": 8.429575071585634e-06, "loss": 0.5874, "mean_token_accuracy": 0.8053198456764221, "num_tokens": 8257536.0, "step": 2626 }, { "epoch": 0.3472112080359503, "grad_norm": 0.9152757150882904, "learning_rate": 8.427998756540942e-06, "loss": 0.6288, "mean_token_accuracy": 0.7939201593399048, "num_tokens": 8323072.0, "step": 2627 }, { "epoch": 0.3473433782712133, "grad_norm": 1.0250101738785335, "learning_rate": 8.42642181818645e-06, "loss": 0.6542, "mean_token_accuracy": 0.7845653891563416, "num_tokens": 8388608.0, "step": 2628 }, { "epoch": 0.34747554850647633, "grad_norm": 0.9336447921162287, "learning_rate": 8.424844256857849e-06, "loss": 0.6102, "mean_token_accuracy": 0.8000854849815369, "num_tokens": 8454144.0, "step": 2629 }, { "epoch": 0.34760771874173935, "grad_norm": 0.9662100846411303, "learning_rate": 8.423266072890973e-06, "loss": 0.6125, "mean_token_accuracy": 0.7969417572021484, "num_tokens": 8519680.0, "step": 2630 }, { "epoch": 0.34773988897700236, "grad_norm": 0.9656465953724489, "learning_rate": 8.421687266621786e-06, "loss": 0.5834, "mean_token_accuracy": 0.8069375157356262, "num_tokens": 8585216.0, "step": 2631 }, { "epoch": 0.3478720592122654, "grad_norm": 0.9247389868612381, "learning_rate": 8.420107838386377e-06, "loss": 0.6004, "mean_token_accuracy": 0.8015810251235962, "num_tokens": 8650752.0, "step": 2632 }, { "epoch": 0.3480042294475284, "grad_norm": 0.8519821084784897, "learning_rate": 8.418527788520976e-06, "loss": 0.6347, "mean_token_accuracy": 0.7913411259651184, "num_tokens": 8716288.0, "step": 2633 }, { "epoch": 0.3481363996827914, "grad_norm": 0.8202436773685877, "learning_rate": 8.416947117361947e-06, "loss": 0.6489, "mean_token_accuracy": 0.7870376110076904, "num_tokens": 8781824.0, "step": 2634 }, { "epoch": 0.34826856991805444, "grad_norm": 0.949671729250387, "learning_rate": 8.415365825245774e-06, "loss": 0.6555, "mean_token_accuracy": 0.7859541177749634, "num_tokens": 8847360.0, "step": 2635 }, { "epoch": 0.34840074015331746, "grad_norm": 0.9298275325110751, "learning_rate": 8.41378391250909e-06, "loss": 0.583, "mean_token_accuracy": 0.8094097375869751, "num_tokens": 8912896.0, "step": 2636 }, { "epoch": 0.3485329103885805, "grad_norm": 0.8717403004459024, "learning_rate": 8.412201379488643e-06, "loss": 0.5913, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 8978432.0, "step": 2637 }, { "epoch": 0.3486650806238435, "grad_norm": 0.835654594971432, "learning_rate": 8.410618226521326e-06, "loss": 0.5885, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 9043968.0, "step": 2638 }, { "epoch": 0.3487972508591065, "grad_norm": 0.8547107431895566, "learning_rate": 8.409034453944159e-06, "loss": 0.5706, "mean_token_accuracy": 0.812248170375824, "num_tokens": 9109504.0, "step": 2639 }, { "epoch": 0.34892942109436953, "grad_norm": 0.8074560199861953, "learning_rate": 8.40745006209429e-06, "loss": 0.5669, "mean_token_accuracy": 0.8134537935256958, "num_tokens": 9175040.0, "step": 2640 }, { "epoch": 0.34906159132963255, "grad_norm": 0.9142081332880966, "learning_rate": 8.405865051309008e-06, "loss": 0.6786, "mean_token_accuracy": 0.7777438759803772, "num_tokens": 9240576.0, "step": 2641 }, { "epoch": 0.34919376156489557, "grad_norm": 0.9573420767358094, "learning_rate": 8.404279421925726e-06, "loss": 0.6046, "mean_token_accuracy": 0.802435576915741, "num_tokens": 9306112.0, "step": 2642 }, { "epoch": 0.3493259318001586, "grad_norm": 0.9250558896205844, "learning_rate": 8.402693174281992e-06, "loss": 0.6031, "mean_token_accuracy": 0.8019930124282837, "num_tokens": 9371648.0, "step": 2643 }, { "epoch": 0.3494581020354216, "grad_norm": 0.8293696805599702, "learning_rate": 8.401106308715486e-06, "loss": 0.6163, "mean_token_accuracy": 0.7965754866600037, "num_tokens": 9437184.0, "step": 2644 }, { "epoch": 0.3495902722706846, "grad_norm": 0.9344298140941584, "learning_rate": 8.399518825564017e-06, "loss": 0.6584, "mean_token_accuracy": 0.783893883228302, "num_tokens": 9502720.0, "step": 2645 }, { "epoch": 0.34972244250594764, "grad_norm": 0.9784391872748965, "learning_rate": 8.397930725165528e-06, "loss": 0.6638, "mean_token_accuracy": 0.784489095211029, "num_tokens": 9568256.0, "step": 2646 }, { "epoch": 0.34985461274121066, "grad_norm": 0.8734148966468284, "learning_rate": 8.396342007858092e-06, "loss": 0.5999, "mean_token_accuracy": 0.8020541071891785, "num_tokens": 9633792.0, "step": 2647 }, { "epoch": 0.3499867829764737, "grad_norm": 0.8196858689534208, "learning_rate": 8.394752673979913e-06, "loss": 0.6108, "mean_token_accuracy": 0.7978421449661255, "num_tokens": 9699328.0, "step": 2648 }, { "epoch": 0.3501189532117367, "grad_norm": 0.8546856247602161, "learning_rate": 8.393162723869328e-06, "loss": 0.597, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 9764864.0, "step": 2649 }, { "epoch": 0.3502511234469997, "grad_norm": 0.8002354330435868, "learning_rate": 8.391572157864803e-06, "loss": 0.5998, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 9830400.0, "step": 2650 }, { "epoch": 0.35038329368226273, "grad_norm": 0.885816471026802, "learning_rate": 8.38998097630494e-06, "loss": 0.6021, "mean_token_accuracy": 0.8043889403343201, "num_tokens": 9895936.0, "step": 2651 }, { "epoch": 0.35051546391752575, "grad_norm": 0.9089881800967381, "learning_rate": 8.388389179528463e-06, "loss": 0.6326, "mean_token_accuracy": 0.7956445813179016, "num_tokens": 9961472.0, "step": 2652 }, { "epoch": 0.35064763415278877, "grad_norm": 0.9494995389146966, "learning_rate": 8.386796767874237e-06, "loss": 0.6209, "mean_token_accuracy": 0.7963313460350037, "num_tokens": 10027008.0, "step": 2653 }, { "epoch": 0.3507798043880518, "grad_norm": 0.9510254682027826, "learning_rate": 8.38520374168125e-06, "loss": 0.6231, "mean_token_accuracy": 0.7957209348678589, "num_tokens": 10092544.0, "step": 2654 }, { "epoch": 0.3509119746233148, "grad_norm": 0.9288212613708847, "learning_rate": 8.383610101288625e-06, "loss": 0.623, "mean_token_accuracy": 0.7927298545837402, "num_tokens": 10158080.0, "step": 2655 }, { "epoch": 0.3510441448585778, "grad_norm": 0.8560193170448361, "learning_rate": 8.382015847035615e-06, "loss": 0.6228, "mean_token_accuracy": 0.7968807220458984, "num_tokens": 10223616.0, "step": 2656 }, { "epoch": 0.35117631509384084, "grad_norm": 0.944752863252182, "learning_rate": 8.380420979261604e-06, "loss": 0.6161, "mean_token_accuracy": 0.7982694506645203, "num_tokens": 10289152.0, "step": 2657 }, { "epoch": 0.35130848532910386, "grad_norm": 0.9495249258083326, "learning_rate": 8.378825498306103e-06, "loss": 0.6404, "mean_token_accuracy": 0.7914937138557434, "num_tokens": 10354688.0, "step": 2658 }, { "epoch": 0.3514406555643669, "grad_norm": 0.9182526354693824, "learning_rate": 8.377229404508762e-06, "loss": 0.6483, "mean_token_accuracy": 0.788884162902832, "num_tokens": 10420224.0, "step": 2659 }, { "epoch": 0.35157282579962995, "grad_norm": 0.9098962562690277, "learning_rate": 8.375632698209352e-06, "loss": 0.6433, "mean_token_accuracy": 0.7893267273902893, "num_tokens": 10485760.0, "step": 2660 }, { "epoch": 0.35170499603489297, "grad_norm": 0.8536554761760546, "learning_rate": 8.37403537974778e-06, "loss": 0.6101, "mean_token_accuracy": 0.7990935444831848, "num_tokens": 10551296.0, "step": 2661 }, { "epoch": 0.351837166270156, "grad_norm": 0.7497162617842362, "learning_rate": 8.372437449464079e-06, "loss": 0.5473, "mean_token_accuracy": 0.8198174834251404, "num_tokens": 10616832.0, "step": 2662 }, { "epoch": 0.351969336505419, "grad_norm": 0.9694851257114392, "learning_rate": 8.37083890769842e-06, "loss": 0.6557, "mean_token_accuracy": 0.7895708680152893, "num_tokens": 10682368.0, "step": 2663 }, { "epoch": 0.352101506740682, "grad_norm": 0.8915301975915826, "learning_rate": 8.369239754791093e-06, "loss": 0.5974, "mean_token_accuracy": 0.8046331405639648, "num_tokens": 10747904.0, "step": 2664 }, { "epoch": 0.35223367697594504, "grad_norm": 0.8114331180912357, "learning_rate": 8.36763999108253e-06, "loss": 0.607, "mean_token_accuracy": 0.8039922118186951, "num_tokens": 10813440.0, "step": 2665 }, { "epoch": 0.35236584721120806, "grad_norm": 0.8674676842317048, "learning_rate": 8.366039616913285e-06, "loss": 0.606, "mean_token_accuracy": 0.8013215661048889, "num_tokens": 10878976.0, "step": 2666 }, { "epoch": 0.3524980174464711, "grad_norm": 0.8739506824354327, "learning_rate": 8.364438632624042e-06, "loss": 0.5934, "mean_token_accuracy": 0.8047399520874023, "num_tokens": 10944512.0, "step": 2667 }, { "epoch": 0.3526301876817341, "grad_norm": 0.8784413448008422, "learning_rate": 8.362837038555624e-06, "loss": 0.5785, "mean_token_accuracy": 0.8109205365180969, "num_tokens": 11010048.0, "step": 2668 }, { "epoch": 0.3527623579169971, "grad_norm": 0.9324655363599806, "learning_rate": 8.361234835048968e-06, "loss": 0.6549, "mean_token_accuracy": 0.7864424586296082, "num_tokens": 11075584.0, "step": 2669 }, { "epoch": 0.35289452815226013, "grad_norm": 0.9349226938902092, "learning_rate": 8.359632022445155e-06, "loss": 0.6175, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 11141120.0, "step": 2670 }, { "epoch": 0.35302669838752315, "grad_norm": 0.9461072758366249, "learning_rate": 8.35802860108539e-06, "loss": 0.6204, "mean_token_accuracy": 0.7962550520896912, "num_tokens": 11206656.0, "step": 2671 }, { "epoch": 0.35315886862278617, "grad_norm": 0.9588276222343258, "learning_rate": 8.356424571311005e-06, "loss": 0.6172, "mean_token_accuracy": 0.7985441088676453, "num_tokens": 11272192.0, "step": 2672 }, { "epoch": 0.3532910388580492, "grad_norm": 1.0117969710858976, "learning_rate": 8.354819933463468e-06, "loss": 0.6575, "mean_token_accuracy": 0.7877700924873352, "num_tokens": 11337728.0, "step": 2673 }, { "epoch": 0.3534232090933122, "grad_norm": 0.8943398386459634, "learning_rate": 8.353214687884372e-06, "loss": 0.66, "mean_token_accuracy": 0.7871291637420654, "num_tokens": 11403264.0, "step": 2674 }, { "epoch": 0.3535553793285752, "grad_norm": 0.8308424399191598, "learning_rate": 8.35160883491544e-06, "loss": 0.6024, "mean_token_accuracy": 0.803427517414093, "num_tokens": 11468800.0, "step": 2675 }, { "epoch": 0.35368754956383824, "grad_norm": 0.941541808935706, "learning_rate": 8.350002374898522e-06, "loss": 0.6318, "mean_token_accuracy": 0.7926992774009705, "num_tokens": 11534336.0, "step": 2676 }, { "epoch": 0.35381971979910126, "grad_norm": 0.9425356145886276, "learning_rate": 8.348395308175605e-06, "loss": 0.6209, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 11599872.0, "step": 2677 }, { "epoch": 0.3539518900343643, "grad_norm": 0.9471909373338293, "learning_rate": 8.346787635088796e-06, "loss": 0.6245, "mean_token_accuracy": 0.7943627238273621, "num_tokens": 11665408.0, "step": 2678 }, { "epoch": 0.3540840602696273, "grad_norm": 0.9101691392459081, "learning_rate": 8.345179355980335e-06, "loss": 0.5916, "mean_token_accuracy": 0.8049841523170471, "num_tokens": 11730944.0, "step": 2679 }, { "epoch": 0.3542162305048903, "grad_norm": 0.9098051320131191, "learning_rate": 8.343570471192594e-06, "loss": 0.6387, "mean_token_accuracy": 0.7883042097091675, "num_tokens": 11796480.0, "step": 2680 }, { "epoch": 0.35434840074015334, "grad_norm": 0.9618849832917813, "learning_rate": 8.341960981068069e-06, "loss": 0.6373, "mean_token_accuracy": 0.7937675714492798, "num_tokens": 11862016.0, "step": 2681 }, { "epoch": 0.35448057097541635, "grad_norm": 0.8910077969744956, "learning_rate": 8.340350885949387e-06, "loss": 0.5739, "mean_token_accuracy": 0.8126602172851562, "num_tokens": 11927552.0, "step": 2682 }, { "epoch": 0.35461274121067937, "grad_norm": 0.8410944010709256, "learning_rate": 8.338740186179302e-06, "loss": 0.6044, "mean_token_accuracy": 0.802069365978241, "num_tokens": 11993088.0, "step": 2683 }, { "epoch": 0.3547449114459424, "grad_norm": 0.8567962001663033, "learning_rate": 8.337128882100701e-06, "loss": 0.6129, "mean_token_accuracy": 0.7965144515037537, "num_tokens": 12058624.0, "step": 2684 }, { "epoch": 0.3548770816812054, "grad_norm": 0.9045343600363934, "learning_rate": 8.335516974056595e-06, "loss": 0.6381, "mean_token_accuracy": 0.7939201593399048, "num_tokens": 12124160.0, "step": 2685 }, { "epoch": 0.3550092519164684, "grad_norm": 0.8332860584880486, "learning_rate": 8.333904462390128e-06, "loss": 0.6176, "mean_token_accuracy": 0.7982541918754578, "num_tokens": 12189696.0, "step": 2686 }, { "epoch": 0.35514142215173145, "grad_norm": 0.8284171821040719, "learning_rate": 8.332291347444569e-06, "loss": 0.5322, "mean_token_accuracy": 0.8239836692810059, "num_tokens": 12255232.0, "step": 2687 }, { "epoch": 0.35527359238699446, "grad_norm": 0.9211894916566266, "learning_rate": 8.330677629563315e-06, "loss": 0.6084, "mean_token_accuracy": 0.8036717176437378, "num_tokens": 12320768.0, "step": 2688 }, { "epoch": 0.3554057626222575, "grad_norm": 0.8761895039556823, "learning_rate": 8.329063309089895e-06, "loss": 0.606, "mean_token_accuracy": 0.8008790016174316, "num_tokens": 12386304.0, "step": 2689 }, { "epoch": 0.3555379328575205, "grad_norm": 0.8491968266179418, "learning_rate": 8.32744838636796e-06, "loss": 0.6098, "mean_token_accuracy": 0.8023898005485535, "num_tokens": 12451840.0, "step": 2690 }, { "epoch": 0.3556701030927835, "grad_norm": 0.9555489819859084, "learning_rate": 8.3258328617413e-06, "loss": 0.6469, "mean_token_accuracy": 0.78789222240448, "num_tokens": 12517376.0, "step": 2691 }, { "epoch": 0.35580227332804654, "grad_norm": 0.8522953997634031, "learning_rate": 8.324216735553822e-06, "loss": 0.5936, "mean_token_accuracy": 0.8044958114624023, "num_tokens": 12582912.0, "step": 2692 }, { "epoch": 0.35593444356330955, "grad_norm": 0.8883676279311865, "learning_rate": 8.322600008149565e-06, "loss": 0.5697, "mean_token_accuracy": 0.816582202911377, "num_tokens": 12648448.0, "step": 2693 }, { "epoch": 0.3560666137985726, "grad_norm": 1.0431489091261683, "learning_rate": 8.3209826798727e-06, "loss": 0.6404, "mean_token_accuracy": 0.7885484099388123, "num_tokens": 12713984.0, "step": 2694 }, { "epoch": 0.3561987840338356, "grad_norm": 0.9637204024687999, "learning_rate": 8.31936475106752e-06, "loss": 0.635, "mean_token_accuracy": 0.7940727472305298, "num_tokens": 12779520.0, "step": 2695 }, { "epoch": 0.3563309542690986, "grad_norm": 0.8956279803133322, "learning_rate": 8.317746222078448e-06, "loss": 0.5495, "mean_token_accuracy": 0.8212367296218872, "num_tokens": 12845056.0, "step": 2696 }, { "epoch": 0.35646312450436163, "grad_norm": 0.8309698701461341, "learning_rate": 8.316127093250035e-06, "loss": 0.5436, "mean_token_accuracy": 0.8234494924545288, "num_tokens": 12910592.0, "step": 2697 }, { "epoch": 0.35659529473962465, "grad_norm": 0.8376986140442753, "learning_rate": 8.314507364926961e-06, "loss": 0.6081, "mean_token_accuracy": 0.8006043434143066, "num_tokens": 12976128.0, "step": 2698 }, { "epoch": 0.35672746497488766, "grad_norm": 0.9184807611091096, "learning_rate": 8.31288703745403e-06, "loss": 0.5876, "mean_token_accuracy": 0.8067543506622314, "num_tokens": 13041664.0, "step": 2699 }, { "epoch": 0.3568596352101507, "grad_norm": 0.9086673148144371, "learning_rate": 8.311266111176181e-06, "loss": 0.6193, "mean_token_accuracy": 0.7988951206207275, "num_tokens": 13107200.0, "step": 2700 }, { "epoch": 0.3569918054454137, "grad_norm": 0.8958444684037814, "learning_rate": 8.30964458643847e-06, "loss": 0.6172, "mean_token_accuracy": 0.8001160025596619, "num_tokens": 13172736.0, "step": 2701 }, { "epoch": 0.3571239756806767, "grad_norm": 0.8482105739857645, "learning_rate": 8.308022463586086e-06, "loss": 0.5885, "mean_token_accuracy": 0.8072274327278137, "num_tokens": 13238272.0, "step": 2702 }, { "epoch": 0.35725614591593974, "grad_norm": 0.8975586651409192, "learning_rate": 8.30639974296435e-06, "loss": 0.5898, "mean_token_accuracy": 0.8080820441246033, "num_tokens": 13303808.0, "step": 2703 }, { "epoch": 0.35738831615120276, "grad_norm": 0.9030741782629323, "learning_rate": 8.304776424918701e-06, "loss": 0.5588, "mean_token_accuracy": 0.8159412741661072, "num_tokens": 13369344.0, "step": 2704 }, { "epoch": 0.3575204863864658, "grad_norm": 0.8581322839800193, "learning_rate": 8.30315250979471e-06, "loss": 0.5981, "mean_token_accuracy": 0.8034886121749878, "num_tokens": 13434880.0, "step": 2705 }, { "epoch": 0.3576526566217288, "grad_norm": 0.9573135240285804, "learning_rate": 8.301527997938076e-06, "loss": 0.6548, "mean_token_accuracy": 0.7841228246688843, "num_tokens": 13500416.0, "step": 2706 }, { "epoch": 0.3577848268569918, "grad_norm": 0.8672018857005329, "learning_rate": 8.299902889694623e-06, "loss": 0.599, "mean_token_accuracy": 0.802313506603241, "num_tokens": 13565952.0, "step": 2707 }, { "epoch": 0.35791699709225483, "grad_norm": 0.936468896572165, "learning_rate": 8.298277185410301e-06, "loss": 0.6528, "mean_token_accuracy": 0.7840770483016968, "num_tokens": 13631488.0, "step": 2708 }, { "epoch": 0.35804916732751785, "grad_norm": 0.8798101979443432, "learning_rate": 8.296650885431192e-06, "loss": 0.625, "mean_token_accuracy": 0.7968349456787109, "num_tokens": 13697024.0, "step": 2709 }, { "epoch": 0.35818133756278087, "grad_norm": 0.7979526744980291, "learning_rate": 8.295023990103497e-06, "loss": 0.5853, "mean_token_accuracy": 0.808387279510498, "num_tokens": 13762560.0, "step": 2710 }, { "epoch": 0.3583135077980439, "grad_norm": 0.8367361289592121, "learning_rate": 8.293396499773552e-06, "loss": 0.5701, "mean_token_accuracy": 0.8149188160896301, "num_tokens": 13828096.0, "step": 2711 }, { "epoch": 0.3584456780333069, "grad_norm": 0.9419330386127697, "learning_rate": 8.291768414787816e-06, "loss": 0.6316, "mean_token_accuracy": 0.7917531728744507, "num_tokens": 13893632.0, "step": 2712 }, { "epoch": 0.3585778482685699, "grad_norm": 0.8503303223071446, "learning_rate": 8.290139735492871e-06, "loss": 0.5834, "mean_token_accuracy": 0.8089824318885803, "num_tokens": 13959168.0, "step": 2713 }, { "epoch": 0.35871001850383294, "grad_norm": 0.9821489606288948, "learning_rate": 8.288510462235428e-06, "loss": 0.642, "mean_token_accuracy": 0.78813636302948, "num_tokens": 14024704.0, "step": 2714 }, { "epoch": 0.35884218873909596, "grad_norm": 0.9198778845201958, "learning_rate": 8.286880595362329e-06, "loss": 0.656, "mean_token_accuracy": 0.7854200005531311, "num_tokens": 14090240.0, "step": 2715 }, { "epoch": 0.358974358974359, "grad_norm": 0.9220700216173483, "learning_rate": 8.285250135220538e-06, "loss": 0.6465, "mean_token_accuracy": 0.7894640564918518, "num_tokens": 14155776.0, "step": 2716 }, { "epoch": 0.359106529209622, "grad_norm": 0.8360597017423399, "learning_rate": 8.283619082157144e-06, "loss": 0.6148, "mean_token_accuracy": 0.7993529438972473, "num_tokens": 14221312.0, "step": 2717 }, { "epoch": 0.359238699444885, "grad_norm": 0.8411495225198162, "learning_rate": 8.281987436519365e-06, "loss": 0.5284, "mean_token_accuracy": 0.8259522914886475, "num_tokens": 14286848.0, "step": 2718 }, { "epoch": 0.35937086968014803, "grad_norm": 1.0219927958184205, "learning_rate": 8.280355198654545e-06, "loss": 0.6253, "mean_token_accuracy": 0.7937523126602173, "num_tokens": 14352384.0, "step": 2719 }, { "epoch": 0.35950303991541105, "grad_norm": 0.9386624768897718, "learning_rate": 8.278722368910153e-06, "loss": 0.6489, "mean_token_accuracy": 0.7861067056655884, "num_tokens": 14417920.0, "step": 2720 }, { "epoch": 0.35963521015067407, "grad_norm": 0.9213040622240265, "learning_rate": 8.277088947633785e-06, "loss": 0.645, "mean_token_accuracy": 0.7894793152809143, "num_tokens": 14483456.0, "step": 2721 }, { "epoch": 0.3597673803859371, "grad_norm": 0.9500195398474733, "learning_rate": 8.27545493517316e-06, "loss": 0.6656, "mean_token_accuracy": 0.7830240726470947, "num_tokens": 14548992.0, "step": 2722 }, { "epoch": 0.3598995506212001, "grad_norm": 0.8703825821764608, "learning_rate": 8.273820331876126e-06, "loss": 0.5671, "mean_token_accuracy": 0.8129807114601135, "num_tokens": 14614528.0, "step": 2723 }, { "epoch": 0.3600317208564631, "grad_norm": 0.9234225539873369, "learning_rate": 8.272185138090659e-06, "loss": 0.6167, "mean_token_accuracy": 0.7990477085113525, "num_tokens": 14680064.0, "step": 2724 }, { "epoch": 0.36016389109172614, "grad_norm": 0.8566599010218665, "learning_rate": 8.270549354164854e-06, "loss": 0.5854, "mean_token_accuracy": 0.8077157735824585, "num_tokens": 14745600.0, "step": 2725 }, { "epoch": 0.36029606132698916, "grad_norm": 0.8990355512762397, "learning_rate": 8.268912980446936e-06, "loss": 0.5991, "mean_token_accuracy": 0.8037022352218628, "num_tokens": 14811136.0, "step": 2726 }, { "epoch": 0.3604282315622522, "grad_norm": 0.9126001120751068, "learning_rate": 8.267276017285256e-06, "loss": 0.6055, "mean_token_accuracy": 0.8009095191955566, "num_tokens": 14876672.0, "step": 2727 }, { "epoch": 0.3605604017975152, "grad_norm": 1.000427972735225, "learning_rate": 8.265638465028288e-06, "loss": 0.6248, "mean_token_accuracy": 0.792714536190033, "num_tokens": 14942208.0, "step": 2728 }, { "epoch": 0.3606925720327782, "grad_norm": 0.8607734579782829, "learning_rate": 8.264000324024633e-06, "loss": 0.5503, "mean_token_accuracy": 0.822900116443634, "num_tokens": 15007744.0, "step": 2729 }, { "epoch": 0.36082474226804123, "grad_norm": 0.930713914176461, "learning_rate": 8.262361594623017e-06, "loss": 0.6391, "mean_token_accuracy": 0.7919210195541382, "num_tokens": 15073280.0, "step": 2730 }, { "epoch": 0.36095691250330425, "grad_norm": 1.0341541517517139, "learning_rate": 8.260722277172291e-06, "loss": 0.644, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 15138816.0, "step": 2731 }, { "epoch": 0.36108908273856727, "grad_norm": 0.9220591115525232, "learning_rate": 8.259082372021434e-06, "loss": 0.6724, "mean_token_accuracy": 0.7813758850097656, "num_tokens": 15204352.0, "step": 2732 }, { "epoch": 0.3612212529738303, "grad_norm": 0.8178069567069007, "learning_rate": 8.257441879519543e-06, "loss": 0.6179, "mean_token_accuracy": 0.798009991645813, "num_tokens": 15269888.0, "step": 2733 }, { "epoch": 0.3613534232090933, "grad_norm": 0.889977305203193, "learning_rate": 8.255800800015849e-06, "loss": 0.6426, "mean_token_accuracy": 0.7917836904525757, "num_tokens": 15335424.0, "step": 2734 }, { "epoch": 0.3614855934443563, "grad_norm": 0.9708747465405112, "learning_rate": 8.2541591338597e-06, "loss": 0.5429, "mean_token_accuracy": 0.8218013644218445, "num_tokens": 15400960.0, "step": 2735 }, { "epoch": 0.36161776367961934, "grad_norm": 0.8831488496879324, "learning_rate": 8.252516881400575e-06, "loss": 0.5668, "mean_token_accuracy": 0.812934935092926, "num_tokens": 15466496.0, "step": 2736 }, { "epoch": 0.36174993391488236, "grad_norm": 0.8881745422518049, "learning_rate": 8.250874042988076e-06, "loss": 0.5612, "mean_token_accuracy": 0.8153461217880249, "num_tokens": 15532032.0, "step": 2737 }, { "epoch": 0.3618821041501454, "grad_norm": 0.8590580912318085, "learning_rate": 8.249230618971926e-06, "loss": 0.558, "mean_token_accuracy": 0.8162922859191895, "num_tokens": 15597568.0, "step": 2738 }, { "epoch": 0.3620142743854084, "grad_norm": 0.8432001827996196, "learning_rate": 8.247586609701975e-06, "loss": 0.57, "mean_token_accuracy": 0.8121413588523865, "num_tokens": 15663104.0, "step": 2739 }, { "epoch": 0.3621464446206714, "grad_norm": 0.8866309089467941, "learning_rate": 8.245942015528203e-06, "loss": 0.6153, "mean_token_accuracy": 0.7991240620613098, "num_tokens": 15728640.0, "step": 2740 }, { "epoch": 0.36227861485593443, "grad_norm": 1.024184035588567, "learning_rate": 8.244296836800705e-06, "loss": 0.6493, "mean_token_accuracy": 0.78819739818573, "num_tokens": 15794176.0, "step": 2741 }, { "epoch": 0.36241078509119745, "grad_norm": 0.946616876960701, "learning_rate": 8.24265107386971e-06, "loss": 0.606, "mean_token_accuracy": 0.8010926842689514, "num_tokens": 15859712.0, "step": 2742 }, { "epoch": 0.36254295532646047, "grad_norm": 0.8685601020590473, "learning_rate": 8.24100472708556e-06, "loss": 0.6144, "mean_token_accuracy": 0.7981016039848328, "num_tokens": 15925248.0, "step": 2743 }, { "epoch": 0.3626751255617235, "grad_norm": 0.9543776848647352, "learning_rate": 8.239357796798735e-06, "loss": 0.6518, "mean_token_accuracy": 0.7868391871452332, "num_tokens": 15990784.0, "step": 2744 }, { "epoch": 0.3628072957969865, "grad_norm": 0.916266228483153, "learning_rate": 8.237710283359827e-06, "loss": 0.6044, "mean_token_accuracy": 0.803458034992218, "num_tokens": 16056320.0, "step": 2745 }, { "epoch": 0.3629394660322495, "grad_norm": 1.0069986202835939, "learning_rate": 8.23606218711956e-06, "loss": 0.6635, "mean_token_accuracy": 0.7816811203956604, "num_tokens": 16121856.0, "step": 2746 }, { "epoch": 0.36307163626751254, "grad_norm": 0.8769156823132666, "learning_rate": 8.234413508428778e-06, "loss": 0.6139, "mean_token_accuracy": 0.7998107671737671, "num_tokens": 16187392.0, "step": 2747 }, { "epoch": 0.36320380650277556, "grad_norm": 0.9207917839348541, "learning_rate": 8.232764247638451e-06, "loss": 0.59, "mean_token_accuracy": 0.8089366555213928, "num_tokens": 16252928.0, "step": 2748 }, { "epoch": 0.3633359767380386, "grad_norm": 0.8989321758052644, "learning_rate": 8.23111440509967e-06, "loss": 0.5936, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 16318464.0, "step": 2749 }, { "epoch": 0.3634681469733016, "grad_norm": 0.813278704706255, "learning_rate": 8.229463981163654e-06, "loss": 0.5494, "mean_token_accuracy": 0.8199090361595154, "num_tokens": 16384000.0, "step": 2750 }, { "epoch": 0.3636003172085646, "grad_norm": 0.8782083130861672, "learning_rate": 8.227812976181745e-06, "loss": 0.6333, "mean_token_accuracy": 0.7951257228851318, "num_tokens": 16449536.0, "step": 2751 }, { "epoch": 0.36373248744382763, "grad_norm": 1.0172112862399605, "learning_rate": 8.226161390505405e-06, "loss": 0.6297, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 16515072.0, "step": 2752 }, { "epoch": 0.36386465767909065, "grad_norm": 0.9624301134251244, "learning_rate": 8.224509224486222e-06, "loss": 0.6383, "mean_token_accuracy": 0.7917225956916809, "num_tokens": 16580608.0, "step": 2753 }, { "epoch": 0.36399682791435367, "grad_norm": 0.8671077879053571, "learning_rate": 8.22285647847591e-06, "loss": 0.6215, "mean_token_accuracy": 0.7938896417617798, "num_tokens": 16646144.0, "step": 2754 }, { "epoch": 0.3641289981496167, "grad_norm": 0.946065761357932, "learning_rate": 8.221203152826304e-06, "loss": 0.6365, "mean_token_accuracy": 0.7898761034011841, "num_tokens": 16711680.0, "step": 2755 }, { "epoch": 0.3642611683848797, "grad_norm": 0.9110685584695599, "learning_rate": 8.21954924788936e-06, "loss": 0.5944, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 16777216.0, "step": 2756 }, { "epoch": 0.3643933386201427, "grad_norm": 0.8832636306517235, "learning_rate": 8.21789476401716e-06, "loss": 0.5934, "mean_token_accuracy": 0.8046331405639648, "num_tokens": 16842752.0, "step": 2757 }, { "epoch": 0.36452550885540574, "grad_norm": 0.9411754705158212, "learning_rate": 8.216239701561912e-06, "loss": 0.6691, "mean_token_accuracy": 0.7797124981880188, "num_tokens": 16908288.0, "step": 2758 }, { "epoch": 0.36465767909066876, "grad_norm": 0.9117104175971673, "learning_rate": 8.214584060875945e-06, "loss": 0.6206, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 16973824.0, "step": 2759 }, { "epoch": 0.3647898493259318, "grad_norm": 0.877324434846399, "learning_rate": 8.212927842311704e-06, "loss": 0.5928, "mean_token_accuracy": 0.8069527745246887, "num_tokens": 17039360.0, "step": 2760 }, { "epoch": 0.3649220195611948, "grad_norm": 1.0094695094236337, "learning_rate": 8.211271046221767e-06, "loss": 0.6771, "mean_token_accuracy": 0.7796514630317688, "num_tokens": 17104896.0, "step": 2761 }, { "epoch": 0.3650541897964578, "grad_norm": 0.882614128146067, "learning_rate": 8.209613672958834e-06, "loss": 0.6249, "mean_token_accuracy": 0.7959498167037964, "num_tokens": 17170432.0, "step": 2762 }, { "epoch": 0.36518636003172084, "grad_norm": 0.9680592685045242, "learning_rate": 8.207955722875722e-06, "loss": 0.6442, "mean_token_accuracy": 0.7861829996109009, "num_tokens": 17235968.0, "step": 2763 }, { "epoch": 0.36531853026698385, "grad_norm": 0.9478989796436131, "learning_rate": 8.206297196325374e-06, "loss": 0.6615, "mean_token_accuracy": 0.7819710373878479, "num_tokens": 17301504.0, "step": 2764 }, { "epoch": 0.3654507005022469, "grad_norm": 0.9041935535988185, "learning_rate": 8.204638093660858e-06, "loss": 0.6393, "mean_token_accuracy": 0.7893267273902893, "num_tokens": 17367040.0, "step": 2765 }, { "epoch": 0.3655828707375099, "grad_norm": 0.9869215219135187, "learning_rate": 8.202978415235358e-06, "loss": 0.6325, "mean_token_accuracy": 0.7919362783432007, "num_tokens": 17432576.0, "step": 2766 }, { "epoch": 0.3657150409727729, "grad_norm": 0.8287728425950147, "learning_rate": 8.201318161402189e-06, "loss": 0.5965, "mean_token_accuracy": 0.8069069981575012, "num_tokens": 17498112.0, "step": 2767 }, { "epoch": 0.3658472112080359, "grad_norm": 0.9328058148409556, "learning_rate": 8.199657332514782e-06, "loss": 0.576, "mean_token_accuracy": 0.8071358799934387, "num_tokens": 17563648.0, "step": 2768 }, { "epoch": 0.36597938144329895, "grad_norm": 1.0128102619979122, "learning_rate": 8.197995928926694e-06, "loss": 0.6636, "mean_token_accuracy": 0.7812690734863281, "num_tokens": 17629184.0, "step": 2769 }, { "epoch": 0.36611155167856196, "grad_norm": 0.9681576828601398, "learning_rate": 8.196333950991605e-06, "loss": 0.6627, "mean_token_accuracy": 0.78265780210495, "num_tokens": 17694720.0, "step": 2770 }, { "epoch": 0.366243721913825, "grad_norm": 0.8267132890347159, "learning_rate": 8.19467139906331e-06, "loss": 0.5607, "mean_token_accuracy": 0.8157886862754822, "num_tokens": 17760256.0, "step": 2771 }, { "epoch": 0.366375892149088, "grad_norm": 0.9199920614019652, "learning_rate": 8.193008273495735e-06, "loss": 0.6415, "mean_token_accuracy": 0.7897998094558716, "num_tokens": 17825792.0, "step": 2772 }, { "epoch": 0.366508062384351, "grad_norm": 0.880106878068834, "learning_rate": 8.191344574642928e-06, "loss": 0.5853, "mean_token_accuracy": 0.8098217844963074, "num_tokens": 17891328.0, "step": 2773 }, { "epoch": 0.36664023261961404, "grad_norm": 1.1837895647159358, "learning_rate": 8.18968030285905e-06, "loss": 0.7035, "mean_token_accuracy": 0.7711970210075378, "num_tokens": 17956864.0, "step": 2774 }, { "epoch": 0.36677240285487706, "grad_norm": 0.9332539047474332, "learning_rate": 8.188015458498392e-06, "loss": 0.5772, "mean_token_accuracy": 0.812034547328949, "num_tokens": 18022400.0, "step": 2775 }, { "epoch": 0.3669045730901401, "grad_norm": 0.9041345024795122, "learning_rate": 8.186350041915364e-06, "loss": 0.5699, "mean_token_accuracy": 0.8113325834274292, "num_tokens": 18087936.0, "step": 2776 }, { "epoch": 0.3670367433254031, "grad_norm": 0.9760302579277642, "learning_rate": 8.1846840534645e-06, "loss": 0.6424, "mean_token_accuracy": 0.7891435623168945, "num_tokens": 18153472.0, "step": 2777 }, { "epoch": 0.3671689135606661, "grad_norm": 0.8310650360967734, "learning_rate": 8.183017493500453e-06, "loss": 0.5899, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 18219008.0, "step": 2778 }, { "epoch": 0.3673010837959292, "grad_norm": 0.8400343776731278, "learning_rate": 8.181350362377999e-06, "loss": 0.5598, "mean_token_accuracy": 0.8169637322425842, "num_tokens": 18284544.0, "step": 2779 }, { "epoch": 0.3674332540311922, "grad_norm": 0.8580990079494961, "learning_rate": 8.179682660452034e-06, "loss": 0.6087, "mean_token_accuracy": 0.8001160025596619, "num_tokens": 18350080.0, "step": 2780 }, { "epoch": 0.3675654242664552, "grad_norm": 0.843396246902864, "learning_rate": 8.17801438807758e-06, "loss": 0.6343, "mean_token_accuracy": 0.7941337823867798, "num_tokens": 18415616.0, "step": 2781 }, { "epoch": 0.36769759450171824, "grad_norm": 0.8867058725762283, "learning_rate": 8.176345545609776e-06, "loss": 0.6325, "mean_token_accuracy": 0.7933860421180725, "num_tokens": 18481152.0, "step": 2782 }, { "epoch": 0.36782976473698126, "grad_norm": 0.8338184837962992, "learning_rate": 8.174676133403884e-06, "loss": 0.573, "mean_token_accuracy": 0.8112562298774719, "num_tokens": 18546688.0, "step": 2783 }, { "epoch": 0.3679619349722443, "grad_norm": 0.9172897401961427, "learning_rate": 8.173006151815283e-06, "loss": 0.5997, "mean_token_accuracy": 0.8009552955627441, "num_tokens": 18612224.0, "step": 2784 }, { "epoch": 0.3680941052075073, "grad_norm": 0.861147464236057, "learning_rate": 8.171335601199485e-06, "loss": 0.6168, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 18677760.0, "step": 2785 }, { "epoch": 0.3682262754427703, "grad_norm": 0.8164022954632272, "learning_rate": 8.169664481912108e-06, "loss": 0.5844, "mean_token_accuracy": 0.808387279510498, "num_tokens": 18743296.0, "step": 2786 }, { "epoch": 0.36835844567803333, "grad_norm": 0.9289509791581361, "learning_rate": 8.1679927943089e-06, "loss": 0.5753, "mean_token_accuracy": 0.8140031695365906, "num_tokens": 18808832.0, "step": 2787 }, { "epoch": 0.36849061591329635, "grad_norm": 0.8524146242123298, "learning_rate": 8.166320538745734e-06, "loss": 0.6225, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 18874368.0, "step": 2788 }, { "epoch": 0.36862278614855937, "grad_norm": 0.8023885186314234, "learning_rate": 8.16464771557859e-06, "loss": 0.5375, "mean_token_accuracy": 0.8247161507606506, "num_tokens": 18939904.0, "step": 2789 }, { "epoch": 0.3687549563838224, "grad_norm": 0.8738224286008437, "learning_rate": 8.162974325163585e-06, "loss": 0.5904, "mean_token_accuracy": 0.8052435517311096, "num_tokens": 19005440.0, "step": 2790 }, { "epoch": 0.3688871266190854, "grad_norm": 0.8718536249239904, "learning_rate": 8.161300367856942e-06, "loss": 0.6313, "mean_token_accuracy": 0.7910206317901611, "num_tokens": 19070976.0, "step": 2791 }, { "epoch": 0.3690192968543484, "grad_norm": 0.8424664964009686, "learning_rate": 8.159625844015017e-06, "loss": 0.5787, "mean_token_accuracy": 0.8113478422164917, "num_tokens": 19136512.0, "step": 2792 }, { "epoch": 0.36915146708961144, "grad_norm": 0.8728965698709412, "learning_rate": 8.157950753994278e-06, "loss": 0.5695, "mean_token_accuracy": 0.8133469820022583, "num_tokens": 19202048.0, "step": 2793 }, { "epoch": 0.36928363732487446, "grad_norm": 0.9474317330154577, "learning_rate": 8.156275098151317e-06, "loss": 0.6513, "mean_token_accuracy": 0.7865187525749207, "num_tokens": 19267584.0, "step": 2794 }, { "epoch": 0.3694158075601375, "grad_norm": 0.887454611007992, "learning_rate": 8.154598876842848e-06, "loss": 0.5875, "mean_token_accuracy": 0.8065407276153564, "num_tokens": 19333120.0, "step": 2795 }, { "epoch": 0.3695479777954005, "grad_norm": 0.9187798167170685, "learning_rate": 8.1529220904257e-06, "loss": 0.5941, "mean_token_accuracy": 0.8047704696655273, "num_tokens": 19398656.0, "step": 2796 }, { "epoch": 0.3696801480306635, "grad_norm": 0.9365313019838103, "learning_rate": 8.151244739256829e-06, "loss": 0.6428, "mean_token_accuracy": 0.7883347868919373, "num_tokens": 19464192.0, "step": 2797 }, { "epoch": 0.36981231826592653, "grad_norm": 0.8158653926008983, "learning_rate": 8.149566823693306e-06, "loss": 0.5913, "mean_token_accuracy": 0.8070138096809387, "num_tokens": 19529728.0, "step": 2798 }, { "epoch": 0.36994448850118955, "grad_norm": 0.8258528447976459, "learning_rate": 8.147888344092326e-06, "loss": 0.5752, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 19595264.0, "step": 2799 }, { "epoch": 0.37007665873645257, "grad_norm": 0.9317388114425198, "learning_rate": 8.146209300811201e-06, "loss": 0.6549, "mean_token_accuracy": 0.7869918346405029, "num_tokens": 19660800.0, "step": 2800 }, { "epoch": 0.3702088289717156, "grad_norm": 0.9349900846920335, "learning_rate": 8.144529694207366e-06, "loss": 0.6081, "mean_token_accuracy": 0.8012300133705139, "num_tokens": 19726336.0, "step": 2801 }, { "epoch": 0.3703409992069786, "grad_norm": 0.8993521744596975, "learning_rate": 8.142849524638372e-06, "loss": 0.5744, "mean_token_accuracy": 0.8092418313026428, "num_tokens": 19791872.0, "step": 2802 }, { "epoch": 0.3704731694422416, "grad_norm": 0.8085380921631775, "learning_rate": 8.141168792461894e-06, "loss": 0.5924, "mean_token_accuracy": 0.8041142821311951, "num_tokens": 19857408.0, "step": 2803 }, { "epoch": 0.37060533967750464, "grad_norm": 0.9533609779884369, "learning_rate": 8.139487498035724e-06, "loss": 0.64, "mean_token_accuracy": 0.7931876182556152, "num_tokens": 19922944.0, "step": 2804 }, { "epoch": 0.37073750991276766, "grad_norm": 0.870456247504486, "learning_rate": 8.137805641717772e-06, "loss": 0.6016, "mean_token_accuracy": 0.8036259412765503, "num_tokens": 19988480.0, "step": 2805 }, { "epoch": 0.3708696801480307, "grad_norm": 0.9214708546276605, "learning_rate": 8.136123223866074e-06, "loss": 0.6007, "mean_token_accuracy": 0.802405059337616, "num_tokens": 20054016.0, "step": 2806 }, { "epoch": 0.3710018503832937, "grad_norm": 0.9188107596257974, "learning_rate": 8.134440244838779e-06, "loss": 0.6922, "mean_token_accuracy": 0.774707019329071, "num_tokens": 20119552.0, "step": 2807 }, { "epoch": 0.3711340206185567, "grad_norm": 0.9739351746028901, "learning_rate": 8.132756704994157e-06, "loss": 0.6625, "mean_token_accuracy": 0.7835429310798645, "num_tokens": 20185088.0, "step": 2808 }, { "epoch": 0.37126619085381973, "grad_norm": 0.8995921657053408, "learning_rate": 8.131072604690605e-06, "loss": 0.6068, "mean_token_accuracy": 0.8028628826141357, "num_tokens": 20250624.0, "step": 2809 }, { "epoch": 0.37139836108908275, "grad_norm": 0.870850351000316, "learning_rate": 8.129387944286626e-06, "loss": 0.6723, "mean_token_accuracy": 0.7807807326316833, "num_tokens": 20316160.0, "step": 2810 }, { "epoch": 0.37153053132434577, "grad_norm": 0.8876113902333307, "learning_rate": 8.127702724140853e-06, "loss": 0.6372, "mean_token_accuracy": 0.7918905019760132, "num_tokens": 20381696.0, "step": 2811 }, { "epoch": 0.3716627015596088, "grad_norm": 0.8644583701895683, "learning_rate": 8.126016944612029e-06, "loss": 0.5769, "mean_token_accuracy": 0.8098217844963074, "num_tokens": 20447232.0, "step": 2812 }, { "epoch": 0.3717948717948718, "grad_norm": 0.8867141652252141, "learning_rate": 8.124330606059028e-06, "loss": 0.5842, "mean_token_accuracy": 0.8046941757202148, "num_tokens": 20512768.0, "step": 2813 }, { "epoch": 0.3719270420301348, "grad_norm": 0.7973330472229624, "learning_rate": 8.122643708840833e-06, "loss": 0.5517, "mean_token_accuracy": 0.8197717070579529, "num_tokens": 20578304.0, "step": 2814 }, { "epoch": 0.37205921226539784, "grad_norm": 0.7938514351981222, "learning_rate": 8.120956253316547e-06, "loss": 0.5635, "mean_token_accuracy": 0.8134080171585083, "num_tokens": 20643840.0, "step": 2815 }, { "epoch": 0.37219138250066086, "grad_norm": 0.9442670195353868, "learning_rate": 8.119268239845396e-06, "loss": 0.6424, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 20709376.0, "step": 2816 }, { "epoch": 0.3723235527359239, "grad_norm": 0.9392797565980398, "learning_rate": 8.117579668786726e-06, "loss": 0.6384, "mean_token_accuracy": 0.7900439500808716, "num_tokens": 20774912.0, "step": 2817 }, { "epoch": 0.3724557229711869, "grad_norm": 0.8843782813812272, "learning_rate": 8.115890540499991e-06, "loss": 0.582, "mean_token_accuracy": 0.8082804083824158, "num_tokens": 20840448.0, "step": 2818 }, { "epoch": 0.3725878932064499, "grad_norm": 0.8984281869013508, "learning_rate": 8.114200855344781e-06, "loss": 0.5872, "mean_token_accuracy": 0.8063117861747742, "num_tokens": 20905984.0, "step": 2819 }, { "epoch": 0.37272006344171293, "grad_norm": 0.9289796143094232, "learning_rate": 8.112510613680788e-06, "loss": 0.588, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 20971520.0, "step": 2820 }, { "epoch": 0.37285223367697595, "grad_norm": 0.9629164892855089, "learning_rate": 8.11081981586783e-06, "loss": 0.6388, "mean_token_accuracy": 0.7915852665901184, "num_tokens": 21037056.0, "step": 2821 }, { "epoch": 0.37298440391223897, "grad_norm": 0.9192220389483864, "learning_rate": 8.109128462265842e-06, "loss": 0.6528, "mean_token_accuracy": 0.7851452827453613, "num_tokens": 21102592.0, "step": 2822 }, { "epoch": 0.373116574147502, "grad_norm": 0.891821490456738, "learning_rate": 8.10743655323488e-06, "loss": 0.6076, "mean_token_accuracy": 0.8025119304656982, "num_tokens": 21168128.0, "step": 2823 }, { "epoch": 0.373248744382765, "grad_norm": 0.8550678037461128, "learning_rate": 8.105744089135116e-06, "loss": 0.579, "mean_token_accuracy": 0.8073037266731262, "num_tokens": 21233664.0, "step": 2824 }, { "epoch": 0.373380914618028, "grad_norm": 0.8957616797811808, "learning_rate": 8.10405107032684e-06, "loss": 0.5727, "mean_token_accuracy": 0.811881959438324, "num_tokens": 21299200.0, "step": 2825 }, { "epoch": 0.37351308485329104, "grad_norm": 0.9124548915054164, "learning_rate": 8.102357497170457e-06, "loss": 0.6107, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 21364736.0, "step": 2826 }, { "epoch": 0.37364525508855406, "grad_norm": 0.8666357792257565, "learning_rate": 8.1006633700265e-06, "loss": 0.6344, "mean_token_accuracy": 0.7942711710929871, "num_tokens": 21430272.0, "step": 2827 }, { "epoch": 0.3737774253238171, "grad_norm": 0.8889518354266615, "learning_rate": 8.098968689255606e-06, "loss": 0.6274, "mean_token_accuracy": 0.7928824424743652, "num_tokens": 21495808.0, "step": 2828 }, { "epoch": 0.3739095955590801, "grad_norm": 1.3011415603734071, "learning_rate": 8.097273455218542e-06, "loss": 0.6712, "mean_token_accuracy": 0.7789799571037292, "num_tokens": 21561344.0, "step": 2829 }, { "epoch": 0.3740417657943431, "grad_norm": 0.8702757514865377, "learning_rate": 8.095577668276186e-06, "loss": 0.5736, "mean_token_accuracy": 0.8108442425727844, "num_tokens": 21626880.0, "step": 2830 }, { "epoch": 0.37417393602960614, "grad_norm": 0.8207807031029547, "learning_rate": 8.093881328789539e-06, "loss": 0.5563, "mean_token_accuracy": 0.8198632597923279, "num_tokens": 21692416.0, "step": 2831 }, { "epoch": 0.37430610626486915, "grad_norm": 0.9660320155918214, "learning_rate": 8.09218443711971e-06, "loss": 0.6331, "mean_token_accuracy": 0.7946984767913818, "num_tokens": 21757952.0, "step": 2832 }, { "epoch": 0.37443827650013217, "grad_norm": 0.9059442860431556, "learning_rate": 8.090486993627936e-06, "loss": 0.6234, "mean_token_accuracy": 0.7966212630271912, "num_tokens": 21823488.0, "step": 2833 }, { "epoch": 0.3745704467353952, "grad_norm": 0.8855626284354287, "learning_rate": 8.088788998675567e-06, "loss": 0.6281, "mean_token_accuracy": 0.7938286066055298, "num_tokens": 21889024.0, "step": 2834 }, { "epoch": 0.3747026169706582, "grad_norm": 0.8599482266646633, "learning_rate": 8.087090452624072e-06, "loss": 0.6182, "mean_token_accuracy": 0.7990171909332275, "num_tokens": 21954560.0, "step": 2835 }, { "epoch": 0.3748347872059212, "grad_norm": 0.9582642141086859, "learning_rate": 8.08539135583503e-06, "loss": 0.6567, "mean_token_accuracy": 0.7874649167060852, "num_tokens": 22020096.0, "step": 2836 }, { "epoch": 0.37496695744118425, "grad_norm": 0.9821319077493987, "learning_rate": 8.08369170867015e-06, "loss": 0.6459, "mean_token_accuracy": 0.7912953495979309, "num_tokens": 22085632.0, "step": 2837 }, { "epoch": 0.37509912767644726, "grad_norm": 0.9680271065387388, "learning_rate": 8.081991511491249e-06, "loss": 0.5858, "mean_token_accuracy": 0.8099896311759949, "num_tokens": 22151168.0, "step": 2838 }, { "epoch": 0.3752312979117103, "grad_norm": 0.7788829721223114, "learning_rate": 8.080290764660263e-06, "loss": 0.5133, "mean_token_accuracy": 0.8328195810317993, "num_tokens": 22216704.0, "step": 2839 }, { "epoch": 0.3753634681469733, "grad_norm": 0.7937702174709167, "learning_rate": 8.078589468539245e-06, "loss": 0.5458, "mean_token_accuracy": 0.8216487765312195, "num_tokens": 22282240.0, "step": 2840 }, { "epoch": 0.3754956383822363, "grad_norm": 1.019047913241933, "learning_rate": 8.076887623490366e-06, "loss": 0.6604, "mean_token_accuracy": 0.7847484946250916, "num_tokens": 22347776.0, "step": 2841 }, { "epoch": 0.37562780861749934, "grad_norm": 0.9226357250859669, "learning_rate": 8.075185229875915e-06, "loss": 0.6458, "mean_token_accuracy": 0.7875106930732727, "num_tokens": 22413312.0, "step": 2842 }, { "epoch": 0.37575997885276236, "grad_norm": 0.9095021760694423, "learning_rate": 8.073482288058294e-06, "loss": 0.636, "mean_token_accuracy": 0.7899371385574341, "num_tokens": 22478848.0, "step": 2843 }, { "epoch": 0.3758921490880254, "grad_norm": 0.8898984686630966, "learning_rate": 8.071778798400025e-06, "loss": 0.5589, "mean_token_accuracy": 0.8187645077705383, "num_tokens": 22544384.0, "step": 2844 }, { "epoch": 0.3760243193232884, "grad_norm": 0.8680845694552982, "learning_rate": 8.070074761263743e-06, "loss": 0.6153, "mean_token_accuracy": 0.7992613911628723, "num_tokens": 22609920.0, "step": 2845 }, { "epoch": 0.3761564895585514, "grad_norm": 0.8965603087993017, "learning_rate": 8.068370177012205e-06, "loss": 0.5994, "mean_token_accuracy": 0.8044347167015076, "num_tokens": 22675456.0, "step": 2846 }, { "epoch": 0.37628865979381443, "grad_norm": 0.9270490713516766, "learning_rate": 8.06666504600828e-06, "loss": 0.6593, "mean_token_accuracy": 0.7839549779891968, "num_tokens": 22740992.0, "step": 2847 }, { "epoch": 0.37642083002907745, "grad_norm": 0.8771066994306327, "learning_rate": 8.064959368614953e-06, "loss": 0.5672, "mean_token_accuracy": 0.8135453462600708, "num_tokens": 22806528.0, "step": 2848 }, { "epoch": 0.37655300026434047, "grad_norm": 0.9932573331935508, "learning_rate": 8.063253145195331e-06, "loss": 0.6769, "mean_token_accuracy": 0.7763398885726929, "num_tokens": 22872064.0, "step": 2849 }, { "epoch": 0.3766851704996035, "grad_norm": 0.9673049377886148, "learning_rate": 8.06154637611263e-06, "loss": 0.6627, "mean_token_accuracy": 0.7821999788284302, "num_tokens": 22937600.0, "step": 2850 }, { "epoch": 0.3768173407348665, "grad_norm": 0.81883599464271, "learning_rate": 8.059839061730188e-06, "loss": 0.5726, "mean_token_accuracy": 0.8149035573005676, "num_tokens": 23003136.0, "step": 2851 }, { "epoch": 0.3769495109701295, "grad_norm": 0.8528851156882021, "learning_rate": 8.058131202411455e-06, "loss": 0.5692, "mean_token_accuracy": 0.8138048052787781, "num_tokens": 23068672.0, "step": 2852 }, { "epoch": 0.37708168120539254, "grad_norm": 0.922669806465549, "learning_rate": 8.056422798519999e-06, "loss": 0.5901, "mean_token_accuracy": 0.8054572343826294, "num_tokens": 23134208.0, "step": 2853 }, { "epoch": 0.37721385144065556, "grad_norm": 0.8154588270330195, "learning_rate": 8.054713850419503e-06, "loss": 0.6171, "mean_token_accuracy": 0.7969570159912109, "num_tokens": 23199744.0, "step": 2854 }, { "epoch": 0.3773460216759186, "grad_norm": 0.9064713531146145, "learning_rate": 8.053004358473767e-06, "loss": 0.5945, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 23265280.0, "step": 2855 }, { "epoch": 0.3774781919111816, "grad_norm": 0.8807488640094703, "learning_rate": 8.051294323046705e-06, "loss": 0.5795, "mean_token_accuracy": 0.8095012903213501, "num_tokens": 23330816.0, "step": 2856 }, { "epoch": 0.3776103621464446, "grad_norm": 0.834831124594728, "learning_rate": 8.04958374450235e-06, "loss": 0.5351, "mean_token_accuracy": 0.826059103012085, "num_tokens": 23396352.0, "step": 2857 }, { "epoch": 0.37774253238170763, "grad_norm": 0.9567907100495269, "learning_rate": 8.047872623204846e-06, "loss": 0.6347, "mean_token_accuracy": 0.7939506769180298, "num_tokens": 23461888.0, "step": 2858 }, { "epoch": 0.37787470261697065, "grad_norm": 0.8922486329141841, "learning_rate": 8.046160959518458e-06, "loss": 0.5817, "mean_token_accuracy": 0.8118361830711365, "num_tokens": 23527424.0, "step": 2859 }, { "epoch": 0.37800687285223367, "grad_norm": 0.938589565798753, "learning_rate": 8.044448753807563e-06, "loss": 0.6597, "mean_token_accuracy": 0.7818489670753479, "num_tokens": 23592960.0, "step": 2860 }, { "epoch": 0.3781390430874967, "grad_norm": 0.8700248845270918, "learning_rate": 8.04273600643665e-06, "loss": 0.5787, "mean_token_accuracy": 0.8088451027870178, "num_tokens": 23658496.0, "step": 2861 }, { "epoch": 0.3782712133227597, "grad_norm": 1.010312600154657, "learning_rate": 8.041022717770329e-06, "loss": 0.6629, "mean_token_accuracy": 0.78265780210495, "num_tokens": 23724032.0, "step": 2862 }, { "epoch": 0.3784033835580227, "grad_norm": 1.0127830529622948, "learning_rate": 8.039308888173325e-06, "loss": 0.6589, "mean_token_accuracy": 0.783802330493927, "num_tokens": 23789568.0, "step": 2863 }, { "epoch": 0.37853555379328574, "grad_norm": 0.9028362553308975, "learning_rate": 8.037594518010476e-06, "loss": 0.6058, "mean_token_accuracy": 0.8010163307189941, "num_tokens": 23855104.0, "step": 2864 }, { "epoch": 0.37866772402854876, "grad_norm": 0.8562711419813985, "learning_rate": 8.035879607646737e-06, "loss": 0.6137, "mean_token_accuracy": 0.7988646030426025, "num_tokens": 23920640.0, "step": 2865 }, { "epoch": 0.3787998942638118, "grad_norm": 0.7988766265064906, "learning_rate": 8.034164157447172e-06, "loss": 0.5611, "mean_token_accuracy": 0.8151019215583801, "num_tokens": 23986176.0, "step": 2866 }, { "epoch": 0.3789320644990748, "grad_norm": 0.8559330197900754, "learning_rate": 8.03244816777697e-06, "loss": 0.6069, "mean_token_accuracy": 0.8008484840393066, "num_tokens": 24051712.0, "step": 2867 }, { "epoch": 0.3790642347343378, "grad_norm": 0.8433133738520864, "learning_rate": 8.030731639001424e-06, "loss": 0.6354, "mean_token_accuracy": 0.7930197715759277, "num_tokens": 24117248.0, "step": 2868 }, { "epoch": 0.37919640496960083, "grad_norm": 0.8277011941292672, "learning_rate": 8.02901457148595e-06, "loss": 0.595, "mean_token_accuracy": 0.8041142821311951, "num_tokens": 24182784.0, "step": 2869 }, { "epoch": 0.37932857520486385, "grad_norm": 0.8872814775109126, "learning_rate": 8.027296965596078e-06, "loss": 0.6062, "mean_token_accuracy": 0.8015962839126587, "num_tokens": 24248320.0, "step": 2870 }, { "epoch": 0.37946074544012687, "grad_norm": 0.8937250052733929, "learning_rate": 8.025578821697448e-06, "loss": 0.6104, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 24313856.0, "step": 2871 }, { "epoch": 0.3795929156753899, "grad_norm": 0.9212671222278358, "learning_rate": 8.023860140155816e-06, "loss": 0.6045, "mean_token_accuracy": 0.7984220385551453, "num_tokens": 24379392.0, "step": 2872 }, { "epoch": 0.3797250859106529, "grad_norm": 0.853819588310984, "learning_rate": 8.022140921337056e-06, "loss": 0.6384, "mean_token_accuracy": 0.7911274433135986, "num_tokens": 24444928.0, "step": 2873 }, { "epoch": 0.3798572561459159, "grad_norm": 0.837444676209231, "learning_rate": 8.020421165607151e-06, "loss": 0.5849, "mean_token_accuracy": 0.8102185130119324, "num_tokens": 24510464.0, "step": 2874 }, { "epoch": 0.37998942638117894, "grad_norm": 0.8447105736853547, "learning_rate": 8.018700873332202e-06, "loss": 0.6228, "mean_token_accuracy": 0.7965907454490662, "num_tokens": 24576000.0, "step": 2875 }, { "epoch": 0.38012159661644196, "grad_norm": 0.910982577430521, "learning_rate": 8.016980044878425e-06, "loss": 0.6259, "mean_token_accuracy": 0.7918142080307007, "num_tokens": 24641536.0, "step": 2876 }, { "epoch": 0.380253766851705, "grad_norm": 0.8954998736353497, "learning_rate": 8.015258680612147e-06, "loss": 0.6261, "mean_token_accuracy": 0.7966365814208984, "num_tokens": 24707072.0, "step": 2877 }, { "epoch": 0.380385937086968, "grad_norm": 0.83935342209176, "learning_rate": 8.01353678089981e-06, "loss": 0.5727, "mean_token_accuracy": 0.8124008178710938, "num_tokens": 24772608.0, "step": 2878 }, { "epoch": 0.380518107322231, "grad_norm": 0.8686074660018699, "learning_rate": 8.011814346107972e-06, "loss": 0.6167, "mean_token_accuracy": 0.7965449690818787, "num_tokens": 24838144.0, "step": 2879 }, { "epoch": 0.38065027755749403, "grad_norm": 0.8588067972746803, "learning_rate": 8.010091376603303e-06, "loss": 0.6235, "mean_token_accuracy": 0.7977811098098755, "num_tokens": 24903680.0, "step": 2880 }, { "epoch": 0.38078244779275705, "grad_norm": 0.87918488548783, "learning_rate": 8.008367872752585e-06, "loss": 0.6027, "mean_token_accuracy": 0.8041447997093201, "num_tokens": 24969216.0, "step": 2881 }, { "epoch": 0.38091461802802007, "grad_norm": 0.9223807855844524, "learning_rate": 8.006643834922719e-06, "loss": 0.6287, "mean_token_accuracy": 0.7921041250228882, "num_tokens": 25034752.0, "step": 2882 }, { "epoch": 0.3810467882632831, "grad_norm": 0.8718909423434328, "learning_rate": 8.004919263480715e-06, "loss": 0.6072, "mean_token_accuracy": 0.8012757897377014, "num_tokens": 25100288.0, "step": 2883 }, { "epoch": 0.3811789584985461, "grad_norm": 0.8628237657202974, "learning_rate": 8.0031941587937e-06, "loss": 0.5899, "mean_token_accuracy": 0.8074105978012085, "num_tokens": 25165824.0, "step": 2884 }, { "epoch": 0.3813111287338091, "grad_norm": 0.9931679183631902, "learning_rate": 8.001468521228912e-06, "loss": 0.6351, "mean_token_accuracy": 0.7926382422447205, "num_tokens": 25231360.0, "step": 2885 }, { "epoch": 0.38144329896907214, "grad_norm": 0.8951506758457012, "learning_rate": 7.999742351153702e-06, "loss": 0.6102, "mean_token_accuracy": 0.7985135912895203, "num_tokens": 25296896.0, "step": 2886 }, { "epoch": 0.38157546920433516, "grad_norm": 0.9022262634066813, "learning_rate": 7.998015648935534e-06, "loss": 0.5961, "mean_token_accuracy": 0.803122341632843, "num_tokens": 25362432.0, "step": 2887 }, { "epoch": 0.3817076394395982, "grad_norm": 0.9168039601767622, "learning_rate": 7.996288414941993e-06, "loss": 0.6183, "mean_token_accuracy": 0.7963313460350037, "num_tokens": 25427968.0, "step": 2888 }, { "epoch": 0.3818398096748612, "grad_norm": 0.886624478121206, "learning_rate": 7.994560649540763e-06, "loss": 0.5622, "mean_token_accuracy": 0.8137590289115906, "num_tokens": 25493504.0, "step": 2889 }, { "epoch": 0.3819719799101242, "grad_norm": 0.8790396493590293, "learning_rate": 7.992832353099655e-06, "loss": 0.5869, "mean_token_accuracy": 0.8071816563606262, "num_tokens": 25559040.0, "step": 2890 }, { "epoch": 0.38210415014538723, "grad_norm": 1.0909120682492317, "learning_rate": 7.991103525986588e-06, "loss": 0.6649, "mean_token_accuracy": 0.7818184494972229, "num_tokens": 25624576.0, "step": 2891 }, { "epoch": 0.38223632038065025, "grad_norm": 0.9483967704887075, "learning_rate": 7.989374168569588e-06, "loss": 0.6069, "mean_token_accuracy": 0.8011079430580139, "num_tokens": 25690112.0, "step": 2892 }, { "epoch": 0.38236849061591327, "grad_norm": 0.9113569531886689, "learning_rate": 7.987644281216804e-06, "loss": 0.6002, "mean_token_accuracy": 0.8051825165748596, "num_tokens": 25755648.0, "step": 2893 }, { "epoch": 0.3825006608511763, "grad_norm": 0.8599876472142376, "learning_rate": 7.98591386429649e-06, "loss": 0.6432, "mean_token_accuracy": 0.7874496579170227, "num_tokens": 25821184.0, "step": 2894 }, { "epoch": 0.3826328310864393, "grad_norm": 0.9302295031804013, "learning_rate": 7.984182918177016e-06, "loss": 0.609, "mean_token_accuracy": 0.7970638275146484, "num_tokens": 25886720.0, "step": 2895 }, { "epoch": 0.3827650013217023, "grad_norm": 0.961916955087518, "learning_rate": 7.982451443226865e-06, "loss": 0.6271, "mean_token_accuracy": 0.7948358058929443, "num_tokens": 25952256.0, "step": 2896 }, { "epoch": 0.3828971715569654, "grad_norm": 0.8624926643661815, "learning_rate": 7.980719439814631e-06, "loss": 0.5605, "mean_token_accuracy": 0.8169637322425842, "num_tokens": 26017792.0, "step": 2897 }, { "epoch": 0.3830293417922284, "grad_norm": 0.8504070673360725, "learning_rate": 7.978986908309023e-06, "loss": 0.592, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 26083328.0, "step": 2898 }, { "epoch": 0.38316151202749144, "grad_norm": 0.9530565841850653, "learning_rate": 7.97725384907886e-06, "loss": 0.6266, "mean_token_accuracy": 0.79337078332901, "num_tokens": 26148864.0, "step": 2899 }, { "epoch": 0.38329368226275445, "grad_norm": 0.9093163071278807, "learning_rate": 7.975520262493072e-06, "loss": 0.6501, "mean_token_accuracy": 0.7833139896392822, "num_tokens": 26214400.0, "step": 2900 }, { "epoch": 0.38342585249801747, "grad_norm": 1.14235278903647, "learning_rate": 7.973786148920706e-06, "loss": 0.6219, "mean_token_accuracy": 0.79783695936203, "num_tokens": 26272752.0, "step": 2901 }, { "epoch": 0.3835580227332805, "grad_norm": 0.9672822334468254, "learning_rate": 7.97205150873092e-06, "loss": 0.6323, "mean_token_accuracy": 0.7911884784698486, "num_tokens": 26338288.0, "step": 2902 }, { "epoch": 0.3836901929685435, "grad_norm": 0.8867302833493254, "learning_rate": 7.970316342292976e-06, "loss": 0.562, "mean_token_accuracy": 0.8157886862754822, "num_tokens": 26403824.0, "step": 2903 }, { "epoch": 0.3838223632038065, "grad_norm": 1.0007166383367736, "learning_rate": 7.968580649976264e-06, "loss": 0.6725, "mean_token_accuracy": 0.7811317443847656, "num_tokens": 26469360.0, "step": 2904 }, { "epoch": 0.38395453343906955, "grad_norm": 0.9266985580979519, "learning_rate": 7.966844432150269e-06, "loss": 0.6043, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 26534896.0, "step": 2905 }, { "epoch": 0.38408670367433256, "grad_norm": 0.9460415246943277, "learning_rate": 7.965107689184599e-06, "loss": 0.5985, "mean_token_accuracy": 0.8059608340263367, "num_tokens": 26600432.0, "step": 2906 }, { "epoch": 0.3842188739095956, "grad_norm": 0.8542461222297911, "learning_rate": 7.96337042144897e-06, "loss": 0.578, "mean_token_accuracy": 0.8113783597946167, "num_tokens": 26665968.0, "step": 2907 }, { "epoch": 0.3843510441448586, "grad_norm": 1.117707539019936, "learning_rate": 7.961632629313209e-06, "loss": 0.6586, "mean_token_accuracy": 0.7854047417640686, "num_tokens": 26731504.0, "step": 2908 }, { "epoch": 0.3844832143801216, "grad_norm": 1.043581257259945, "learning_rate": 7.959894313147256e-06, "loss": 0.6275, "mean_token_accuracy": 0.7919362783432007, "num_tokens": 26797040.0, "step": 2909 }, { "epoch": 0.38461538461538464, "grad_norm": 0.9403573214274736, "learning_rate": 7.958155473321164e-06, "loss": 0.6137, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 26862576.0, "step": 2910 }, { "epoch": 0.38474755485064766, "grad_norm": 0.9748429498759764, "learning_rate": 7.956416110205095e-06, "loss": 0.5876, "mean_token_accuracy": 0.8068153858184814, "num_tokens": 26928112.0, "step": 2911 }, { "epoch": 0.3848797250859107, "grad_norm": 0.870012903298413, "learning_rate": 7.95467622416932e-06, "loss": 0.563, "mean_token_accuracy": 0.8136368989944458, "num_tokens": 26993648.0, "step": 2912 }, { "epoch": 0.3850118953211737, "grad_norm": 1.0061825016430268, "learning_rate": 7.952935815584229e-06, "loss": 0.6452, "mean_token_accuracy": 0.7873886227607727, "num_tokens": 27059184.0, "step": 2913 }, { "epoch": 0.3851440655564367, "grad_norm": 1.0759562745125641, "learning_rate": 7.951194884820315e-06, "loss": 0.6316, "mean_token_accuracy": 0.7919057607650757, "num_tokens": 27124720.0, "step": 2914 }, { "epoch": 0.38527623579169973, "grad_norm": 0.8215151518941646, "learning_rate": 7.949453432248188e-06, "loss": 0.5817, "mean_token_accuracy": 0.8093639612197876, "num_tokens": 27190256.0, "step": 2915 }, { "epoch": 0.38540840602696275, "grad_norm": 0.9660228774695009, "learning_rate": 7.947711458238567e-06, "loss": 0.6387, "mean_token_accuracy": 0.7904254794120789, "num_tokens": 27255792.0, "step": 2916 }, { "epoch": 0.38554057626222576, "grad_norm": 0.9217629677357678, "learning_rate": 7.94596896316228e-06, "loss": 0.5977, "mean_token_accuracy": 0.802130401134491, "num_tokens": 27321328.0, "step": 2917 }, { "epoch": 0.3856727464974888, "grad_norm": 0.8970272489466931, "learning_rate": 7.944225947390269e-06, "loss": 0.5649, "mean_token_accuracy": 0.8148272633552551, "num_tokens": 27386864.0, "step": 2918 }, { "epoch": 0.3858049167327518, "grad_norm": 0.9041048670777124, "learning_rate": 7.942482411293585e-06, "loss": 0.6144, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 27452400.0, "step": 2919 }, { "epoch": 0.3859370869680148, "grad_norm": 0.8671377350495207, "learning_rate": 7.940738355243393e-06, "loss": 0.5384, "mean_token_accuracy": 0.8232206106185913, "num_tokens": 27517936.0, "step": 2920 }, { "epoch": 0.38606925720327784, "grad_norm": 0.9597283665078521, "learning_rate": 7.938993779610962e-06, "loss": 0.6158, "mean_token_accuracy": 0.7996124029159546, "num_tokens": 27583472.0, "step": 2921 }, { "epoch": 0.38620142743854086, "grad_norm": 0.9778952124152084, "learning_rate": 7.937248684767678e-06, "loss": 0.6371, "mean_token_accuracy": 0.79324871301651, "num_tokens": 27649008.0, "step": 2922 }, { "epoch": 0.3863335976738039, "grad_norm": 0.9329894633451546, "learning_rate": 7.935503071085035e-06, "loss": 0.5833, "mean_token_accuracy": 0.8083414435386658, "num_tokens": 27714544.0, "step": 2923 }, { "epoch": 0.3864657679090669, "grad_norm": 0.921028408641885, "learning_rate": 7.93375693893464e-06, "loss": 0.6256, "mean_token_accuracy": 0.7951409816741943, "num_tokens": 27780080.0, "step": 2924 }, { "epoch": 0.3865979381443299, "grad_norm": 0.982336155788853, "learning_rate": 7.932010288688204e-06, "loss": 0.6634, "mean_token_accuracy": 0.7840312719345093, "num_tokens": 27845616.0, "step": 2925 }, { "epoch": 0.38673010837959293, "grad_norm": 1.0447039266774898, "learning_rate": 7.930263120717555e-06, "loss": 0.6533, "mean_token_accuracy": 0.7861677408218384, "num_tokens": 27911152.0, "step": 2926 }, { "epoch": 0.38686227861485595, "grad_norm": 0.8337595000274226, "learning_rate": 7.928515435394626e-06, "loss": 0.5403, "mean_token_accuracy": 0.821603000164032, "num_tokens": 27976688.0, "step": 2927 }, { "epoch": 0.38699444885011897, "grad_norm": 0.9090983646341378, "learning_rate": 7.926767233091466e-06, "loss": 0.5888, "mean_token_accuracy": 0.8090587258338928, "num_tokens": 28042224.0, "step": 2928 }, { "epoch": 0.387126619085382, "grad_norm": 0.8985381328172255, "learning_rate": 7.92501851418023e-06, "loss": 0.5912, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 28107760.0, "step": 2929 }, { "epoch": 0.387258789320645, "grad_norm": 0.8820412702602852, "learning_rate": 7.923269279033183e-06, "loss": 0.5318, "mean_token_accuracy": 0.8246245980262756, "num_tokens": 28173296.0, "step": 2930 }, { "epoch": 0.387390959555908, "grad_norm": 0.9403354416164231, "learning_rate": 7.9215195280227e-06, "loss": 0.6126, "mean_token_accuracy": 0.7974911332130432, "num_tokens": 28238832.0, "step": 2931 }, { "epoch": 0.38752312979117104, "grad_norm": 0.8753726227768726, "learning_rate": 7.919769261521267e-06, "loss": 0.6315, "mean_token_accuracy": 0.7924856543540955, "num_tokens": 28304368.0, "step": 2932 }, { "epoch": 0.38765530002643406, "grad_norm": 0.8918392190957719, "learning_rate": 7.918018479901479e-06, "loss": 0.6375, "mean_token_accuracy": 0.7933555245399475, "num_tokens": 28369904.0, "step": 2933 }, { "epoch": 0.3877874702616971, "grad_norm": 1.024468586007697, "learning_rate": 7.916267183536043e-06, "loss": 0.6202, "mean_token_accuracy": 0.7947442531585693, "num_tokens": 28435440.0, "step": 2934 }, { "epoch": 0.3879196404969601, "grad_norm": 0.9930796637884947, "learning_rate": 7.91451537279777e-06, "loss": 0.6322, "mean_token_accuracy": 0.7957361936569214, "num_tokens": 28500976.0, "step": 2935 }, { "epoch": 0.3880518107322231, "grad_norm": 0.9443873627457742, "learning_rate": 7.912763048059586e-06, "loss": 0.6302, "mean_token_accuracy": 0.7962245345115662, "num_tokens": 28566512.0, "step": 2936 }, { "epoch": 0.38818398096748613, "grad_norm": 0.9698840426143011, "learning_rate": 7.911010209694524e-06, "loss": 0.6248, "mean_token_accuracy": 0.7964991927146912, "num_tokens": 28632048.0, "step": 2937 }, { "epoch": 0.38831615120274915, "grad_norm": 1.0315196777892017, "learning_rate": 7.909256858075728e-06, "loss": 0.7105, "mean_token_accuracy": 0.7700982689857483, "num_tokens": 28697584.0, "step": 2938 }, { "epoch": 0.38844832143801217, "grad_norm": 1.0421254448076767, "learning_rate": 7.907502993576446e-06, "loss": 0.6259, "mean_token_accuracy": 0.7955835461616516, "num_tokens": 28763120.0, "step": 2939 }, { "epoch": 0.3885804916732752, "grad_norm": 0.9892062104983648, "learning_rate": 7.905748616570044e-06, "loss": 0.6342, "mean_token_accuracy": 0.7931418418884277, "num_tokens": 28828656.0, "step": 2940 }, { "epoch": 0.3887126619085382, "grad_norm": 0.8277139274873626, "learning_rate": 7.90399372742999e-06, "loss": 0.5353, "mean_token_accuracy": 0.8256165385246277, "num_tokens": 28894192.0, "step": 2941 }, { "epoch": 0.3888448321438012, "grad_norm": 1.059805196656546, "learning_rate": 7.90223832652986e-06, "loss": 0.6355, "mean_token_accuracy": 0.7915700078010559, "num_tokens": 28959728.0, "step": 2942 }, { "epoch": 0.38897700237906424, "grad_norm": 0.8777861309057841, "learning_rate": 7.900482414243349e-06, "loss": 0.6094, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 29025264.0, "step": 2943 }, { "epoch": 0.38910917261432726, "grad_norm": 0.8755964206142584, "learning_rate": 7.898725990944248e-06, "loss": 0.5521, "mean_token_accuracy": 0.8222286701202393, "num_tokens": 29090800.0, "step": 2944 }, { "epoch": 0.3892413428495903, "grad_norm": 0.8314452892076328, "learning_rate": 7.896969057006467e-06, "loss": 0.5631, "mean_token_accuracy": 0.8156360387802124, "num_tokens": 29156336.0, "step": 2945 }, { "epoch": 0.3893735130848533, "grad_norm": 0.9450415100561133, "learning_rate": 7.895211612804018e-06, "loss": 0.653, "mean_token_accuracy": 0.7853283882141113, "num_tokens": 29221872.0, "step": 2946 }, { "epoch": 0.3895056833201163, "grad_norm": 0.9214154084059581, "learning_rate": 7.893453658711026e-06, "loss": 0.5929, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 29287408.0, "step": 2947 }, { "epoch": 0.38963785355537933, "grad_norm": 0.8678193942618766, "learning_rate": 7.891695195101719e-06, "loss": 0.5072, "mean_token_accuracy": 0.831476628780365, "num_tokens": 29352944.0, "step": 2948 }, { "epoch": 0.38977002379064235, "grad_norm": 0.9761743799743868, "learning_rate": 7.889936222350442e-06, "loss": 0.6859, "mean_token_accuracy": 0.7743865251541138, "num_tokens": 29418480.0, "step": 2949 }, { "epoch": 0.38990219402590537, "grad_norm": 0.8069772828042696, "learning_rate": 7.88817674083164e-06, "loss": 0.5731, "mean_token_accuracy": 0.8138810992240906, "num_tokens": 29484016.0, "step": 2950 }, { "epoch": 0.3900343642611684, "grad_norm": 0.8672513038310863, "learning_rate": 7.886416750919875e-06, "loss": 0.5721, "mean_token_accuracy": 0.8125991821289062, "num_tokens": 29549552.0, "step": 2951 }, { "epoch": 0.3901665344964314, "grad_norm": 0.8805350187233105, "learning_rate": 7.884656252989806e-06, "loss": 0.6058, "mean_token_accuracy": 0.8026797771453857, "num_tokens": 29615088.0, "step": 2952 }, { "epoch": 0.3902987047316944, "grad_norm": 1.020176066281352, "learning_rate": 7.88289524741621e-06, "loss": 0.66, "mean_token_accuracy": 0.784458577632904, "num_tokens": 29680624.0, "step": 2953 }, { "epoch": 0.39043087496695744, "grad_norm": 0.9268835666838879, "learning_rate": 7.881133734573967e-06, "loss": 0.596, "mean_token_accuracy": 0.8045721054077148, "num_tokens": 29746160.0, "step": 2954 }, { "epoch": 0.39056304520222046, "grad_norm": 0.8528313345518762, "learning_rate": 7.879371714838065e-06, "loss": 0.5784, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 29811696.0, "step": 2955 }, { "epoch": 0.3906952154374835, "grad_norm": 0.8822711225176308, "learning_rate": 7.877609188583604e-06, "loss": 0.6003, "mean_token_accuracy": 0.8015199899673462, "num_tokens": 29877232.0, "step": 2956 }, { "epoch": 0.3908273856727465, "grad_norm": 0.9402500556739801, "learning_rate": 7.875846156185787e-06, "loss": 0.633, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 29942768.0, "step": 2957 }, { "epoch": 0.3909595559080095, "grad_norm": 0.874173795228464, "learning_rate": 7.874082618019926e-06, "loss": 0.554, "mean_token_accuracy": 0.8177115321159363, "num_tokens": 30008304.0, "step": 2958 }, { "epoch": 0.39109172614327253, "grad_norm": 0.8891423808692848, "learning_rate": 7.872318574461445e-06, "loss": 0.5987, "mean_token_accuracy": 0.803396999835968, "num_tokens": 30073840.0, "step": 2959 }, { "epoch": 0.39122389637853555, "grad_norm": 0.9381731276539343, "learning_rate": 7.870554025885869e-06, "loss": 0.6034, "mean_token_accuracy": 0.8004516959190369, "num_tokens": 30139376.0, "step": 2960 }, { "epoch": 0.39135606661379857, "grad_norm": 0.885278699976976, "learning_rate": 7.868788972668835e-06, "loss": 0.6136, "mean_token_accuracy": 0.7975826859474182, "num_tokens": 30204912.0, "step": 2961 }, { "epoch": 0.3914882368490616, "grad_norm": 0.9307524820986034, "learning_rate": 7.867023415186087e-06, "loss": 0.6295, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 30270448.0, "step": 2962 }, { "epoch": 0.3916204070843246, "grad_norm": 0.8913912203823419, "learning_rate": 7.86525735381347e-06, "loss": 0.6147, "mean_token_accuracy": 0.7974300980567932, "num_tokens": 30335984.0, "step": 2963 }, { "epoch": 0.3917525773195876, "grad_norm": 0.954718899944762, "learning_rate": 7.863490788926949e-06, "loss": 0.6729, "mean_token_accuracy": 0.7809180617332458, "num_tokens": 30401520.0, "step": 2964 }, { "epoch": 0.39188474755485064, "grad_norm": 0.9755940260567065, "learning_rate": 7.861723720902583e-06, "loss": 0.6054, "mean_token_accuracy": 0.8028781414031982, "num_tokens": 30467056.0, "step": 2965 }, { "epoch": 0.39201691779011366, "grad_norm": 0.8808763926785996, "learning_rate": 7.859956150116546e-06, "loss": 0.6091, "mean_token_accuracy": 0.8029850125312805, "num_tokens": 30532592.0, "step": 2966 }, { "epoch": 0.3921490880253767, "grad_norm": 0.9591214447758912, "learning_rate": 7.858188076945118e-06, "loss": 0.5345, "mean_token_accuracy": 0.8238767981529236, "num_tokens": 30598128.0, "step": 2967 }, { "epoch": 0.3922812582606397, "grad_norm": 0.8934182693999211, "learning_rate": 7.856419501764685e-06, "loss": 0.6012, "mean_token_accuracy": 0.8020846247673035, "num_tokens": 30663664.0, "step": 2968 }, { "epoch": 0.3924134284959027, "grad_norm": 0.9301590085878119, "learning_rate": 7.854650424951735e-06, "loss": 0.6005, "mean_token_accuracy": 0.8019930124282837, "num_tokens": 30729200.0, "step": 2969 }, { "epoch": 0.39254559873116573, "grad_norm": 0.8607644620308609, "learning_rate": 7.852880846882874e-06, "loss": 0.5554, "mean_token_accuracy": 0.8196801543235779, "num_tokens": 30794736.0, "step": 2970 }, { "epoch": 0.39267776896642875, "grad_norm": 0.8111933090732176, "learning_rate": 7.851110767934802e-06, "loss": 0.5473, "mean_token_accuracy": 0.8191612958908081, "num_tokens": 30860272.0, "step": 2971 }, { "epoch": 0.39280993920169177, "grad_norm": 0.8824785998485195, "learning_rate": 7.84934018848434e-06, "loss": 0.5583, "mean_token_accuracy": 0.8159717917442322, "num_tokens": 30925808.0, "step": 2972 }, { "epoch": 0.3929421094369548, "grad_norm": 0.8214606856197599, "learning_rate": 7.847569108908399e-06, "loss": 0.5523, "mean_token_accuracy": 0.8161091208457947, "num_tokens": 30991344.0, "step": 2973 }, { "epoch": 0.3930742796722178, "grad_norm": 0.9136932242339428, "learning_rate": 7.84579752958401e-06, "loss": 0.6235, "mean_token_accuracy": 0.797796368598938, "num_tokens": 31056880.0, "step": 2974 }, { "epoch": 0.3932064499074808, "grad_norm": 0.8512302197077857, "learning_rate": 7.844025450888304e-06, "loss": 0.5776, "mean_token_accuracy": 0.8113631010055542, "num_tokens": 31122416.0, "step": 2975 }, { "epoch": 0.39333862014274384, "grad_norm": 0.9376851622256952, "learning_rate": 7.842252873198518e-06, "loss": 0.638, "mean_token_accuracy": 0.79324871301651, "num_tokens": 31187952.0, "step": 2976 }, { "epoch": 0.39347079037800686, "grad_norm": 0.8422251316811186, "learning_rate": 7.840479796892e-06, "loss": 0.586, "mean_token_accuracy": 0.8094555139541626, "num_tokens": 31253488.0, "step": 2977 }, { "epoch": 0.3936029606132699, "grad_norm": 0.9251001311885142, "learning_rate": 7.838706222346196e-06, "loss": 0.6033, "mean_token_accuracy": 0.8016725778579712, "num_tokens": 31319024.0, "step": 2978 }, { "epoch": 0.3937351308485329, "grad_norm": 0.9636263921974588, "learning_rate": 7.836932149938667e-06, "loss": 0.6331, "mean_token_accuracy": 0.7928061485290527, "num_tokens": 31384560.0, "step": 2979 }, { "epoch": 0.3938673010837959, "grad_norm": 0.9023039918535273, "learning_rate": 7.835157580047076e-06, "loss": 0.6268, "mean_token_accuracy": 0.7918447256088257, "num_tokens": 31450096.0, "step": 2980 }, { "epoch": 0.39399947131905894, "grad_norm": 0.8802445056330208, "learning_rate": 7.833382513049192e-06, "loss": 0.6036, "mean_token_accuracy": 0.8026950359344482, "num_tokens": 31515632.0, "step": 2981 }, { "epoch": 0.39413164155432195, "grad_norm": 0.8410283197360049, "learning_rate": 7.831606949322885e-06, "loss": 0.5914, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 31581168.0, "step": 2982 }, { "epoch": 0.394263811789585, "grad_norm": 0.9434277987000117, "learning_rate": 7.82983088924614e-06, "loss": 0.6229, "mean_token_accuracy": 0.7953394055366516, "num_tokens": 31646704.0, "step": 2983 }, { "epoch": 0.394395982024848, "grad_norm": 0.9127365103232714, "learning_rate": 7.828054333197043e-06, "loss": 0.5399, "mean_token_accuracy": 0.8229764103889465, "num_tokens": 31712240.0, "step": 2984 }, { "epoch": 0.394528152260111, "grad_norm": 0.8734556460138027, "learning_rate": 7.826277281553784e-06, "loss": 0.5825, "mean_token_accuracy": 0.808478832244873, "num_tokens": 31777776.0, "step": 2985 }, { "epoch": 0.394660322495374, "grad_norm": 0.9222573577221135, "learning_rate": 7.824499734694662e-06, "loss": 0.5845, "mean_token_accuracy": 0.8080057501792908, "num_tokens": 31843312.0, "step": 2986 }, { "epoch": 0.39479249273063705, "grad_norm": 1.0482630655580363, "learning_rate": 7.822721692998078e-06, "loss": 0.6347, "mean_token_accuracy": 0.7908527851104736, "num_tokens": 31908848.0, "step": 2987 }, { "epoch": 0.39492466296590006, "grad_norm": 0.9512499520603813, "learning_rate": 7.820943156842539e-06, "loss": 0.6654, "mean_token_accuracy": 0.7818031907081604, "num_tokens": 31974384.0, "step": 2988 }, { "epoch": 0.3950568332011631, "grad_norm": 0.8652316713040432, "learning_rate": 7.819164126606661e-06, "loss": 0.6084, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 32039920.0, "step": 2989 }, { "epoch": 0.3951890034364261, "grad_norm": 0.8657600682387148, "learning_rate": 7.817384602669162e-06, "loss": 0.6041, "mean_token_accuracy": 0.803244411945343, "num_tokens": 32105456.0, "step": 2990 }, { "epoch": 0.3953211736716891, "grad_norm": 0.83407503334168, "learning_rate": 7.815604585408864e-06, "loss": 0.6007, "mean_token_accuracy": 0.8025271892547607, "num_tokens": 32170992.0, "step": 2991 }, { "epoch": 0.39545334390695214, "grad_norm": 0.9256861032242475, "learning_rate": 7.813824075204695e-06, "loss": 0.6109, "mean_token_accuracy": 0.7985899448394775, "num_tokens": 32236528.0, "step": 2992 }, { "epoch": 0.39558551414221516, "grad_norm": 0.9787453650619319, "learning_rate": 7.812043072435691e-06, "loss": 0.6333, "mean_token_accuracy": 0.7890520095825195, "num_tokens": 32302064.0, "step": 2993 }, { "epoch": 0.3957176843774782, "grad_norm": 0.8619087387423692, "learning_rate": 7.81026157748099e-06, "loss": 0.5815, "mean_token_accuracy": 0.8092876076698303, "num_tokens": 32367600.0, "step": 2994 }, { "epoch": 0.3958498546127412, "grad_norm": 0.7930788309646591, "learning_rate": 7.808479590719831e-06, "loss": 0.5401, "mean_token_accuracy": 0.8233426809310913, "num_tokens": 32433136.0, "step": 2995 }, { "epoch": 0.3959820248480042, "grad_norm": 0.9067147015227316, "learning_rate": 7.806697112531566e-06, "loss": 0.5809, "mean_token_accuracy": 0.8113631010055542, "num_tokens": 32498672.0, "step": 2996 }, { "epoch": 0.39611419508326723, "grad_norm": 0.8821939516678841, "learning_rate": 7.804914143295646e-06, "loss": 0.5906, "mean_token_accuracy": 0.8073495030403137, "num_tokens": 32564208.0, "step": 2997 }, { "epoch": 0.39624636531853025, "grad_norm": 0.9002775491665923, "learning_rate": 7.803130683391627e-06, "loss": 0.6228, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 32629744.0, "step": 2998 }, { "epoch": 0.39637853555379327, "grad_norm": 0.9886831776172031, "learning_rate": 7.801346733199173e-06, "loss": 0.6179, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 32695280.0, "step": 2999 }, { "epoch": 0.3965107057890563, "grad_norm": 0.8048050236978057, "learning_rate": 7.799562293098047e-06, "loss": 0.5858, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 32760816.0, "step": 3000 }, { "epoch": 0.3966428760243193, "grad_norm": 1.0376004432908585, "learning_rate": 7.797777363468121e-06, "loss": 0.6078, "mean_token_accuracy": 0.8003601431846619, "num_tokens": 32826352.0, "step": 3001 }, { "epoch": 0.3967750462595823, "grad_norm": 0.9441439016663631, "learning_rate": 7.795991944689368e-06, "loss": 0.6579, "mean_token_accuracy": 0.7840617895126343, "num_tokens": 32891888.0, "step": 3002 }, { "epoch": 0.39690721649484534, "grad_norm": 0.9384328520093913, "learning_rate": 7.794206037141867e-06, "loss": 0.6127, "mean_token_accuracy": 0.7989256381988525, "num_tokens": 32957424.0, "step": 3003 }, { "epoch": 0.39703938673010836, "grad_norm": 0.8992937053480993, "learning_rate": 7.7924196412058e-06, "loss": 0.6004, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 33022960.0, "step": 3004 }, { "epoch": 0.3971715569653714, "grad_norm": 1.208579070305832, "learning_rate": 7.790632757261455e-06, "loss": 0.6566, "mean_token_accuracy": 0.7873886227607727, "num_tokens": 33088496.0, "step": 3005 }, { "epoch": 0.3973037272006344, "grad_norm": 0.8552550346017884, "learning_rate": 7.78884538568922e-06, "loss": 0.5353, "mean_token_accuracy": 0.8217403292655945, "num_tokens": 33154032.0, "step": 3006 }, { "epoch": 0.3974358974358974, "grad_norm": 0.9549591108851779, "learning_rate": 7.78705752686959e-06, "loss": 0.5845, "mean_token_accuracy": 0.8078836798667908, "num_tokens": 33219568.0, "step": 3007 }, { "epoch": 0.39756806767116043, "grad_norm": 0.8909818088455582, "learning_rate": 7.785269181183163e-06, "loss": 0.5892, "mean_token_accuracy": 0.8091045022010803, "num_tokens": 33285104.0, "step": 3008 }, { "epoch": 0.39770023790642345, "grad_norm": 0.9081153026558916, "learning_rate": 7.783480349010639e-06, "loss": 0.6251, "mean_token_accuracy": 0.7975063920021057, "num_tokens": 33350640.0, "step": 3009 }, { "epoch": 0.39783240814168647, "grad_norm": 0.9213533239375248, "learning_rate": 7.781691030732826e-06, "loss": 0.6382, "mean_token_accuracy": 0.7914174199104309, "num_tokens": 33416176.0, "step": 3010 }, { "epoch": 0.3979645783769495, "grad_norm": 0.9040891159870719, "learning_rate": 7.77990122673063e-06, "loss": 0.5829, "mean_token_accuracy": 0.8105237483978271, "num_tokens": 33481712.0, "step": 3011 }, { "epoch": 0.3980967486122125, "grad_norm": 0.9016009503746868, "learning_rate": 7.778110937385066e-06, "loss": 0.5687, "mean_token_accuracy": 0.8151476979255676, "num_tokens": 33547248.0, "step": 3012 }, { "epoch": 0.3982289188474755, "grad_norm": 0.8664470823745781, "learning_rate": 7.776320163077244e-06, "loss": 0.6003, "mean_token_accuracy": 0.8015352487564087, "num_tokens": 33612784.0, "step": 3013 }, { "epoch": 0.39836108908273854, "grad_norm": 0.8772186155635201, "learning_rate": 7.774528904188386e-06, "loss": 0.6331, "mean_token_accuracy": 0.7929892539978027, "num_tokens": 33678320.0, "step": 3014 }, { "epoch": 0.3984932593180016, "grad_norm": 0.9103793108843916, "learning_rate": 7.772737161099814e-06, "loss": 0.6146, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 33743856.0, "step": 3015 }, { "epoch": 0.39862542955326463, "grad_norm": 0.8759224769552714, "learning_rate": 7.770944934192952e-06, "loss": 0.6383, "mean_token_accuracy": 0.7900744676589966, "num_tokens": 33809392.0, "step": 3016 }, { "epoch": 0.39875759978852765, "grad_norm": 0.9895204114102378, "learning_rate": 7.769152223849326e-06, "loss": 0.6494, "mean_token_accuracy": 0.7883653044700623, "num_tokens": 33874928.0, "step": 3017 }, { "epoch": 0.39888977002379067, "grad_norm": 0.9211303558541668, "learning_rate": 7.76735903045057e-06, "loss": 0.6883, "mean_token_accuracy": 0.7745696306228638, "num_tokens": 33940464.0, "step": 3018 }, { "epoch": 0.3990219402590537, "grad_norm": 0.980248827825292, "learning_rate": 7.765565354378414e-06, "loss": 0.5978, "mean_token_accuracy": 0.8041600584983826, "num_tokens": 34006000.0, "step": 3019 }, { "epoch": 0.3991541104943167, "grad_norm": 0.8798519802460064, "learning_rate": 7.763771196014694e-06, "loss": 0.6083, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 34071536.0, "step": 3020 }, { "epoch": 0.3992862807295797, "grad_norm": 0.8431673057494491, "learning_rate": 7.761976555741353e-06, "loss": 0.5948, "mean_token_accuracy": 0.8026797771453857, "num_tokens": 34137072.0, "step": 3021 }, { "epoch": 0.39941845096484274, "grad_norm": 0.9406686110156768, "learning_rate": 7.76018143394043e-06, "loss": 0.6297, "mean_token_accuracy": 0.792348325252533, "num_tokens": 34202608.0, "step": 3022 }, { "epoch": 0.39955062120010576, "grad_norm": 0.8958057473296831, "learning_rate": 7.758385830994067e-06, "loss": 0.5875, "mean_token_accuracy": 0.8051519989967346, "num_tokens": 34268144.0, "step": 3023 }, { "epoch": 0.3996827914353688, "grad_norm": 0.8736526274291566, "learning_rate": 7.756589747284514e-06, "loss": 0.6009, "mean_token_accuracy": 0.8055030107498169, "num_tokens": 34333680.0, "step": 3024 }, { "epoch": 0.3998149616706318, "grad_norm": 0.904143699335065, "learning_rate": 7.754793183194116e-06, "loss": 0.6127, "mean_token_accuracy": 0.7985899448394775, "num_tokens": 34399216.0, "step": 3025 }, { "epoch": 0.3999471319058948, "grad_norm": 0.8640008305900881, "learning_rate": 7.752996139105328e-06, "loss": 0.6031, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 34464752.0, "step": 3026 }, { "epoch": 0.40007930214115783, "grad_norm": 0.9059285117444803, "learning_rate": 7.751198615400701e-06, "loss": 0.613, "mean_token_accuracy": 0.7984525561332703, "num_tokens": 34530288.0, "step": 3027 }, { "epoch": 0.40021147237642085, "grad_norm": 1.0020846097147158, "learning_rate": 7.749400612462889e-06, "loss": 0.6529, "mean_token_accuracy": 0.7866408228874207, "num_tokens": 34595824.0, "step": 3028 }, { "epoch": 0.40034364261168387, "grad_norm": 0.8125382017900008, "learning_rate": 7.747602130674653e-06, "loss": 0.6521, "mean_token_accuracy": 0.7862440347671509, "num_tokens": 34661360.0, "step": 3029 }, { "epoch": 0.4004758128469469, "grad_norm": 0.8928703669229976, "learning_rate": 7.74580317041885e-06, "loss": 0.6024, "mean_token_accuracy": 0.802008330821991, "num_tokens": 34726896.0, "step": 3030 }, { "epoch": 0.4006079830822099, "grad_norm": 0.8135660291702115, "learning_rate": 7.744003732078439e-06, "loss": 0.5983, "mean_token_accuracy": 0.8072121739387512, "num_tokens": 34792432.0, "step": 3031 }, { "epoch": 0.4007401533174729, "grad_norm": 0.9281885028157927, "learning_rate": 7.742203816036485e-06, "loss": 0.5729, "mean_token_accuracy": 0.8120192885398865, "num_tokens": 34857968.0, "step": 3032 }, { "epoch": 0.40087232355273594, "grad_norm": 0.8514683856003119, "learning_rate": 7.740403422676152e-06, "loss": 0.6031, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 34923504.0, "step": 3033 }, { "epoch": 0.40100449378799896, "grad_norm": 0.9491483443549286, "learning_rate": 7.738602552380708e-06, "loss": 0.6595, "mean_token_accuracy": 0.7814064025878906, "num_tokens": 34989040.0, "step": 3034 }, { "epoch": 0.401136664023262, "grad_norm": 0.8888393130794988, "learning_rate": 7.73680120553352e-06, "loss": 0.5954, "mean_token_accuracy": 0.8037480115890503, "num_tokens": 35054576.0, "step": 3035 }, { "epoch": 0.401268834258525, "grad_norm": 0.874255372104847, "learning_rate": 7.734999382518055e-06, "loss": 0.5937, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 35120112.0, "step": 3036 }, { "epoch": 0.401401004493788, "grad_norm": 0.9546377104753284, "learning_rate": 7.733197083717885e-06, "loss": 0.6188, "mean_token_accuracy": 0.7981473803520203, "num_tokens": 35185648.0, "step": 3037 }, { "epoch": 0.40153317472905103, "grad_norm": 1.019397867629862, "learning_rate": 7.731394309516681e-06, "loss": 0.637, "mean_token_accuracy": 0.7929739952087402, "num_tokens": 35251184.0, "step": 3038 }, { "epoch": 0.40166534496431405, "grad_norm": 0.9173860356642831, "learning_rate": 7.729591060298219e-06, "loss": 0.631, "mean_token_accuracy": 0.7971401810646057, "num_tokens": 35316720.0, "step": 3039 }, { "epoch": 0.40179751519957707, "grad_norm": 0.9219153487997902, "learning_rate": 7.72778733644637e-06, "loss": 0.6269, "mean_token_accuracy": 0.7947595119476318, "num_tokens": 35382256.0, "step": 3040 }, { "epoch": 0.4019296854348401, "grad_norm": 0.9559181870160599, "learning_rate": 7.725983138345107e-06, "loss": 0.6066, "mean_token_accuracy": 0.8009247779846191, "num_tokens": 35447792.0, "step": 3041 }, { "epoch": 0.4020618556701031, "grad_norm": 0.9416771235766201, "learning_rate": 7.724178466378512e-06, "loss": 0.6309, "mean_token_accuracy": 0.7909290790557861, "num_tokens": 35513328.0, "step": 3042 }, { "epoch": 0.4021940259053661, "grad_norm": 0.8447413883389375, "learning_rate": 7.722373320930759e-06, "loss": 0.5826, "mean_token_accuracy": 0.8058387041091919, "num_tokens": 35578864.0, "step": 3043 }, { "epoch": 0.40232619614062914, "grad_norm": 0.8978403353609485, "learning_rate": 7.720567702386122e-06, "loss": 0.5952, "mean_token_accuracy": 0.8018556833267212, "num_tokens": 35644400.0, "step": 3044 }, { "epoch": 0.40245836637589216, "grad_norm": 0.9002348561488662, "learning_rate": 7.718761611128987e-06, "loss": 0.6219, "mean_token_accuracy": 0.7947289943695068, "num_tokens": 35709936.0, "step": 3045 }, { "epoch": 0.4025905366111552, "grad_norm": 0.9855770759898658, "learning_rate": 7.71695504754383e-06, "loss": 0.6337, "mean_token_accuracy": 0.7930808067321777, "num_tokens": 35775472.0, "step": 3046 }, { "epoch": 0.4027227068464182, "grad_norm": 0.9114615633906724, "learning_rate": 7.715148012015228e-06, "loss": 0.6548, "mean_token_accuracy": 0.7850689888000488, "num_tokens": 35841008.0, "step": 3047 }, { "epoch": 0.4028548770816812, "grad_norm": 0.8713865667531104, "learning_rate": 7.713340504927863e-06, "loss": 0.6063, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 35906544.0, "step": 3048 }, { "epoch": 0.40298704731694424, "grad_norm": 0.9565489794411614, "learning_rate": 7.711532526666518e-06, "loss": 0.6397, "mean_token_accuracy": 0.7910053730010986, "num_tokens": 35972080.0, "step": 3049 }, { "epoch": 0.40311921755220725, "grad_norm": 0.9412587160204532, "learning_rate": 7.709724077616069e-06, "loss": 0.5879, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 36037616.0, "step": 3050 }, { "epoch": 0.4032513877874703, "grad_norm": 0.9063285298679623, "learning_rate": 7.707915158161503e-06, "loss": 0.6393, "mean_token_accuracy": 0.7915242314338684, "num_tokens": 36103152.0, "step": 3051 }, { "epoch": 0.4033835580227333, "grad_norm": 0.8958326846687054, "learning_rate": 7.706105768687894e-06, "loss": 0.5989, "mean_token_accuracy": 0.8029086589813232, "num_tokens": 36168688.0, "step": 3052 }, { "epoch": 0.4035157282579963, "grad_norm": 1.0064429489623126, "learning_rate": 7.70429590958043e-06, "loss": 0.6517, "mean_token_accuracy": 0.7839854955673218, "num_tokens": 36234224.0, "step": 3053 }, { "epoch": 0.4036478984932593, "grad_norm": 0.8706898935288645, "learning_rate": 7.702485581224386e-06, "loss": 0.638, "mean_token_accuracy": 0.7916615605354309, "num_tokens": 36299760.0, "step": 3054 }, { "epoch": 0.40378006872852235, "grad_norm": 0.8736410967019707, "learning_rate": 7.700674784005147e-06, "loss": 0.6647, "mean_token_accuracy": 0.7827035784721375, "num_tokens": 36365296.0, "step": 3055 }, { "epoch": 0.40391223896378536, "grad_norm": 0.9139276254743091, "learning_rate": 7.698863518308196e-06, "loss": 0.6503, "mean_token_accuracy": 0.7860456705093384, "num_tokens": 36430832.0, "step": 3056 }, { "epoch": 0.4040444091990484, "grad_norm": 0.9294602702217996, "learning_rate": 7.69705178451911e-06, "loss": 0.6417, "mean_token_accuracy": 0.7894182801246643, "num_tokens": 36496368.0, "step": 3057 }, { "epoch": 0.4041765794343114, "grad_norm": 0.8758151148449125, "learning_rate": 7.695239583023568e-06, "loss": 0.5511, "mean_token_accuracy": 0.8202295303344727, "num_tokens": 36561904.0, "step": 3058 }, { "epoch": 0.4043087496695744, "grad_norm": 0.9096052836614942, "learning_rate": 7.693426914207354e-06, "loss": 0.6344, "mean_token_accuracy": 0.7918599843978882, "num_tokens": 36627440.0, "step": 3059 }, { "epoch": 0.40444091990483744, "grad_norm": 0.8224136170926726, "learning_rate": 7.691613778456347e-06, "loss": 0.6123, "mean_token_accuracy": 0.8017183542251587, "num_tokens": 36692976.0, "step": 3060 }, { "epoch": 0.40457309014010046, "grad_norm": 0.9116650633739508, "learning_rate": 7.68980017615652e-06, "loss": 0.5751, "mean_token_accuracy": 0.8132401704788208, "num_tokens": 36758512.0, "step": 3061 }, { "epoch": 0.4047052603753635, "grad_norm": 0.8954792562753704, "learning_rate": 7.687986107693956e-06, "loss": 0.6162, "mean_token_accuracy": 0.7957514524459839, "num_tokens": 36824048.0, "step": 3062 }, { "epoch": 0.4048374306106265, "grad_norm": 0.9089085758056377, "learning_rate": 7.686171573454834e-06, "loss": 0.5869, "mean_token_accuracy": 0.8062354922294617, "num_tokens": 36889584.0, "step": 3063 }, { "epoch": 0.4049696008458895, "grad_norm": 0.8474113836236166, "learning_rate": 7.684356573825426e-06, "loss": 0.628, "mean_token_accuracy": 0.7932028770446777, "num_tokens": 36955120.0, "step": 3064 }, { "epoch": 0.40510177108115253, "grad_norm": 0.8387085446480116, "learning_rate": 7.68254110919211e-06, "loss": 0.5817, "mean_token_accuracy": 0.8091807961463928, "num_tokens": 37020656.0, "step": 3065 }, { "epoch": 0.40523394131641555, "grad_norm": 0.9165927433536118, "learning_rate": 7.680725179941361e-06, "loss": 0.6084, "mean_token_accuracy": 0.8013673424720764, "num_tokens": 37086192.0, "step": 3066 }, { "epoch": 0.40536611155167857, "grad_norm": 0.9120943641181117, "learning_rate": 7.678908786459751e-06, "loss": 0.6165, "mean_token_accuracy": 0.7982541918754578, "num_tokens": 37151728.0, "step": 3067 }, { "epoch": 0.4054982817869416, "grad_norm": 0.9700441993442085, "learning_rate": 7.677091929133954e-06, "loss": 0.6825, "mean_token_accuracy": 0.7771029472351074, "num_tokens": 37217264.0, "step": 3068 }, { "epoch": 0.4056304520222046, "grad_norm": 0.9409473061120418, "learning_rate": 7.675274608350738e-06, "loss": 0.6563, "mean_token_accuracy": 0.7850079536437988, "num_tokens": 37282800.0, "step": 3069 }, { "epoch": 0.4057626222574676, "grad_norm": 0.8583555192856276, "learning_rate": 7.673456824496975e-06, "loss": 0.587, "mean_token_accuracy": 0.8076852560043335, "num_tokens": 37348336.0, "step": 3070 }, { "epoch": 0.40589479249273064, "grad_norm": 0.94946154339193, "learning_rate": 7.671638577959632e-06, "loss": 0.6056, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 37413872.0, "step": 3071 }, { "epoch": 0.40602696272799366, "grad_norm": 0.823351402356534, "learning_rate": 7.669819869125777e-06, "loss": 0.5487, "mean_token_accuracy": 0.8204278945922852, "num_tokens": 37479408.0, "step": 3072 }, { "epoch": 0.4061591329632567, "grad_norm": 1.0547177890621096, "learning_rate": 7.668000698382573e-06, "loss": 0.6134, "mean_token_accuracy": 0.8006653785705566, "num_tokens": 37544944.0, "step": 3073 }, { "epoch": 0.4062913031985197, "grad_norm": 0.8389569584744346, "learning_rate": 7.666181066117287e-06, "loss": 0.5752, "mean_token_accuracy": 0.808478832244873, "num_tokens": 37610480.0, "step": 3074 }, { "epoch": 0.4064234734337827, "grad_norm": 0.888798523456669, "learning_rate": 7.664360972717276e-06, "loss": 0.5861, "mean_token_accuracy": 0.8075631856918335, "num_tokens": 37676016.0, "step": 3075 }, { "epoch": 0.40655564366904573, "grad_norm": 0.9460352431881898, "learning_rate": 7.662540418570006e-06, "loss": 0.6045, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 37741552.0, "step": 3076 }, { "epoch": 0.40668781390430875, "grad_norm": 0.8836616892584802, "learning_rate": 7.660719404063028e-06, "loss": 0.5771, "mean_token_accuracy": 0.8100964426994324, "num_tokens": 37807088.0, "step": 3077 }, { "epoch": 0.40681998413957177, "grad_norm": 0.8934818624260569, "learning_rate": 7.658897929584003e-06, "loss": 0.6073, "mean_token_accuracy": 0.803091824054718, "num_tokens": 37872624.0, "step": 3078 }, { "epoch": 0.4069521543748348, "grad_norm": 0.9597700113584926, "learning_rate": 7.657075995520682e-06, "loss": 0.5872, "mean_token_accuracy": 0.8059913516044617, "num_tokens": 37938160.0, "step": 3079 }, { "epoch": 0.4070843246100978, "grad_norm": 0.8203129843091449, "learning_rate": 7.655253602260921e-06, "loss": 0.5889, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 38003696.0, "step": 3080 }, { "epoch": 0.4072164948453608, "grad_norm": 0.7850193027761616, "learning_rate": 7.653430750192664e-06, "loss": 0.5311, "mean_token_accuracy": 0.8290044069290161, "num_tokens": 38069232.0, "step": 3081 }, { "epoch": 0.40734866508062384, "grad_norm": 1.0652868292525373, "learning_rate": 7.651607439703962e-06, "loss": 0.6111, "mean_token_accuracy": 0.802038848400116, "num_tokens": 38134768.0, "step": 3082 }, { "epoch": 0.40748083531588686, "grad_norm": 1.0036443720772772, "learning_rate": 7.649783671182957e-06, "loss": 0.61, "mean_token_accuracy": 0.8005542159080505, "num_tokens": 38197565.0, "step": 3083 }, { "epoch": 0.4076130055511499, "grad_norm": 0.8851647710445163, "learning_rate": 7.647959445017895e-06, "loss": 0.5775, "mean_token_accuracy": 0.8117751479148865, "num_tokens": 38263101.0, "step": 3084 }, { "epoch": 0.4077451757864129, "grad_norm": 0.9602256436430217, "learning_rate": 7.646134761597112e-06, "loss": 0.6664, "mean_token_accuracy": 0.784641683101654, "num_tokens": 38328637.0, "step": 3085 }, { "epoch": 0.4078773460216759, "grad_norm": 0.8915555687112207, "learning_rate": 7.644309621309045e-06, "loss": 0.6114, "mean_token_accuracy": 0.7987425327301025, "num_tokens": 38394173.0, "step": 3086 }, { "epoch": 0.40800951625693893, "grad_norm": 0.9496659443677521, "learning_rate": 7.642484024542232e-06, "loss": 0.6313, "mean_token_accuracy": 0.7945916056632996, "num_tokens": 38459709.0, "step": 3087 }, { "epoch": 0.40814168649220195, "grad_norm": 0.8715754978174469, "learning_rate": 7.6406579716853e-06, "loss": 0.6168, "mean_token_accuracy": 0.7998565435409546, "num_tokens": 38525245.0, "step": 3088 }, { "epoch": 0.40827385672746497, "grad_norm": 1.069853145926252, "learning_rate": 7.63883146312698e-06, "loss": 0.6548, "mean_token_accuracy": 0.7875564694404602, "num_tokens": 38590781.0, "step": 3089 }, { "epoch": 0.408406026962728, "grad_norm": 0.9113268449151061, "learning_rate": 7.637004499256099e-06, "loss": 0.6347, "mean_token_accuracy": 0.7930655479431152, "num_tokens": 38656317.0, "step": 3090 }, { "epoch": 0.408538197197991, "grad_norm": 0.9615362507230011, "learning_rate": 7.635177080461575e-06, "loss": 0.6419, "mean_token_accuracy": 0.7878769636154175, "num_tokens": 38721853.0, "step": 3091 }, { "epoch": 0.408670367433254, "grad_norm": 0.9017001069008315, "learning_rate": 7.633349207132432e-06, "loss": 0.5717, "mean_token_accuracy": 0.8109052777290344, "num_tokens": 38787389.0, "step": 3092 }, { "epoch": 0.40880253766851704, "grad_norm": 0.8393832838746871, "learning_rate": 7.631520879657783e-06, "loss": 0.6201, "mean_token_accuracy": 0.7967739105224609, "num_tokens": 38852925.0, "step": 3093 }, { "epoch": 0.40893470790378006, "grad_norm": 0.9464351917570715, "learning_rate": 7.62969209842684e-06, "loss": 0.5978, "mean_token_accuracy": 0.8045415878295898, "num_tokens": 38918461.0, "step": 3094 }, { "epoch": 0.4090668781390431, "grad_norm": 0.9128865165918822, "learning_rate": 7.6278628638289165e-06, "loss": 0.6219, "mean_token_accuracy": 0.7971706986427307, "num_tokens": 38983997.0, "step": 3095 }, { "epoch": 0.4091990483743061, "grad_norm": 0.8356474974667912, "learning_rate": 7.626033176253415e-06, "loss": 0.5457, "mean_token_accuracy": 0.8210688829421997, "num_tokens": 39049533.0, "step": 3096 }, { "epoch": 0.4093312186095691, "grad_norm": 0.905900376090036, "learning_rate": 7.624203036089838e-06, "loss": 0.6091, "mean_token_accuracy": 0.7994902729988098, "num_tokens": 39115069.0, "step": 3097 }, { "epoch": 0.40946338884483213, "grad_norm": 0.8984332902467616, "learning_rate": 7.622372443727786e-06, "loss": 0.6128, "mean_token_accuracy": 0.7979947328567505, "num_tokens": 39180605.0, "step": 3098 }, { "epoch": 0.40959555908009515, "grad_norm": 0.960566651000325, "learning_rate": 7.620541399556951e-06, "loss": 0.6106, "mean_token_accuracy": 0.7978116273880005, "num_tokens": 39246141.0, "step": 3099 }, { "epoch": 0.40972772931535817, "grad_norm": 0.9064680296914819, "learning_rate": 7.6187099039671274e-06, "loss": 0.6215, "mean_token_accuracy": 0.7946832180023193, "num_tokens": 39311677.0, "step": 3100 }, { "epoch": 0.4098598995506212, "grad_norm": 0.9620517891072256, "learning_rate": 7.6168779573482e-06, "loss": 0.6482, "mean_token_accuracy": 0.7872054576873779, "num_tokens": 39377213.0, "step": 3101 }, { "epoch": 0.4099920697858842, "grad_norm": 0.7902298199783856, "learning_rate": 7.6150455600901514e-06, "loss": 0.539, "mean_token_accuracy": 0.8262879848480225, "num_tokens": 39442749.0, "step": 3102 }, { "epoch": 0.4101242400211472, "grad_norm": 0.8527543233477216, "learning_rate": 7.613212712583064e-06, "loss": 0.6165, "mean_token_accuracy": 0.7981473803520203, "num_tokens": 39508285.0, "step": 3103 }, { "epoch": 0.41025641025641024, "grad_norm": 0.8895857536309574, "learning_rate": 7.611379415217111e-06, "loss": 0.5946, "mean_token_accuracy": 0.8063423037528992, "num_tokens": 39573821.0, "step": 3104 }, { "epoch": 0.41038858049167326, "grad_norm": 0.9682183345560429, "learning_rate": 7.609545668382562e-06, "loss": 0.6111, "mean_token_accuracy": 0.8018251657485962, "num_tokens": 39639357.0, "step": 3105 }, { "epoch": 0.4105207507269363, "grad_norm": 0.8399091587615124, "learning_rate": 7.607711472469784e-06, "loss": 0.6426, "mean_token_accuracy": 0.7899066209793091, "num_tokens": 39704893.0, "step": 3106 }, { "epoch": 0.4106529209621993, "grad_norm": 0.8520885905281446, "learning_rate": 7.60587682786924e-06, "loss": 0.5695, "mean_token_accuracy": 0.8138048052787781, "num_tokens": 39770429.0, "step": 3107 }, { "epoch": 0.4107850911974623, "grad_norm": 0.8784220652172753, "learning_rate": 7.604041734971487e-06, "loss": 0.6306, "mean_token_accuracy": 0.79330974817276, "num_tokens": 39835965.0, "step": 3108 }, { "epoch": 0.41091726143272533, "grad_norm": 0.9721750977375809, "learning_rate": 7.602206194167177e-06, "loss": 0.6147, "mean_token_accuracy": 0.7986509799957275, "num_tokens": 39901501.0, "step": 3109 }, { "epoch": 0.41104943166798835, "grad_norm": 0.7878201284802893, "learning_rate": 7.6003702058470595e-06, "loss": 0.5765, "mean_token_accuracy": 0.808692455291748, "num_tokens": 39967037.0, "step": 3110 }, { "epoch": 0.41118160190325137, "grad_norm": 0.9008350946798332, "learning_rate": 7.598533770401979e-06, "loss": 0.635, "mean_token_accuracy": 0.7961024045944214, "num_tokens": 40032573.0, "step": 3111 }, { "epoch": 0.4113137721385144, "grad_norm": 0.9004298705896351, "learning_rate": 7.596696888222872e-06, "loss": 0.6345, "mean_token_accuracy": 0.7926077246665955, "num_tokens": 40098109.0, "step": 3112 }, { "epoch": 0.4114459423737774, "grad_norm": 0.8862805209324015, "learning_rate": 7.594859559700774e-06, "loss": 0.6006, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 40163645.0, "step": 3113 }, { "epoch": 0.4115781126090404, "grad_norm": 0.956555658277742, "learning_rate": 7.593021785226811e-06, "loss": 0.6723, "mean_token_accuracy": 0.7813606262207031, "num_tokens": 40229181.0, "step": 3114 }, { "epoch": 0.41171028284430344, "grad_norm": 0.8873923787509788, "learning_rate": 7.59118356519221e-06, "loss": 0.6068, "mean_token_accuracy": 0.7994139790534973, "num_tokens": 40294717.0, "step": 3115 }, { "epoch": 0.41184245307956646, "grad_norm": 0.9553139340868986, "learning_rate": 7.589344899988292e-06, "loss": 0.5805, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 40360253.0, "step": 3116 }, { "epoch": 0.4119746233148295, "grad_norm": 0.8891263835284333, "learning_rate": 7.587505790006463e-06, "loss": 0.635, "mean_token_accuracy": 0.7923941016197205, "num_tokens": 40425789.0, "step": 3117 }, { "epoch": 0.4121067935500925, "grad_norm": 0.8867317974774698, "learning_rate": 7.585666235638237e-06, "loss": 0.5666, "mean_token_accuracy": 0.8116072416305542, "num_tokens": 40491325.0, "step": 3118 }, { "epoch": 0.4122389637853555, "grad_norm": 0.8441146665097432, "learning_rate": 7.583826237275216e-06, "loss": 0.5969, "mean_token_accuracy": 0.8025119304656982, "num_tokens": 40556861.0, "step": 3119 }, { "epoch": 0.41237113402061853, "grad_norm": 0.8491849969230632, "learning_rate": 7.581985795309094e-06, "loss": 0.5919, "mean_token_accuracy": 0.8047704696655273, "num_tokens": 40622397.0, "step": 3120 }, { "epoch": 0.41250330425588155, "grad_norm": 0.9496054300533712, "learning_rate": 7.580144910131666e-06, "loss": 0.6178, "mean_token_accuracy": 0.7963466048240662, "num_tokens": 40687933.0, "step": 3121 }, { "epoch": 0.41263547449114457, "grad_norm": 0.9358400608683253, "learning_rate": 7.578303582134818e-06, "loss": 0.6116, "mean_token_accuracy": 0.8003601431846619, "num_tokens": 40753469.0, "step": 3122 }, { "epoch": 0.4127676447264076, "grad_norm": 1.0543234296643573, "learning_rate": 7.576461811710527e-06, "loss": 0.6636, "mean_token_accuracy": 0.78277987241745, "num_tokens": 40819005.0, "step": 3123 }, { "epoch": 0.4128998149616706, "grad_norm": 0.9737233240476455, "learning_rate": 7.574619599250872e-06, "loss": 0.6921, "mean_token_accuracy": 0.7740660309791565, "num_tokens": 40884541.0, "step": 3124 }, { "epoch": 0.4130319851969336, "grad_norm": 0.7787332222324115, "learning_rate": 7.572776945148017e-06, "loss": 0.5613, "mean_token_accuracy": 0.8164143562316895, "num_tokens": 40950077.0, "step": 3125 }, { "epoch": 0.41316415543219664, "grad_norm": 0.886827910091689, "learning_rate": 7.570933849794227e-06, "loss": 0.5792, "mean_token_accuracy": 0.8093487024307251, "num_tokens": 41015613.0, "step": 3126 }, { "epoch": 0.41329632566745966, "grad_norm": 0.809881105072587, "learning_rate": 7.569090313581861e-06, "loss": 0.5849, "mean_token_accuracy": 0.8085246086120605, "num_tokens": 41081149.0, "step": 3127 }, { "epoch": 0.4134284959027227, "grad_norm": 0.8859115358910898, "learning_rate": 7.567246336903365e-06, "loss": 0.6239, "mean_token_accuracy": 0.7964991927146912, "num_tokens": 41146685.0, "step": 3128 }, { "epoch": 0.4135606661379857, "grad_norm": 0.9321313612217359, "learning_rate": 7.565401920151285e-06, "loss": 0.5775, "mean_token_accuracy": 0.8105390071868896, "num_tokens": 41212221.0, "step": 3129 }, { "epoch": 0.4136928363732487, "grad_norm": 0.8178757833497631, "learning_rate": 7.56355706371826e-06, "loss": 0.5574, "mean_token_accuracy": 0.818321943283081, "num_tokens": 41277757.0, "step": 3130 }, { "epoch": 0.41382500660851174, "grad_norm": 0.9164926199215059, "learning_rate": 7.561711767997018e-06, "loss": 0.5963, "mean_token_accuracy": 0.8047246932983398, "num_tokens": 41343293.0, "step": 3131 }, { "epoch": 0.41395717684377475, "grad_norm": 0.9296356489751461, "learning_rate": 7.559866033380387e-06, "loss": 0.6585, "mean_token_accuracy": 0.7833902835845947, "num_tokens": 41408829.0, "step": 3132 }, { "epoch": 0.41408934707903783, "grad_norm": 0.8929238679833522, "learning_rate": 7.558019860261283e-06, "loss": 0.5689, "mean_token_accuracy": 0.813087522983551, "num_tokens": 41474365.0, "step": 3133 }, { "epoch": 0.41422151731430085, "grad_norm": 0.8649927509213765, "learning_rate": 7.556173249032721e-06, "loss": 0.6011, "mean_token_accuracy": 0.8032901883125305, "num_tokens": 41539901.0, "step": 3134 }, { "epoch": 0.41435368754956386, "grad_norm": 0.8585800171175982, "learning_rate": 7.5543262000878034e-06, "loss": 0.5855, "mean_token_accuracy": 0.807669997215271, "num_tokens": 41605437.0, "step": 3135 }, { "epoch": 0.4144858577848269, "grad_norm": 0.9438912782464203, "learning_rate": 7.55247871381973e-06, "loss": 0.6422, "mean_token_accuracy": 0.7906696200370789, "num_tokens": 41670973.0, "step": 3136 }, { "epoch": 0.4146180280200899, "grad_norm": 0.8433044704496159, "learning_rate": 7.550630790621789e-06, "loss": 0.5806, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 41736509.0, "step": 3137 }, { "epoch": 0.4147501982553529, "grad_norm": 0.8635319374243418, "learning_rate": 7.548782430887368e-06, "loss": 0.5451, "mean_token_accuracy": 0.8209772706031799, "num_tokens": 41802045.0, "step": 3138 }, { "epoch": 0.41488236849061594, "grad_norm": 0.8511056591339393, "learning_rate": 7.546933635009942e-06, "loss": 0.5947, "mean_token_accuracy": 0.8025424480438232, "num_tokens": 41867581.0, "step": 3139 }, { "epoch": 0.41501453872587896, "grad_norm": 0.8765485016840098, "learning_rate": 7.545084403383084e-06, "loss": 0.6259, "mean_token_accuracy": 0.7952020764350891, "num_tokens": 41933117.0, "step": 3140 }, { "epoch": 0.415146708961142, "grad_norm": 0.8490672820677349, "learning_rate": 7.543234736400453e-06, "loss": 0.5864, "mean_token_accuracy": 0.8082346320152283, "num_tokens": 41998653.0, "step": 3141 }, { "epoch": 0.415278879196405, "grad_norm": 1.0179976981447092, "learning_rate": 7.541384634455807e-06, "loss": 0.641, "mean_token_accuracy": 0.7877243161201477, "num_tokens": 42064189.0, "step": 3142 }, { "epoch": 0.415411049431668, "grad_norm": 0.867450993459796, "learning_rate": 7.539534097942993e-06, "loss": 0.6086, "mean_token_accuracy": 0.8017946481704712, "num_tokens": 42129725.0, "step": 3143 }, { "epoch": 0.41554321966693103, "grad_norm": 0.9069933292485772, "learning_rate": 7.537683127255952e-06, "loss": 0.6292, "mean_token_accuracy": 0.7948815822601318, "num_tokens": 42195261.0, "step": 3144 }, { "epoch": 0.41567538990219405, "grad_norm": 0.838518812829915, "learning_rate": 7.535831722788718e-06, "loss": 0.6137, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 42260797.0, "step": 3145 }, { "epoch": 0.41580756013745707, "grad_norm": 0.8444189084855036, "learning_rate": 7.533979884935414e-06, "loss": 0.5413, "mean_token_accuracy": 0.8209925293922424, "num_tokens": 42326333.0, "step": 3146 }, { "epoch": 0.4159397303727201, "grad_norm": 0.8484596714982816, "learning_rate": 7.53212761409026e-06, "loss": 0.6241, "mean_token_accuracy": 0.7953241467475891, "num_tokens": 42391869.0, "step": 3147 }, { "epoch": 0.4160719006079831, "grad_norm": 0.9990665728358495, "learning_rate": 7.530274910647565e-06, "loss": 0.6741, "mean_token_accuracy": 0.779193639755249, "num_tokens": 42457405.0, "step": 3148 }, { "epoch": 0.4162040708432461, "grad_norm": 0.9528219718208898, "learning_rate": 7.528421775001729e-06, "loss": 0.6486, "mean_token_accuracy": 0.7885178923606873, "num_tokens": 42522941.0, "step": 3149 }, { "epoch": 0.41633624107850914, "grad_norm": 0.9096042266114485, "learning_rate": 7.526568207547248e-06, "loss": 0.5932, "mean_token_accuracy": 0.8016420602798462, "num_tokens": 42588477.0, "step": 3150 }, { "epoch": 0.41646841131377216, "grad_norm": 0.8628085945493648, "learning_rate": 7.52471420867871e-06, "loss": 0.623, "mean_token_accuracy": 0.7952325940132141, "num_tokens": 42654013.0, "step": 3151 }, { "epoch": 0.4166005815490352, "grad_norm": 0.9324801650324972, "learning_rate": 7.5228597787907866e-06, "loss": 0.6189, "mean_token_accuracy": 0.7967739105224609, "num_tokens": 42719549.0, "step": 3152 }, { "epoch": 0.4167327517842982, "grad_norm": 0.8512112125307313, "learning_rate": 7.521004918278252e-06, "loss": 0.6098, "mean_token_accuracy": 0.7991088032722473, "num_tokens": 42785085.0, "step": 3153 }, { "epoch": 0.4168649220195612, "grad_norm": 0.8421478486930777, "learning_rate": 7.519149627535965e-06, "loss": 0.5666, "mean_token_accuracy": 0.8144152164459229, "num_tokens": 42850621.0, "step": 3154 }, { "epoch": 0.41699709225482423, "grad_norm": 0.8584324081500558, "learning_rate": 7.517293906958879e-06, "loss": 0.6047, "mean_token_accuracy": 0.8027865886688232, "num_tokens": 42916157.0, "step": 3155 }, { "epoch": 0.41712926249008725, "grad_norm": 0.8477370422223846, "learning_rate": 7.515437756942039e-06, "loss": 0.5887, "mean_token_accuracy": 0.8047094345092773, "num_tokens": 42981693.0, "step": 3156 }, { "epoch": 0.41726143272535027, "grad_norm": 0.9659184161407645, "learning_rate": 7.51358117788058e-06, "loss": 0.6697, "mean_token_accuracy": 0.7801703214645386, "num_tokens": 43047229.0, "step": 3157 }, { "epoch": 0.4173936029606133, "grad_norm": 0.8722553764924459, "learning_rate": 7.511724170169727e-06, "loss": 0.652, "mean_token_accuracy": 0.7841380834579468, "num_tokens": 43112765.0, "step": 3158 }, { "epoch": 0.4175257731958763, "grad_norm": 0.8359433170041688, "learning_rate": 7.509866734204801e-06, "loss": 0.5908, "mean_token_accuracy": 0.807425856590271, "num_tokens": 43178301.0, "step": 3159 }, { "epoch": 0.4176579434311393, "grad_norm": 0.8074491080227677, "learning_rate": 7.5080088703812116e-06, "loss": 0.5636, "mean_token_accuracy": 0.8149188160896301, "num_tokens": 43243837.0, "step": 3160 }, { "epoch": 0.41779011366640234, "grad_norm": 0.9077893664812896, "learning_rate": 7.506150579094457e-06, "loss": 0.6028, "mean_token_accuracy": 0.8017183542251587, "num_tokens": 43309373.0, "step": 3161 }, { "epoch": 0.41792228390166536, "grad_norm": 0.8995445764928478, "learning_rate": 7.50429186074013e-06, "loss": 0.6222, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 43374909.0, "step": 3162 }, { "epoch": 0.4180544541369284, "grad_norm": 0.9090846021529781, "learning_rate": 7.502432715713911e-06, "loss": 0.6492, "mean_token_accuracy": 0.7849316596984863, "num_tokens": 43440445.0, "step": 3163 }, { "epoch": 0.4181866243721914, "grad_norm": 0.884468104363864, "learning_rate": 7.500573144411576e-06, "loss": 0.6496, "mean_token_accuracy": 0.7858014702796936, "num_tokens": 43505981.0, "step": 3164 }, { "epoch": 0.4183187946074544, "grad_norm": 0.8198835905447049, "learning_rate": 7.498713147228986e-06, "loss": 0.5531, "mean_token_accuracy": 0.8192070722579956, "num_tokens": 43571517.0, "step": 3165 }, { "epoch": 0.41845096484271743, "grad_norm": 0.8424041834900651, "learning_rate": 7.496852724562099e-06, "loss": 0.6164, "mean_token_accuracy": 0.7985899448394775, "num_tokens": 43637053.0, "step": 3166 }, { "epoch": 0.41858313507798045, "grad_norm": 0.8422497340632875, "learning_rate": 7.494991876806957e-06, "loss": 0.598, "mean_token_accuracy": 0.8021761775016785, "num_tokens": 43702589.0, "step": 3167 }, { "epoch": 0.41871530531324347, "grad_norm": 0.9435718688632557, "learning_rate": 7.493130604359698e-06, "loss": 0.6308, "mean_token_accuracy": 0.7956293225288391, "num_tokens": 43768125.0, "step": 3168 }, { "epoch": 0.4188474755485065, "grad_norm": 0.9202355554386816, "learning_rate": 7.4912689076165465e-06, "loss": 0.6466, "mean_token_accuracy": 0.7866408228874207, "num_tokens": 43833661.0, "step": 3169 }, { "epoch": 0.4189796457837695, "grad_norm": 0.9765146822160318, "learning_rate": 7.489406786973818e-06, "loss": 0.6425, "mean_token_accuracy": 0.7891740798950195, "num_tokens": 43899197.0, "step": 3170 }, { "epoch": 0.4191118160190325, "grad_norm": 0.8332899528847161, "learning_rate": 7.487544242827922e-06, "loss": 0.5745, "mean_token_accuracy": 0.8093639612197876, "num_tokens": 43964733.0, "step": 3171 }, { "epoch": 0.41924398625429554, "grad_norm": 0.7956538261909492, "learning_rate": 7.485681275575352e-06, "loss": 0.5718, "mean_token_accuracy": 0.8123703002929688, "num_tokens": 44030269.0, "step": 3172 }, { "epoch": 0.41937615648955856, "grad_norm": 0.8635445523527859, "learning_rate": 7.4838178856126985e-06, "loss": 0.6535, "mean_token_accuracy": 0.7865492701530457, "num_tokens": 44095805.0, "step": 3173 }, { "epoch": 0.4195083267248216, "grad_norm": 0.8783442770994335, "learning_rate": 7.481954073336635e-06, "loss": 0.6109, "mean_token_accuracy": 0.8037785291671753, "num_tokens": 44161341.0, "step": 3174 }, { "epoch": 0.4196404969600846, "grad_norm": 0.8550878226542789, "learning_rate": 7.480089839143929e-06, "loss": 0.6212, "mean_token_accuracy": 0.7959498167037964, "num_tokens": 44226877.0, "step": 3175 }, { "epoch": 0.4197726671953476, "grad_norm": 0.9748556304847932, "learning_rate": 7.478225183431439e-06, "loss": 0.6772, "mean_token_accuracy": 0.7789799571037292, "num_tokens": 44292413.0, "step": 3176 }, { "epoch": 0.41990483743061063, "grad_norm": 0.8473823424441518, "learning_rate": 7.476360106596109e-06, "loss": 0.5383, "mean_token_accuracy": 0.8221523761749268, "num_tokens": 44357949.0, "step": 3177 }, { "epoch": 0.42003700766587365, "grad_norm": 0.9450456438446863, "learning_rate": 7.474494609034976e-06, "loss": 0.6571, "mean_token_accuracy": 0.7837870717048645, "num_tokens": 44423485.0, "step": 3178 }, { "epoch": 0.42016917790113667, "grad_norm": 0.878532361059284, "learning_rate": 7.472628691145164e-06, "loss": 0.6151, "mean_token_accuracy": 0.7960871458053589, "num_tokens": 44489021.0, "step": 3179 }, { "epoch": 0.4203013481363997, "grad_norm": 0.8587615432691137, "learning_rate": 7.470762353323891e-06, "loss": 0.5448, "mean_token_accuracy": 0.8237242102622986, "num_tokens": 44554557.0, "step": 3180 }, { "epoch": 0.4204335183716627, "grad_norm": 0.9398295208805921, "learning_rate": 7.4688955959684585e-06, "loss": 0.6139, "mean_token_accuracy": 0.7975369095802307, "num_tokens": 44620093.0, "step": 3181 }, { "epoch": 0.4205656886069257, "grad_norm": 0.8804039881794801, "learning_rate": 7.467028419476261e-06, "loss": 0.62, "mean_token_accuracy": 0.7958277463912964, "num_tokens": 44685629.0, "step": 3182 }, { "epoch": 0.42069785884218874, "grad_norm": 0.9106123063532509, "learning_rate": 7.465160824244782e-06, "loss": 0.6084, "mean_token_accuracy": 0.8018251657485962, "num_tokens": 44751165.0, "step": 3183 }, { "epoch": 0.42083002907745176, "grad_norm": 0.7597478075048829, "learning_rate": 7.463292810671592e-06, "loss": 0.5267, "mean_token_accuracy": 0.827081561088562, "num_tokens": 44816701.0, "step": 3184 }, { "epoch": 0.4209621993127148, "grad_norm": 0.9637124618065243, "learning_rate": 7.461424379154356e-06, "loss": 0.6674, "mean_token_accuracy": 0.7786595225334167, "num_tokens": 44882237.0, "step": 3185 }, { "epoch": 0.4210943695479778, "grad_norm": 0.889334763981788, "learning_rate": 7.459555530090819e-06, "loss": 0.585, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 44947773.0, "step": 3186 }, { "epoch": 0.4212265397832408, "grad_norm": 0.8599647127045608, "learning_rate": 7.4576862638788225e-06, "loss": 0.5575, "mean_token_accuracy": 0.8179556727409363, "num_tokens": 45013309.0, "step": 3187 }, { "epoch": 0.42135871001850383, "grad_norm": 0.821963799878799, "learning_rate": 7.455816580916295e-06, "loss": 0.5788, "mean_token_accuracy": 0.8093334436416626, "num_tokens": 45078845.0, "step": 3188 }, { "epoch": 0.42149088025376685, "grad_norm": 1.0229605010386416, "learning_rate": 7.4539464816012505e-06, "loss": 0.6466, "mean_token_accuracy": 0.7893572449684143, "num_tokens": 45144381.0, "step": 3189 }, { "epoch": 0.42162305048902987, "grad_norm": 0.9450733993225318, "learning_rate": 7.452075966331798e-06, "loss": 0.6445, "mean_token_accuracy": 0.7891740798950195, "num_tokens": 45209917.0, "step": 3190 }, { "epoch": 0.4217552207242929, "grad_norm": 0.9407312111240194, "learning_rate": 7.450205035506126e-06, "loss": 0.621, "mean_token_accuracy": 0.7957209348678589, "num_tokens": 45275453.0, "step": 3191 }, { "epoch": 0.4218873909595559, "grad_norm": 0.9197895301616912, "learning_rate": 7.44833368952252e-06, "loss": 0.6438, "mean_token_accuracy": 0.7878311276435852, "num_tokens": 45340989.0, "step": 3192 }, { "epoch": 0.4220195611948189, "grad_norm": 0.9230314692397849, "learning_rate": 7.4464619287793505e-06, "loss": 0.5923, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 45406525.0, "step": 3193 }, { "epoch": 0.42215173143008194, "grad_norm": 0.9310592113124474, "learning_rate": 7.444589753675074e-06, "loss": 0.6183, "mean_token_accuracy": 0.7948968410491943, "num_tokens": 45472061.0, "step": 3194 }, { "epoch": 0.42228390166534496, "grad_norm": 0.8790111131127996, "learning_rate": 7.442717164608241e-06, "loss": 0.6333, "mean_token_accuracy": 0.7909443378448486, "num_tokens": 45537597.0, "step": 3195 }, { "epoch": 0.422416071900608, "grad_norm": 0.9391985212666758, "learning_rate": 7.440844161977483e-06, "loss": 0.6359, "mean_token_accuracy": 0.7908985614776611, "num_tokens": 45603133.0, "step": 3196 }, { "epoch": 0.422548242135871, "grad_norm": 0.9507740886346863, "learning_rate": 7.438970746181527e-06, "loss": 0.6368, "mean_token_accuracy": 0.7924551367759705, "num_tokens": 45668669.0, "step": 3197 }, { "epoch": 0.422680412371134, "grad_norm": 0.9798006935024506, "learning_rate": 7.4370969176191794e-06, "loss": 0.5745, "mean_token_accuracy": 0.8105847835540771, "num_tokens": 45734205.0, "step": 3198 }, { "epoch": 0.42281258260639704, "grad_norm": 0.8662958554957652, "learning_rate": 7.4352226766893434e-06, "loss": 0.5972, "mean_token_accuracy": 0.8035954236984253, "num_tokens": 45799741.0, "step": 3199 }, { "epoch": 0.42294475284166005, "grad_norm": 0.9565007135112885, "learning_rate": 7.433348023791002e-06, "loss": 0.6347, "mean_token_accuracy": 0.7935081124305725, "num_tokens": 45865277.0, "step": 3200 }, { "epoch": 0.4230769230769231, "grad_norm": 0.8181874355568522, "learning_rate": 7.4314729593232335e-06, "loss": 0.5486, "mean_token_accuracy": 0.8186729550361633, "num_tokens": 45930813.0, "step": 3201 }, { "epoch": 0.4232090933121861, "grad_norm": 0.9190010561619991, "learning_rate": 7.4295974836851966e-06, "loss": 0.6526, "mean_token_accuracy": 0.7853589057922363, "num_tokens": 45996349.0, "step": 3202 }, { "epoch": 0.4233412635474491, "grad_norm": 0.9640294508495187, "learning_rate": 7.427721597276145e-06, "loss": 0.6306, "mean_token_accuracy": 0.7958887815475464, "num_tokens": 46061885.0, "step": 3203 }, { "epoch": 0.4234734337827121, "grad_norm": 0.8626940099988525, "learning_rate": 7.425845300495412e-06, "loss": 0.5532, "mean_token_accuracy": 0.8177267909049988, "num_tokens": 46127421.0, "step": 3204 }, { "epoch": 0.42360560401797515, "grad_norm": 0.8742667798033981, "learning_rate": 7.423968593742423e-06, "loss": 0.5872, "mean_token_accuracy": 0.8093334436416626, "num_tokens": 46192957.0, "step": 3205 }, { "epoch": 0.42373777425323816, "grad_norm": 0.9002147912348325, "learning_rate": 7.422091477416692e-06, "loss": 0.5449, "mean_token_accuracy": 0.8210077881813049, "num_tokens": 46258493.0, "step": 3206 }, { "epoch": 0.4238699444885012, "grad_norm": 0.9181797151642292, "learning_rate": 7.420213951917814e-06, "loss": 0.5764, "mean_token_accuracy": 0.8108137249946594, "num_tokens": 46324029.0, "step": 3207 }, { "epoch": 0.4240021147237642, "grad_norm": 0.9122992157963882, "learning_rate": 7.418336017645477e-06, "loss": 0.6272, "mean_token_accuracy": 0.7922261953353882, "num_tokens": 46389565.0, "step": 3208 }, { "epoch": 0.4241342849590272, "grad_norm": 0.874267847126754, "learning_rate": 7.416457674999454e-06, "loss": 0.6503, "mean_token_accuracy": 0.7859541177749634, "num_tokens": 46455101.0, "step": 3209 }, { "epoch": 0.42426645519429024, "grad_norm": 0.8701591869126158, "learning_rate": 7.414578924379605e-06, "loss": 0.5867, "mean_token_accuracy": 0.8059760928153992, "num_tokens": 46520637.0, "step": 3210 }, { "epoch": 0.42439862542955326, "grad_norm": 0.8462294215918171, "learning_rate": 7.412699766185876e-06, "loss": 0.5691, "mean_token_accuracy": 0.8133927583694458, "num_tokens": 46586173.0, "step": 3211 }, { "epoch": 0.4245307956648163, "grad_norm": 0.9362046095166814, "learning_rate": 7.410820200818301e-06, "loss": 0.5721, "mean_token_accuracy": 0.8144457340240479, "num_tokens": 46651709.0, "step": 3212 }, { "epoch": 0.4246629659000793, "grad_norm": 0.8595019606264444, "learning_rate": 7.408940228677e-06, "loss": 0.638, "mean_token_accuracy": 0.7929129600524902, "num_tokens": 46717245.0, "step": 3213 }, { "epoch": 0.4247951361353423, "grad_norm": 0.8436777186584715, "learning_rate": 7.40705985016218e-06, "loss": 0.5583, "mean_token_accuracy": 0.8185203075408936, "num_tokens": 46782781.0, "step": 3214 }, { "epoch": 0.42492730637060533, "grad_norm": 0.8845455180182628, "learning_rate": 7.405179065674135e-06, "loss": 0.5877, "mean_token_accuracy": 0.806678056716919, "num_tokens": 46848317.0, "step": 3215 }, { "epoch": 0.42505947660586835, "grad_norm": 0.8080988882040683, "learning_rate": 7.403297875613244e-06, "loss": 0.531, "mean_token_accuracy": 0.8256317973136902, "num_tokens": 46913853.0, "step": 3216 }, { "epoch": 0.42519164684113137, "grad_norm": 0.9114373901102021, "learning_rate": 7.401416280379974e-06, "loss": 0.5875, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 46979389.0, "step": 3217 }, { "epoch": 0.4253238170763944, "grad_norm": 1.022129248506183, "learning_rate": 7.399534280374876e-06, "loss": 0.6642, "mean_token_accuracy": 0.7820473909378052, "num_tokens": 47044925.0, "step": 3218 }, { "epoch": 0.4254559873116574, "grad_norm": 0.8356025631012817, "learning_rate": 7.3976518759985895e-06, "loss": 0.5699, "mean_token_accuracy": 0.8111036419868469, "num_tokens": 47110461.0, "step": 3219 }, { "epoch": 0.4255881575469204, "grad_norm": 0.8926016356875441, "learning_rate": 7.395769067651838e-06, "loss": 0.6148, "mean_token_accuracy": 0.7958124876022339, "num_tokens": 47175997.0, "step": 3220 }, { "epoch": 0.42572032778218344, "grad_norm": 0.9422283150839569, "learning_rate": 7.3938858557354335e-06, "loss": 0.6008, "mean_token_accuracy": 0.802252471446991, "num_tokens": 47241533.0, "step": 3221 }, { "epoch": 0.42585249801744646, "grad_norm": 0.9484772052537235, "learning_rate": 7.3920022406502725e-06, "loss": 0.696, "mean_token_accuracy": 0.7757599949836731, "num_tokens": 47307069.0, "step": 3222 }, { "epoch": 0.4259846682527095, "grad_norm": 0.8583125002588047, "learning_rate": 7.390118222797334e-06, "loss": 0.5904, "mean_token_accuracy": 0.8049994111061096, "num_tokens": 47372605.0, "step": 3223 }, { "epoch": 0.4261168384879725, "grad_norm": 0.8450230022916693, "learning_rate": 7.388233802577691e-06, "loss": 0.5631, "mean_token_accuracy": 0.8140794634819031, "num_tokens": 47438141.0, "step": 3224 }, { "epoch": 0.4262490087232355, "grad_norm": 0.9604610891089574, "learning_rate": 7.386348980392495e-06, "loss": 0.6243, "mean_token_accuracy": 0.7954156994819641, "num_tokens": 47503677.0, "step": 3225 }, { "epoch": 0.42638117895849853, "grad_norm": 0.8909117793852808, "learning_rate": 7.384463756642985e-06, "loss": 0.6361, "mean_token_accuracy": 0.7907611727714539, "num_tokens": 47569213.0, "step": 3226 }, { "epoch": 0.42651334919376155, "grad_norm": 0.9780189832339631, "learning_rate": 7.382578131730484e-06, "loss": 0.6441, "mean_token_accuracy": 0.7929587364196777, "num_tokens": 47634749.0, "step": 3227 }, { "epoch": 0.42664551942902457, "grad_norm": 0.8327135730898415, "learning_rate": 7.380692106056408e-06, "loss": 0.5886, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 47700285.0, "step": 3228 }, { "epoch": 0.4267776896642876, "grad_norm": 0.8895959081969241, "learning_rate": 7.3788056800222455e-06, "loss": 0.6302, "mean_token_accuracy": 0.7942101359367371, "num_tokens": 47765821.0, "step": 3229 }, { "epoch": 0.4269098598995506, "grad_norm": 0.9221627918196749, "learning_rate": 7.376918854029582e-06, "loss": 0.6625, "mean_token_accuracy": 0.783466637134552, "num_tokens": 47831357.0, "step": 3230 }, { "epoch": 0.4270420301348136, "grad_norm": 0.9009990206840757, "learning_rate": 7.375031628480079e-06, "loss": 0.6082, "mean_token_accuracy": 0.7995360493659973, "num_tokens": 47896893.0, "step": 3231 }, { "epoch": 0.42717420037007664, "grad_norm": 0.7652696641959382, "learning_rate": 7.373144003775493e-06, "loss": 0.563, "mean_token_accuracy": 0.8147814869880676, "num_tokens": 47962429.0, "step": 3232 }, { "epoch": 0.42730637060533966, "grad_norm": 0.8367469946815885, "learning_rate": 7.371255980317656e-06, "loss": 0.5776, "mean_token_accuracy": 0.8101117014884949, "num_tokens": 48027965.0, "step": 3233 }, { "epoch": 0.4274385408406027, "grad_norm": 0.9110210909837116, "learning_rate": 7.36936755850849e-06, "loss": 0.6522, "mean_token_accuracy": 0.7870223522186279, "num_tokens": 48093501.0, "step": 3234 }, { "epoch": 0.4275707110758657, "grad_norm": 0.7950593476075176, "learning_rate": 7.36747873875e-06, "loss": 0.5332, "mean_token_accuracy": 0.8254181146621704, "num_tokens": 48159037.0, "step": 3235 }, { "epoch": 0.4277028813111287, "grad_norm": 0.9760405728514999, "learning_rate": 7.365589521444277e-06, "loss": 0.6541, "mean_token_accuracy": 0.7859236001968384, "num_tokens": 48224573.0, "step": 3236 }, { "epoch": 0.42783505154639173, "grad_norm": 0.916926450791915, "learning_rate": 7.363699906993496e-06, "loss": 0.6088, "mean_token_accuracy": 0.802374541759491, "num_tokens": 48290109.0, "step": 3237 }, { "epoch": 0.42796722178165475, "grad_norm": 0.8729722430988867, "learning_rate": 7.3618098957999154e-06, "loss": 0.6009, "mean_token_accuracy": 0.8036412000656128, "num_tokens": 48355645.0, "step": 3238 }, { "epoch": 0.42809939201691777, "grad_norm": 0.9091656044340323, "learning_rate": 7.3599194882658786e-06, "loss": 0.6007, "mean_token_accuracy": 0.8035649061203003, "num_tokens": 48421181.0, "step": 3239 }, { "epoch": 0.4282315622521808, "grad_norm": 0.9427094448577488, "learning_rate": 7.358028684793816e-06, "loss": 0.6432, "mean_token_accuracy": 0.7887467741966248, "num_tokens": 48486717.0, "step": 3240 }, { "epoch": 0.4283637324874438, "grad_norm": 0.8395721893787187, "learning_rate": 7.3561374857862415e-06, "loss": 0.516, "mean_token_accuracy": 0.8308662176132202, "num_tokens": 48552253.0, "step": 3241 }, { "epoch": 0.4284959027227068, "grad_norm": 0.9353271116172789, "learning_rate": 7.354245891645747e-06, "loss": 0.647, "mean_token_accuracy": 0.7861982583999634, "num_tokens": 48617789.0, "step": 3242 }, { "epoch": 0.42862807295796984, "grad_norm": 0.911381574437715, "learning_rate": 7.3523539027750165e-06, "loss": 0.628, "mean_token_accuracy": 0.7919210195541382, "num_tokens": 48683325.0, "step": 3243 }, { "epoch": 0.42876024319323286, "grad_norm": 0.8327723918038475, "learning_rate": 7.350461519576816e-06, "loss": 0.5638, "mean_token_accuracy": 0.8169179558753967, "num_tokens": 48748861.0, "step": 3244 }, { "epoch": 0.4288924134284959, "grad_norm": 0.8698871646757206, "learning_rate": 7.348568742453991e-06, "loss": 0.592, "mean_token_accuracy": 0.8071206212043762, "num_tokens": 48814397.0, "step": 3245 }, { "epoch": 0.4290245836637589, "grad_norm": 0.9414579809448202, "learning_rate": 7.346675571809476e-06, "loss": 0.6337, "mean_token_accuracy": 0.7915394902229309, "num_tokens": 48879933.0, "step": 3246 }, { "epoch": 0.4291567538990219, "grad_norm": 0.928761428311003, "learning_rate": 7.34478200804629e-06, "loss": 0.6367, "mean_token_accuracy": 0.7906391024589539, "num_tokens": 48945469.0, "step": 3247 }, { "epoch": 0.42928892413428493, "grad_norm": 0.8583601515032622, "learning_rate": 7.342888051567529e-06, "loss": 0.6256, "mean_token_accuracy": 0.7947747707366943, "num_tokens": 49011005.0, "step": 3248 }, { "epoch": 0.42942109436954795, "grad_norm": 0.8503982541055327, "learning_rate": 7.340993702776379e-06, "loss": 0.5882, "mean_token_accuracy": 0.808448314666748, "num_tokens": 49076541.0, "step": 3249 }, { "epoch": 0.42955326460481097, "grad_norm": 0.8178282204956824, "learning_rate": 7.339098962076104e-06, "loss": 0.5475, "mean_token_accuracy": 0.8204126358032227, "num_tokens": 49142077.0, "step": 3250 }, { "epoch": 0.429685434840074, "grad_norm": 0.8567732223893999, "learning_rate": 7.337203829870061e-06, "loss": 0.5461, "mean_token_accuracy": 0.8219539523124695, "num_tokens": 49207613.0, "step": 3251 }, { "epoch": 0.42981760507533706, "grad_norm": 0.9370389143792145, "learning_rate": 7.33530830656168e-06, "loss": 0.6426, "mean_token_accuracy": 0.78831946849823, "num_tokens": 49273149.0, "step": 3252 }, { "epoch": 0.4299497753106001, "grad_norm": 0.8870414061757393, "learning_rate": 7.333412392554478e-06, "loss": 0.6267, "mean_token_accuracy": 0.7920888662338257, "num_tokens": 49338685.0, "step": 3253 }, { "epoch": 0.4300819455458631, "grad_norm": 0.8971529481794029, "learning_rate": 7.3315160882520575e-06, "loss": 0.6147, "mean_token_accuracy": 0.79891037940979, "num_tokens": 49404221.0, "step": 3254 }, { "epoch": 0.4302141157811261, "grad_norm": 0.8842780775711664, "learning_rate": 7.3296193940581005e-06, "loss": 0.6137, "mean_token_accuracy": 0.7980252504348755, "num_tokens": 49469757.0, "step": 3255 }, { "epoch": 0.43034628601638913, "grad_norm": 0.8756287683571835, "learning_rate": 7.327722310376374e-06, "loss": 0.6222, "mean_token_accuracy": 0.7951715588569641, "num_tokens": 49535293.0, "step": 3256 }, { "epoch": 0.43047845625165215, "grad_norm": 0.8800143661365081, "learning_rate": 7.32582483761073e-06, "loss": 0.6225, "mean_token_accuracy": 0.7955530285835266, "num_tokens": 49600829.0, "step": 3257 }, { "epoch": 0.43061062648691517, "grad_norm": 0.9230168125405839, "learning_rate": 7.323926976165095e-06, "loss": 0.6244, "mean_token_accuracy": 0.7961176633834839, "num_tokens": 49666365.0, "step": 3258 }, { "epoch": 0.4307427967221782, "grad_norm": 0.9886162913233241, "learning_rate": 7.322028726443491e-06, "loss": 0.6855, "mean_token_accuracy": 0.7807349562644958, "num_tokens": 49731901.0, "step": 3259 }, { "epoch": 0.4308749669574412, "grad_norm": 0.857027481903737, "learning_rate": 7.320130088850009e-06, "loss": 0.6144, "mean_token_accuracy": 0.79891037940979, "num_tokens": 49797437.0, "step": 3260 }, { "epoch": 0.4310071371927042, "grad_norm": 0.940841848879208, "learning_rate": 7.318231063788834e-06, "loss": 0.6325, "mean_token_accuracy": 0.7936607003211975, "num_tokens": 49862973.0, "step": 3261 }, { "epoch": 0.43113930742796724, "grad_norm": 0.8328689848260521, "learning_rate": 7.316331651664229e-06, "loss": 0.5792, "mean_token_accuracy": 0.808631420135498, "num_tokens": 49928509.0, "step": 3262 }, { "epoch": 0.43127147766323026, "grad_norm": 0.8980703741630227, "learning_rate": 7.314431852880537e-06, "loss": 0.6242, "mean_token_accuracy": 0.7943779826164246, "num_tokens": 49994045.0, "step": 3263 }, { "epoch": 0.4314036478984933, "grad_norm": 0.9244320589881436, "learning_rate": 7.312531667842185e-06, "loss": 0.6115, "mean_token_accuracy": 0.7973385453224182, "num_tokens": 50059581.0, "step": 3264 }, { "epoch": 0.4315358181337563, "grad_norm": 1.0086965818043898, "learning_rate": 7.310631096953685e-06, "loss": 0.6239, "mean_token_accuracy": 0.7944390177726746, "num_tokens": 50125117.0, "step": 3265 }, { "epoch": 0.4316679883690193, "grad_norm": 0.8385226348049457, "learning_rate": 7.308730140619628e-06, "loss": 0.5639, "mean_token_accuracy": 0.816521167755127, "num_tokens": 50190653.0, "step": 3266 }, { "epoch": 0.43180015860428234, "grad_norm": 0.9178904130832771, "learning_rate": 7.3068287992446875e-06, "loss": 0.5918, "mean_token_accuracy": 0.8045110702514648, "num_tokens": 50256189.0, "step": 3267 }, { "epoch": 0.43193232883954535, "grad_norm": 0.9436324682464055, "learning_rate": 7.304927073233619e-06, "loss": 0.6009, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 50321725.0, "step": 3268 }, { "epoch": 0.4320644990748084, "grad_norm": 0.9257673560522758, "learning_rate": 7.30302496299126e-06, "loss": 0.6429, "mean_token_accuracy": 0.7911579608917236, "num_tokens": 50387261.0, "step": 3269 }, { "epoch": 0.4321966693100714, "grad_norm": 0.8479669707575513, "learning_rate": 7.3011224689225324e-06, "loss": 0.5395, "mean_token_accuracy": 0.822930634021759, "num_tokens": 50452797.0, "step": 3270 }, { "epoch": 0.4323288395453344, "grad_norm": 0.9819996838689864, "learning_rate": 7.299219591432434e-06, "loss": 0.6472, "mean_token_accuracy": 0.7857556939125061, "num_tokens": 50518333.0, "step": 3271 }, { "epoch": 0.4324610097805974, "grad_norm": 0.8613142088277091, "learning_rate": 7.297316330926051e-06, "loss": 0.6101, "mean_token_accuracy": 0.7981473803520203, "num_tokens": 50583869.0, "step": 3272 }, { "epoch": 0.43259318001586045, "grad_norm": 0.8547356167993062, "learning_rate": 7.295412687808546e-06, "loss": 0.6233, "mean_token_accuracy": 0.7946068644523621, "num_tokens": 50649405.0, "step": 3273 }, { "epoch": 0.43272535025112346, "grad_norm": 0.8540635053086867, "learning_rate": 7.293508662485164e-06, "loss": 0.5971, "mean_token_accuracy": 0.8063270449638367, "num_tokens": 50714941.0, "step": 3274 }, { "epoch": 0.4328575204863865, "grad_norm": 0.7644827532489102, "learning_rate": 7.2916042553612355e-06, "loss": 0.5322, "mean_token_accuracy": 0.8263185024261475, "num_tokens": 50780477.0, "step": 3275 }, { "epoch": 0.4329896907216495, "grad_norm": 0.826963083167471, "learning_rate": 7.289699466842166e-06, "loss": 0.5514, "mean_token_accuracy": 0.8195886015892029, "num_tokens": 50846013.0, "step": 3276 }, { "epoch": 0.4331218609569125, "grad_norm": 0.8544561729509167, "learning_rate": 7.287794297333446e-06, "loss": 0.5996, "mean_token_accuracy": 0.8006958961486816, "num_tokens": 50911549.0, "step": 3277 }, { "epoch": 0.43325403119217554, "grad_norm": 0.8415312292611742, "learning_rate": 7.285888747240647e-06, "loss": 0.5902, "mean_token_accuracy": 0.8060829043388367, "num_tokens": 50977085.0, "step": 3278 }, { "epoch": 0.43338620142743856, "grad_norm": 0.8727829587786501, "learning_rate": 7.28398281696942e-06, "loss": 0.6129, "mean_token_accuracy": 0.7991240620613098, "num_tokens": 51042621.0, "step": 3279 }, { "epoch": 0.4335183716627016, "grad_norm": 0.8904426490148666, "learning_rate": 7.2820765069255e-06, "loss": 0.6235, "mean_token_accuracy": 0.7964991927146912, "num_tokens": 51108157.0, "step": 3280 }, { "epoch": 0.4336505418979646, "grad_norm": 0.8054889253748257, "learning_rate": 7.280169817514697e-06, "loss": 0.5793, "mean_token_accuracy": 0.8092113137245178, "num_tokens": 51173693.0, "step": 3281 }, { "epoch": 0.4337827121332276, "grad_norm": 0.8812219955275465, "learning_rate": 7.278262749142907e-06, "loss": 0.5996, "mean_token_accuracy": 0.803000271320343, "num_tokens": 51239229.0, "step": 3282 }, { "epoch": 0.43391488236849063, "grad_norm": 0.8400961852827032, "learning_rate": 7.276355302216106e-06, "loss": 0.6154, "mean_token_accuracy": 0.7953699231147766, "num_tokens": 51304765.0, "step": 3283 }, { "epoch": 0.43404705260375365, "grad_norm": 0.8893444084884393, "learning_rate": 7.274447477140349e-06, "loss": 0.5799, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 51370301.0, "step": 3284 }, { "epoch": 0.43417922283901667, "grad_norm": 0.8662332455240549, "learning_rate": 7.272539274321772e-06, "loss": 0.6626, "mean_token_accuracy": 0.7822457551956177, "num_tokens": 51435837.0, "step": 3285 }, { "epoch": 0.4343113930742797, "grad_norm": 0.9032024162316961, "learning_rate": 7.27063069416659e-06, "loss": 0.5783, "mean_token_accuracy": 0.8111036419868469, "num_tokens": 51501373.0, "step": 3286 }, { "epoch": 0.4344435633095427, "grad_norm": 0.9302480220466504, "learning_rate": 7.2687217370811025e-06, "loss": 0.6465, "mean_token_accuracy": 0.7913258671760559, "num_tokens": 51566909.0, "step": 3287 }, { "epoch": 0.4345757335448057, "grad_norm": 0.8670676217645967, "learning_rate": 7.266812403471686e-06, "loss": 0.5979, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 51632445.0, "step": 3288 }, { "epoch": 0.43470790378006874, "grad_norm": 0.9108980480487467, "learning_rate": 7.264902693744795e-06, "loss": 0.6199, "mean_token_accuracy": 0.7944085001945496, "num_tokens": 51697981.0, "step": 3289 }, { "epoch": 0.43484007401533176, "grad_norm": 0.8438164345326469, "learning_rate": 7.26299260830697e-06, "loss": 0.6507, "mean_token_accuracy": 0.7859693765640259, "num_tokens": 51763517.0, "step": 3290 }, { "epoch": 0.4349722442505948, "grad_norm": 0.8201745328190515, "learning_rate": 7.261082147564828e-06, "loss": 0.545, "mean_token_accuracy": 0.8205347061157227, "num_tokens": 51829053.0, "step": 3291 }, { "epoch": 0.4351044144858578, "grad_norm": 0.8814132256558048, "learning_rate": 7.259171311925065e-06, "loss": 0.631, "mean_token_accuracy": 0.7946678996086121, "num_tokens": 51894589.0, "step": 3292 }, { "epoch": 0.4352365847211208, "grad_norm": 0.8642792104462151, "learning_rate": 7.257260101794458e-06, "loss": 0.602, "mean_token_accuracy": 0.8043889403343201, "num_tokens": 51960125.0, "step": 3293 }, { "epoch": 0.43536875495638383, "grad_norm": 0.9333830844729066, "learning_rate": 7.2553485175798656e-06, "loss": 0.6126, "mean_token_accuracy": 0.7978116273880005, "num_tokens": 52025661.0, "step": 3294 }, { "epoch": 0.43550092519164685, "grad_norm": 0.847475353871511, "learning_rate": 7.253436559688223e-06, "loss": 0.6118, "mean_token_accuracy": 0.8014436364173889, "num_tokens": 52091197.0, "step": 3295 }, { "epoch": 0.43563309542690987, "grad_norm": 0.8580187487217393, "learning_rate": 7.2515242285265455e-06, "loss": 0.5783, "mean_token_accuracy": 0.8102642893791199, "num_tokens": 52156733.0, "step": 3296 }, { "epoch": 0.4357652656621729, "grad_norm": 0.9429754322215065, "learning_rate": 7.249611524501927e-06, "loss": 0.5998, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 52222269.0, "step": 3297 }, { "epoch": 0.4358974358974359, "grad_norm": 0.7891573113941555, "learning_rate": 7.247698448021546e-06, "loss": 0.5701, "mean_token_accuracy": 0.8109663128852844, "num_tokens": 52287805.0, "step": 3298 }, { "epoch": 0.4360296061326989, "grad_norm": 0.8762288433696022, "learning_rate": 7.245784999492655e-06, "loss": 0.596, "mean_token_accuracy": 0.8047094345092773, "num_tokens": 52353341.0, "step": 3299 }, { "epoch": 0.43616177636796194, "grad_norm": 0.9404566986442688, "learning_rate": 7.243871179322586e-06, "loss": 0.5998, "mean_token_accuracy": 0.8063423037528992, "num_tokens": 52418877.0, "step": 3300 }, { "epoch": 0.43629394660322496, "grad_norm": 0.9339823003280159, "learning_rate": 7.241956987918753e-06, "loss": 0.6233, "mean_token_accuracy": 0.7948968410491943, "num_tokens": 52484413.0, "step": 3301 }, { "epoch": 0.436426116838488, "grad_norm": 0.8499220224909407, "learning_rate": 7.240042425688647e-06, "loss": 0.5908, "mean_token_accuracy": 0.806708574295044, "num_tokens": 52549949.0, "step": 3302 }, { "epoch": 0.436558287073751, "grad_norm": 0.8699356444498649, "learning_rate": 7.238127493039837e-06, "loss": 0.5572, "mean_token_accuracy": 0.8191612958908081, "num_tokens": 52615485.0, "step": 3303 }, { "epoch": 0.436690457309014, "grad_norm": 0.9373209176451032, "learning_rate": 7.2362121903799746e-06, "loss": 0.6405, "mean_token_accuracy": 0.7902423143386841, "num_tokens": 52681021.0, "step": 3304 }, { "epoch": 0.43682262754427703, "grad_norm": 0.8387168244422138, "learning_rate": 7.234296518116785e-06, "loss": 0.5999, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 52746557.0, "step": 3305 }, { "epoch": 0.43695479777954005, "grad_norm": 0.896700449399951, "learning_rate": 7.232380476658077e-06, "loss": 0.579, "mean_token_accuracy": 0.8097606897354126, "num_tokens": 52812093.0, "step": 3306 }, { "epoch": 0.43708696801480307, "grad_norm": 0.8619835414439275, "learning_rate": 7.2304640664117355e-06, "loss": 0.5811, "mean_token_accuracy": 0.808753490447998, "num_tokens": 52877629.0, "step": 3307 }, { "epoch": 0.4372191382500661, "grad_norm": 0.9930698040592455, "learning_rate": 7.228547287785722e-06, "loss": 0.6597, "mean_token_accuracy": 0.7840160131454468, "num_tokens": 52943165.0, "step": 3308 }, { "epoch": 0.4373513084853291, "grad_norm": 0.8763817809040735, "learning_rate": 7.226630141188082e-06, "loss": 0.6181, "mean_token_accuracy": 0.7961787581443787, "num_tokens": 53008701.0, "step": 3309 }, { "epoch": 0.4374834787205921, "grad_norm": 0.902127565093498, "learning_rate": 7.224712627026934e-06, "loss": 0.5734, "mean_token_accuracy": 0.8100506663322449, "num_tokens": 53074237.0, "step": 3310 }, { "epoch": 0.43761564895585514, "grad_norm": 1.0229652445264408, "learning_rate": 7.222794745710476e-06, "loss": 0.68, "mean_token_accuracy": 0.7800024151802063, "num_tokens": 53139773.0, "step": 3311 }, { "epoch": 0.43774781919111816, "grad_norm": 1.2828537338019943, "learning_rate": 7.220876497646988e-06, "loss": 0.6775, "mean_token_accuracy": 0.7794530391693115, "num_tokens": 53205309.0, "step": 3312 }, { "epoch": 0.4378799894263812, "grad_norm": 0.9429344502900653, "learning_rate": 7.218957883244822e-06, "loss": 0.6217, "mean_token_accuracy": 0.7945916056632996, "num_tokens": 53270845.0, "step": 3313 }, { "epoch": 0.4380121596616442, "grad_norm": 0.8657575065720714, "learning_rate": 7.217038902912411e-06, "loss": 0.5846, "mean_token_accuracy": 0.8117598295211792, "num_tokens": 53336381.0, "step": 3314 }, { "epoch": 0.4381443298969072, "grad_norm": 0.8880969722508484, "learning_rate": 7.2151195570582685e-06, "loss": 0.6178, "mean_token_accuracy": 0.7971859574317932, "num_tokens": 53401917.0, "step": 3315 }, { "epoch": 0.43827650013217023, "grad_norm": 0.9303463112767072, "learning_rate": 7.213199846090979e-06, "loss": 0.5928, "mean_token_accuracy": 0.8047399520874023, "num_tokens": 53467453.0, "step": 3316 }, { "epoch": 0.43840867036743325, "grad_norm": 0.9497636037447116, "learning_rate": 7.211279770419213e-06, "loss": 0.5682, "mean_token_accuracy": 0.8160023093223572, "num_tokens": 53532989.0, "step": 3317 }, { "epoch": 0.43854084060269627, "grad_norm": 0.9634866699550997, "learning_rate": 7.209359330451713e-06, "loss": 0.6102, "mean_token_accuracy": 0.7993071675300598, "num_tokens": 53598525.0, "step": 3318 }, { "epoch": 0.4386730108379593, "grad_norm": 0.8773214209783862, "learning_rate": 7.207438526597299e-06, "loss": 0.5967, "mean_token_accuracy": 0.8069680333137512, "num_tokens": 53664061.0, "step": 3319 }, { "epoch": 0.4388051810732223, "grad_norm": 0.796247525177291, "learning_rate": 7.205517359264874e-06, "loss": 0.5387, "mean_token_accuracy": 0.8241972923278809, "num_tokens": 53729597.0, "step": 3320 }, { "epoch": 0.4389373513084853, "grad_norm": 0.9420999186923504, "learning_rate": 7.20359582886341e-06, "loss": 0.6618, "mean_token_accuracy": 0.7821694612503052, "num_tokens": 53795133.0, "step": 3321 }, { "epoch": 0.43906952154374834, "grad_norm": 0.9565070023910979, "learning_rate": 7.201673935801964e-06, "loss": 0.5762, "mean_token_accuracy": 0.8127670884132385, "num_tokens": 53860669.0, "step": 3322 }, { "epoch": 0.43920169177901136, "grad_norm": 0.8397696824919305, "learning_rate": 7.199751680489663e-06, "loss": 0.525, "mean_token_accuracy": 0.8272494077682495, "num_tokens": 53926205.0, "step": 3323 }, { "epoch": 0.4393338620142744, "grad_norm": 0.9439011883717713, "learning_rate": 7.197829063335721e-06, "loss": 0.6065, "mean_token_accuracy": 0.8015352487564087, "num_tokens": 53991741.0, "step": 3324 }, { "epoch": 0.4394660322495374, "grad_norm": 0.8799713810954557, "learning_rate": 7.195906084749415e-06, "loss": 0.6275, "mean_token_accuracy": 0.7950646877288818, "num_tokens": 54057277.0, "step": 3325 }, { "epoch": 0.4395982024848004, "grad_norm": 0.9412188303824959, "learning_rate": 7.193982745140116e-06, "loss": 0.6232, "mean_token_accuracy": 0.7957819700241089, "num_tokens": 54122813.0, "step": 3326 }, { "epoch": 0.43973037272006343, "grad_norm": 0.8588635832761465, "learning_rate": 7.192059044917255e-06, "loss": 0.5509, "mean_token_accuracy": 0.817360520362854, "num_tokens": 54188349.0, "step": 3327 }, { "epoch": 0.43986254295532645, "grad_norm": 0.9369229349231825, "learning_rate": 7.190134984490352e-06, "loss": 0.6361, "mean_token_accuracy": 0.7902117967605591, "num_tokens": 54253885.0, "step": 3328 }, { "epoch": 0.43999471319058947, "grad_norm": 0.9027192756786171, "learning_rate": 7.188210564268999e-06, "loss": 0.6219, "mean_token_accuracy": 0.7960718870162964, "num_tokens": 54319421.0, "step": 3329 }, { "epoch": 0.4401268834258525, "grad_norm": 0.9450481890722389, "learning_rate": 7.186285784662863e-06, "loss": 0.6064, "mean_token_accuracy": 0.7984983325004578, "num_tokens": 54384957.0, "step": 3330 }, { "epoch": 0.4402590536611155, "grad_norm": 0.7855601911208429, "learning_rate": 7.184360646081692e-06, "loss": 0.5643, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 54450493.0, "step": 3331 }, { "epoch": 0.4403912238963785, "grad_norm": 0.8047379609831865, "learning_rate": 7.182435148935305e-06, "loss": 0.5866, "mean_token_accuracy": 0.807761549949646, "num_tokens": 54516029.0, "step": 3332 }, { "epoch": 0.44052339413164154, "grad_norm": 0.8252847585895299, "learning_rate": 7.180509293633602e-06, "loss": 0.5747, "mean_token_accuracy": 0.8098523020744324, "num_tokens": 54581565.0, "step": 3333 }, { "epoch": 0.44065556436690456, "grad_norm": 0.9395944452777638, "learning_rate": 7.1785830805865566e-06, "loss": 0.6065, "mean_token_accuracy": 0.8005127310752869, "num_tokens": 54647101.0, "step": 3334 }, { "epoch": 0.4407877346021676, "grad_norm": 0.9027879785429453, "learning_rate": 7.17665651020422e-06, "loss": 0.5999, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 54712637.0, "step": 3335 }, { "epoch": 0.4409199048374306, "grad_norm": 0.8532516299257782, "learning_rate": 7.174729582896719e-06, "loss": 0.6149, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 54778173.0, "step": 3336 }, { "epoch": 0.4410520750726936, "grad_norm": 0.8526910756798224, "learning_rate": 7.172802299074255e-06, "loss": 0.5807, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 54843709.0, "step": 3337 }, { "epoch": 0.44118424530795664, "grad_norm": 0.8120401430013261, "learning_rate": 7.170874659147107e-06, "loss": 0.5709, "mean_token_accuracy": 0.8126602172851562, "num_tokens": 54909245.0, "step": 3338 }, { "epoch": 0.44131641554321965, "grad_norm": 0.9458287807990129, "learning_rate": 7.16894666352563e-06, "loss": 0.6406, "mean_token_accuracy": 0.7895708680152893, "num_tokens": 54974781.0, "step": 3339 }, { "epoch": 0.44144858577848267, "grad_norm": 0.8054838076151511, "learning_rate": 7.167018312620252e-06, "loss": 0.5809, "mean_token_accuracy": 0.8093181252479553, "num_tokens": 55040317.0, "step": 3340 }, { "epoch": 0.4415807560137457, "grad_norm": 0.9046583351126541, "learning_rate": 7.165089606841482e-06, "loss": 0.611, "mean_token_accuracy": 0.8023287653923035, "num_tokens": 55105853.0, "step": 3341 }, { "epoch": 0.4417129262490087, "grad_norm": 0.9040160846760112, "learning_rate": 7.1631605465998985e-06, "loss": 0.5967, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 55171389.0, "step": 3342 }, { "epoch": 0.4418450964842717, "grad_norm": 0.910968917388711, "learning_rate": 7.161231132306159e-06, "loss": 0.6006, "mean_token_accuracy": 0.8010469079017639, "num_tokens": 55236925.0, "step": 3343 }, { "epoch": 0.44197726671953474, "grad_norm": 0.9359530742633445, "learning_rate": 7.159301364370995e-06, "loss": 0.6455, "mean_token_accuracy": 0.7893572449684143, "num_tokens": 55302461.0, "step": 3344 }, { "epoch": 0.44210943695479776, "grad_norm": 0.9604148683215966, "learning_rate": 7.157371243205215e-06, "loss": 0.6239, "mean_token_accuracy": 0.7961634993553162, "num_tokens": 55367997.0, "step": 3345 }, { "epoch": 0.4422416071900608, "grad_norm": 0.8852048695940378, "learning_rate": 7.1554407692197e-06, "loss": 0.5563, "mean_token_accuracy": 0.8167806267738342, "num_tokens": 55433533.0, "step": 3346 }, { "epoch": 0.4423737774253238, "grad_norm": 0.8747289125408739, "learning_rate": 7.15350994282541e-06, "loss": 0.5703, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 55499069.0, "step": 3347 }, { "epoch": 0.4425059476605868, "grad_norm": 0.8208590029719847, "learning_rate": 7.151578764433374e-06, "loss": 0.6095, "mean_token_accuracy": 0.798009991645813, "num_tokens": 55564605.0, "step": 3348 }, { "epoch": 0.44263811789584984, "grad_norm": 0.778383578649842, "learning_rate": 7.149647234454702e-06, "loss": 0.5326, "mean_token_accuracy": 0.8267305493354797, "num_tokens": 55630141.0, "step": 3349 }, { "epoch": 0.44277028813111285, "grad_norm": 1.0109329246749468, "learning_rate": 7.147715353300575e-06, "loss": 0.6491, "mean_token_accuracy": 0.7856641411781311, "num_tokens": 55695677.0, "step": 3350 }, { "epoch": 0.4429024583663759, "grad_norm": 0.9144642995826161, "learning_rate": 7.145783121382253e-06, "loss": 0.582, "mean_token_accuracy": 0.8076242208480835, "num_tokens": 55761213.0, "step": 3351 }, { "epoch": 0.4430346286016389, "grad_norm": 0.9545759928674372, "learning_rate": 7.143850539111064e-06, "loss": 0.5536, "mean_token_accuracy": 0.8155597448348999, "num_tokens": 55826749.0, "step": 3352 }, { "epoch": 0.4431667988369019, "grad_norm": 0.9112206202308553, "learning_rate": 7.141917606898419e-06, "loss": 0.6476, "mean_token_accuracy": 0.7877548336982727, "num_tokens": 55892285.0, "step": 3353 }, { "epoch": 0.44329896907216493, "grad_norm": 0.831114658959035, "learning_rate": 7.139984325155793e-06, "loss": 0.5719, "mean_token_accuracy": 0.8112714886665344, "num_tokens": 55957821.0, "step": 3354 }, { "epoch": 0.44343113930742795, "grad_norm": 0.8664361256611341, "learning_rate": 7.138050694294748e-06, "loss": 0.5578, "mean_token_accuracy": 0.8177115321159363, "num_tokens": 56023357.0, "step": 3355 }, { "epoch": 0.44356330954269096, "grad_norm": 0.8933374480283032, "learning_rate": 7.136116714726908e-06, "loss": 0.5974, "mean_token_accuracy": 0.8024508357048035, "num_tokens": 56088893.0, "step": 3356 }, { "epoch": 0.443695479777954, "grad_norm": 0.8520900324437026, "learning_rate": 7.134182386863979e-06, "loss": 0.5722, "mean_token_accuracy": 0.8111189007759094, "num_tokens": 56154429.0, "step": 3357 }, { "epoch": 0.443827650013217, "grad_norm": 0.9007792812117027, "learning_rate": 7.13224771111774e-06, "loss": 0.6324, "mean_token_accuracy": 0.7958887815475464, "num_tokens": 56219965.0, "step": 3358 }, { "epoch": 0.44395982024848, "grad_norm": 0.8661973469786134, "learning_rate": 7.130312687900042e-06, "loss": 0.6149, "mean_token_accuracy": 0.8014283776283264, "num_tokens": 56285501.0, "step": 3359 }, { "epoch": 0.44409199048374304, "grad_norm": 1.1716179684818315, "learning_rate": 7.1283773176228115e-06, "loss": 0.6523, "mean_token_accuracy": 0.7861372232437134, "num_tokens": 56351037.0, "step": 3360 }, { "epoch": 0.44422416071900606, "grad_norm": 0.8148293478513609, "learning_rate": 7.126441600698048e-06, "loss": 0.584, "mean_token_accuracy": 0.8105847835540771, "num_tokens": 56416573.0, "step": 3361 }, { "epoch": 0.4443563309542691, "grad_norm": 0.9505589696274509, "learning_rate": 7.124505537537825e-06, "loss": 0.6191, "mean_token_accuracy": 0.7975369095802307, "num_tokens": 56482109.0, "step": 3362 }, { "epoch": 0.4444885011895321, "grad_norm": 0.8855302726440574, "learning_rate": 7.12256912855429e-06, "loss": 0.5874, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 56547645.0, "step": 3363 }, { "epoch": 0.4446206714247951, "grad_norm": 0.9505188127726578, "learning_rate": 7.120632374159665e-06, "loss": 0.6432, "mean_token_accuracy": 0.7915394902229309, "num_tokens": 56613181.0, "step": 3364 }, { "epoch": 0.44475284166005813, "grad_norm": 0.961883102905653, "learning_rate": 7.118695274766244e-06, "loss": 0.6417, "mean_token_accuracy": 0.7909138202667236, "num_tokens": 56678717.0, "step": 3365 }, { "epoch": 0.44488501189532115, "grad_norm": 0.9204282172667281, "learning_rate": 7.116757830786395e-06, "loss": 0.657, "mean_token_accuracy": 0.7830851078033447, "num_tokens": 56744253.0, "step": 3366 }, { "epoch": 0.44501718213058417, "grad_norm": 0.8484255990270116, "learning_rate": 7.114820042632559e-06, "loss": 0.5814, "mean_token_accuracy": 0.8090434670448303, "num_tokens": 56809789.0, "step": 3367 }, { "epoch": 0.4451493523658472, "grad_norm": 0.8803705067112977, "learning_rate": 7.1128819107172505e-06, "loss": 0.6202, "mean_token_accuracy": 0.7968044281005859, "num_tokens": 56875325.0, "step": 3368 }, { "epoch": 0.4452815226011102, "grad_norm": 0.8407493355658211, "learning_rate": 7.110943435453057e-06, "loss": 0.5569, "mean_token_accuracy": 0.816612720489502, "num_tokens": 56940861.0, "step": 3369 }, { "epoch": 0.4454136928363733, "grad_norm": 0.8598566956180362, "learning_rate": 7.109004617252641e-06, "loss": 0.6596, "mean_token_accuracy": 0.7842143774032593, "num_tokens": 57006397.0, "step": 3370 }, { "epoch": 0.4455458630716363, "grad_norm": 0.7998423889842344, "learning_rate": 7.107065456528737e-06, "loss": 0.5441, "mean_token_accuracy": 0.8207026124000549, "num_tokens": 57071933.0, "step": 3371 }, { "epoch": 0.4456780333068993, "grad_norm": 0.8518234501785688, "learning_rate": 7.105125953694149e-06, "loss": 0.6224, "mean_token_accuracy": 0.7927756309509277, "num_tokens": 57137469.0, "step": 3372 }, { "epoch": 0.44581020354216233, "grad_norm": 0.7652595157469801, "learning_rate": 7.103186109161759e-06, "loss": 0.5596, "mean_token_accuracy": 0.8136063814163208, "num_tokens": 57203005.0, "step": 3373 }, { "epoch": 0.44594237377742535, "grad_norm": 0.7863070148080893, "learning_rate": 7.101245923344518e-06, "loss": 0.5547, "mean_token_accuracy": 0.8170552849769592, "num_tokens": 57268541.0, "step": 3374 }, { "epoch": 0.44607454401268837, "grad_norm": 0.9134539358935263, "learning_rate": 7.099305396655451e-06, "loss": 0.5796, "mean_token_accuracy": 0.8109968304634094, "num_tokens": 57334077.0, "step": 3375 }, { "epoch": 0.4462067142479514, "grad_norm": 0.8863843540015118, "learning_rate": 7.097364529507657e-06, "loss": 0.6068, "mean_token_accuracy": 0.7986509799957275, "num_tokens": 57399613.0, "step": 3376 }, { "epoch": 0.4463388844832144, "grad_norm": 0.8036685884738248, "learning_rate": 7.095423322314303e-06, "loss": 0.562, "mean_token_accuracy": 0.8130417466163635, "num_tokens": 57465149.0, "step": 3377 }, { "epoch": 0.4464710547184774, "grad_norm": 0.8896457074995522, "learning_rate": 7.093481775488637e-06, "loss": 0.6182, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 57530685.0, "step": 3378 }, { "epoch": 0.44660322495374044, "grad_norm": 0.9056177955642021, "learning_rate": 7.091539889443968e-06, "loss": 0.5772, "mean_token_accuracy": 0.8095928430557251, "num_tokens": 57596221.0, "step": 3379 }, { "epoch": 0.44673539518900346, "grad_norm": 0.8735042309777248, "learning_rate": 7.089597664593686e-06, "loss": 0.5701, "mean_token_accuracy": 0.8106458187103271, "num_tokens": 57661757.0, "step": 3380 }, { "epoch": 0.4468675654242665, "grad_norm": 0.9019183013954404, "learning_rate": 7.0876551013512514e-06, "loss": 0.6442, "mean_token_accuracy": 0.7893725037574768, "num_tokens": 57727293.0, "step": 3381 }, { "epoch": 0.4469997356595295, "grad_norm": 0.8492234820844092, "learning_rate": 7.0857122001301915e-06, "loss": 0.5723, "mean_token_accuracy": 0.8122787475585938, "num_tokens": 57792829.0, "step": 3382 }, { "epoch": 0.4471319058947925, "grad_norm": 0.9330830496390086, "learning_rate": 7.083768961344112e-06, "loss": 0.6138, "mean_token_accuracy": 0.7975216507911682, "num_tokens": 57858365.0, "step": 3383 }, { "epoch": 0.44726407613005553, "grad_norm": 0.8656457294605653, "learning_rate": 7.081825385406689e-06, "loss": 0.571, "mean_token_accuracy": 0.8118667006492615, "num_tokens": 57923901.0, "step": 3384 }, { "epoch": 0.44739624636531855, "grad_norm": 0.8746299757858282, "learning_rate": 7.079881472731666e-06, "loss": 0.5931, "mean_token_accuracy": 0.8040532469749451, "num_tokens": 57989437.0, "step": 3385 }, { "epoch": 0.44752841660058157, "grad_norm": 0.7794902007809151, "learning_rate": 7.077937223732863e-06, "loss": 0.5662, "mean_token_accuracy": 0.8144609928131104, "num_tokens": 58054973.0, "step": 3386 }, { "epoch": 0.4476605868358446, "grad_norm": 0.7685786362143567, "learning_rate": 7.075992638824169e-06, "loss": 0.5274, "mean_token_accuracy": 0.8270663022994995, "num_tokens": 58120509.0, "step": 3387 }, { "epoch": 0.4477927570711076, "grad_norm": 0.8722058771549791, "learning_rate": 7.074047718419547e-06, "loss": 0.6189, "mean_token_accuracy": 0.7977200746536255, "num_tokens": 58186045.0, "step": 3388 }, { "epoch": 0.4479249273063706, "grad_norm": 0.8848039345562646, "learning_rate": 7.072102462933032e-06, "loss": 0.5911, "mean_token_accuracy": 0.8025577068328857, "num_tokens": 58251581.0, "step": 3389 }, { "epoch": 0.44805709754163364, "grad_norm": 0.9058880674824606, "learning_rate": 7.070156872778723e-06, "loss": 0.6333, "mean_token_accuracy": 0.7911427021026611, "num_tokens": 58317117.0, "step": 3390 }, { "epoch": 0.44818926777689666, "grad_norm": 0.9483481563331045, "learning_rate": 7.0682109483707995e-06, "loss": 0.6059, "mean_token_accuracy": 0.8000854849815369, "num_tokens": 58382653.0, "step": 3391 }, { "epoch": 0.4483214380121597, "grad_norm": 0.936404317471259, "learning_rate": 7.066264690123508e-06, "loss": 0.6561, "mean_token_accuracy": 0.7842143774032593, "num_tokens": 58448189.0, "step": 3392 }, { "epoch": 0.4484536082474227, "grad_norm": 0.8453848698414546, "learning_rate": 7.0643180984511624e-06, "loss": 0.5859, "mean_token_accuracy": 0.8090892434120178, "num_tokens": 58513725.0, "step": 3393 }, { "epoch": 0.4485857784826857, "grad_norm": 0.8869193350441169, "learning_rate": 7.062371173768156e-06, "loss": 0.6005, "mean_token_accuracy": 0.8023592829704285, "num_tokens": 58579261.0, "step": 3394 }, { "epoch": 0.44871794871794873, "grad_norm": 0.9582223813145001, "learning_rate": 7.0604239164889465e-06, "loss": 0.6541, "mean_token_accuracy": 0.7854505181312561, "num_tokens": 58644797.0, "step": 3395 }, { "epoch": 0.44885011895321175, "grad_norm": 0.8343484796223914, "learning_rate": 7.058476327028064e-06, "loss": 0.5257, "mean_token_accuracy": 0.8237547278404236, "num_tokens": 58710333.0, "step": 3396 }, { "epoch": 0.44898228918847477, "grad_norm": 0.8858129980267447, "learning_rate": 7.056528405800111e-06, "loss": 0.6359, "mean_token_accuracy": 0.7901965379714966, "num_tokens": 58775869.0, "step": 3397 }, { "epoch": 0.4491144594237378, "grad_norm": 0.8882645259106112, "learning_rate": 7.054580153219758e-06, "loss": 0.5891, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 58841405.0, "step": 3398 }, { "epoch": 0.4492466296590008, "grad_norm": 0.9527012230295027, "learning_rate": 7.052631569701748e-06, "loss": 0.5977, "mean_token_accuracy": 0.8034733533859253, "num_tokens": 58906941.0, "step": 3399 }, { "epoch": 0.4493787998942638, "grad_norm": 0.8491591408469795, "learning_rate": 7.050682655660893e-06, "loss": 0.5811, "mean_token_accuracy": 0.811790406703949, "num_tokens": 58972477.0, "step": 3400 }, { "epoch": 0.44951097012952684, "grad_norm": 0.7613973149012728, "learning_rate": 7.048733411512077e-06, "loss": 0.5504, "mean_token_accuracy": 0.8193291425704956, "num_tokens": 59038013.0, "step": 3401 }, { "epoch": 0.44964314036478986, "grad_norm": 0.8189173695467769, "learning_rate": 7.046783837670253e-06, "loss": 0.5597, "mean_token_accuracy": 0.8150408864021301, "num_tokens": 59103549.0, "step": 3402 }, { "epoch": 0.4497753106000529, "grad_norm": 0.8131343311088466, "learning_rate": 7.044833934550445e-06, "loss": 0.5825, "mean_token_accuracy": 0.8125076293945312, "num_tokens": 59169085.0, "step": 3403 }, { "epoch": 0.4499074808353159, "grad_norm": 0.8315156610720712, "learning_rate": 7.042883702567744e-06, "loss": 0.6138, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 59234621.0, "step": 3404 }, { "epoch": 0.4500396510705789, "grad_norm": 0.9330127555224799, "learning_rate": 7.040933142137318e-06, "loss": 0.6154, "mean_token_accuracy": 0.7970790863037109, "num_tokens": 59300157.0, "step": 3405 }, { "epoch": 0.45017182130584193, "grad_norm": 0.8788176460757212, "learning_rate": 7.0389822536743955e-06, "loss": 0.5433, "mean_token_accuracy": 0.8205041885375977, "num_tokens": 59365693.0, "step": 3406 }, { "epoch": 0.45030399154110495, "grad_norm": 0.9020379038730321, "learning_rate": 7.0370310375942855e-06, "loss": 0.6562, "mean_token_accuracy": 0.7841228246688843, "num_tokens": 59431229.0, "step": 3407 }, { "epoch": 0.45043616177636797, "grad_norm": 0.9838637259129992, "learning_rate": 7.035079494312358e-06, "loss": 0.6032, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 59496765.0, "step": 3408 }, { "epoch": 0.450568332011631, "grad_norm": 0.9455802093610423, "learning_rate": 7.033127624244055e-06, "loss": 0.7003, "mean_token_accuracy": 0.7711665034294128, "num_tokens": 59562301.0, "step": 3409 }, { "epoch": 0.450700502246894, "grad_norm": 0.9578505576864474, "learning_rate": 7.031175427804892e-06, "loss": 0.606, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 59627837.0, "step": 3410 }, { "epoch": 0.450832672482157, "grad_norm": 0.9152169691622407, "learning_rate": 7.0292229054104475e-06, "loss": 0.624, "mean_token_accuracy": 0.7956293225288391, "num_tokens": 59693373.0, "step": 3411 }, { "epoch": 0.45096484271742004, "grad_norm": 0.8610389540054643, "learning_rate": 7.027270057476375e-06, "loss": 0.5438, "mean_token_accuracy": 0.8221065998077393, "num_tokens": 59758909.0, "step": 3412 }, { "epoch": 0.45109701295268306, "grad_norm": 0.873932430793716, "learning_rate": 7.025316884418395e-06, "loss": 0.5744, "mean_token_accuracy": 0.8087382316589355, "num_tokens": 59824445.0, "step": 3413 }, { "epoch": 0.4512291831879461, "grad_norm": 0.997077614629601, "learning_rate": 7.023363386652295e-06, "loss": 0.6095, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 59889981.0, "step": 3414 }, { "epoch": 0.4513613534232091, "grad_norm": 0.8534709612221784, "learning_rate": 7.021409564593938e-06, "loss": 0.5833, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 59955517.0, "step": 3415 }, { "epoch": 0.4514935236584721, "grad_norm": 0.936090551296066, "learning_rate": 7.019455418659248e-06, "loss": 0.625, "mean_token_accuracy": 0.7972317337989807, "num_tokens": 60021053.0, "step": 3416 }, { "epoch": 0.45162569389373514, "grad_norm": 0.8480618221446676, "learning_rate": 7.0175009492642236e-06, "loss": 0.5741, "mean_token_accuracy": 0.8102642893791199, "num_tokens": 60086589.0, "step": 3417 }, { "epoch": 0.45175786412899815, "grad_norm": 0.8827474909710797, "learning_rate": 7.01554615682493e-06, "loss": 0.6195, "mean_token_accuracy": 0.7945458292961121, "num_tokens": 60152125.0, "step": 3418 }, { "epoch": 0.4518900343642612, "grad_norm": 1.007943182050888, "learning_rate": 7.013591041757502e-06, "loss": 0.631, "mean_token_accuracy": 0.7919667959213257, "num_tokens": 60217661.0, "step": 3419 }, { "epoch": 0.4520222045995242, "grad_norm": 0.9656860532197789, "learning_rate": 7.011635604478143e-06, "loss": 0.6458, "mean_token_accuracy": 0.7869307994842529, "num_tokens": 60283197.0, "step": 3420 }, { "epoch": 0.4521543748347872, "grad_norm": 0.8825450237344272, "learning_rate": 7.009679845403124e-06, "loss": 0.587, "mean_token_accuracy": 0.8072579503059387, "num_tokens": 60348733.0, "step": 3421 }, { "epoch": 0.4522865450700502, "grad_norm": 0.9266947485494035, "learning_rate": 7.007723764948786e-06, "loss": 0.5866, "mean_token_accuracy": 0.807608962059021, "num_tokens": 60414269.0, "step": 3422 }, { "epoch": 0.45241871530531325, "grad_norm": 0.897454919995236, "learning_rate": 7.005767363531537e-06, "loss": 0.5742, "mean_token_accuracy": 0.8113936185836792, "num_tokens": 60479805.0, "step": 3423 }, { "epoch": 0.45255088554057626, "grad_norm": 0.9381853936804898, "learning_rate": 7.003810641567856e-06, "loss": 0.6382, "mean_token_accuracy": 0.7906085848808289, "num_tokens": 60545341.0, "step": 3424 }, { "epoch": 0.4526830557758393, "grad_norm": 0.8636214058569712, "learning_rate": 7.001853599474284e-06, "loss": 0.5779, "mean_token_accuracy": 0.808631420135498, "num_tokens": 60610877.0, "step": 3425 }, { "epoch": 0.4528152260111023, "grad_norm": 1.058478648871374, "learning_rate": 6.999896237667439e-06, "loss": 0.6885, "mean_token_accuracy": 0.7767584323883057, "num_tokens": 60672688.0, "step": 3426 }, { "epoch": 0.4529473962463653, "grad_norm": 0.8365368421715248, "learning_rate": 6.997938556564001e-06, "loss": 0.5808, "mean_token_accuracy": 0.8116072416305542, "num_tokens": 60738224.0, "step": 3427 }, { "epoch": 0.45307956648162834, "grad_norm": 0.8483549765495222, "learning_rate": 6.995980556580717e-06, "loss": 0.5502, "mean_token_accuracy": 0.8191002607345581, "num_tokens": 60803760.0, "step": 3428 }, { "epoch": 0.45321173671689136, "grad_norm": 0.8772423022818377, "learning_rate": 6.994022238134408e-06, "loss": 0.6333, "mean_token_accuracy": 0.7937828302383423, "num_tokens": 60869296.0, "step": 3429 }, { "epoch": 0.4533439069521544, "grad_norm": 0.8492857246317391, "learning_rate": 6.992063601641956e-06, "loss": 0.5661, "mean_token_accuracy": 0.8147509694099426, "num_tokens": 60934832.0, "step": 3430 }, { "epoch": 0.4534760771874174, "grad_norm": 0.9135182939411906, "learning_rate": 6.990104647520319e-06, "loss": 0.6349, "mean_token_accuracy": 0.7917225956916809, "num_tokens": 61000368.0, "step": 3431 }, { "epoch": 0.4536082474226804, "grad_norm": 0.8473099696506442, "learning_rate": 6.988145376186509e-06, "loss": 0.5806, "mean_token_accuracy": 0.8070901036262512, "num_tokens": 61065904.0, "step": 3432 }, { "epoch": 0.45374041765794343, "grad_norm": 0.9017604004979927, "learning_rate": 6.986185788057621e-06, "loss": 0.5961, "mean_token_accuracy": 0.8023898005485535, "num_tokens": 61131440.0, "step": 3433 }, { "epoch": 0.45387258789320645, "grad_norm": 0.8633528248150059, "learning_rate": 6.984225883550808e-06, "loss": 0.5973, "mean_token_accuracy": 0.8018556833267212, "num_tokens": 61196976.0, "step": 3434 }, { "epoch": 0.45400475812846947, "grad_norm": 0.9081331979066047, "learning_rate": 6.982265663083291e-06, "loss": 0.609, "mean_token_accuracy": 0.7997344732284546, "num_tokens": 61262512.0, "step": 3435 }, { "epoch": 0.4541369283637325, "grad_norm": 0.8277055748934014, "learning_rate": 6.980305127072363e-06, "loss": 0.5636, "mean_token_accuracy": 0.8129807114601135, "num_tokens": 61328048.0, "step": 3436 }, { "epoch": 0.4542690985989955, "grad_norm": 0.904758006132997, "learning_rate": 6.978344275935381e-06, "loss": 0.6073, "mean_token_accuracy": 0.802069365978241, "num_tokens": 61393584.0, "step": 3437 }, { "epoch": 0.4544012688342585, "grad_norm": 0.8308020897479247, "learning_rate": 6.976383110089768e-06, "loss": 0.5906, "mean_token_accuracy": 0.808570384979248, "num_tokens": 61459120.0, "step": 3438 }, { "epoch": 0.45453343906952154, "grad_norm": 0.9763105407458625, "learning_rate": 6.9744216299530155e-06, "loss": 0.616, "mean_token_accuracy": 0.797887921333313, "num_tokens": 61524656.0, "step": 3439 }, { "epoch": 0.45466560930478456, "grad_norm": 0.7800470204040059, "learning_rate": 6.972459835942681e-06, "loss": 0.5347, "mean_token_accuracy": 0.8221523761749268, "num_tokens": 61590192.0, "step": 3440 }, { "epoch": 0.4547977795400476, "grad_norm": 0.8885806985986868, "learning_rate": 6.970497728476391e-06, "loss": 0.5824, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 61655728.0, "step": 3441 }, { "epoch": 0.4549299497753106, "grad_norm": 0.8203084106858186, "learning_rate": 6.968535307971835e-06, "loss": 0.5676, "mean_token_accuracy": 0.8134690523147583, "num_tokens": 61721264.0, "step": 3442 }, { "epoch": 0.4550621200105736, "grad_norm": 0.9050197779454735, "learning_rate": 6.966572574846774e-06, "loss": 0.6091, "mean_token_accuracy": 0.8002991080284119, "num_tokens": 61786800.0, "step": 3443 }, { "epoch": 0.45519429024583663, "grad_norm": 0.8529928172541176, "learning_rate": 6.964609529519031e-06, "loss": 0.5511, "mean_token_accuracy": 0.818596601486206, "num_tokens": 61852336.0, "step": 3444 }, { "epoch": 0.45532646048109965, "grad_norm": 0.8416322926174409, "learning_rate": 6.962646172406499e-06, "loss": 0.5836, "mean_token_accuracy": 0.8068917393684387, "num_tokens": 61917872.0, "step": 3445 }, { "epoch": 0.45545863071636267, "grad_norm": 0.9367079142741862, "learning_rate": 6.960682503927133e-06, "loss": 0.5998, "mean_token_accuracy": 0.8001617789268494, "num_tokens": 61983408.0, "step": 3446 }, { "epoch": 0.4555908009516257, "grad_norm": 0.9019173237420435, "learning_rate": 6.95871852449896e-06, "loss": 0.5687, "mean_token_accuracy": 0.8127059936523438, "num_tokens": 62048944.0, "step": 3447 }, { "epoch": 0.4557229711868887, "grad_norm": 0.9178970448424302, "learning_rate": 6.956754234540069e-06, "loss": 0.5836, "mean_token_accuracy": 0.808600902557373, "num_tokens": 62114480.0, "step": 3448 }, { "epoch": 0.4558551414221517, "grad_norm": 0.9160036436888737, "learning_rate": 6.954789634468615e-06, "loss": 0.5965, "mean_token_accuracy": 0.8017946481704712, "num_tokens": 62180016.0, "step": 3449 }, { "epoch": 0.45598731165741474, "grad_norm": 0.8984395510056308, "learning_rate": 6.952824724702822e-06, "loss": 0.6178, "mean_token_accuracy": 0.7958277463912964, "num_tokens": 62245552.0, "step": 3450 }, { "epoch": 0.45611948189267776, "grad_norm": 1.0018620647088925, "learning_rate": 6.950859505660979e-06, "loss": 0.6247, "mean_token_accuracy": 0.7952478528022766, "num_tokens": 62311088.0, "step": 3451 }, { "epoch": 0.4562516521279408, "grad_norm": 0.8786080927325617, "learning_rate": 6.948893977761439e-06, "loss": 0.5998, "mean_token_accuracy": 0.8003143668174744, "num_tokens": 62376624.0, "step": 3452 }, { "epoch": 0.4563838223632038, "grad_norm": 0.918402332137423, "learning_rate": 6.946928141422621e-06, "loss": 0.5964, "mean_token_accuracy": 0.8052740693092346, "num_tokens": 62442160.0, "step": 3453 }, { "epoch": 0.4565159925984668, "grad_norm": 0.9169633707379914, "learning_rate": 6.944961997063011e-06, "loss": 0.6307, "mean_token_accuracy": 0.792653501033783, "num_tokens": 62507696.0, "step": 3454 }, { "epoch": 0.45664816283372983, "grad_norm": 0.8156772117932986, "learning_rate": 6.9429955451011625e-06, "loss": 0.5572, "mean_token_accuracy": 0.8136063814163208, "num_tokens": 62573232.0, "step": 3455 }, { "epoch": 0.45678033306899285, "grad_norm": 0.7767819386124196, "learning_rate": 6.941028785955688e-06, "loss": 0.5341, "mean_token_accuracy": 0.8259064555168152, "num_tokens": 62638768.0, "step": 3456 }, { "epoch": 0.45691250330425587, "grad_norm": 1.0842052391428136, "learning_rate": 6.939061720045272e-06, "loss": 0.6619, "mean_token_accuracy": 0.783680260181427, "num_tokens": 62704304.0, "step": 3457 }, { "epoch": 0.4570446735395189, "grad_norm": 0.8726294249117452, "learning_rate": 6.937094347788662e-06, "loss": 0.6225, "mean_token_accuracy": 0.7953699231147766, "num_tokens": 62769840.0, "step": 3458 }, { "epoch": 0.4571768437747819, "grad_norm": 0.8663930278763297, "learning_rate": 6.935126669604668e-06, "loss": 0.6085, "mean_token_accuracy": 0.8017946481704712, "num_tokens": 62835376.0, "step": 3459 }, { "epoch": 0.4573090140100449, "grad_norm": 0.9250678718610392, "learning_rate": 6.933158685912172e-06, "loss": 0.6165, "mean_token_accuracy": 0.7942253947257996, "num_tokens": 62900912.0, "step": 3460 }, { "epoch": 0.45744118424530794, "grad_norm": 0.911663875225032, "learning_rate": 6.931190397130113e-06, "loss": 0.5886, "mean_token_accuracy": 0.806708574295044, "num_tokens": 62966448.0, "step": 3461 }, { "epoch": 0.45757335448057096, "grad_norm": 1.2622461653672492, "learning_rate": 6.9292218036774985e-06, "loss": 0.6037, "mean_token_accuracy": 0.8071663975715637, "num_tokens": 63031984.0, "step": 3462 }, { "epoch": 0.457705524715834, "grad_norm": 0.9094436905586634, "learning_rate": 6.927252905973403e-06, "loss": 0.6148, "mean_token_accuracy": 0.7983304858207703, "num_tokens": 63097520.0, "step": 3463 }, { "epoch": 0.457837694951097, "grad_norm": 0.9932319656362807, "learning_rate": 6.925283704436963e-06, "loss": 0.6008, "mean_token_accuracy": 0.8013520836830139, "num_tokens": 63163056.0, "step": 3464 }, { "epoch": 0.45796986518636, "grad_norm": 0.9421415423476927, "learning_rate": 6.923314199487378e-06, "loss": 0.6324, "mean_token_accuracy": 0.7924856543540955, "num_tokens": 63228592.0, "step": 3465 }, { "epoch": 0.45810203542162303, "grad_norm": 0.8447218772247598, "learning_rate": 6.9213443915439196e-06, "loss": 0.5764, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 63294128.0, "step": 3466 }, { "epoch": 0.45823420565688605, "grad_norm": 0.9035632053556418, "learning_rate": 6.919374281025914e-06, "loss": 0.5692, "mean_token_accuracy": 0.8124313354492188, "num_tokens": 63359664.0, "step": 3467 }, { "epoch": 0.45836637589214907, "grad_norm": 0.9352461754771605, "learning_rate": 6.917403868352756e-06, "loss": 0.6451, "mean_token_accuracy": 0.7894182801246643, "num_tokens": 63425200.0, "step": 3468 }, { "epoch": 0.4584985461274121, "grad_norm": 0.8106617211845019, "learning_rate": 6.915433153943911e-06, "loss": 0.57, "mean_token_accuracy": 0.8161702156066895, "num_tokens": 63490736.0, "step": 3469 }, { "epoch": 0.4586307163626751, "grad_norm": 0.8596703555916163, "learning_rate": 6.913462138218899e-06, "loss": 0.5672, "mean_token_accuracy": 0.8146898746490479, "num_tokens": 63556272.0, "step": 3470 }, { "epoch": 0.4587628865979381, "grad_norm": 1.025164934761193, "learning_rate": 6.9114908215973084e-06, "loss": 0.6014, "mean_token_accuracy": 0.8011994957923889, "num_tokens": 63621808.0, "step": 3471 }, { "epoch": 0.45889505683320114, "grad_norm": 0.91269399595204, "learning_rate": 6.9095192044987934e-06, "loss": 0.5293, "mean_token_accuracy": 0.8259217143058777, "num_tokens": 63687344.0, "step": 3472 }, { "epoch": 0.45902722706846416, "grad_norm": 0.8271939244651868, "learning_rate": 6.907547287343066e-06, "loss": 0.5795, "mean_token_accuracy": 0.8127059936523438, "num_tokens": 63752880.0, "step": 3473 }, { "epoch": 0.4591593973037272, "grad_norm": 0.8858326257587058, "learning_rate": 6.90557507054991e-06, "loss": 0.6067, "mean_token_accuracy": 0.8004974722862244, "num_tokens": 63818416.0, "step": 3474 }, { "epoch": 0.4592915675389902, "grad_norm": 0.9073696628857848, "learning_rate": 6.903602554539166e-06, "loss": 0.6013, "mean_token_accuracy": 0.8025882244110107, "num_tokens": 63883952.0, "step": 3475 }, { "epoch": 0.4594237377742532, "grad_norm": 0.8487185497965252, "learning_rate": 6.901629739730744e-06, "loss": 0.5894, "mean_token_accuracy": 0.8047246932983398, "num_tokens": 63949488.0, "step": 3476 }, { "epoch": 0.45955590800951623, "grad_norm": 0.8712432928306204, "learning_rate": 6.899656626544616e-06, "loss": 0.5759, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 64015024.0, "step": 3477 }, { "epoch": 0.45968807824477925, "grad_norm": 0.9575963074795784, "learning_rate": 6.897683215400813e-06, "loss": 0.6328, "mean_token_accuracy": 0.7896166443824768, "num_tokens": 64080560.0, "step": 3478 }, { "epoch": 0.45982024848004227, "grad_norm": 0.8391978767420358, "learning_rate": 6.895709506719437e-06, "loss": 0.6125, "mean_token_accuracy": 0.7990935444831848, "num_tokens": 64146096.0, "step": 3479 }, { "epoch": 0.4599524187153053, "grad_norm": 0.7912178137119488, "learning_rate": 6.893735500920646e-06, "loss": 0.5929, "mean_token_accuracy": 0.8059150576591492, "num_tokens": 64211632.0, "step": 3480 }, { "epoch": 0.4600845889505683, "grad_norm": 0.8825139944901793, "learning_rate": 6.891761198424666e-06, "loss": 0.6043, "mean_token_accuracy": 0.8015657663345337, "num_tokens": 64277168.0, "step": 3481 }, { "epoch": 0.4602167591858313, "grad_norm": 0.7619095173904681, "learning_rate": 6.889786599651786e-06, "loss": 0.5296, "mean_token_accuracy": 0.8248382210731506, "num_tokens": 64342704.0, "step": 3482 }, { "epoch": 0.46034892942109434, "grad_norm": 0.9548141243939867, "learning_rate": 6.887811705022352e-06, "loss": 0.6206, "mean_token_accuracy": 0.7968807220458984, "num_tokens": 64408240.0, "step": 3483 }, { "epoch": 0.46048109965635736, "grad_norm": 0.8413015371040177, "learning_rate": 6.885836514956785e-06, "loss": 0.574, "mean_token_accuracy": 0.8080820441246033, "num_tokens": 64473776.0, "step": 3484 }, { "epoch": 0.4606132698916204, "grad_norm": 0.8026503019154501, "learning_rate": 6.883861029875557e-06, "loss": 0.5719, "mean_token_accuracy": 0.812782347202301, "num_tokens": 64539312.0, "step": 3485 }, { "epoch": 0.4607454401268834, "grad_norm": 0.9022597550876099, "learning_rate": 6.881885250199208e-06, "loss": 0.6183, "mean_token_accuracy": 0.7975216507911682, "num_tokens": 64604848.0, "step": 3486 }, { "epoch": 0.4608776103621464, "grad_norm": 1.0104240508855633, "learning_rate": 6.879909176348342e-06, "loss": 0.6216, "mean_token_accuracy": 0.7946221232414246, "num_tokens": 64670384.0, "step": 3487 }, { "epoch": 0.4610097805974095, "grad_norm": 0.8685846574945396, "learning_rate": 6.877932808743622e-06, "loss": 0.5991, "mean_token_accuracy": 0.802466094493866, "num_tokens": 64735920.0, "step": 3488 }, { "epoch": 0.4611419508326725, "grad_norm": 0.8066317066925424, "learning_rate": 6.875956147805777e-06, "loss": 0.6091, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 64801456.0, "step": 3489 }, { "epoch": 0.4612741210679355, "grad_norm": 0.8566514992686907, "learning_rate": 6.8739791939555954e-06, "loss": 0.6106, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 64866992.0, "step": 3490 }, { "epoch": 0.46140629130319855, "grad_norm": 0.8912899324403748, "learning_rate": 6.87200194761393e-06, "loss": 0.5842, "mean_token_accuracy": 0.8061591982841492, "num_tokens": 64932528.0, "step": 3491 }, { "epoch": 0.46153846153846156, "grad_norm": 1.0018039492114421, "learning_rate": 6.870024409201696e-06, "loss": 0.6289, "mean_token_accuracy": 0.7939659357070923, "num_tokens": 64998064.0, "step": 3492 }, { "epoch": 0.4616706317737246, "grad_norm": 0.8554716775733238, "learning_rate": 6.868046579139868e-06, "loss": 0.5774, "mean_token_accuracy": 0.8104474544525146, "num_tokens": 65063600.0, "step": 3493 }, { "epoch": 0.4618028020089876, "grad_norm": 0.8337266148270764, "learning_rate": 6.866068457849486e-06, "loss": 0.5419, "mean_token_accuracy": 0.8191307783126831, "num_tokens": 65129136.0, "step": 3494 }, { "epoch": 0.4619349722442506, "grad_norm": 0.9733154417613465, "learning_rate": 6.864090045751652e-06, "loss": 0.6328, "mean_token_accuracy": 0.792348325252533, "num_tokens": 65194672.0, "step": 3495 }, { "epoch": 0.46206714247951364, "grad_norm": 0.9005642452434083, "learning_rate": 6.862111343267527e-06, "loss": 0.5647, "mean_token_accuracy": 0.8116530179977417, "num_tokens": 65260208.0, "step": 3496 }, { "epoch": 0.46219931271477666, "grad_norm": 1.0123335131960134, "learning_rate": 6.860132350818336e-06, "loss": 0.6773, "mean_token_accuracy": 0.7767214179039001, "num_tokens": 65325744.0, "step": 3497 }, { "epoch": 0.4623314829500397, "grad_norm": 0.8103773587053728, "learning_rate": 6.858153068825365e-06, "loss": 0.5597, "mean_token_accuracy": 0.8165669441223145, "num_tokens": 65391280.0, "step": 3498 }, { "epoch": 0.4624636531853027, "grad_norm": 0.814101195729227, "learning_rate": 6.856173497709961e-06, "loss": 0.5443, "mean_token_accuracy": 0.8169789910316467, "num_tokens": 65456816.0, "step": 3499 }, { "epoch": 0.4625958234205657, "grad_norm": 0.8629267535301542, "learning_rate": 6.854193637893536e-06, "loss": 0.6112, "mean_token_accuracy": 0.8004364371299744, "num_tokens": 65522352.0, "step": 3500 }, { "epoch": 0.46272799365582873, "grad_norm": 0.8452961602862066, "learning_rate": 6.852213489797558e-06, "loss": 0.6006, "mean_token_accuracy": 0.8014894127845764, "num_tokens": 65587888.0, "step": 3501 }, { "epoch": 0.46286016389109175, "grad_norm": 0.9390988312077695, "learning_rate": 6.8502330538435605e-06, "loss": 0.6071, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 65653424.0, "step": 3502 }, { "epoch": 0.46299233412635477, "grad_norm": 0.848332371526305, "learning_rate": 6.848252330453137e-06, "loss": 0.53, "mean_token_accuracy": 0.826303243637085, "num_tokens": 65718960.0, "step": 3503 }, { "epoch": 0.4631245043616178, "grad_norm": 0.8843939398013908, "learning_rate": 6.8462713200479405e-06, "loss": 0.6306, "mean_token_accuracy": 0.7943779826164246, "num_tokens": 65784496.0, "step": 3504 }, { "epoch": 0.4632566745968808, "grad_norm": 0.9454488481983364, "learning_rate": 6.844290023049689e-06, "loss": 0.6631, "mean_token_accuracy": 0.7827340960502625, "num_tokens": 65850032.0, "step": 3505 }, { "epoch": 0.4633888448321438, "grad_norm": 0.8636286903077987, "learning_rate": 6.842308439880157e-06, "loss": 0.552, "mean_token_accuracy": 0.8179709315299988, "num_tokens": 65915568.0, "step": 3506 }, { "epoch": 0.46352101506740684, "grad_norm": 0.9378621905053373, "learning_rate": 6.840326570961186e-06, "loss": 0.6423, "mean_token_accuracy": 0.7868391871452332, "num_tokens": 65981104.0, "step": 3507 }, { "epoch": 0.46365318530266986, "grad_norm": 0.8130100647080603, "learning_rate": 6.838344416714671e-06, "loss": 0.5735, "mean_token_accuracy": 0.8109968304634094, "num_tokens": 66046640.0, "step": 3508 }, { "epoch": 0.4637853555379329, "grad_norm": 0.8512000323205642, "learning_rate": 6.836361977562571e-06, "loss": 0.5853, "mean_token_accuracy": 0.8044041991233826, "num_tokens": 66112176.0, "step": 3509 }, { "epoch": 0.4639175257731959, "grad_norm": 0.9383175344271111, "learning_rate": 6.834379253926908e-06, "loss": 0.6092, "mean_token_accuracy": 0.7992766499519348, "num_tokens": 66177712.0, "step": 3510 }, { "epoch": 0.4640496960084589, "grad_norm": 0.9877270231418243, "learning_rate": 6.832396246229762e-06, "loss": 0.6407, "mean_token_accuracy": 0.7892656326293945, "num_tokens": 66243248.0, "step": 3511 }, { "epoch": 0.46418186624372193, "grad_norm": 0.8226540894081472, "learning_rate": 6.8304129548932725e-06, "loss": 0.536, "mean_token_accuracy": 0.822961151599884, "num_tokens": 66308784.0, "step": 3512 }, { "epoch": 0.46431403647898495, "grad_norm": 0.8455893647020432, "learning_rate": 6.828429380339643e-06, "loss": 0.5886, "mean_token_accuracy": 0.8069985508918762, "num_tokens": 66374320.0, "step": 3513 }, { "epoch": 0.46444620671424797, "grad_norm": 0.8138285610421664, "learning_rate": 6.826445522991135e-06, "loss": 0.5096, "mean_token_accuracy": 0.8309119939804077, "num_tokens": 66439856.0, "step": 3514 }, { "epoch": 0.464578376949511, "grad_norm": 0.8325724306004995, "learning_rate": 6.824461383270067e-06, "loss": 0.6068, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 66505392.0, "step": 3515 }, { "epoch": 0.464710547184774, "grad_norm": 0.8750893407089688, "learning_rate": 6.822476961598825e-06, "loss": 0.5717, "mean_token_accuracy": 0.8124923706054688, "num_tokens": 66570928.0, "step": 3516 }, { "epoch": 0.464842717420037, "grad_norm": 0.901719813082413, "learning_rate": 6.820492258399848e-06, "loss": 0.5849, "mean_token_accuracy": 0.808661937713623, "num_tokens": 66636464.0, "step": 3517 }, { "epoch": 0.46497488765530004, "grad_norm": 0.892804608242912, "learning_rate": 6.818507274095642e-06, "loss": 0.6166, "mean_token_accuracy": 0.7945153117179871, "num_tokens": 66702000.0, "step": 3518 }, { "epoch": 0.46510705789056306, "grad_norm": 0.8698261809020174, "learning_rate": 6.816522009108761e-06, "loss": 0.592, "mean_token_accuracy": 0.8064186573028564, "num_tokens": 66767536.0, "step": 3519 }, { "epoch": 0.4652392281258261, "grad_norm": 0.8456206873468528, "learning_rate": 6.814536463861835e-06, "loss": 0.6061, "mean_token_accuracy": 0.8019930124282837, "num_tokens": 66833072.0, "step": 3520 }, { "epoch": 0.4653713983610891, "grad_norm": 0.8188903510732854, "learning_rate": 6.8125506387775395e-06, "loss": 0.5702, "mean_token_accuracy": 0.8128281235694885, "num_tokens": 66898608.0, "step": 3521 }, { "epoch": 0.4655035685963521, "grad_norm": 0.8869872942969927, "learning_rate": 6.81056453427862e-06, "loss": 0.5985, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 66964144.0, "step": 3522 }, { "epoch": 0.46563573883161513, "grad_norm": 0.8962058239698407, "learning_rate": 6.808578150787871e-06, "loss": 0.6296, "mean_token_accuracy": 0.7931876182556152, "num_tokens": 67029680.0, "step": 3523 }, { "epoch": 0.46576790906687815, "grad_norm": 0.9327235638340002, "learning_rate": 6.806591488728156e-06, "loss": 0.6476, "mean_token_accuracy": 0.7860914468765259, "num_tokens": 67095216.0, "step": 3524 }, { "epoch": 0.46590007930214117, "grad_norm": 0.8889658715432823, "learning_rate": 6.804604548522391e-06, "loss": 0.5833, "mean_token_accuracy": 0.8080667853355408, "num_tokens": 67160752.0, "step": 3525 }, { "epoch": 0.4660322495374042, "grad_norm": 0.8696031901476787, "learning_rate": 6.802617330593557e-06, "loss": 0.609, "mean_token_accuracy": 0.798696756362915, "num_tokens": 67226288.0, "step": 3526 }, { "epoch": 0.4661644197726672, "grad_norm": 0.8032675990723217, "learning_rate": 6.8006298353646905e-06, "loss": 0.5787, "mean_token_accuracy": 0.8102795481681824, "num_tokens": 67291824.0, "step": 3527 }, { "epoch": 0.4662965900079302, "grad_norm": 0.934516680662857, "learning_rate": 6.7986420632588865e-06, "loss": 0.5349, "mean_token_accuracy": 0.8252044916152954, "num_tokens": 67357360.0, "step": 3528 }, { "epoch": 0.46642876024319324, "grad_norm": 0.8206853006672438, "learning_rate": 6.796654014699301e-06, "loss": 0.5731, "mean_token_accuracy": 0.8093181252479553, "num_tokens": 67422896.0, "step": 3529 }, { "epoch": 0.46656093047845626, "grad_norm": 0.8513814527703129, "learning_rate": 6.794665690109148e-06, "loss": 0.6063, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 67488432.0, "step": 3530 }, { "epoch": 0.4666931007137193, "grad_norm": 0.8931961825867909, "learning_rate": 6.7926770899117e-06, "loss": 0.5578, "mean_token_accuracy": 0.8153918981552124, "num_tokens": 67553968.0, "step": 3531 }, { "epoch": 0.4668252709489823, "grad_norm": 0.8418857890501519, "learning_rate": 6.79068821453029e-06, "loss": 0.6521, "mean_token_accuracy": 0.7848705649375916, "num_tokens": 67619504.0, "step": 3532 }, { "epoch": 0.4669574411842453, "grad_norm": 0.8301406374067088, "learning_rate": 6.788699064388304e-06, "loss": 0.5996, "mean_token_accuracy": 0.8040837645530701, "num_tokens": 67685040.0, "step": 3533 }, { "epoch": 0.46708961141950833, "grad_norm": 0.8733364999229862, "learning_rate": 6.7867096399091955e-06, "loss": 0.6324, "mean_token_accuracy": 0.7912189960479736, "num_tokens": 67750576.0, "step": 3534 }, { "epoch": 0.46722178165477135, "grad_norm": 0.8662881845811432, "learning_rate": 6.78471994151647e-06, "loss": 0.5905, "mean_token_accuracy": 0.8070443272590637, "num_tokens": 67816112.0, "step": 3535 }, { "epoch": 0.46735395189003437, "grad_norm": 0.8613834382088724, "learning_rate": 6.78272996963369e-06, "loss": 0.6395, "mean_token_accuracy": 0.7916005253791809, "num_tokens": 67881648.0, "step": 3536 }, { "epoch": 0.4674861221252974, "grad_norm": 0.8918369045641592, "learning_rate": 6.780739724684482e-06, "loss": 0.5847, "mean_token_accuracy": 0.8070290684700012, "num_tokens": 67947184.0, "step": 3537 }, { "epoch": 0.4676182923605604, "grad_norm": 0.8349336277244903, "learning_rate": 6.778749207092525e-06, "loss": 0.5639, "mean_token_accuracy": 0.8166279792785645, "num_tokens": 68012720.0, "step": 3538 }, { "epoch": 0.4677504625958234, "grad_norm": 0.8481263059624612, "learning_rate": 6.776758417281563e-06, "loss": 0.5778, "mean_token_accuracy": 0.8092265725135803, "num_tokens": 68078256.0, "step": 3539 }, { "epoch": 0.46788263283108644, "grad_norm": 0.904248309917287, "learning_rate": 6.77476735567539e-06, "loss": 0.6484, "mean_token_accuracy": 0.7876480221748352, "num_tokens": 68143792.0, "step": 3540 }, { "epoch": 0.46801480306634946, "grad_norm": 0.8574895874131022, "learning_rate": 6.772776022697861e-06, "loss": 0.6152, "mean_token_accuracy": 0.7996276617050171, "num_tokens": 68209328.0, "step": 3541 }, { "epoch": 0.4681469733016125, "grad_norm": 0.9528473912740454, "learning_rate": 6.77078441877289e-06, "loss": 0.6967, "mean_token_accuracy": 0.7713344097137451, "num_tokens": 68274864.0, "step": 3542 }, { "epoch": 0.4682791435368755, "grad_norm": 0.9059268683603626, "learning_rate": 6.76879254432445e-06, "loss": 0.6204, "mean_token_accuracy": 0.7983152270317078, "num_tokens": 68340400.0, "step": 3543 }, { "epoch": 0.4684113137721385, "grad_norm": 0.8299991891149856, "learning_rate": 6.766800399776566e-06, "loss": 0.5872, "mean_token_accuracy": 0.8070748448371887, "num_tokens": 68405936.0, "step": 3544 }, { "epoch": 0.46854348400740153, "grad_norm": 0.8865022882155493, "learning_rate": 6.764807985553328e-06, "loss": 0.6481, "mean_token_accuracy": 0.7859998941421509, "num_tokens": 68471472.0, "step": 3545 }, { "epoch": 0.46867565424266455, "grad_norm": 0.8744164876515186, "learning_rate": 6.762815302078875e-06, "loss": 0.5883, "mean_token_accuracy": 0.8046789169311523, "num_tokens": 68537008.0, "step": 3546 }, { "epoch": 0.46880782447792757, "grad_norm": 0.8076059831213587, "learning_rate": 6.7608223497774104e-06, "loss": 0.5758, "mean_token_accuracy": 0.8102795481681824, "num_tokens": 68602544.0, "step": 3547 }, { "epoch": 0.4689399947131906, "grad_norm": 0.8769175971293244, "learning_rate": 6.7588291290731935e-06, "loss": 0.6006, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 68668080.0, "step": 3548 }, { "epoch": 0.4690721649484536, "grad_norm": 0.7994821642626422, "learning_rate": 6.756835640390535e-06, "loss": 0.5402, "mean_token_accuracy": 0.8217098116874695, "num_tokens": 68733616.0, "step": 3549 }, { "epoch": 0.4692043351837166, "grad_norm": 0.8674512229483968, "learning_rate": 6.75484188415381e-06, "loss": 0.6133, "mean_token_accuracy": 0.7981931567192078, "num_tokens": 68799152.0, "step": 3550 }, { "epoch": 0.46933650541897964, "grad_norm": 0.8619355544640867, "learning_rate": 6.752847860787448e-06, "loss": 0.6018, "mean_token_accuracy": 0.8032901883125305, "num_tokens": 68864688.0, "step": 3551 }, { "epoch": 0.46946867565424266, "grad_norm": 0.8575127679126879, "learning_rate": 6.7508535707159325e-06, "loss": 0.5962, "mean_token_accuracy": 0.8036717176437378, "num_tokens": 68930224.0, "step": 3552 }, { "epoch": 0.4696008458895057, "grad_norm": 0.8609474418260152, "learning_rate": 6.74885901436381e-06, "loss": 0.606, "mean_token_accuracy": 0.8038548231124878, "num_tokens": 68995760.0, "step": 3553 }, { "epoch": 0.4697330161247687, "grad_norm": 0.9272993527258957, "learning_rate": 6.7468641921556765e-06, "loss": 0.6068, "mean_token_accuracy": 0.8011384606361389, "num_tokens": 69061296.0, "step": 3554 }, { "epoch": 0.4698651863600317, "grad_norm": 0.9058806126614265, "learning_rate": 6.744869104516189e-06, "loss": 0.6207, "mean_token_accuracy": 0.797704815864563, "num_tokens": 69126832.0, "step": 3555 }, { "epoch": 0.46999735659529474, "grad_norm": 0.8539660486638689, "learning_rate": 6.742873751870061e-06, "loss": 0.5917, "mean_token_accuracy": 0.8047399520874023, "num_tokens": 69192368.0, "step": 3556 }, { "epoch": 0.47012952683055775, "grad_norm": 0.9117797772448974, "learning_rate": 6.740878134642058e-06, "loss": 0.6813, "mean_token_accuracy": 0.7733030319213867, "num_tokens": 69257904.0, "step": 3557 }, { "epoch": 0.47026169706582077, "grad_norm": 0.8976012216148884, "learning_rate": 6.73888225325701e-06, "loss": 0.6121, "mean_token_accuracy": 0.7994902729988098, "num_tokens": 69323440.0, "step": 3558 }, { "epoch": 0.4703938673010838, "grad_norm": 0.9645729745147877, "learning_rate": 6.736886108139797e-06, "loss": 0.618, "mean_token_accuracy": 0.7946984767913818, "num_tokens": 69388976.0, "step": 3559 }, { "epoch": 0.4705260375363468, "grad_norm": 1.03267218251324, "learning_rate": 6.734889699715354e-06, "loss": 0.6522, "mean_token_accuracy": 0.7869613170623779, "num_tokens": 69454512.0, "step": 3560 }, { "epoch": 0.4706582077716098, "grad_norm": 0.8449003480575953, "learning_rate": 6.732893028408678e-06, "loss": 0.5563, "mean_token_accuracy": 0.8156360387802124, "num_tokens": 69520048.0, "step": 3561 }, { "epoch": 0.47079037800687284, "grad_norm": 1.0105597951180445, "learning_rate": 6.7308960946448155e-06, "loss": 0.6223, "mean_token_accuracy": 0.7940574884414673, "num_tokens": 69585584.0, "step": 3562 }, { "epoch": 0.47092254824213586, "grad_norm": 0.7685888781383177, "learning_rate": 6.7288988988488745e-06, "loss": 0.5383, "mean_token_accuracy": 0.8221218585968018, "num_tokens": 69651120.0, "step": 3563 }, { "epoch": 0.4710547184773989, "grad_norm": 0.943019573929045, "learning_rate": 6.726901441446017e-06, "loss": 0.5651, "mean_token_accuracy": 0.8138963580131531, "num_tokens": 69716656.0, "step": 3564 }, { "epoch": 0.4711868887126619, "grad_norm": 0.851590616650845, "learning_rate": 6.7249037228614566e-06, "loss": 0.5782, "mean_token_accuracy": 0.8105084896087646, "num_tokens": 69782192.0, "step": 3565 }, { "epoch": 0.4713190589479249, "grad_norm": 0.8029285999235369, "learning_rate": 6.7229057435204685e-06, "loss": 0.5177, "mean_token_accuracy": 0.8275851607322693, "num_tokens": 69847728.0, "step": 3566 }, { "epoch": 0.47145122918318794, "grad_norm": 0.8576447817251072, "learning_rate": 6.72090750384838e-06, "loss": 0.61, "mean_token_accuracy": 0.797704815864563, "num_tokens": 69913264.0, "step": 3567 }, { "epoch": 0.47158339941845095, "grad_norm": 0.8834237817173269, "learning_rate": 6.718909004270573e-06, "loss": 0.6528, "mean_token_accuracy": 0.7861829996109009, "num_tokens": 69978800.0, "step": 3568 }, { "epoch": 0.471715569653714, "grad_norm": 0.7957995423168315, "learning_rate": 6.71691024521249e-06, "loss": 0.5604, "mean_token_accuracy": 0.8152240514755249, "num_tokens": 70044336.0, "step": 3569 }, { "epoch": 0.471847739888977, "grad_norm": 0.8406741677420496, "learning_rate": 6.714911227099623e-06, "loss": 0.5772, "mean_token_accuracy": 0.808570384979248, "num_tokens": 70109872.0, "step": 3570 }, { "epoch": 0.47197991012424, "grad_norm": 0.828487384990894, "learning_rate": 6.71291195035752e-06, "loss": 0.5701, "mean_token_accuracy": 0.8104169368743896, "num_tokens": 70175408.0, "step": 3571 }, { "epoch": 0.47211208035950303, "grad_norm": 0.8546674782337668, "learning_rate": 6.710912415411789e-06, "loss": 0.5944, "mean_token_accuracy": 0.8026645183563232, "num_tokens": 70240944.0, "step": 3572 }, { "epoch": 0.47224425059476605, "grad_norm": 0.9762572026096943, "learning_rate": 6.708912622688083e-06, "loss": 0.6704, "mean_token_accuracy": 0.7802923917770386, "num_tokens": 70306480.0, "step": 3573 }, { "epoch": 0.47237642083002906, "grad_norm": 0.9110321061783947, "learning_rate": 6.706912572612123e-06, "loss": 0.6848, "mean_token_accuracy": 0.7762635946273804, "num_tokens": 70372016.0, "step": 3574 }, { "epoch": 0.4725085910652921, "grad_norm": 0.8653919985320847, "learning_rate": 6.704912265609675e-06, "loss": 0.6615, "mean_token_accuracy": 0.7805060148239136, "num_tokens": 70437552.0, "step": 3575 }, { "epoch": 0.4726407613005551, "grad_norm": 0.9073316616387782, "learning_rate": 6.70291170210656e-06, "loss": 0.6073, "mean_token_accuracy": 0.79872727394104, "num_tokens": 70503088.0, "step": 3576 }, { "epoch": 0.4727729315358181, "grad_norm": 0.9299413222338934, "learning_rate": 6.70091088252866e-06, "loss": 0.5798, "mean_token_accuracy": 0.8072579503059387, "num_tokens": 70568624.0, "step": 3577 }, { "epoch": 0.47290510177108114, "grad_norm": 0.9265753075492801, "learning_rate": 6.698909807301908e-06, "loss": 0.6288, "mean_token_accuracy": 0.7951409816741943, "num_tokens": 70634160.0, "step": 3578 }, { "epoch": 0.47303727200634416, "grad_norm": 0.9766551601837383, "learning_rate": 6.696908476852286e-06, "loss": 0.6692, "mean_token_accuracy": 0.7802618741989136, "num_tokens": 70699696.0, "step": 3579 }, { "epoch": 0.4731694422416072, "grad_norm": 0.875459026997553, "learning_rate": 6.694906891605841e-06, "loss": 0.6188, "mean_token_accuracy": 0.7951104640960693, "num_tokens": 70765232.0, "step": 3580 }, { "epoch": 0.4733016124768702, "grad_norm": 0.8979740828641398, "learning_rate": 6.692905051988666e-06, "loss": 0.6028, "mean_token_accuracy": 0.8035496473312378, "num_tokens": 70830768.0, "step": 3581 }, { "epoch": 0.4734337827121332, "grad_norm": 0.8819277545908717, "learning_rate": 6.690902958426913e-06, "loss": 0.5696, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 70896304.0, "step": 3582 }, { "epoch": 0.47356595294739623, "grad_norm": 0.834882158588291, "learning_rate": 6.688900611346782e-06, "loss": 0.5942, "mean_token_accuracy": 0.8029544353485107, "num_tokens": 70961840.0, "step": 3583 }, { "epoch": 0.47369812318265925, "grad_norm": 0.8567158621518117, "learning_rate": 6.6868980111745345e-06, "loss": 0.59, "mean_token_accuracy": 0.8069222569465637, "num_tokens": 71027376.0, "step": 3584 }, { "epoch": 0.47383029341792227, "grad_norm": 0.83463808258995, "learning_rate": 6.684895158336481e-06, "loss": 0.5471, "mean_token_accuracy": 0.8213130235671997, "num_tokens": 71092912.0, "step": 3585 }, { "epoch": 0.4739624636531853, "grad_norm": 0.9481463535569057, "learning_rate": 6.682892053258987e-06, "loss": 0.6216, "mean_token_accuracy": 0.7954919934272766, "num_tokens": 71158448.0, "step": 3586 }, { "epoch": 0.4740946338884483, "grad_norm": 0.8399494446377583, "learning_rate": 6.680888696368472e-06, "loss": 0.5819, "mean_token_accuracy": 0.8082346320152283, "num_tokens": 71223984.0, "step": 3587 }, { "epoch": 0.4742268041237113, "grad_norm": 0.9637763899394722, "learning_rate": 6.678885088091409e-06, "loss": 0.6606, "mean_token_accuracy": 0.7807959914207458, "num_tokens": 71289520.0, "step": 3588 }, { "epoch": 0.47435897435897434, "grad_norm": 0.8529836696904841, "learning_rate": 6.676881228854323e-06, "loss": 0.5841, "mean_token_accuracy": 0.8058997392654419, "num_tokens": 71355056.0, "step": 3589 }, { "epoch": 0.47449114459423736, "grad_norm": 0.8435498537664532, "learning_rate": 6.674877119083796e-06, "loss": 0.5777, "mean_token_accuracy": 0.8126449584960938, "num_tokens": 71420592.0, "step": 3590 }, { "epoch": 0.4746233148295004, "grad_norm": 0.861111907018234, "learning_rate": 6.67287275920646e-06, "loss": 0.6019, "mean_token_accuracy": 0.803061306476593, "num_tokens": 71486128.0, "step": 3591 }, { "epoch": 0.4747554850647634, "grad_norm": 0.8656081856584257, "learning_rate": 6.670868149649001e-06, "loss": 0.5614, "mean_token_accuracy": 0.8160633444786072, "num_tokens": 71551664.0, "step": 3592 }, { "epoch": 0.4748876553000264, "grad_norm": 0.8352500454003772, "learning_rate": 6.668863290838159e-06, "loss": 0.5861, "mean_token_accuracy": 0.8066322803497314, "num_tokens": 71617200.0, "step": 3593 }, { "epoch": 0.47501982553528943, "grad_norm": 0.8051195823628892, "learning_rate": 6.666858183200727e-06, "loss": 0.5481, "mean_token_accuracy": 0.8193444013595581, "num_tokens": 71682736.0, "step": 3594 }, { "epoch": 0.47515199577055245, "grad_norm": 0.9002243223440946, "learning_rate": 6.664852827163549e-06, "loss": 0.6146, "mean_token_accuracy": 0.7972622513771057, "num_tokens": 71748272.0, "step": 3595 }, { "epoch": 0.47528416600581547, "grad_norm": 0.8619283988691275, "learning_rate": 6.662847223153525e-06, "loss": 0.6101, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 71813808.0, "step": 3596 }, { "epoch": 0.4754163362410785, "grad_norm": 0.8698405453457164, "learning_rate": 6.660841371597607e-06, "loss": 0.6302, "mean_token_accuracy": 0.7904254794120789, "num_tokens": 71879344.0, "step": 3597 }, { "epoch": 0.4755485064763415, "grad_norm": 0.8816234244784139, "learning_rate": 6.6588352729227955e-06, "loss": 0.6283, "mean_token_accuracy": 0.7903797030448914, "num_tokens": 71944880.0, "step": 3598 }, { "epoch": 0.4756806767116045, "grad_norm": 0.9596050026109243, "learning_rate": 6.656828927556151e-06, "loss": 0.6864, "mean_token_accuracy": 0.7758362889289856, "num_tokens": 72010416.0, "step": 3599 }, { "epoch": 0.47581284694686754, "grad_norm": 0.9039536379563298, "learning_rate": 6.654822335924779e-06, "loss": 0.6417, "mean_token_accuracy": 0.7870986461639404, "num_tokens": 72075952.0, "step": 3600 }, { "epoch": 0.47594501718213056, "grad_norm": 0.8138481156383773, "learning_rate": 6.652815498455844e-06, "loss": 0.5866, "mean_token_accuracy": 0.808387279510498, "num_tokens": 72141488.0, "step": 3601 }, { "epoch": 0.4760771874173936, "grad_norm": 0.8481717845963174, "learning_rate": 6.650808415576557e-06, "loss": 0.6111, "mean_token_accuracy": 0.7999023199081421, "num_tokens": 72207024.0, "step": 3602 }, { "epoch": 0.4762093576526566, "grad_norm": 0.8755749440490747, "learning_rate": 6.648801087714185e-06, "loss": 0.6007, "mean_token_accuracy": 0.8039616942405701, "num_tokens": 72272560.0, "step": 3603 }, { "epoch": 0.4763415278879196, "grad_norm": 0.7637478673313335, "learning_rate": 6.646793515296048e-06, "loss": 0.5166, "mean_token_accuracy": 0.8288212418556213, "num_tokens": 72338096.0, "step": 3604 }, { "epoch": 0.47647369812318263, "grad_norm": 0.8252973037019243, "learning_rate": 6.644785698749515e-06, "loss": 0.5929, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 72403632.0, "step": 3605 }, { "epoch": 0.4766058683584457, "grad_norm": 0.9205349698415375, "learning_rate": 6.642777638502009e-06, "loss": 0.626, "mean_token_accuracy": 0.7944695353507996, "num_tokens": 72469168.0, "step": 3606 }, { "epoch": 0.4767380385937087, "grad_norm": 0.8180992654575246, "learning_rate": 6.640769334981001e-06, "loss": 0.5646, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 72534704.0, "step": 3607 }, { "epoch": 0.47687020882897174, "grad_norm": 0.9088690392591195, "learning_rate": 6.6387607886140205e-06, "loss": 0.6173, "mean_token_accuracy": 0.7960871458053589, "num_tokens": 72600240.0, "step": 3608 }, { "epoch": 0.47700237906423476, "grad_norm": 0.845863004181701, "learning_rate": 6.6367519998286445e-06, "loss": 0.6104, "mean_token_accuracy": 0.797735333442688, "num_tokens": 72665776.0, "step": 3609 }, { "epoch": 0.4771345492994978, "grad_norm": 0.8606077276101335, "learning_rate": 6.634742969052499e-06, "loss": 0.5807, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 72731312.0, "step": 3610 }, { "epoch": 0.4772667195347608, "grad_norm": 0.9096494852874376, "learning_rate": 6.632733696713268e-06, "loss": 0.6065, "mean_token_accuracy": 0.7997497320175171, "num_tokens": 72796848.0, "step": 3611 }, { "epoch": 0.4773988897700238, "grad_norm": 0.9575195110492013, "learning_rate": 6.630724183238685e-06, "loss": 0.6123, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 72862384.0, "step": 3612 }, { "epoch": 0.47753106000528683, "grad_norm": 0.9404943876148665, "learning_rate": 6.628714429056529e-06, "loss": 0.6588, "mean_token_accuracy": 0.7852978706359863, "num_tokens": 72927920.0, "step": 3613 }, { "epoch": 0.47766323024054985, "grad_norm": 0.9402314683203185, "learning_rate": 6.626704434594638e-06, "loss": 0.6173, "mean_token_accuracy": 0.7963923811912537, "num_tokens": 72993456.0, "step": 3614 }, { "epoch": 0.47779540047581287, "grad_norm": 0.8873744388667392, "learning_rate": 6.624694200280894e-06, "loss": 0.6432, "mean_token_accuracy": 0.7892961502075195, "num_tokens": 73058992.0, "step": 3615 }, { "epoch": 0.4779275707110759, "grad_norm": 0.9720371968770949, "learning_rate": 6.62268372654324e-06, "loss": 0.5948, "mean_token_accuracy": 0.807486891746521, "num_tokens": 73124528.0, "step": 3616 }, { "epoch": 0.4780597409463389, "grad_norm": 0.9438743965526881, "learning_rate": 6.620673013809659e-06, "loss": 0.6589, "mean_token_accuracy": 0.7830545902252197, "num_tokens": 73190064.0, "step": 3617 }, { "epoch": 0.4781919111816019, "grad_norm": 0.9378728990199726, "learning_rate": 6.618662062508192e-06, "loss": 0.5808, "mean_token_accuracy": 0.8069832921028137, "num_tokens": 73255600.0, "step": 3618 }, { "epoch": 0.47832408141686494, "grad_norm": 0.8518218640950219, "learning_rate": 6.616650873066928e-06, "loss": 0.5643, "mean_token_accuracy": 0.8148120045661926, "num_tokens": 73321136.0, "step": 3619 }, { "epoch": 0.47845625165212796, "grad_norm": 0.8014246722227703, "learning_rate": 6.614639445914007e-06, "loss": 0.6069, "mean_token_accuracy": 0.8028323650360107, "num_tokens": 73386672.0, "step": 3620 }, { "epoch": 0.478588421887391, "grad_norm": 0.794306396219509, "learning_rate": 6.61262778147762e-06, "loss": 0.5499, "mean_token_accuracy": 0.8194664716720581, "num_tokens": 73452208.0, "step": 3621 }, { "epoch": 0.478720592122654, "grad_norm": 0.9358605742508121, "learning_rate": 6.610615880186009e-06, "loss": 0.5998, "mean_token_accuracy": 0.7999938726425171, "num_tokens": 73517744.0, "step": 3622 }, { "epoch": 0.478852762357917, "grad_norm": 0.9607124142199799, "learning_rate": 6.608603742467463e-06, "loss": 0.6017, "mean_token_accuracy": 0.8009858131408691, "num_tokens": 73583280.0, "step": 3623 }, { "epoch": 0.47898493259318004, "grad_norm": 0.8787361925617072, "learning_rate": 6.606591368750327e-06, "loss": 0.5787, "mean_token_accuracy": 0.8094860315322876, "num_tokens": 73648816.0, "step": 3624 }, { "epoch": 0.47911710282844305, "grad_norm": 0.7789810851979073, "learning_rate": 6.604578759462993e-06, "loss": 0.5453, "mean_token_accuracy": 0.8217403292655945, "num_tokens": 73714352.0, "step": 3625 }, { "epoch": 0.47924927306370607, "grad_norm": 0.7902120482332452, "learning_rate": 6.602565915033902e-06, "loss": 0.5575, "mean_token_accuracy": 0.8151019215583801, "num_tokens": 73779888.0, "step": 3626 }, { "epoch": 0.4793814432989691, "grad_norm": 0.9704449989005456, "learning_rate": 6.600552835891548e-06, "loss": 0.6295, "mean_token_accuracy": 0.792439877986908, "num_tokens": 73845424.0, "step": 3627 }, { "epoch": 0.4795136135342321, "grad_norm": 0.9398402140586409, "learning_rate": 6.598539522464474e-06, "loss": 0.6202, "mean_token_accuracy": 0.7974606156349182, "num_tokens": 73910960.0, "step": 3628 }, { "epoch": 0.4796457837694951, "grad_norm": 1.0009304430499246, "learning_rate": 6.596525975181272e-06, "loss": 0.6029, "mean_token_accuracy": 0.7999938726425171, "num_tokens": 73976496.0, "step": 3629 }, { "epoch": 0.47977795400475814, "grad_norm": 6.790303611423124, "learning_rate": 6.594512194470583e-06, "loss": 0.5633, "mean_token_accuracy": 0.8135148286819458, "num_tokens": 74042032.0, "step": 3630 }, { "epoch": 0.47991012424002116, "grad_norm": 0.8787233770695305, "learning_rate": 6.592498180761099e-06, "loss": 0.5743, "mean_token_accuracy": 0.8114851713180542, "num_tokens": 74107568.0, "step": 3631 }, { "epoch": 0.4800422944752842, "grad_norm": 0.9257211396552639, "learning_rate": 6.590483934481562e-06, "loss": 0.6324, "mean_token_accuracy": 0.7928366661071777, "num_tokens": 74173104.0, "step": 3632 }, { "epoch": 0.4801744647105472, "grad_norm": 0.8848252186595634, "learning_rate": 6.5884694560607635e-06, "loss": 0.6397, "mean_token_accuracy": 0.7899829149246216, "num_tokens": 74238640.0, "step": 3633 }, { "epoch": 0.4803066349458102, "grad_norm": 0.9428508808447106, "learning_rate": 6.586454745927544e-06, "loss": 0.6518, "mean_token_accuracy": 0.7874954342842102, "num_tokens": 74304176.0, "step": 3634 }, { "epoch": 0.48043880518107324, "grad_norm": 0.796078001491415, "learning_rate": 6.584439804510791e-06, "loss": 0.554, "mean_token_accuracy": 0.816643238067627, "num_tokens": 74369712.0, "step": 3635 }, { "epoch": 0.48057097541633625, "grad_norm": 0.929945741883558, "learning_rate": 6.5824246322394455e-06, "loss": 0.59, "mean_token_accuracy": 0.8022372126579285, "num_tokens": 74435248.0, "step": 3636 }, { "epoch": 0.4807031456515993, "grad_norm": 1.0347653400475223, "learning_rate": 6.580409229542497e-06, "loss": 0.6469, "mean_token_accuracy": 0.7836039662361145, "num_tokens": 74500784.0, "step": 3637 }, { "epoch": 0.4808353158868623, "grad_norm": 0.8666205158568417, "learning_rate": 6.578393596848981e-06, "loss": 0.5617, "mean_token_accuracy": 0.8170095086097717, "num_tokens": 74566320.0, "step": 3638 }, { "epoch": 0.4809674861221253, "grad_norm": 0.8214187354993109, "learning_rate": 6.576377734587982e-06, "loss": 0.6156, "mean_token_accuracy": 0.7964534163475037, "num_tokens": 74631856.0, "step": 3639 }, { "epoch": 0.48109965635738833, "grad_norm": 0.8535719192731372, "learning_rate": 6.574361643188636e-06, "loss": 0.5954, "mean_token_accuracy": 0.8032901883125305, "num_tokens": 74697392.0, "step": 3640 }, { "epoch": 0.48123182659265135, "grad_norm": 0.9003261022392753, "learning_rate": 6.572345323080128e-06, "loss": 0.537, "mean_token_accuracy": 0.8252044916152954, "num_tokens": 74762928.0, "step": 3641 }, { "epoch": 0.48136399682791436, "grad_norm": 0.9445851827128154, "learning_rate": 6.570328774691689e-06, "loss": 0.6226, "mean_token_accuracy": 0.7942559123039246, "num_tokens": 74828464.0, "step": 3642 }, { "epoch": 0.4814961670631774, "grad_norm": 0.939208607226865, "learning_rate": 6.568311998452601e-06, "loss": 0.6147, "mean_token_accuracy": 0.7987344861030579, "num_tokens": 74892477.0, "step": 3643 }, { "epoch": 0.4816283372984404, "grad_norm": 0.9370185240699794, "learning_rate": 6.566294994792194e-06, "loss": 0.6679, "mean_token_accuracy": 0.7811775207519531, "num_tokens": 74958013.0, "step": 3644 }, { "epoch": 0.4817605075337034, "grad_norm": 0.8224126766675325, "learning_rate": 6.5642777641398445e-06, "loss": 0.5797, "mean_token_accuracy": 0.8096538782119751, "num_tokens": 75023549.0, "step": 3645 }, { "epoch": 0.48189267776896644, "grad_norm": 0.8856896558556772, "learning_rate": 6.56226030692498e-06, "loss": 0.5975, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 75089085.0, "step": 3646 }, { "epoch": 0.48202484800422946, "grad_norm": 0.8885125246613025, "learning_rate": 6.560242623577073e-06, "loss": 0.5805, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 75154621.0, "step": 3647 }, { "epoch": 0.4821570182394925, "grad_norm": 0.8496864583906193, "learning_rate": 6.55822471452565e-06, "loss": 0.6286, "mean_token_accuracy": 0.7921499013900757, "num_tokens": 75220157.0, "step": 3648 }, { "epoch": 0.4822891884747555, "grad_norm": 0.8211373484499561, "learning_rate": 6.556206580200278e-06, "loss": 0.5652, "mean_token_accuracy": 0.8134690523147583, "num_tokens": 75285693.0, "step": 3649 }, { "epoch": 0.4824213587100185, "grad_norm": 0.9345073598075023, "learning_rate": 6.554188221030577e-06, "loss": 0.6586, "mean_token_accuracy": 0.784458577632904, "num_tokens": 75351229.0, "step": 3650 }, { "epoch": 0.48255352894528153, "grad_norm": 0.8477100054885411, "learning_rate": 6.552169637446215e-06, "loss": 0.5598, "mean_token_accuracy": 0.8151171803474426, "num_tokens": 75416765.0, "step": 3651 }, { "epoch": 0.48268569918054455, "grad_norm": 0.9189638587477674, "learning_rate": 6.550150829876903e-06, "loss": 0.6276, "mean_token_accuracy": 0.7900134325027466, "num_tokens": 75482301.0, "step": 3652 }, { "epoch": 0.48281786941580757, "grad_norm": 0.9004622858634566, "learning_rate": 6.548131798752406e-06, "loss": 0.6272, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 75547837.0, "step": 3653 }, { "epoch": 0.4829500396510706, "grad_norm": 0.9674919844812392, "learning_rate": 6.546112544502534e-06, "loss": 0.656, "mean_token_accuracy": 0.7854810357093811, "num_tokens": 75613373.0, "step": 3654 }, { "epoch": 0.4830822098863336, "grad_norm": 0.8489527296206909, "learning_rate": 6.54409306755714e-06, "loss": 0.5919, "mean_token_accuracy": 0.8036259412765503, "num_tokens": 75678909.0, "step": 3655 }, { "epoch": 0.4832143801215966, "grad_norm": 0.87379309535289, "learning_rate": 6.542073368346133e-06, "loss": 0.6111, "mean_token_accuracy": 0.8000091314315796, "num_tokens": 75744445.0, "step": 3656 }, { "epoch": 0.48334655035685964, "grad_norm": 0.8855486125775907, "learning_rate": 6.540053447299461e-06, "loss": 0.6048, "mean_token_accuracy": 0.8005127310752869, "num_tokens": 75809981.0, "step": 3657 }, { "epoch": 0.48347872059212266, "grad_norm": 0.9759988190918955, "learning_rate": 6.5380333048471266e-06, "loss": 0.6552, "mean_token_accuracy": 0.7839397192001343, "num_tokens": 75875517.0, "step": 3658 }, { "epoch": 0.4836108908273857, "grad_norm": 0.8966863482166115, "learning_rate": 6.536012941419173e-06, "loss": 0.6453, "mean_token_accuracy": 0.7897692322731018, "num_tokens": 75941053.0, "step": 3659 }, { "epoch": 0.4837430610626487, "grad_norm": 0.9611992720754727, "learning_rate": 6.533992357445694e-06, "loss": 0.6379, "mean_token_accuracy": 0.7913106083869934, "num_tokens": 76006589.0, "step": 3660 }, { "epoch": 0.4838752312979117, "grad_norm": 0.9695132754158722, "learning_rate": 6.53197155335683e-06, "loss": 0.6333, "mean_token_accuracy": 0.7895708680152893, "num_tokens": 76072125.0, "step": 3661 }, { "epoch": 0.48400740153317473, "grad_norm": 0.882641024511253, "learning_rate": 6.529950529582771e-06, "loss": 0.6029, "mean_token_accuracy": 0.8014588952064514, "num_tokens": 76137661.0, "step": 3662 }, { "epoch": 0.48413957176843775, "grad_norm": 0.9696690610014344, "learning_rate": 6.527929286553746e-06, "loss": 0.6151, "mean_token_accuracy": 0.800596296787262, "num_tokens": 76198375.0, "step": 3663 }, { "epoch": 0.48427174200370077, "grad_norm": 0.8531036102482047, "learning_rate": 6.525907824700036e-06, "loss": 0.5871, "mean_token_accuracy": 0.8040532469749451, "num_tokens": 76263911.0, "step": 3664 }, { "epoch": 0.4844039122389638, "grad_norm": 0.8293822321087959, "learning_rate": 6.523886144451973e-06, "loss": 0.5806, "mean_token_accuracy": 0.8112868070602417, "num_tokens": 76329447.0, "step": 3665 }, { "epoch": 0.4845360824742268, "grad_norm": 0.8627402876615956, "learning_rate": 6.521864246239925e-06, "loss": 0.6401, "mean_token_accuracy": 0.7910816669464111, "num_tokens": 76394983.0, "step": 3666 }, { "epoch": 0.4846682527094898, "grad_norm": 0.8495986929252901, "learning_rate": 6.519842130494315e-06, "loss": 0.5645, "mean_token_accuracy": 0.8118361830711365, "num_tokens": 76460519.0, "step": 3667 }, { "epoch": 0.48480042294475284, "grad_norm": 0.9403642461528758, "learning_rate": 6.517819797645607e-06, "loss": 0.592, "mean_token_accuracy": 0.8055945634841919, "num_tokens": 76526055.0, "step": 3668 }, { "epoch": 0.48493259318001586, "grad_norm": 0.8652075946555932, "learning_rate": 6.5157972481243145e-06, "loss": 0.6341, "mean_token_accuracy": 0.7921041250228882, "num_tokens": 76591591.0, "step": 3669 }, { "epoch": 0.4850647634152789, "grad_norm": 0.8795403605564712, "learning_rate": 6.5137744823609984e-06, "loss": 0.6047, "mean_token_accuracy": 0.8026950359344482, "num_tokens": 76657127.0, "step": 3670 }, { "epoch": 0.4851969336505419, "grad_norm": 0.8200254983422899, "learning_rate": 6.51175150078626e-06, "loss": 0.5968, "mean_token_accuracy": 0.8036259412765503, "num_tokens": 76722663.0, "step": 3671 }, { "epoch": 0.4853291038858049, "grad_norm": 0.8725109960620131, "learning_rate": 6.50972830383075e-06, "loss": 0.635, "mean_token_accuracy": 0.7919057607650757, "num_tokens": 76788199.0, "step": 3672 }, { "epoch": 0.48546127412106793, "grad_norm": 0.8349053396832923, "learning_rate": 6.507704891925166e-06, "loss": 0.5694, "mean_token_accuracy": 0.811851441860199, "num_tokens": 76853735.0, "step": 3673 }, { "epoch": 0.48559344435633095, "grad_norm": 0.8998295683686989, "learning_rate": 6.505681265500251e-06, "loss": 0.6224, "mean_token_accuracy": 0.7959955930709839, "num_tokens": 76919271.0, "step": 3674 }, { "epoch": 0.48572561459159397, "grad_norm": 0.8631501966979175, "learning_rate": 6.50365742498679e-06, "loss": 0.6598, "mean_token_accuracy": 0.7807044386863708, "num_tokens": 76984807.0, "step": 3675 }, { "epoch": 0.485857784826857, "grad_norm": 0.9523156936639859, "learning_rate": 6.50163337081562e-06, "loss": 0.5984, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 77050343.0, "step": 3676 }, { "epoch": 0.48598995506212, "grad_norm": 0.8582496732275174, "learning_rate": 6.499609103417615e-06, "loss": 0.5741, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 77115879.0, "step": 3677 }, { "epoch": 0.486122125297383, "grad_norm": 0.8356668741761747, "learning_rate": 6.497584623223706e-06, "loss": 0.6115, "mean_token_accuracy": 0.797674298286438, "num_tokens": 77181415.0, "step": 3678 }, { "epoch": 0.48625429553264604, "grad_norm": 0.9807510078785195, "learning_rate": 6.495559930664857e-06, "loss": 0.6573, "mean_token_accuracy": 0.7845348715782166, "num_tokens": 77246951.0, "step": 3679 }, { "epoch": 0.48638646576790906, "grad_norm": 0.8444551936259824, "learning_rate": 6.493535026172083e-06, "loss": 0.5766, "mean_token_accuracy": 0.806678056716919, "num_tokens": 77312487.0, "step": 3680 }, { "epoch": 0.4865186360031721, "grad_norm": 0.925950412936906, "learning_rate": 6.491509910176448e-06, "loss": 0.6293, "mean_token_accuracy": 0.7910206317901611, "num_tokens": 77378023.0, "step": 3681 }, { "epoch": 0.4866508062384351, "grad_norm": 0.9667222085368435, "learning_rate": 6.489484583109053e-06, "loss": 0.631, "mean_token_accuracy": 0.7901507616043091, "num_tokens": 77443559.0, "step": 3682 }, { "epoch": 0.4867829764736981, "grad_norm": 0.8935344302698126, "learning_rate": 6.48745904540105e-06, "loss": 0.602, "mean_token_accuracy": 0.8038243055343628, "num_tokens": 77509095.0, "step": 3683 }, { "epoch": 0.48691514670896113, "grad_norm": 0.9846818312760265, "learning_rate": 6.485433297483632e-06, "loss": 0.6227, "mean_token_accuracy": 0.7929434776306152, "num_tokens": 77574631.0, "step": 3684 }, { "epoch": 0.48704731694422415, "grad_norm": 0.8625909208764666, "learning_rate": 6.483407339788039e-06, "loss": 0.6026, "mean_token_accuracy": 0.8014131188392639, "num_tokens": 77640167.0, "step": 3685 }, { "epoch": 0.48717948717948717, "grad_norm": 0.8310277740975021, "learning_rate": 6.481381172745557e-06, "loss": 0.564, "mean_token_accuracy": 0.813148558139801, "num_tokens": 77705703.0, "step": 3686 }, { "epoch": 0.4873116574147502, "grad_norm": 0.9072213540415599, "learning_rate": 6.479354796787511e-06, "loss": 0.6551, "mean_token_accuracy": 0.7840160131454468, "num_tokens": 77771239.0, "step": 3687 }, { "epoch": 0.4874438276500132, "grad_norm": 0.8550644563118155, "learning_rate": 6.477328212345278e-06, "loss": 0.5252, "mean_token_accuracy": 0.8263795375823975, "num_tokens": 77836775.0, "step": 3688 }, { "epoch": 0.4875759978852762, "grad_norm": 0.8645793558455486, "learning_rate": 6.475301419850273e-06, "loss": 0.5597, "mean_token_accuracy": 0.8167653679847717, "num_tokens": 77902311.0, "step": 3689 }, { "epoch": 0.48770816812053924, "grad_norm": 0.8572268558459242, "learning_rate": 6.473274419733958e-06, "loss": 0.5917, "mean_token_accuracy": 0.806525468826294, "num_tokens": 77967847.0, "step": 3690 }, { "epoch": 0.48784033835580226, "grad_norm": 0.9091965179204328, "learning_rate": 6.471247212427841e-06, "loss": 0.6346, "mean_token_accuracy": 0.7940880060195923, "num_tokens": 78033383.0, "step": 3691 }, { "epoch": 0.4879725085910653, "grad_norm": 0.8740374048394027, "learning_rate": 6.469219798363469e-06, "loss": 0.608, "mean_token_accuracy": 0.7992766499519348, "num_tokens": 78098919.0, "step": 3692 }, { "epoch": 0.4881046788263283, "grad_norm": 0.8799427491757966, "learning_rate": 6.46719217797244e-06, "loss": 0.5657, "mean_token_accuracy": 0.8117598295211792, "num_tokens": 78164455.0, "step": 3693 }, { "epoch": 0.4882368490615913, "grad_norm": 0.8479861632793836, "learning_rate": 6.465164351686388e-06, "loss": 0.5971, "mean_token_accuracy": 0.8038090467453003, "num_tokens": 78229991.0, "step": 3694 }, { "epoch": 0.48836901929685433, "grad_norm": 0.846521940331842, "learning_rate": 6.463136319936998e-06, "loss": 0.5624, "mean_token_accuracy": 0.811912477016449, "num_tokens": 78295527.0, "step": 3695 }, { "epoch": 0.48850118953211735, "grad_norm": 0.8274665368032247, "learning_rate": 6.461108083155993e-06, "loss": 0.544, "mean_token_accuracy": 0.8197106719017029, "num_tokens": 78361063.0, "step": 3696 }, { "epoch": 0.48863335976738037, "grad_norm": 0.8817734163375296, "learning_rate": 6.459079641775144e-06, "loss": 0.6083, "mean_token_accuracy": 0.7981931567192078, "num_tokens": 78426599.0, "step": 3697 }, { "epoch": 0.4887655300026434, "grad_norm": 0.8365384622863227, "learning_rate": 6.457050996226264e-06, "loss": 0.5508, "mean_token_accuracy": 0.816582202911377, "num_tokens": 78492135.0, "step": 3698 }, { "epoch": 0.4888977002379064, "grad_norm": 0.840606865061193, "learning_rate": 6.455022146941209e-06, "loss": 0.5729, "mean_token_accuracy": 0.8118667006492615, "num_tokens": 78557671.0, "step": 3699 }, { "epoch": 0.4890298704731694, "grad_norm": 0.8117531179520155, "learning_rate": 6.452993094351876e-06, "loss": 0.5668, "mean_token_accuracy": 0.8135606050491333, "num_tokens": 78623207.0, "step": 3700 }, { "epoch": 0.48916204070843244, "grad_norm": 0.8660629484750196, "learning_rate": 6.450963838890212e-06, "loss": 0.6027, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 78688743.0, "step": 3701 }, { "epoch": 0.48929421094369546, "grad_norm": 0.8337813102199377, "learning_rate": 6.448934380988202e-06, "loss": 0.5379, "mean_token_accuracy": 0.8220149874687195, "num_tokens": 78754279.0, "step": 3702 }, { "epoch": 0.4894263811789585, "grad_norm": 0.8983388601287585, "learning_rate": 6.446904721077873e-06, "loss": 0.6051, "mean_token_accuracy": 0.79866623878479, "num_tokens": 78819815.0, "step": 3703 }, { "epoch": 0.4895585514142215, "grad_norm": 0.8284012629503544, "learning_rate": 6.444874859591301e-06, "loss": 0.5841, "mean_token_accuracy": 0.8091960549354553, "num_tokens": 78885351.0, "step": 3704 }, { "epoch": 0.4896907216494845, "grad_norm": 0.8845273862659044, "learning_rate": 6.442844796960597e-06, "loss": 0.6, "mean_token_accuracy": 0.8016725778579712, "num_tokens": 78950887.0, "step": 3705 }, { "epoch": 0.48982289188474754, "grad_norm": 0.9184650364306493, "learning_rate": 6.440814533617921e-06, "loss": 0.6231, "mean_token_accuracy": 0.7970638275146484, "num_tokens": 79016423.0, "step": 3706 }, { "epoch": 0.48995506212001055, "grad_norm": 0.842570218466682, "learning_rate": 6.438784069995474e-06, "loss": 0.5766, "mean_token_accuracy": 0.8113631010055542, "num_tokens": 79081959.0, "step": 3707 }, { "epoch": 0.49008723235527357, "grad_norm": 0.9869995705247968, "learning_rate": 6.436753406525499e-06, "loss": 0.6551, "mean_token_accuracy": 0.784489095211029, "num_tokens": 79147495.0, "step": 3708 }, { "epoch": 0.4902194025905366, "grad_norm": 0.8682503772604125, "learning_rate": 6.434722543640281e-06, "loss": 0.591, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 79213031.0, "step": 3709 }, { "epoch": 0.4903515728257996, "grad_norm": 0.8275557983206441, "learning_rate": 6.432691481772151e-06, "loss": 0.6021, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 79278567.0, "step": 3710 }, { "epoch": 0.4904837430610626, "grad_norm": 0.8223059509503462, "learning_rate": 6.430660221353477e-06, "loss": 0.5657, "mean_token_accuracy": 0.8146135807037354, "num_tokens": 79344103.0, "step": 3711 }, { "epoch": 0.49061591329632565, "grad_norm": 0.9769435453141625, "learning_rate": 6.428628762816671e-06, "loss": 0.6158, "mean_token_accuracy": 0.7963466048240662, "num_tokens": 79409639.0, "step": 3712 }, { "epoch": 0.49074808353158866, "grad_norm": 0.886650542973576, "learning_rate": 6.426597106594191e-06, "loss": 0.5956, "mean_token_accuracy": 0.8019458055496216, "num_tokens": 79470701.0, "step": 3713 }, { "epoch": 0.4908802537668517, "grad_norm": 0.8810588090837154, "learning_rate": 6.424565253118533e-06, "loss": 0.6299, "mean_token_accuracy": 0.7920888662338257, "num_tokens": 79536237.0, "step": 3714 }, { "epoch": 0.4910124240021147, "grad_norm": 0.891565156462399, "learning_rate": 6.422533202822237e-06, "loss": 0.6275, "mean_token_accuracy": 0.7982236742973328, "num_tokens": 79601773.0, "step": 3715 }, { "epoch": 0.4911445942373777, "grad_norm": 0.9251819637221953, "learning_rate": 6.420500956137883e-06, "loss": 0.6487, "mean_token_accuracy": 0.7866255640983582, "num_tokens": 79667309.0, "step": 3716 }, { "epoch": 0.49127676447264074, "grad_norm": 0.9183141782353466, "learning_rate": 6.418468513498093e-06, "loss": 0.5596, "mean_token_accuracy": 0.8168264031410217, "num_tokens": 79732845.0, "step": 3717 }, { "epoch": 0.49140893470790376, "grad_norm": 0.8334907430765023, "learning_rate": 6.4164358753355325e-06, "loss": 0.5543, "mean_token_accuracy": 0.8171926736831665, "num_tokens": 79798381.0, "step": 3718 }, { "epoch": 0.4915411049431668, "grad_norm": 0.8438335953268149, "learning_rate": 6.41440304208291e-06, "loss": 0.5964, "mean_token_accuracy": 0.8048162460327148, "num_tokens": 79863917.0, "step": 3719 }, { "epoch": 0.4916732751784298, "grad_norm": 0.8384243864352835, "learning_rate": 6.41237001417297e-06, "loss": 0.5847, "mean_token_accuracy": 0.8094402551651001, "num_tokens": 79929453.0, "step": 3720 }, { "epoch": 0.4918054454136928, "grad_norm": 0.8781529556830827, "learning_rate": 6.410336792038502e-06, "loss": 0.6262, "mean_token_accuracy": 0.7960261106491089, "num_tokens": 79994989.0, "step": 3721 }, { "epoch": 0.49193761564895583, "grad_norm": 0.8724475858715118, "learning_rate": 6.408303376112337e-06, "loss": 0.5861, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 80060525.0, "step": 3722 }, { "epoch": 0.49206978588421885, "grad_norm": 0.9279507877368831, "learning_rate": 6.4062697668273485e-06, "loss": 0.6064, "mean_token_accuracy": 0.79866623878479, "num_tokens": 80126061.0, "step": 3723 }, { "epoch": 0.4922019561194819, "grad_norm": 0.922643023623641, "learning_rate": 6.404235964616448e-06, "loss": 0.6354, "mean_token_accuracy": 0.792409360408783, "num_tokens": 80191597.0, "step": 3724 }, { "epoch": 0.49233412635474494, "grad_norm": 0.8536317948943448, "learning_rate": 6.402201969912589e-06, "loss": 0.5765, "mean_token_accuracy": 0.8111494183540344, "num_tokens": 80257133.0, "step": 3725 }, { "epoch": 0.49246629659000796, "grad_norm": 0.7940712842690357, "learning_rate": 6.4001677831487685e-06, "loss": 0.5755, "mean_token_accuracy": 0.808478832244873, "num_tokens": 80322669.0, "step": 3726 }, { "epoch": 0.492598466825271, "grad_norm": 0.9016815629784843, "learning_rate": 6.39813340475802e-06, "loss": 0.5985, "mean_token_accuracy": 0.8044347167015076, "num_tokens": 80388205.0, "step": 3727 }, { "epoch": 0.492730637060534, "grad_norm": 0.7834889973017068, "learning_rate": 6.396098835173423e-06, "loss": 0.522, "mean_token_accuracy": 0.8266084790229797, "num_tokens": 80453741.0, "step": 3728 }, { "epoch": 0.492862807295797, "grad_norm": 0.8740980738734893, "learning_rate": 6.394064074828093e-06, "loss": 0.6159, "mean_token_accuracy": 0.7977200746536255, "num_tokens": 80519277.0, "step": 3729 }, { "epoch": 0.49299497753106003, "grad_norm": 0.8408460020124136, "learning_rate": 6.392029124155187e-06, "loss": 0.6049, "mean_token_accuracy": 0.8025729656219482, "num_tokens": 80584813.0, "step": 3730 }, { "epoch": 0.49312714776632305, "grad_norm": 0.862263307372192, "learning_rate": 6.389993983587904e-06, "loss": 0.6257, "mean_token_accuracy": 0.7967739105224609, "num_tokens": 80650349.0, "step": 3731 }, { "epoch": 0.49325931800158607, "grad_norm": 0.9680680204829363, "learning_rate": 6.387958653559484e-06, "loss": 0.5918, "mean_token_accuracy": 0.8051367402076721, "num_tokens": 80715885.0, "step": 3732 }, { "epoch": 0.4933914882368491, "grad_norm": 0.8685506489288606, "learning_rate": 6.385923134503208e-06, "loss": 0.5612, "mean_token_accuracy": 0.8159260153770447, "num_tokens": 80781421.0, "step": 3733 }, { "epoch": 0.4935236584721121, "grad_norm": 0.8731041395901766, "learning_rate": 6.383887426852393e-06, "loss": 0.6113, "mean_token_accuracy": 0.7999023199081421, "num_tokens": 80846957.0, "step": 3734 }, { "epoch": 0.4936558287073751, "grad_norm": 0.9266207332272397, "learning_rate": 6.381851531040397e-06, "loss": 0.6684, "mean_token_accuracy": 0.7809180617332458, "num_tokens": 80912493.0, "step": 3735 }, { "epoch": 0.49378799894263814, "grad_norm": 0.8474700788931874, "learning_rate": 6.379815447500624e-06, "loss": 0.623, "mean_token_accuracy": 0.7942101359367371, "num_tokens": 80978029.0, "step": 3736 }, { "epoch": 0.49392016917790116, "grad_norm": 0.8593758190886697, "learning_rate": 6.377779176666509e-06, "loss": 0.5869, "mean_token_accuracy": 0.8040379881858826, "num_tokens": 81043565.0, "step": 3737 }, { "epoch": 0.4940523394131642, "grad_norm": 0.8711460628552549, "learning_rate": 6.375742718971535e-06, "loss": 0.5685, "mean_token_accuracy": 0.8135148286819458, "num_tokens": 81109101.0, "step": 3738 }, { "epoch": 0.4941845096484272, "grad_norm": 0.9092065005336699, "learning_rate": 6.373706074849221e-06, "loss": 0.6055, "mean_token_accuracy": 0.8032291531562805, "num_tokens": 81174637.0, "step": 3739 }, { "epoch": 0.4943166798836902, "grad_norm": 0.9354879828379282, "learning_rate": 6.371669244733124e-06, "loss": 0.6416, "mean_token_accuracy": 0.7881516218185425, "num_tokens": 81240173.0, "step": 3740 }, { "epoch": 0.49444885011895323, "grad_norm": 0.9024843734029317, "learning_rate": 6.369632229056844e-06, "loss": 0.6183, "mean_token_accuracy": 0.7983152270317078, "num_tokens": 81305709.0, "step": 3741 }, { "epoch": 0.49458102035421625, "grad_norm": 0.8561670535733814, "learning_rate": 6.367595028254021e-06, "loss": 0.617, "mean_token_accuracy": 0.7982084155082703, "num_tokens": 81371245.0, "step": 3742 }, { "epoch": 0.49471319058947927, "grad_norm": 0.889539515508935, "learning_rate": 6.36555764275833e-06, "loss": 0.6249, "mean_token_accuracy": 0.7964686751365662, "num_tokens": 81436781.0, "step": 3743 }, { "epoch": 0.4948453608247423, "grad_norm": 0.8967302404823615, "learning_rate": 6.363520073003488e-06, "loss": 0.5945, "mean_token_accuracy": 0.8049077987670898, "num_tokens": 81502317.0, "step": 3744 }, { "epoch": 0.4949775310600053, "grad_norm": 0.8232168477846045, "learning_rate": 6.36148231942325e-06, "loss": 0.6163, "mean_token_accuracy": 0.7998718023300171, "num_tokens": 81567853.0, "step": 3745 }, { "epoch": 0.4951097012952683, "grad_norm": 1.000656181993364, "learning_rate": 6.359444382451414e-06, "loss": 0.6419, "mean_token_accuracy": 0.7905933260917664, "num_tokens": 81633389.0, "step": 3746 }, { "epoch": 0.49524187153053134, "grad_norm": 0.8510943127222537, "learning_rate": 6.357406262521813e-06, "loss": 0.5804, "mean_token_accuracy": 0.8098675608634949, "num_tokens": 81698925.0, "step": 3747 }, { "epoch": 0.49537404176579436, "grad_norm": 0.8349640701258959, "learning_rate": 6.355367960068321e-06, "loss": 0.5909, "mean_token_accuracy": 0.806647539138794, "num_tokens": 81764461.0, "step": 3748 }, { "epoch": 0.4955062120010574, "grad_norm": 0.8616823477913809, "learning_rate": 6.353329475524846e-06, "loss": 0.6174, "mean_token_accuracy": 0.7966976165771484, "num_tokens": 81829997.0, "step": 3749 }, { "epoch": 0.4956383822363204, "grad_norm": 0.9569505536524359, "learning_rate": 6.351290809325344e-06, "loss": 0.6653, "mean_token_accuracy": 0.7795140743255615, "num_tokens": 81895533.0, "step": 3750 }, { "epoch": 0.4957705524715834, "grad_norm": 0.8049142447681896, "learning_rate": 6.349251961903802e-06, "loss": 0.5918, "mean_token_accuracy": 0.8041905760765076, "num_tokens": 81961069.0, "step": 3751 }, { "epoch": 0.49590272270684643, "grad_norm": 0.7698189374609969, "learning_rate": 6.3472129336942504e-06, "loss": 0.5557, "mean_token_accuracy": 0.8163533210754395, "num_tokens": 82026605.0, "step": 3752 }, { "epoch": 0.49603489294210945, "grad_norm": 0.8632293381678535, "learning_rate": 6.345173725130752e-06, "loss": 0.5607, "mean_token_accuracy": 0.8154681921005249, "num_tokens": 82092141.0, "step": 3753 }, { "epoch": 0.49616706317737247, "grad_norm": 0.9001325916798971, "learning_rate": 6.343134336647413e-06, "loss": 0.5492, "mean_token_accuracy": 0.8196648955345154, "num_tokens": 82157677.0, "step": 3754 }, { "epoch": 0.4962992334126355, "grad_norm": 0.9377629889251723, "learning_rate": 6.341094768678378e-06, "loss": 0.6105, "mean_token_accuracy": 0.7975521683692932, "num_tokens": 82223213.0, "step": 3755 }, { "epoch": 0.4964314036478985, "grad_norm": 0.8329176315763639, "learning_rate": 6.339055021657827e-06, "loss": 0.6145, "mean_token_accuracy": 0.7955682873725891, "num_tokens": 82288749.0, "step": 3756 }, { "epoch": 0.4965635738831615, "grad_norm": 0.8849242881434528, "learning_rate": 6.337015096019981e-06, "loss": 0.6073, "mean_token_accuracy": 0.7977505922317505, "num_tokens": 82354285.0, "step": 3757 }, { "epoch": 0.49669574411842454, "grad_norm": 0.7945866788709666, "learning_rate": 6.334974992199095e-06, "loss": 0.557, "mean_token_accuracy": 0.8155597448348999, "num_tokens": 82419821.0, "step": 3758 }, { "epoch": 0.49682791435368756, "grad_norm": 0.9174477132297696, "learning_rate": 6.332934710629467e-06, "loss": 0.6256, "mean_token_accuracy": 0.7937980890274048, "num_tokens": 82485357.0, "step": 3759 }, { "epoch": 0.4969600845889506, "grad_norm": 0.8564870319735471, "learning_rate": 6.33089425174543e-06, "loss": 0.6153, "mean_token_accuracy": 0.7986509799957275, "num_tokens": 82550893.0, "step": 3760 }, { "epoch": 0.4970922548242136, "grad_norm": 0.9163911278577522, "learning_rate": 6.3288536159813525e-06, "loss": 0.6125, "mean_token_accuracy": 0.7999938726425171, "num_tokens": 82616429.0, "step": 3761 }, { "epoch": 0.4972244250594766, "grad_norm": 0.9118477086453036, "learning_rate": 6.326812803771648e-06, "loss": 0.6357, "mean_token_accuracy": 0.7896776795387268, "num_tokens": 82681965.0, "step": 3762 }, { "epoch": 0.49735659529473963, "grad_norm": 0.8869868962495809, "learning_rate": 6.32477181555076e-06, "loss": 0.5858, "mean_token_accuracy": 0.806861162185669, "num_tokens": 82747501.0, "step": 3763 }, { "epoch": 0.49748876553000265, "grad_norm": 0.8722251395127344, "learning_rate": 6.322730651753171e-06, "loss": 0.5744, "mean_token_accuracy": 0.8106763362884521, "num_tokens": 82813037.0, "step": 3764 }, { "epoch": 0.49762093576526567, "grad_norm": 0.9118486424065734, "learning_rate": 6.320689312813405e-06, "loss": 0.5976, "mean_token_accuracy": 0.803091824054718, "num_tokens": 82878573.0, "step": 3765 }, { "epoch": 0.4977531060005287, "grad_norm": 0.848459168084872, "learning_rate": 6.318647799166018e-06, "loss": 0.6106, "mean_token_accuracy": 0.7991088032722473, "num_tokens": 82944109.0, "step": 3766 }, { "epoch": 0.4978852762357917, "grad_norm": 0.8970897549973339, "learning_rate": 6.316606111245607e-06, "loss": 0.655, "mean_token_accuracy": 0.7863508462905884, "num_tokens": 83009645.0, "step": 3767 }, { "epoch": 0.4980174464710547, "grad_norm": 0.8686438581434328, "learning_rate": 6.314564249486804e-06, "loss": 0.6232, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 83075181.0, "step": 3768 }, { "epoch": 0.49814961670631774, "grad_norm": 0.9829301260796871, "learning_rate": 6.312522214324279e-06, "loss": 0.6404, "mean_token_accuracy": 0.7908527851104736, "num_tokens": 83140717.0, "step": 3769 }, { "epoch": 0.49828178694158076, "grad_norm": 0.8721995091247222, "learning_rate": 6.31048000619274e-06, "loss": 0.5797, "mean_token_accuracy": 0.8100964426994324, "num_tokens": 83206253.0, "step": 3770 }, { "epoch": 0.4984139571768438, "grad_norm": 0.8493298167815281, "learning_rate": 6.30843762552693e-06, "loss": 0.5752, "mean_token_accuracy": 0.8107678890228271, "num_tokens": 83271789.0, "step": 3771 }, { "epoch": 0.4985461274121068, "grad_norm": 0.8522112589991521, "learning_rate": 6.306395072761628e-06, "loss": 0.6133, "mean_token_accuracy": 0.7975216507911682, "num_tokens": 83337325.0, "step": 3772 }, { "epoch": 0.4986782976473698, "grad_norm": 0.8101273213114502, "learning_rate": 6.3043523483316495e-06, "loss": 0.5526, "mean_token_accuracy": 0.818138837814331, "num_tokens": 83402861.0, "step": 3773 }, { "epoch": 0.49881046788263284, "grad_norm": 0.8771700377233198, "learning_rate": 6.302309452671852e-06, "loss": 0.6144, "mean_token_accuracy": 0.7963771224021912, "num_tokens": 83468397.0, "step": 3774 }, { "epoch": 0.49894263811789585, "grad_norm": 0.860917964266101, "learning_rate": 6.300266386217121e-06, "loss": 0.6231, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 83533933.0, "step": 3775 }, { "epoch": 0.49907480835315887, "grad_norm": 0.8237751285940308, "learning_rate": 6.298223149402387e-06, "loss": 0.6131, "mean_token_accuracy": 0.7996124029159546, "num_tokens": 83599469.0, "step": 3776 }, { "epoch": 0.4992069785884219, "grad_norm": 0.8666628412891524, "learning_rate": 6.296179742662607e-06, "loss": 0.5715, "mean_token_accuracy": 0.8124313354492188, "num_tokens": 83665005.0, "step": 3777 }, { "epoch": 0.4993391488236849, "grad_norm": 0.7622123877073328, "learning_rate": 6.294136166432785e-06, "loss": 0.5212, "mean_token_accuracy": 0.8274173140525818, "num_tokens": 83730541.0, "step": 3778 }, { "epoch": 0.4994713190589479, "grad_norm": 0.8846992404421289, "learning_rate": 6.292092421147954e-06, "loss": 0.6, "mean_token_accuracy": 0.8014283776283264, "num_tokens": 83796077.0, "step": 3779 }, { "epoch": 0.49960348929421095, "grad_norm": 0.9092365292632062, "learning_rate": 6.290048507243183e-06, "loss": 0.6254, "mean_token_accuracy": 0.7930502891540527, "num_tokens": 83861613.0, "step": 3780 }, { "epoch": 0.49973565952947396, "grad_norm": 0.9760967917341973, "learning_rate": 6.288004425153581e-06, "loss": 0.6112, "mean_token_accuracy": 0.797735333442688, "num_tokens": 83927149.0, "step": 3781 }, { "epoch": 0.499867829764737, "grad_norm": 0.8003550433744991, "learning_rate": 6.285960175314287e-06, "loss": 0.5497, "mean_token_accuracy": 0.8186576962471008, "num_tokens": 83992685.0, "step": 3782 }, { "epoch": 0.5, "grad_norm": 0.8535177487029763, "learning_rate": 6.283915758160483e-06, "loss": 0.6049, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 84058221.0, "step": 3783 }, { "epoch": 0.500132170235263, "grad_norm": 0.9223257591500644, "learning_rate": 6.281871174127381e-06, "loss": 0.6112, "mean_token_accuracy": 0.7982389330863953, "num_tokens": 84123757.0, "step": 3784 }, { "epoch": 0.500264340470526, "grad_norm": 0.8406756153494465, "learning_rate": 6.27982642365023e-06, "loss": 0.5904, "mean_token_accuracy": 0.8050757050514221, "num_tokens": 84189293.0, "step": 3785 }, { "epoch": 0.500396510705789, "grad_norm": 0.8443773344015328, "learning_rate": 6.277781507164315e-06, "loss": 0.5662, "mean_token_accuracy": 0.811973512172699, "num_tokens": 84254829.0, "step": 3786 }, { "epoch": 0.5005286809410521, "grad_norm": 0.8369547154854149, "learning_rate": 6.275736425104957e-06, "loss": 0.5767, "mean_token_accuracy": 0.8086161613464355, "num_tokens": 84320365.0, "step": 3787 }, { "epoch": 0.5006608511763151, "grad_norm": 0.8698511141188897, "learning_rate": 6.273691177907508e-06, "loss": 0.6124, "mean_token_accuracy": 0.7981321215629578, "num_tokens": 84385901.0, "step": 3788 }, { "epoch": 0.5007930214115781, "grad_norm": 0.8641954599986883, "learning_rate": 6.271645766007364e-06, "loss": 0.5469, "mean_token_accuracy": 0.8199395537376404, "num_tokens": 84451437.0, "step": 3789 }, { "epoch": 0.5009251916468411, "grad_norm": 0.8421248741694276, "learning_rate": 6.269600189839945e-06, "loss": 0.5898, "mean_token_accuracy": 0.806678056716919, "num_tokens": 84516973.0, "step": 3790 }, { "epoch": 0.5010573618821041, "grad_norm": 0.8398464813204648, "learning_rate": 6.267554449840715e-06, "loss": 0.6178, "mean_token_accuracy": 0.7956598997116089, "num_tokens": 84582509.0, "step": 3791 }, { "epoch": 0.5011895321173672, "grad_norm": 0.8422250772184023, "learning_rate": 6.265508546445166e-06, "loss": 0.6059, "mean_token_accuracy": 0.7999175786972046, "num_tokens": 84648045.0, "step": 3792 }, { "epoch": 0.5013217023526302, "grad_norm": 0.8911598672521363, "learning_rate": 6.263462480088832e-06, "loss": 0.6018, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 84713581.0, "step": 3793 }, { "epoch": 0.5014538725878932, "grad_norm": 0.835345321490545, "learning_rate": 6.261416251207274e-06, "loss": 0.5476, "mean_token_accuracy": 0.8202905654907227, "num_tokens": 84779117.0, "step": 3794 }, { "epoch": 0.5015860428231562, "grad_norm": 0.8824542416700659, "learning_rate": 6.259369860236095e-06, "loss": 0.618, "mean_token_accuracy": 0.7962855696678162, "num_tokens": 84844653.0, "step": 3795 }, { "epoch": 0.5017182130584192, "grad_norm": 0.8750949804051652, "learning_rate": 6.257323307610928e-06, "loss": 0.6249, "mean_token_accuracy": 0.7935081124305725, "num_tokens": 84910189.0, "step": 3796 }, { "epoch": 0.5018503832936823, "grad_norm": 0.8060138065675142, "learning_rate": 6.25527659376744e-06, "loss": 0.5905, "mean_token_accuracy": 0.8056555986404419, "num_tokens": 84975725.0, "step": 3797 }, { "epoch": 0.5019825535289453, "grad_norm": 0.8584572409072566, "learning_rate": 6.253229719141333e-06, "loss": 0.6456, "mean_token_accuracy": 0.7849010825157166, "num_tokens": 85041261.0, "step": 3798 }, { "epoch": 0.5021147237642083, "grad_norm": 0.8656700982926033, "learning_rate": 6.251182684168345e-06, "loss": 0.5874, "mean_token_accuracy": 0.8066017627716064, "num_tokens": 85106797.0, "step": 3799 }, { "epoch": 0.5022468939994713, "grad_norm": 0.843423657037279, "learning_rate": 6.2491354892842485e-06, "loss": 0.5627, "mean_token_accuracy": 0.8129196763038635, "num_tokens": 85172333.0, "step": 3800 }, { "epoch": 0.5023790642347343, "grad_norm": 0.9932187163183829, "learning_rate": 6.247088134924847e-06, "loss": 0.646, "mean_token_accuracy": 0.7868391871452332, "num_tokens": 85237869.0, "step": 3801 }, { "epoch": 0.5025112344699973, "grad_norm": 0.8555086651267179, "learning_rate": 6.245040621525979e-06, "loss": 0.5526, "mean_token_accuracy": 0.8169789910316467, "num_tokens": 85303405.0, "step": 3802 }, { "epoch": 0.5026434047052604, "grad_norm": 0.8980726401227684, "learning_rate": 6.242992949523518e-06, "loss": 0.5864, "mean_token_accuracy": 0.8040379881858826, "num_tokens": 85368941.0, "step": 3803 }, { "epoch": 0.5027755749405234, "grad_norm": 0.8169533059870576, "learning_rate": 6.24094511935337e-06, "loss": 0.5139, "mean_token_accuracy": 0.8297216296195984, "num_tokens": 85434477.0, "step": 3804 }, { "epoch": 0.5029077451757864, "grad_norm": 0.8428933113118616, "learning_rate": 6.2388971314514766e-06, "loss": 0.5653, "mean_token_accuracy": 0.8134995698928833, "num_tokens": 85500013.0, "step": 3805 }, { "epoch": 0.5030399154110494, "grad_norm": 0.7898484457106898, "learning_rate": 6.2368489862538115e-06, "loss": 0.5543, "mean_token_accuracy": 0.8152850866317749, "num_tokens": 85565549.0, "step": 3806 }, { "epoch": 0.5031720856463124, "grad_norm": 0.9473502868122661, "learning_rate": 6.234800684196381e-06, "loss": 0.6092, "mean_token_accuracy": 0.797887921333313, "num_tokens": 85631085.0, "step": 3807 }, { "epoch": 0.5033042558815755, "grad_norm": 0.831852731288633, "learning_rate": 6.2327522257152275e-06, "loss": 0.5877, "mean_token_accuracy": 0.8055945634841919, "num_tokens": 85696621.0, "step": 3808 }, { "epoch": 0.5034364261168385, "grad_norm": 0.9330496823737063, "learning_rate": 6.230703611246421e-06, "loss": 0.6211, "mean_token_accuracy": 0.7969722747802734, "num_tokens": 85762157.0, "step": 3809 }, { "epoch": 0.5035685963521015, "grad_norm": 0.9652575712148317, "learning_rate": 6.228654841226075e-06, "loss": 0.6571, "mean_token_accuracy": 0.783649742603302, "num_tokens": 85827693.0, "step": 3810 }, { "epoch": 0.5037007665873645, "grad_norm": 0.9121362354995215, "learning_rate": 6.226605916090324e-06, "loss": 0.6239, "mean_token_accuracy": 0.7968654632568359, "num_tokens": 85893229.0, "step": 3811 }, { "epoch": 0.5038329368226275, "grad_norm": 0.9568839681606871, "learning_rate": 6.224556836275346e-06, "loss": 0.5953, "mean_token_accuracy": 0.8041142821311951, "num_tokens": 85958765.0, "step": 3812 }, { "epoch": 0.5039651070578905, "grad_norm": 0.939981849359976, "learning_rate": 6.222507602217345e-06, "loss": 0.5958, "mean_token_accuracy": 0.8057471513748169, "num_tokens": 86024301.0, "step": 3813 }, { "epoch": 0.5040972772931536, "grad_norm": 0.9211489772421045, "learning_rate": 6.220458214352559e-06, "loss": 0.6053, "mean_token_accuracy": 0.7997802495956421, "num_tokens": 86089837.0, "step": 3814 }, { "epoch": 0.5042294475284166, "grad_norm": 0.8723322051374182, "learning_rate": 6.218408673117261e-06, "loss": 0.5811, "mean_token_accuracy": 0.8055487871170044, "num_tokens": 86155373.0, "step": 3815 }, { "epoch": 0.5043616177636796, "grad_norm": 0.9248070880010089, "learning_rate": 6.216358978947757e-06, "loss": 0.5661, "mean_token_accuracy": 0.8134995698928833, "num_tokens": 86220909.0, "step": 3816 }, { "epoch": 0.5044937879989426, "grad_norm": 0.8859424482195472, "learning_rate": 6.214309132280383e-06, "loss": 0.6476, "mean_token_accuracy": 0.7847179770469666, "num_tokens": 86286445.0, "step": 3817 }, { "epoch": 0.5046259582342056, "grad_norm": 0.855205769893485, "learning_rate": 6.212259133551507e-06, "loss": 0.5808, "mean_token_accuracy": 0.8086771965026855, "num_tokens": 86351981.0, "step": 3818 }, { "epoch": 0.5047581284694687, "grad_norm": 0.8689278913708259, "learning_rate": 6.2102089831975335e-06, "loss": 0.5893, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 86417517.0, "step": 3819 }, { "epoch": 0.5048902987047317, "grad_norm": 0.8838389537693924, "learning_rate": 6.208158681654896e-06, "loss": 0.6385, "mean_token_accuracy": 0.7894793152809143, "num_tokens": 86483053.0, "step": 3820 }, { "epoch": 0.5050224689399947, "grad_norm": 0.8505985634947045, "learning_rate": 6.206108229360059e-06, "loss": 0.6261, "mean_token_accuracy": 0.7953851819038391, "num_tokens": 86548589.0, "step": 3821 }, { "epoch": 0.5051546391752577, "grad_norm": 0.8642842129511752, "learning_rate": 6.204057626749523e-06, "loss": 0.5976, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 86614125.0, "step": 3822 }, { "epoch": 0.5052868094105207, "grad_norm": 0.8672948857706974, "learning_rate": 6.202006874259818e-06, "loss": 0.565, "mean_token_accuracy": 0.8133469820022583, "num_tokens": 86679661.0, "step": 3823 }, { "epoch": 0.5054189796457838, "grad_norm": 0.7961921561342193, "learning_rate": 6.199955972327507e-06, "loss": 0.5413, "mean_token_accuracy": 0.821725070476532, "num_tokens": 86745197.0, "step": 3824 }, { "epoch": 0.5055511498810468, "grad_norm": 0.9103229239569839, "learning_rate": 6.197904921389183e-06, "loss": 0.6148, "mean_token_accuracy": 0.797948956489563, "num_tokens": 86810733.0, "step": 3825 }, { "epoch": 0.5056833201163098, "grad_norm": 0.8406301138792378, "learning_rate": 6.195853721881473e-06, "loss": 0.5673, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 86876269.0, "step": 3826 }, { "epoch": 0.5058154903515728, "grad_norm": 0.8071549794767525, "learning_rate": 6.193802374241034e-06, "loss": 0.5505, "mean_token_accuracy": 0.8181540966033936, "num_tokens": 86941805.0, "step": 3827 }, { "epoch": 0.5059476605868358, "grad_norm": 0.8475397411802297, "learning_rate": 6.191750878904553e-06, "loss": 0.5624, "mean_token_accuracy": 0.8125076293945312, "num_tokens": 87007341.0, "step": 3828 }, { "epoch": 0.5060798308220988, "grad_norm": 0.8825304733360548, "learning_rate": 6.189699236308754e-06, "loss": 0.6122, "mean_token_accuracy": 0.799001932144165, "num_tokens": 87072877.0, "step": 3829 }, { "epoch": 0.5062120010573619, "grad_norm": 0.8758538229202841, "learning_rate": 6.187647446890387e-06, "loss": 0.6112, "mean_token_accuracy": 0.7979031801223755, "num_tokens": 87138413.0, "step": 3830 }, { "epoch": 0.5063441712926249, "grad_norm": 0.8287049314398021, "learning_rate": 6.185595511086237e-06, "loss": 0.5843, "mean_token_accuracy": 0.807517409324646, "num_tokens": 87203949.0, "step": 3831 }, { "epoch": 0.5064763415278879, "grad_norm": 0.8989966224690624, "learning_rate": 6.183543429333118e-06, "loss": 0.6073, "mean_token_accuracy": 0.8004211783409119, "num_tokens": 87269485.0, "step": 3832 }, { "epoch": 0.5066085117631509, "grad_norm": 0.8407706363650984, "learning_rate": 6.181491202067871e-06, "loss": 0.6117, "mean_token_accuracy": 0.7982999682426453, "num_tokens": 87335021.0, "step": 3833 }, { "epoch": 0.5067406819984139, "grad_norm": 0.8919259394321206, "learning_rate": 6.1794388297273775e-06, "loss": 0.5927, "mean_token_accuracy": 0.8054113984107971, "num_tokens": 87400557.0, "step": 3834 }, { "epoch": 0.506872852233677, "grad_norm": 0.873647321844657, "learning_rate": 6.1773863127485415e-06, "loss": 0.5896, "mean_token_accuracy": 0.8065712451934814, "num_tokens": 87466093.0, "step": 3835 }, { "epoch": 0.50700502246894, "grad_norm": 0.8999466050863655, "learning_rate": 6.1753336515683025e-06, "loss": 0.5847, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 87531629.0, "step": 3836 }, { "epoch": 0.507137192704203, "grad_norm": 0.8816603776205908, "learning_rate": 6.1732808466236306e-06, "loss": 0.6289, "mean_token_accuracy": 0.7917531728744507, "num_tokens": 87597165.0, "step": 3837 }, { "epoch": 0.507269362939466, "grad_norm": 0.8069281382015745, "learning_rate": 6.171227898351522e-06, "loss": 0.5686, "mean_token_accuracy": 0.8125076293945312, "num_tokens": 87662701.0, "step": 3838 }, { "epoch": 0.507401533174729, "grad_norm": 0.917514517619111, "learning_rate": 6.1691748071890086e-06, "loss": 0.6307, "mean_token_accuracy": 0.7928977012634277, "num_tokens": 87728237.0, "step": 3839 }, { "epoch": 0.507533703409992, "grad_norm": 0.8747477245681254, "learning_rate": 6.167121573573151e-06, "loss": 0.5878, "mean_token_accuracy": 0.8063575625419617, "num_tokens": 87793773.0, "step": 3840 }, { "epoch": 0.5076658736452551, "grad_norm": 0.9583189982105176, "learning_rate": 6.165068197941037e-06, "loss": 0.6229, "mean_token_accuracy": 0.798757791519165, "num_tokens": 87859309.0, "step": 3841 }, { "epoch": 0.5077980438805181, "grad_norm": 0.8667208790004498, "learning_rate": 6.163014680729791e-06, "loss": 0.5745, "mean_token_accuracy": 0.8075631856918335, "num_tokens": 87924845.0, "step": 3842 }, { "epoch": 0.5079302141157811, "grad_norm": 0.8364198961001929, "learning_rate": 6.1609610223765614e-06, "loss": 0.5933, "mean_token_accuracy": 0.8049841523170471, "num_tokens": 87990381.0, "step": 3843 }, { "epoch": 0.5080623843510441, "grad_norm": 0.793073996456685, "learning_rate": 6.15890722331853e-06, "loss": 0.5589, "mean_token_accuracy": 0.8137285113334656, "num_tokens": 88055917.0, "step": 3844 }, { "epoch": 0.5081945545863071, "grad_norm": 0.8335652819987962, "learning_rate": 6.156853283992908e-06, "loss": 0.5915, "mean_token_accuracy": 0.806861162185669, "num_tokens": 88121453.0, "step": 3845 }, { "epoch": 0.5083267248215702, "grad_norm": 0.8272462845778656, "learning_rate": 6.1547992048369355e-06, "loss": 0.6359, "mean_token_accuracy": 0.7900286912918091, "num_tokens": 88186989.0, "step": 3846 }, { "epoch": 0.5084588950568332, "grad_norm": 0.8375513587994173, "learning_rate": 6.152744986287884e-06, "loss": 0.5891, "mean_token_accuracy": 0.8043736815452576, "num_tokens": 88252525.0, "step": 3847 }, { "epoch": 0.5085910652920962, "grad_norm": 0.8893142819640146, "learning_rate": 6.150690628783054e-06, "loss": 0.6063, "mean_token_accuracy": 0.7992003560066223, "num_tokens": 88318061.0, "step": 3848 }, { "epoch": 0.5087232355273592, "grad_norm": 0.8438500731558292, "learning_rate": 6.148636132759772e-06, "loss": 0.6015, "mean_token_accuracy": 0.8022982478141785, "num_tokens": 88383597.0, "step": 3849 }, { "epoch": 0.5088554057626222, "grad_norm": 0.7956875338054609, "learning_rate": 6.146581498655403e-06, "loss": 0.5892, "mean_token_accuracy": 0.8059913516044617, "num_tokens": 88449133.0, "step": 3850 }, { "epoch": 0.5089875759978852, "grad_norm": 0.8554250629667038, "learning_rate": 6.1445267269073315e-06, "loss": 0.5756, "mean_token_accuracy": 0.808784008026123, "num_tokens": 88514669.0, "step": 3851 }, { "epoch": 0.5091197462331483, "grad_norm": 0.8688094297682277, "learning_rate": 6.142471817952975e-06, "loss": 0.6031, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 88580205.0, "step": 3852 }, { "epoch": 0.5092519164684113, "grad_norm": 0.8317607442984275, "learning_rate": 6.140416772229785e-06, "loss": 0.5225, "mean_token_accuracy": 0.8269442319869995, "num_tokens": 88645741.0, "step": 3853 }, { "epoch": 0.5093840867036743, "grad_norm": 0.9367789257780246, "learning_rate": 6.138361590175233e-06, "loss": 0.6503, "mean_token_accuracy": 0.7867476344108582, "num_tokens": 88711277.0, "step": 3854 }, { "epoch": 0.5095162569389373, "grad_norm": 0.8585466960276855, "learning_rate": 6.136306272226826e-06, "loss": 0.6256, "mean_token_accuracy": 0.7931265830993652, "num_tokens": 88776813.0, "step": 3855 }, { "epoch": 0.5096484271742003, "grad_norm": 0.9175368458520786, "learning_rate": 6.1342508188221005e-06, "loss": 0.6371, "mean_token_accuracy": 0.7897234559059143, "num_tokens": 88842349.0, "step": 3856 }, { "epoch": 0.5097805974094634, "grad_norm": 0.8131721328438405, "learning_rate": 6.132195230398617e-06, "loss": 0.5525, "mean_token_accuracy": 0.8157734274864197, "num_tokens": 88907885.0, "step": 3857 }, { "epoch": 0.5099127676447264, "grad_norm": 0.7817235546163375, "learning_rate": 6.130139507393969e-06, "loss": 0.5387, "mean_token_accuracy": 0.8232206106185913, "num_tokens": 88973421.0, "step": 3858 }, { "epoch": 0.5100449378799894, "grad_norm": 0.9560811776309276, "learning_rate": 6.128083650245776e-06, "loss": 0.5915, "mean_token_accuracy": 0.8053656220436096, "num_tokens": 89038957.0, "step": 3859 }, { "epoch": 0.5101771081152524, "grad_norm": 0.9688890154851624, "learning_rate": 6.126027659391686e-06, "loss": 0.6504, "mean_token_accuracy": 0.7880600690841675, "num_tokens": 89104493.0, "step": 3860 }, { "epoch": 0.5103092783505154, "grad_norm": 0.9956419544394365, "learning_rate": 6.123971535269379e-06, "loss": 0.6561, "mean_token_accuracy": 0.7861219644546509, "num_tokens": 89170029.0, "step": 3861 }, { "epoch": 0.5104414485857784, "grad_norm": 0.8515631152953549, "learning_rate": 6.12191527831656e-06, "loss": 0.5372, "mean_token_accuracy": 0.8223049640655518, "num_tokens": 89235565.0, "step": 3862 }, { "epoch": 0.5105736188210415, "grad_norm": 0.9270251755362144, "learning_rate": 6.119858888970963e-06, "loss": 0.6153, "mean_token_accuracy": 0.7970943450927734, "num_tokens": 89301101.0, "step": 3863 }, { "epoch": 0.5107057890563045, "grad_norm": 0.8318161214043208, "learning_rate": 6.11780236767035e-06, "loss": 0.527, "mean_token_accuracy": 0.8276461958885193, "num_tokens": 89366637.0, "step": 3864 }, { "epoch": 0.5108379592915675, "grad_norm": 0.8400922620617001, "learning_rate": 6.115745714852511e-06, "loss": 0.5455, "mean_token_accuracy": 0.8201227188110352, "num_tokens": 89432173.0, "step": 3865 }, { "epoch": 0.5109701295268305, "grad_norm": 0.8869328455339214, "learning_rate": 6.113688930955265e-06, "loss": 0.6158, "mean_token_accuracy": 0.7975826859474182, "num_tokens": 89497709.0, "step": 3866 }, { "epoch": 0.5111022997620935, "grad_norm": 0.8703184184925121, "learning_rate": 6.111632016416459e-06, "loss": 0.5718, "mean_token_accuracy": 0.8114851713180542, "num_tokens": 89563245.0, "step": 3867 }, { "epoch": 0.5112344699973566, "grad_norm": 0.8987217411325232, "learning_rate": 6.109574971673966e-06, "loss": 0.571, "mean_token_accuracy": 0.8119582533836365, "num_tokens": 89628781.0, "step": 3868 }, { "epoch": 0.5113666402326196, "grad_norm": 0.8803450180012576, "learning_rate": 6.107517797165691e-06, "loss": 0.612, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 89694317.0, "step": 3869 }, { "epoch": 0.5114988104678826, "grad_norm": 0.9264934812547968, "learning_rate": 6.105460493329559e-06, "loss": 0.626, "mean_token_accuracy": 0.7936607003211975, "num_tokens": 89759853.0, "step": 3870 }, { "epoch": 0.5116309807031456, "grad_norm": 0.9403158253832979, "learning_rate": 6.103403060603529e-06, "loss": 0.6313, "mean_token_accuracy": 0.792409360408783, "num_tokens": 89825389.0, "step": 3871 }, { "epoch": 0.5117631509384086, "grad_norm": 0.8991639917449211, "learning_rate": 6.101345499425587e-06, "loss": 0.5842, "mean_token_accuracy": 0.8058081865310669, "num_tokens": 89890925.0, "step": 3872 }, { "epoch": 0.5118953211736716, "grad_norm": 0.8941271686404122, "learning_rate": 6.099287810233743e-06, "loss": 0.5894, "mean_token_accuracy": 0.8069680333137512, "num_tokens": 89956461.0, "step": 3873 }, { "epoch": 0.5120274914089347, "grad_norm": 0.8773178107948142, "learning_rate": 6.097229993466036e-06, "loss": 0.5944, "mean_token_accuracy": 0.8027560710906982, "num_tokens": 90021997.0, "step": 3874 }, { "epoch": 0.5121596616441977, "grad_norm": 0.9493310924229256, "learning_rate": 6.095172049560532e-06, "loss": 0.6257, "mean_token_accuracy": 0.7947137355804443, "num_tokens": 90087533.0, "step": 3875 }, { "epoch": 0.5122918318794607, "grad_norm": 0.8653557118533449, "learning_rate": 6.093113978955324e-06, "loss": 0.614, "mean_token_accuracy": 0.7955225110054016, "num_tokens": 90153069.0, "step": 3876 }, { "epoch": 0.5124240021147237, "grad_norm": 0.8359462739513976, "learning_rate": 6.091055782088535e-06, "loss": 0.5663, "mean_token_accuracy": 0.8146746158599854, "num_tokens": 90218605.0, "step": 3877 }, { "epoch": 0.5125561723499867, "grad_norm": 0.8806855129981054, "learning_rate": 6.08899745939831e-06, "loss": 0.6039, "mean_token_accuracy": 0.8016268014907837, "num_tokens": 90284141.0, "step": 3878 }, { "epoch": 0.5126883425852498, "grad_norm": 0.8697974433449197, "learning_rate": 6.086939011322824e-06, "loss": 0.5894, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 90349677.0, "step": 3879 }, { "epoch": 0.5128205128205128, "grad_norm": 0.8781374228947192, "learning_rate": 6.084880438300276e-06, "loss": 0.579, "mean_token_accuracy": 0.808722972869873, "num_tokens": 90415213.0, "step": 3880 }, { "epoch": 0.5129526830557758, "grad_norm": 0.9643875506518735, "learning_rate": 6.082821740768894e-06, "loss": 0.6963, "mean_token_accuracy": 0.7711207270622253, "num_tokens": 90480749.0, "step": 3881 }, { "epoch": 0.5130848532910388, "grad_norm": 0.9710282307205067, "learning_rate": 6.0807629191669325e-06, "loss": 0.6158, "mean_token_accuracy": 0.7971554398536682, "num_tokens": 90546285.0, "step": 3882 }, { "epoch": 0.5132170235263018, "grad_norm": 0.8124124889969622, "learning_rate": 6.0787039739326695e-06, "loss": 0.5222, "mean_token_accuracy": 0.8236173987388611, "num_tokens": 90611821.0, "step": 3883 }, { "epoch": 0.5133491937615648, "grad_norm": 0.8604901733162228, "learning_rate": 6.076644905504414e-06, "loss": 0.6166, "mean_token_accuracy": 0.7962550520896912, "num_tokens": 90677357.0, "step": 3884 }, { "epoch": 0.5134813639968279, "grad_norm": 0.9944541265256278, "learning_rate": 6.0745857143204985e-06, "loss": 0.6326, "mean_token_accuracy": 0.7920583486557007, "num_tokens": 90742893.0, "step": 3885 }, { "epoch": 0.5136135342320909, "grad_norm": 0.9458958396588485, "learning_rate": 6.072526400819279e-06, "loss": 0.6478, "mean_token_accuracy": 0.7865187525749207, "num_tokens": 90808429.0, "step": 3886 }, { "epoch": 0.5137457044673539, "grad_norm": 0.9050356407463145, "learning_rate": 6.070466965439144e-06, "loss": 0.5702, "mean_token_accuracy": 0.8116072416305542, "num_tokens": 90873965.0, "step": 3887 }, { "epoch": 0.5138778747026169, "grad_norm": 0.8726012594691404, "learning_rate": 6.068407408618502e-06, "loss": 0.5866, "mean_token_accuracy": 0.8096386194229126, "num_tokens": 90939501.0, "step": 3888 }, { "epoch": 0.5140100449378799, "grad_norm": 0.8452360115549553, "learning_rate": 6.06634773079579e-06, "loss": 0.6043, "mean_token_accuracy": 0.8001312613487244, "num_tokens": 91005037.0, "step": 3889 }, { "epoch": 0.514142215173143, "grad_norm": 0.8409324116998808, "learning_rate": 6.0642879324094705e-06, "loss": 0.5886, "mean_token_accuracy": 0.8042363524436951, "num_tokens": 91070573.0, "step": 3890 }, { "epoch": 0.514274385408406, "grad_norm": 0.8791638236676167, "learning_rate": 6.062228013898032e-06, "loss": 0.5608, "mean_token_accuracy": 0.816582202911377, "num_tokens": 91136109.0, "step": 3891 }, { "epoch": 0.514406555643669, "grad_norm": 0.8556532796550933, "learning_rate": 6.060167975699987e-06, "loss": 0.5867, "mean_token_accuracy": 0.8068459033966064, "num_tokens": 91201645.0, "step": 3892 }, { "epoch": 0.514538725878932, "grad_norm": 0.9356091711547443, "learning_rate": 6.058107818253877e-06, "loss": 0.6097, "mean_token_accuracy": 0.7992461323738098, "num_tokens": 91267181.0, "step": 3893 }, { "epoch": 0.514670896114195, "grad_norm": 0.8636391985605385, "learning_rate": 6.056047541998264e-06, "loss": 0.6162, "mean_token_accuracy": 0.7964076399803162, "num_tokens": 91332717.0, "step": 3894 }, { "epoch": 0.514803066349458, "grad_norm": 0.9611879689738336, "learning_rate": 6.053987147371738e-06, "loss": 0.5712, "mean_token_accuracy": 0.8108595013618469, "num_tokens": 91398253.0, "step": 3895 }, { "epoch": 0.5149352365847211, "grad_norm": 0.8788281828136526, "learning_rate": 6.051926634812915e-06, "loss": 0.5771, "mean_token_accuracy": 0.808814525604248, "num_tokens": 91463789.0, "step": 3896 }, { "epoch": 0.5150674068199841, "grad_norm": 0.892239520441945, "learning_rate": 6.049866004760435e-06, "loss": 0.6232, "mean_token_accuracy": 0.7946068644523621, "num_tokens": 91529325.0, "step": 3897 }, { "epoch": 0.5151995770552471, "grad_norm": 0.8390085438288827, "learning_rate": 6.047805257652965e-06, "loss": 0.5987, "mean_token_accuracy": 0.8026797771453857, "num_tokens": 91594861.0, "step": 3898 }, { "epoch": 0.5153317472905101, "grad_norm": 0.926734113814064, "learning_rate": 6.04574439392919e-06, "loss": 0.6361, "mean_token_accuracy": 0.7899676561355591, "num_tokens": 91660397.0, "step": 3899 }, { "epoch": 0.5154639175257731, "grad_norm": 0.9859465029411701, "learning_rate": 6.043683414027829e-06, "loss": 0.5849, "mean_token_accuracy": 0.8056861162185669, "num_tokens": 91725933.0, "step": 3900 }, { "epoch": 0.5155960877610362, "grad_norm": 0.8647089086965573, "learning_rate": 6.04162231838762e-06, "loss": 0.5797, "mean_token_accuracy": 0.8095165491104126, "num_tokens": 91791469.0, "step": 3901 }, { "epoch": 0.5157282579962993, "grad_norm": 0.9091301642883186, "learning_rate": 6.0395611074473285e-06, "loss": 0.6162, "mean_token_accuracy": 0.7956751585006714, "num_tokens": 91857005.0, "step": 3902 }, { "epoch": 0.5158604282315623, "grad_norm": 0.8832805918867404, "learning_rate": 6.0374997816457425e-06, "loss": 0.5811, "mean_token_accuracy": 0.8071358799934387, "num_tokens": 91922541.0, "step": 3903 }, { "epoch": 0.5159925984668253, "grad_norm": 0.864648638575903, "learning_rate": 6.035438341421674e-06, "loss": 0.6182, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 91988077.0, "step": 3904 }, { "epoch": 0.5161247687020883, "grad_norm": 0.8445169426910282, "learning_rate": 6.0333767872139625e-06, "loss": 0.597, "mean_token_accuracy": 0.8016115427017212, "num_tokens": 92053613.0, "step": 3905 }, { "epoch": 0.5162569389373514, "grad_norm": 0.7882436865450064, "learning_rate": 6.031315119461469e-06, "loss": 0.5488, "mean_token_accuracy": 0.818382978439331, "num_tokens": 92119149.0, "step": 3906 }, { "epoch": 0.5163891091726144, "grad_norm": 0.9165656122570501, "learning_rate": 6.029253338603081e-06, "loss": 0.6374, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 92184685.0, "step": 3907 }, { "epoch": 0.5165212794078774, "grad_norm": 0.9092321871988984, "learning_rate": 6.027191445077707e-06, "loss": 0.5683, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 92250221.0, "step": 3908 }, { "epoch": 0.5166534496431404, "grad_norm": 0.8643155137055757, "learning_rate": 6.025129439324279e-06, "loss": 0.532, "mean_token_accuracy": 0.8255860209465027, "num_tokens": 92315757.0, "step": 3909 }, { "epoch": 0.5167856198784034, "grad_norm": 0.8306640240365641, "learning_rate": 6.023067321781759e-06, "loss": 0.5926, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 92381293.0, "step": 3910 }, { "epoch": 0.5169177901136665, "grad_norm": 0.7833046144177535, "learning_rate": 6.021005092889127e-06, "loss": 0.5858, "mean_token_accuracy": 0.8040074706077576, "num_tokens": 92446829.0, "step": 3911 }, { "epoch": 0.5170499603489295, "grad_norm": 1.0223553899178346, "learning_rate": 6.018942753085388e-06, "loss": 0.6734, "mean_token_accuracy": 0.7763551473617554, "num_tokens": 92512365.0, "step": 3912 }, { "epoch": 0.5171821305841925, "grad_norm": 0.8570987463946024, "learning_rate": 6.016880302809571e-06, "loss": 0.6379, "mean_token_accuracy": 0.7915547490119934, "num_tokens": 92577901.0, "step": 3913 }, { "epoch": 0.5173143008194555, "grad_norm": 0.9197377942216874, "learning_rate": 6.014817742500732e-06, "loss": 0.5767, "mean_token_accuracy": 0.8082514405250549, "num_tokens": 92641607.0, "step": 3914 }, { "epoch": 0.5174464710547185, "grad_norm": 0.8222216212047309, "learning_rate": 6.012755072597941e-06, "loss": 0.5511, "mean_token_accuracy": 0.8208399415016174, "num_tokens": 92707143.0, "step": 3915 }, { "epoch": 0.5175786412899815, "grad_norm": 0.789648800528564, "learning_rate": 6.010692293540301e-06, "loss": 0.5903, "mean_token_accuracy": 0.8058387041091919, "num_tokens": 92772679.0, "step": 3916 }, { "epoch": 0.5177108115252446, "grad_norm": 0.8139718631098718, "learning_rate": 6.008629405766937e-06, "loss": 0.5715, "mean_token_accuracy": 0.8094249963760376, "num_tokens": 92838215.0, "step": 3917 }, { "epoch": 0.5178429817605076, "grad_norm": 0.8816685388106659, "learning_rate": 6.006566409716989e-06, "loss": 0.5903, "mean_token_accuracy": 0.8030765652656555, "num_tokens": 92903751.0, "step": 3918 }, { "epoch": 0.5179751519957706, "grad_norm": 0.8171800189659707, "learning_rate": 6.00450330582963e-06, "loss": 0.5684, "mean_token_accuracy": 0.8118361830711365, "num_tokens": 92969287.0, "step": 3919 }, { "epoch": 0.5181073222310336, "grad_norm": 0.8254166392378713, "learning_rate": 6.0024400945440496e-06, "loss": 0.5847, "mean_token_accuracy": 0.8069069981575012, "num_tokens": 93034823.0, "step": 3920 }, { "epoch": 0.5182394924662966, "grad_norm": 0.89027100324545, "learning_rate": 6.000376776299463e-06, "loss": 0.6093, "mean_token_accuracy": 0.7956140637397766, "num_tokens": 93100359.0, "step": 3921 }, { "epoch": 0.5183716627015597, "grad_norm": 0.8924147984900093, "learning_rate": 5.99831335153511e-06, "loss": 0.6315, "mean_token_accuracy": 0.7915394902229309, "num_tokens": 93165895.0, "step": 3922 }, { "epoch": 0.5185038329368227, "grad_norm": 0.900764003137067, "learning_rate": 5.996249820690247e-06, "loss": 0.5763, "mean_token_accuracy": 0.8096691370010376, "num_tokens": 93231431.0, "step": 3923 }, { "epoch": 0.5186360031720857, "grad_norm": 0.8248688919930791, "learning_rate": 5.994186184204158e-06, "loss": 0.568, "mean_token_accuracy": 0.8123550415039062, "num_tokens": 93296967.0, "step": 3924 }, { "epoch": 0.5187681734073487, "grad_norm": 0.8009640156657012, "learning_rate": 5.9921224425161485e-06, "loss": 0.5324, "mean_token_accuracy": 0.8256317973136902, "num_tokens": 93362503.0, "step": 3925 }, { "epoch": 0.5189003436426117, "grad_norm": 0.8463554368808472, "learning_rate": 5.990058596065546e-06, "loss": 0.6259, "mean_token_accuracy": 0.7946984767913818, "num_tokens": 93428039.0, "step": 3926 }, { "epoch": 0.5190325138778747, "grad_norm": 0.8865126310757149, "learning_rate": 5.9879946452917e-06, "loss": 0.6063, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 93493575.0, "step": 3927 }, { "epoch": 0.5191646841131378, "grad_norm": 0.9303964228526929, "learning_rate": 5.985930590633983e-06, "loss": 0.5989, "mean_token_accuracy": 0.8021761775016785, "num_tokens": 93559111.0, "step": 3928 }, { "epoch": 0.5192968543484008, "grad_norm": 0.9474162648150999, "learning_rate": 5.983866432531788e-06, "loss": 0.5967, "mean_token_accuracy": 0.8030460476875305, "num_tokens": 93624647.0, "step": 3929 }, { "epoch": 0.5194290245836638, "grad_norm": 0.8697206717401957, "learning_rate": 5.9818021714245325e-06, "loss": 0.5501, "mean_token_accuracy": 0.8163618445396423, "num_tokens": 93684795.0, "step": 3930 }, { "epoch": 0.5195611948189268, "grad_norm": 0.7996278418714302, "learning_rate": 5.979737807751655e-06, "loss": 0.5885, "mean_token_accuracy": 0.8054572343826294, "num_tokens": 93750331.0, "step": 3931 }, { "epoch": 0.5196933650541898, "grad_norm": 0.8287034234788484, "learning_rate": 5.977673341952612e-06, "loss": 0.5858, "mean_token_accuracy": 0.8059455752372742, "num_tokens": 93815867.0, "step": 3932 }, { "epoch": 0.5198255352894529, "grad_norm": 1.0386044831405354, "learning_rate": 5.975608774466889e-06, "loss": 0.6173, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 93881403.0, "step": 3933 }, { "epoch": 0.5199577055247159, "grad_norm": 0.8657806846427819, "learning_rate": 5.973544105733987e-06, "loss": 0.6131, "mean_token_accuracy": 0.7956445813179016, "num_tokens": 93946939.0, "step": 3934 }, { "epoch": 0.5200898757599789, "grad_norm": 0.8691387153204321, "learning_rate": 5.9714793361934334e-06, "loss": 0.5934, "mean_token_accuracy": 0.806708574295044, "num_tokens": 94012475.0, "step": 3935 }, { "epoch": 0.5202220459952419, "grad_norm": 0.8040041620194419, "learning_rate": 5.9694144662847705e-06, "loss": 0.5689, "mean_token_accuracy": 0.8126449584960938, "num_tokens": 94078011.0, "step": 3936 }, { "epoch": 0.5203542162305049, "grad_norm": 0.835751635524524, "learning_rate": 5.96734949644757e-06, "loss": 0.5534, "mean_token_accuracy": 0.8150103688240051, "num_tokens": 94143547.0, "step": 3937 }, { "epoch": 0.520486386465768, "grad_norm": 0.7706374254396637, "learning_rate": 5.9652844271214194e-06, "loss": 0.5235, "mean_token_accuracy": 0.8280124664306641, "num_tokens": 94209083.0, "step": 3938 }, { "epoch": 0.520618556701031, "grad_norm": 0.8770952771562027, "learning_rate": 5.963219258745928e-06, "loss": 0.5893, "mean_token_accuracy": 0.8060829043388367, "num_tokens": 94274619.0, "step": 3939 }, { "epoch": 0.520750726936294, "grad_norm": 0.7764361611810062, "learning_rate": 5.961153991760729e-06, "loss": 0.5387, "mean_token_accuracy": 0.8213893175125122, "num_tokens": 94340155.0, "step": 3940 }, { "epoch": 0.520882897171557, "grad_norm": 0.8551603043565047, "learning_rate": 5.959088626605471e-06, "loss": 0.6159, "mean_token_accuracy": 0.7970027923583984, "num_tokens": 94405691.0, "step": 3941 }, { "epoch": 0.52101506740682, "grad_norm": 0.8427476713241849, "learning_rate": 5.957023163719832e-06, "loss": 0.572, "mean_token_accuracy": 0.8099896311759949, "num_tokens": 94471227.0, "step": 3942 }, { "epoch": 0.521147237642083, "grad_norm": 0.8314916438649098, "learning_rate": 5.954957603543502e-06, "loss": 0.619, "mean_token_accuracy": 0.793645441532135, "num_tokens": 94536763.0, "step": 3943 }, { "epoch": 0.5212794078773461, "grad_norm": 0.8803592083029389, "learning_rate": 5.9528919465161976e-06, "loss": 0.6238, "mean_token_accuracy": 0.7943322062492371, "num_tokens": 94602299.0, "step": 3944 }, { "epoch": 0.5214115781126091, "grad_norm": 0.8822957445811701, "learning_rate": 5.950826193077654e-06, "loss": 0.5961, "mean_token_accuracy": 0.8051519989967346, "num_tokens": 94667835.0, "step": 3945 }, { "epoch": 0.5215437483478721, "grad_norm": 0.8894389718188542, "learning_rate": 5.948760343667626e-06, "loss": 0.6145, "mean_token_accuracy": 0.8002380728721619, "num_tokens": 94733371.0, "step": 3946 }, { "epoch": 0.5216759185831351, "grad_norm": 0.7995942332412767, "learning_rate": 5.946694398725888e-06, "loss": 0.56, "mean_token_accuracy": 0.8148425221443176, "num_tokens": 94798907.0, "step": 3947 }, { "epoch": 0.5218080888183981, "grad_norm": 0.9095891806167953, "learning_rate": 5.944628358692242e-06, "loss": 0.5716, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 94864443.0, "step": 3948 }, { "epoch": 0.5219402590536611, "grad_norm": 0.8452846311776975, "learning_rate": 5.942562224006502e-06, "loss": 0.5471, "mean_token_accuracy": 0.8208552002906799, "num_tokens": 94929979.0, "step": 3949 }, { "epoch": 0.5220724292889242, "grad_norm": 0.8463315334465508, "learning_rate": 5.940495995108502e-06, "loss": 0.5796, "mean_token_accuracy": 0.8054724931716919, "num_tokens": 94995515.0, "step": 3950 }, { "epoch": 0.5222045995241872, "grad_norm": 0.8477060750665879, "learning_rate": 5.938429672438103e-06, "loss": 0.5596, "mean_token_accuracy": 0.8193901777267456, "num_tokens": 95061051.0, "step": 3951 }, { "epoch": 0.5223367697594502, "grad_norm": 0.7995240396771172, "learning_rate": 5.936363256435179e-06, "loss": 0.5349, "mean_token_accuracy": 0.821969211101532, "num_tokens": 95126587.0, "step": 3952 }, { "epoch": 0.5224689399947132, "grad_norm": 0.8217982569458111, "learning_rate": 5.93429674753963e-06, "loss": 0.6183, "mean_token_accuracy": 0.7954309582710266, "num_tokens": 95192123.0, "step": 3953 }, { "epoch": 0.5226011102299762, "grad_norm": 0.8039702546570008, "learning_rate": 5.93223014619137e-06, "loss": 0.5904, "mean_token_accuracy": 0.8037174940109253, "num_tokens": 95257659.0, "step": 3954 }, { "epoch": 0.5227332804652393, "grad_norm": 0.8931500975529085, "learning_rate": 5.930163452830335e-06, "loss": 0.553, "mean_token_accuracy": 0.8187645077705383, "num_tokens": 95323195.0, "step": 3955 }, { "epoch": 0.5228654507005023, "grad_norm": 0.8687397728580646, "learning_rate": 5.928096667896486e-06, "loss": 0.6353, "mean_token_accuracy": 0.78789222240448, "num_tokens": 95388731.0, "step": 3956 }, { "epoch": 0.5229976209357653, "grad_norm": 0.8944097222745506, "learning_rate": 5.926029791829791e-06, "loss": 0.5711, "mean_token_accuracy": 0.8124465942382812, "num_tokens": 95454267.0, "step": 3957 }, { "epoch": 0.5231297911710283, "grad_norm": 0.8544192294991584, "learning_rate": 5.923962825070248e-06, "loss": 0.6127, "mean_token_accuracy": 0.7956751585006714, "num_tokens": 95519803.0, "step": 3958 }, { "epoch": 0.5232619614062913, "grad_norm": 0.7739893445004975, "learning_rate": 5.9218957680578726e-06, "loss": 0.584, "mean_token_accuracy": 0.8076547384262085, "num_tokens": 95585339.0, "step": 3959 }, { "epoch": 0.5233941316415543, "grad_norm": 0.7957602310552494, "learning_rate": 5.919828621232697e-06, "loss": 0.5849, "mean_token_accuracy": 0.8070748448371887, "num_tokens": 95650875.0, "step": 3960 }, { "epoch": 0.5235263018768174, "grad_norm": 0.9463046565170873, "learning_rate": 5.917761385034773e-06, "loss": 0.6904, "mean_token_accuracy": 0.7731046080589294, "num_tokens": 95716411.0, "step": 3961 }, { "epoch": 0.5236584721120804, "grad_norm": 0.8393611755522842, "learning_rate": 5.915694059904173e-06, "loss": 0.544, "mean_token_accuracy": 0.8201379776000977, "num_tokens": 95781947.0, "step": 3962 }, { "epoch": 0.5237906423473434, "grad_norm": 0.8808126041520201, "learning_rate": 5.9136266462809844e-06, "loss": 0.5914, "mean_token_accuracy": 0.803396999835968, "num_tokens": 95847483.0, "step": 3963 }, { "epoch": 0.5239228125826064, "grad_norm": 0.84275251984948, "learning_rate": 5.911559144605321e-06, "loss": 0.5967, "mean_token_accuracy": 0.8069680333137512, "num_tokens": 95913019.0, "step": 3964 }, { "epoch": 0.5240549828178694, "grad_norm": 0.9174564831934704, "learning_rate": 5.909491555317306e-06, "loss": 0.6118, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 95978555.0, "step": 3965 }, { "epoch": 0.5241871530531325, "grad_norm": 0.8355362534255383, "learning_rate": 5.9074238788570914e-06, "loss": 0.5538, "mean_token_accuracy": 0.8158954977989197, "num_tokens": 96044091.0, "step": 3966 }, { "epoch": 0.5243193232883955, "grad_norm": 0.8669723668099595, "learning_rate": 5.905356115664837e-06, "loss": 0.622, "mean_token_accuracy": 0.7957209348678589, "num_tokens": 96109627.0, "step": 3967 }, { "epoch": 0.5244514935236585, "grad_norm": 0.8551741335098482, "learning_rate": 5.903288266180731e-06, "loss": 0.5972, "mean_token_accuracy": 0.8048162460327148, "num_tokens": 96175163.0, "step": 3968 }, { "epoch": 0.5245836637589215, "grad_norm": 0.9315594575231482, "learning_rate": 5.9012203308449715e-06, "loss": 0.6115, "mean_token_accuracy": 0.8014741539955139, "num_tokens": 96240699.0, "step": 3969 }, { "epoch": 0.5247158339941845, "grad_norm": 0.967423376483974, "learning_rate": 5.899152310097782e-06, "loss": 0.5945, "mean_token_accuracy": 0.8042668700218201, "num_tokens": 96306235.0, "step": 3970 }, { "epoch": 0.5248480042294476, "grad_norm": 0.8616660864507685, "learning_rate": 5.897084204379398e-06, "loss": 0.553, "mean_token_accuracy": 0.8167653679847717, "num_tokens": 96371771.0, "step": 3971 }, { "epoch": 0.5249801744647106, "grad_norm": 0.8462544048700827, "learning_rate": 5.895016014130078e-06, "loss": 0.577, "mean_token_accuracy": 0.812843382358551, "num_tokens": 96437307.0, "step": 3972 }, { "epoch": 0.5251123446999736, "grad_norm": 0.84012251720269, "learning_rate": 5.892947739790096e-06, "loss": 0.5924, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 96502843.0, "step": 3973 }, { "epoch": 0.5252445149352366, "grad_norm": 0.8699025131377587, "learning_rate": 5.8908793817997425e-06, "loss": 0.612, "mean_token_accuracy": 0.8005738258361816, "num_tokens": 96568379.0, "step": 3974 }, { "epoch": 0.5253766851704996, "grad_norm": 0.8689300099326437, "learning_rate": 5.888810940599332e-06, "loss": 0.577, "mean_token_accuracy": 0.8097454309463501, "num_tokens": 96633915.0, "step": 3975 }, { "epoch": 0.5255088554057626, "grad_norm": 0.9231530191032459, "learning_rate": 5.886742416629188e-06, "loss": 0.6405, "mean_token_accuracy": 0.7871139049530029, "num_tokens": 96699451.0, "step": 3976 }, { "epoch": 0.5256410256410257, "grad_norm": 0.799217693382149, "learning_rate": 5.88467381032966e-06, "loss": 0.5553, "mean_token_accuracy": 0.8170247673988342, "num_tokens": 96764987.0, "step": 3977 }, { "epoch": 0.5257731958762887, "grad_norm": 0.8727028975280757, "learning_rate": 5.882605122141109e-06, "loss": 0.5521, "mean_token_accuracy": 0.8168569207191467, "num_tokens": 96830523.0, "step": 3978 }, { "epoch": 0.5259053661115517, "grad_norm": 0.8247074959743417, "learning_rate": 5.880536352503915e-06, "loss": 0.5591, "mean_token_accuracy": 0.8142321109771729, "num_tokens": 96896059.0, "step": 3979 }, { "epoch": 0.5260375363468147, "grad_norm": 0.8226674329668203, "learning_rate": 5.878467501858477e-06, "loss": 0.5263, "mean_token_accuracy": 0.8290044069290161, "num_tokens": 96961595.0, "step": 3980 }, { "epoch": 0.5261697065820777, "grad_norm": 0.8160052349563314, "learning_rate": 5.876398570645209e-06, "loss": 0.5589, "mean_token_accuracy": 0.8158954977989197, "num_tokens": 97027131.0, "step": 3981 }, { "epoch": 0.5263018768173408, "grad_norm": 0.8778337252560667, "learning_rate": 5.874329559304544e-06, "loss": 0.5992, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 97092667.0, "step": 3982 }, { "epoch": 0.5264340470526038, "grad_norm": 0.7928968275068355, "learning_rate": 5.872260468276932e-06, "loss": 0.525, "mean_token_accuracy": 0.8254028558731079, "num_tokens": 97158203.0, "step": 3983 }, { "epoch": 0.5265662172878668, "grad_norm": 0.8475614375896589, "learning_rate": 5.870191298002837e-06, "loss": 0.5851, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 97223739.0, "step": 3984 }, { "epoch": 0.5266983875231298, "grad_norm": 0.8549944533140658, "learning_rate": 5.868122048922747e-06, "loss": 0.5873, "mean_token_accuracy": 0.807853102684021, "num_tokens": 97289275.0, "step": 3985 }, { "epoch": 0.5268305577583928, "grad_norm": 0.9231482230674719, "learning_rate": 5.866052721477156e-06, "loss": 0.6114, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 97354811.0, "step": 3986 }, { "epoch": 0.5269627279936558, "grad_norm": 0.9374575589036646, "learning_rate": 5.863983316106584e-06, "loss": 0.6347, "mean_token_accuracy": 0.7894487977027893, "num_tokens": 97420347.0, "step": 3987 }, { "epoch": 0.5270948982289189, "grad_norm": 0.8109249365716658, "learning_rate": 5.861913833251562e-06, "loss": 0.5599, "mean_token_accuracy": 0.8152087926864624, "num_tokens": 97485883.0, "step": 3988 }, { "epoch": 0.5272270684641819, "grad_norm": 0.8526632780368296, "learning_rate": 5.859844273352644e-06, "loss": 0.6107, "mean_token_accuracy": 0.7985135912895203, "num_tokens": 97551419.0, "step": 3989 }, { "epoch": 0.5273592386994449, "grad_norm": 0.8240010212508042, "learning_rate": 5.8577746368503905e-06, "loss": 0.5875, "mean_token_accuracy": 0.8040685057640076, "num_tokens": 97616955.0, "step": 3990 }, { "epoch": 0.5274914089347079, "grad_norm": 0.8739962615000013, "learning_rate": 5.855704924185388e-06, "loss": 0.6033, "mean_token_accuracy": 0.8006958961486816, "num_tokens": 97682491.0, "step": 3991 }, { "epoch": 0.5276235791699709, "grad_norm": 0.8529998182040549, "learning_rate": 5.853635135798232e-06, "loss": 0.619, "mean_token_accuracy": 0.797613263130188, "num_tokens": 97748027.0, "step": 3992 }, { "epoch": 0.527755749405234, "grad_norm": 0.8118594602830023, "learning_rate": 5.851565272129542e-06, "loss": 0.6025, "mean_token_accuracy": 0.7992156147956848, "num_tokens": 97813563.0, "step": 3993 }, { "epoch": 0.527887919640497, "grad_norm": 0.8634959488532006, "learning_rate": 5.849495333619943e-06, "loss": 0.6173, "mean_token_accuracy": 0.7961787581443787, "num_tokens": 97879099.0, "step": 3994 }, { "epoch": 0.52802008987576, "grad_norm": 0.8515971527194515, "learning_rate": 5.847425320710086e-06, "loss": 0.5783, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 97944635.0, "step": 3995 }, { "epoch": 0.528152260111023, "grad_norm": 0.9044599059991031, "learning_rate": 5.845355233840632e-06, "loss": 0.5952, "mean_token_accuracy": 0.8022677302360535, "num_tokens": 98010171.0, "step": 3996 }, { "epoch": 0.528284430346286, "grad_norm": 0.8120043720965069, "learning_rate": 5.843285073452257e-06, "loss": 0.5313, "mean_token_accuracy": 0.8243956565856934, "num_tokens": 98075707.0, "step": 3997 }, { "epoch": 0.528416600581549, "grad_norm": 0.9090638376209574, "learning_rate": 5.8412148399856594e-06, "loss": 0.6553, "mean_token_accuracy": 0.7832529544830322, "num_tokens": 98141243.0, "step": 3998 }, { "epoch": 0.5285487708168121, "grad_norm": 0.9533865959150333, "learning_rate": 5.839144533881546e-06, "loss": 0.5966, "mean_token_accuracy": 0.802435576915741, "num_tokens": 98206779.0, "step": 3999 }, { "epoch": 0.5286809410520751, "grad_norm": 0.8437616177421388, "learning_rate": 5.837074155580641e-06, "loss": 0.5817, "mean_token_accuracy": 0.8082346320152283, "num_tokens": 98272315.0, "step": 4000 }, { "epoch": 0.5288131112873381, "grad_norm": 0.886980807581164, "learning_rate": 5.835003705523688e-06, "loss": 0.6348, "mean_token_accuracy": 0.793584406375885, "num_tokens": 98337851.0, "step": 4001 }, { "epoch": 0.5289452815226011, "grad_norm": 0.8716362535606783, "learning_rate": 5.832933184151439e-06, "loss": 0.6216, "mean_token_accuracy": 0.7941795587539673, "num_tokens": 98403387.0, "step": 4002 }, { "epoch": 0.5290774517578641, "grad_norm": 0.9686307395614595, "learning_rate": 5.830862591904666e-06, "loss": 0.6024, "mean_token_accuracy": 0.8000549674034119, "num_tokens": 98468923.0, "step": 4003 }, { "epoch": 0.5292096219931272, "grad_norm": 0.8414338789146288, "learning_rate": 5.828791929224156e-06, "loss": 0.5535, "mean_token_accuracy": 0.8176199197769165, "num_tokens": 98534459.0, "step": 4004 }, { "epoch": 0.5293417922283902, "grad_norm": 0.9343418621009116, "learning_rate": 5.8267211965507085e-06, "loss": 0.6091, "mean_token_accuracy": 0.7987120151519775, "num_tokens": 98599995.0, "step": 4005 }, { "epoch": 0.5294739624636532, "grad_norm": 0.9008493954159859, "learning_rate": 5.82465039432514e-06, "loss": 0.5897, "mean_token_accuracy": 0.8040379881858826, "num_tokens": 98665531.0, "step": 4006 }, { "epoch": 0.5296061326989162, "grad_norm": 0.8748853861618024, "learning_rate": 5.82257952298828e-06, "loss": 0.6101, "mean_token_accuracy": 0.7972317337989807, "num_tokens": 98731067.0, "step": 4007 }, { "epoch": 0.5297383029341792, "grad_norm": 0.914019389883012, "learning_rate": 5.820508582980976e-06, "loss": 0.6047, "mean_token_accuracy": 0.7972164750099182, "num_tokens": 98796603.0, "step": 4008 }, { "epoch": 0.5298704731694422, "grad_norm": 0.9287761270656847, "learning_rate": 5.818437574744086e-06, "loss": 0.6328, "mean_token_accuracy": 0.7918294668197632, "num_tokens": 98862139.0, "step": 4009 }, { "epoch": 0.5300026434047053, "grad_norm": 0.9709857184467293, "learning_rate": 5.816366498718483e-06, "loss": 0.5998, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 98927675.0, "step": 4010 }, { "epoch": 0.5301348136399683, "grad_norm": 0.7892479209165753, "learning_rate": 5.814295355345059e-06, "loss": 0.579, "mean_token_accuracy": 0.807731032371521, "num_tokens": 98993211.0, "step": 4011 }, { "epoch": 0.5302669838752313, "grad_norm": 0.7650365056001257, "learning_rate": 5.8122241450647165e-06, "loss": 0.5084, "mean_token_accuracy": 0.832560122013092, "num_tokens": 99058747.0, "step": 4012 }, { "epoch": 0.5303991541104943, "grad_norm": 0.9040356108455159, "learning_rate": 5.8101528683183705e-06, "loss": 0.6213, "mean_token_accuracy": 0.7943932414054871, "num_tokens": 99124283.0, "step": 4013 }, { "epoch": 0.5305313243457573, "grad_norm": 0.895065674406422, "learning_rate": 5.8080815255469556e-06, "loss": 0.6382, "mean_token_accuracy": 0.7921499013900757, "num_tokens": 99189819.0, "step": 4014 }, { "epoch": 0.5306634945810204, "grad_norm": 0.9193549651828187, "learning_rate": 5.806010117191416e-06, "loss": 0.5963, "mean_token_accuracy": 0.802374541759491, "num_tokens": 99255355.0, "step": 4015 }, { "epoch": 0.5307956648162834, "grad_norm": 0.9272246077512686, "learning_rate": 5.803938643692711e-06, "loss": 0.627, "mean_token_accuracy": 0.7929129600524902, "num_tokens": 99320891.0, "step": 4016 }, { "epoch": 0.5309278350515464, "grad_norm": 0.8964220128731877, "learning_rate": 5.801867105491816e-06, "loss": 0.6406, "mean_token_accuracy": 0.789219856262207, "num_tokens": 99386427.0, "step": 4017 }, { "epoch": 0.5310600052868094, "grad_norm": 0.912067232487307, "learning_rate": 5.799795503029716e-06, "loss": 0.6229, "mean_token_accuracy": 0.7953851819038391, "num_tokens": 99451963.0, "step": 4018 }, { "epoch": 0.5311921755220724, "grad_norm": 0.8362300338173833, "learning_rate": 5.797723836747411e-06, "loss": 0.5656, "mean_token_accuracy": 0.8137285113334656, "num_tokens": 99517499.0, "step": 4019 }, { "epoch": 0.5313243457573354, "grad_norm": 0.9521669068767544, "learning_rate": 5.79565210708592e-06, "loss": 0.6192, "mean_token_accuracy": 0.7960871458053589, "num_tokens": 99583035.0, "step": 4020 }, { "epoch": 0.5314565159925985, "grad_norm": 0.8202768887208906, "learning_rate": 5.793580314486267e-06, "loss": 0.5576, "mean_token_accuracy": 0.8160480856895447, "num_tokens": 99648571.0, "step": 4021 }, { "epoch": 0.5315886862278615, "grad_norm": 0.875280546774811, "learning_rate": 5.7915084593894935e-06, "loss": 0.607, "mean_token_accuracy": 0.8027408123016357, "num_tokens": 99714107.0, "step": 4022 }, { "epoch": 0.5317208564631245, "grad_norm": 0.9056852540263007, "learning_rate": 5.789436542236657e-06, "loss": 0.5974, "mean_token_accuracy": 0.8015046715736389, "num_tokens": 99779643.0, "step": 4023 }, { "epoch": 0.5318530266983875, "grad_norm": 0.8316520974142675, "learning_rate": 5.787364563468823e-06, "loss": 0.5649, "mean_token_accuracy": 0.8091960549354553, "num_tokens": 99845179.0, "step": 4024 }, { "epoch": 0.5319851969336505, "grad_norm": 0.9472353233605942, "learning_rate": 5.785292523527073e-06, "loss": 0.6483, "mean_token_accuracy": 0.7859693765640259, "num_tokens": 99910715.0, "step": 4025 }, { "epoch": 0.5321173671689136, "grad_norm": 0.8417732083470935, "learning_rate": 5.783220422852501e-06, "loss": 0.5538, "mean_token_accuracy": 0.818199872970581, "num_tokens": 99976251.0, "step": 4026 }, { "epoch": 0.5322495374041766, "grad_norm": 0.8579408504089776, "learning_rate": 5.781148261886215e-06, "loss": 0.578, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 100041787.0, "step": 4027 }, { "epoch": 0.5323817076394396, "grad_norm": 0.9029687143735083, "learning_rate": 5.779076041069334e-06, "loss": 0.6233, "mean_token_accuracy": 0.7966365814208984, "num_tokens": 100107323.0, "step": 4028 }, { "epoch": 0.5325138778747026, "grad_norm": 0.7939926770452113, "learning_rate": 5.77700376084299e-06, "loss": 0.4964, "mean_token_accuracy": 0.8371230363845825, "num_tokens": 100172859.0, "step": 4029 }, { "epoch": 0.5326460481099656, "grad_norm": 0.8103353933794742, "learning_rate": 5.774931421648329e-06, "loss": 0.5537, "mean_token_accuracy": 0.8155597448348999, "num_tokens": 100238395.0, "step": 4030 }, { "epoch": 0.5327782183452286, "grad_norm": 0.9057384676024438, "learning_rate": 5.772859023926508e-06, "loss": 0.5834, "mean_token_accuracy": 0.8089519143104553, "num_tokens": 100303931.0, "step": 4031 }, { "epoch": 0.5329103885804917, "grad_norm": 0.9055258743993067, "learning_rate": 5.770786568118697e-06, "loss": 0.6134, "mean_token_accuracy": 0.7964991927146912, "num_tokens": 100369467.0, "step": 4032 }, { "epoch": 0.5330425588157547, "grad_norm": 1.1147236311483482, "learning_rate": 5.76871405466608e-06, "loss": 0.6221, "mean_token_accuracy": 0.7950036525726318, "num_tokens": 100435003.0, "step": 4033 }, { "epoch": 0.5331747290510177, "grad_norm": 0.9227437313044438, "learning_rate": 5.766641484009852e-06, "loss": 0.5661, "mean_token_accuracy": 0.8131993412971497, "num_tokens": 100497302.0, "step": 4034 }, { "epoch": 0.5333068992862807, "grad_norm": 0.7667455795284358, "learning_rate": 5.7645688565912175e-06, "loss": 0.5386, "mean_token_accuracy": 0.8222744464874268, "num_tokens": 100562838.0, "step": 4035 }, { "epoch": 0.5334390695215437, "grad_norm": 0.7663008063848891, "learning_rate": 5.762496172851398e-06, "loss": 0.5435, "mean_token_accuracy": 0.8208857178688049, "num_tokens": 100628374.0, "step": 4036 }, { "epoch": 0.5335712397568068, "grad_norm": 0.8165531504896741, "learning_rate": 5.760423433231624e-06, "loss": 0.601, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 100693910.0, "step": 4037 }, { "epoch": 0.5337034099920698, "grad_norm": 0.8598712828349319, "learning_rate": 5.758350638173139e-06, "loss": 0.6352, "mean_token_accuracy": 0.7899676561355591, "num_tokens": 100759446.0, "step": 4038 }, { "epoch": 0.5338355802273328, "grad_norm": 0.8779029224953382, "learning_rate": 5.756277788117196e-06, "loss": 0.6054, "mean_token_accuracy": 0.7984983325004578, "num_tokens": 100824982.0, "step": 4039 }, { "epoch": 0.5339677504625958, "grad_norm": 0.8820896987157514, "learning_rate": 5.754204883505062e-06, "loss": 0.6246, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 100890518.0, "step": 4040 }, { "epoch": 0.5340999206978588, "grad_norm": 0.9449661609272809, "learning_rate": 5.752131924778017e-06, "loss": 0.618, "mean_token_accuracy": 0.7950189113616943, "num_tokens": 100956054.0, "step": 4041 }, { "epoch": 0.5342320909331219, "grad_norm": 0.8711140435248801, "learning_rate": 5.750058912377348e-06, "loss": 0.5494, "mean_token_accuracy": 0.8205805420875549, "num_tokens": 101021590.0, "step": 4042 }, { "epoch": 0.5343642611683849, "grad_norm": 0.8219862172424661, "learning_rate": 5.747985846744359e-06, "loss": 0.6074, "mean_token_accuracy": 0.7978421449661255, "num_tokens": 101087126.0, "step": 4043 }, { "epoch": 0.5344964314036479, "grad_norm": 0.8247134038831048, "learning_rate": 5.745912728320362e-06, "loss": 0.5867, "mean_token_accuracy": 0.807425856590271, "num_tokens": 101152662.0, "step": 4044 }, { "epoch": 0.5346286016389109, "grad_norm": 0.9472939440645961, "learning_rate": 5.743839557546678e-06, "loss": 0.6382, "mean_token_accuracy": 0.7871291637420654, "num_tokens": 101218198.0, "step": 4045 }, { "epoch": 0.5347607718741739, "grad_norm": 0.7856851111023094, "learning_rate": 5.741766334864643e-06, "loss": 0.576, "mean_token_accuracy": 0.8109205365180969, "num_tokens": 101283734.0, "step": 4046 }, { "epoch": 0.5348929421094369, "grad_norm": 0.8632075776566851, "learning_rate": 5.739693060715603e-06, "loss": 0.5948, "mean_token_accuracy": 0.8048772811889648, "num_tokens": 101349270.0, "step": 4047 }, { "epoch": 0.5350251123447, "grad_norm": 0.8487955820479386, "learning_rate": 5.737619735540914e-06, "loss": 0.6078, "mean_token_accuracy": 0.7998107671737671, "num_tokens": 101414806.0, "step": 4048 }, { "epoch": 0.535157282579963, "grad_norm": 0.8646533614221227, "learning_rate": 5.735546359781946e-06, "loss": 0.5619, "mean_token_accuracy": 0.8154071569442749, "num_tokens": 101480342.0, "step": 4049 }, { "epoch": 0.535289452815226, "grad_norm": 0.8831835880706845, "learning_rate": 5.733472933880074e-06, "loss": 0.5796, "mean_token_accuracy": 0.8093181252479553, "num_tokens": 101545878.0, "step": 4050 }, { "epoch": 0.535421623050489, "grad_norm": 0.8249670086031268, "learning_rate": 5.7313994582766895e-06, "loss": 0.5831, "mean_token_accuracy": 0.8098370432853699, "num_tokens": 101611414.0, "step": 4051 }, { "epoch": 0.535553793285752, "grad_norm": 0.8230947093698656, "learning_rate": 5.729325933413193e-06, "loss": 0.5764, "mean_token_accuracy": 0.8094097375869751, "num_tokens": 101676950.0, "step": 4052 }, { "epoch": 0.535685963521015, "grad_norm": 0.7901658668118089, "learning_rate": 5.727252359730993e-06, "loss": 0.5264, "mean_token_accuracy": 0.8274478316307068, "num_tokens": 101742486.0, "step": 4053 }, { "epoch": 0.5358181337562781, "grad_norm": 0.8946555300001156, "learning_rate": 5.725178737671509e-06, "loss": 0.5974, "mean_token_accuracy": 0.802160918712616, "num_tokens": 101808022.0, "step": 4054 }, { "epoch": 0.5359503039915411, "grad_norm": 0.8859150171762076, "learning_rate": 5.723105067676175e-06, "loss": 0.6093, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 101873558.0, "step": 4055 }, { "epoch": 0.5360824742268041, "grad_norm": 0.9742339211834282, "learning_rate": 5.721031350186428e-06, "loss": 0.6004, "mean_token_accuracy": 0.8021761775016785, "num_tokens": 101939094.0, "step": 4056 }, { "epoch": 0.5362146444620671, "grad_norm": 1.0251548579231275, "learning_rate": 5.718957585643722e-06, "loss": 0.6846, "mean_token_accuracy": 0.7772097587585449, "num_tokens": 102004630.0, "step": 4057 }, { "epoch": 0.5363468146973301, "grad_norm": 0.8346989564627312, "learning_rate": 5.716883774489518e-06, "loss": 0.6141, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 102070166.0, "step": 4058 }, { "epoch": 0.5364789849325932, "grad_norm": 0.7713506146022069, "learning_rate": 5.714809917165286e-06, "loss": 0.5281, "mean_token_accuracy": 0.8244109153747559, "num_tokens": 102135702.0, "step": 4059 }, { "epoch": 0.5366111551678562, "grad_norm": 0.9588633631819139, "learning_rate": 5.712736014112509e-06, "loss": 0.6119, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 102201238.0, "step": 4060 }, { "epoch": 0.5367433254031192, "grad_norm": 0.926699969911382, "learning_rate": 5.710662065772674e-06, "loss": 0.6407, "mean_token_accuracy": 0.7891740798950195, "num_tokens": 102266774.0, "step": 4061 }, { "epoch": 0.5368754956383822, "grad_norm": 0.8355322462311173, "learning_rate": 5.708588072587284e-06, "loss": 0.5559, "mean_token_accuracy": 0.8173452615737915, "num_tokens": 102332310.0, "step": 4062 }, { "epoch": 0.5370076658736452, "grad_norm": 0.8628146228528296, "learning_rate": 5.70651403499785e-06, "loss": 0.6028, "mean_token_accuracy": 0.8020235896110535, "num_tokens": 102397846.0, "step": 4063 }, { "epoch": 0.5371398361089083, "grad_norm": 0.8325332297905955, "learning_rate": 5.7044399534458895e-06, "loss": 0.5219, "mean_token_accuracy": 0.8268068432807922, "num_tokens": 102463382.0, "step": 4064 }, { "epoch": 0.5372720063441713, "grad_norm": 0.9311688782996661, "learning_rate": 5.70236582837293e-06, "loss": 0.5973, "mean_token_accuracy": 0.8001465201377869, "num_tokens": 102528918.0, "step": 4065 }, { "epoch": 0.5374041765794343, "grad_norm": 0.7946969760771716, "learning_rate": 5.700291660220512e-06, "loss": 0.553, "mean_token_accuracy": 0.818291425704956, "num_tokens": 102594454.0, "step": 4066 }, { "epoch": 0.5375363468146973, "grad_norm": 0.8389966078155386, "learning_rate": 5.69821744943018e-06, "loss": 0.5824, "mean_token_accuracy": 0.8058844804763794, "num_tokens": 102659990.0, "step": 4067 }, { "epoch": 0.5376685170499603, "grad_norm": 0.8991884253760892, "learning_rate": 5.696143196443492e-06, "loss": 0.5765, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 102725526.0, "step": 4068 }, { "epoch": 0.5378006872852233, "grad_norm": 0.8584679834358471, "learning_rate": 5.6940689017020124e-06, "loss": 0.5893, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 102791062.0, "step": 4069 }, { "epoch": 0.5379328575204864, "grad_norm": 0.9451656665885432, "learning_rate": 5.6919945656473165e-06, "loss": 0.6063, "mean_token_accuracy": 0.8009552955627441, "num_tokens": 102856598.0, "step": 4070 }, { "epoch": 0.5380650277557494, "grad_norm": 0.8638784339725519, "learning_rate": 5.689920188720985e-06, "loss": 0.5837, "mean_token_accuracy": 0.806586503982544, "num_tokens": 102922134.0, "step": 4071 }, { "epoch": 0.5381971979910124, "grad_norm": 0.8582662184230613, "learning_rate": 5.6878457713646105e-06, "loss": 0.6017, "mean_token_accuracy": 0.8012300133705139, "num_tokens": 102987670.0, "step": 4072 }, { "epoch": 0.5383293682262754, "grad_norm": 0.8316032295365504, "learning_rate": 5.685771314019795e-06, "loss": 0.6, "mean_token_accuracy": 0.8015046715736389, "num_tokens": 103053206.0, "step": 4073 }, { "epoch": 0.5384615384615384, "grad_norm": 0.9352512535296572, "learning_rate": 5.683696817128143e-06, "loss": 0.5786, "mean_token_accuracy": 0.806769609451294, "num_tokens": 103118742.0, "step": 4074 }, { "epoch": 0.5385937086968015, "grad_norm": 0.8462184666614648, "learning_rate": 5.681622281131275e-06, "loss": 0.5614, "mean_token_accuracy": 0.8165669441223145, "num_tokens": 103184278.0, "step": 4075 }, { "epoch": 0.5387258789320645, "grad_norm": 0.9250785021296715, "learning_rate": 5.679547706470816e-06, "loss": 0.5866, "mean_token_accuracy": 0.8058387041091919, "num_tokens": 103249814.0, "step": 4076 }, { "epoch": 0.5388580491673275, "grad_norm": 0.8970661010043848, "learning_rate": 5.677473093588398e-06, "loss": 0.5925, "mean_token_accuracy": 0.8044652938842773, "num_tokens": 103315350.0, "step": 4077 }, { "epoch": 0.5389902194025905, "grad_norm": 0.8534688951507464, "learning_rate": 5.675398442925664e-06, "loss": 0.5883, "mean_token_accuracy": 0.8073800802230835, "num_tokens": 103380886.0, "step": 4078 }, { "epoch": 0.5391223896378535, "grad_norm": 0.8607236853632693, "learning_rate": 5.673323754924264e-06, "loss": 0.5409, "mean_token_accuracy": 0.8221828937530518, "num_tokens": 103446422.0, "step": 4079 }, { "epoch": 0.5392545598731165, "grad_norm": 0.8743927542796935, "learning_rate": 5.671249030025855e-06, "loss": 0.5838, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 103511958.0, "step": 4080 }, { "epoch": 0.5393867301083796, "grad_norm": 0.8224571978958739, "learning_rate": 5.669174268672103e-06, "loss": 0.5431, "mean_token_accuracy": 0.8192986249923706, "num_tokens": 103577494.0, "step": 4081 }, { "epoch": 0.5395189003436426, "grad_norm": 0.912806093043946, "learning_rate": 5.667099471304681e-06, "loss": 0.565, "mean_token_accuracy": 0.8148120045661926, "num_tokens": 103643030.0, "step": 4082 }, { "epoch": 0.5396510705789056, "grad_norm": 0.8332512544639143, "learning_rate": 5.665024638365271e-06, "loss": 0.6015, "mean_token_accuracy": 0.79878830909729, "num_tokens": 103708566.0, "step": 4083 }, { "epoch": 0.5397832408141686, "grad_norm": 0.905691603376225, "learning_rate": 5.662949770295559e-06, "loss": 0.5277, "mean_token_accuracy": 0.8244720101356506, "num_tokens": 103774102.0, "step": 4084 }, { "epoch": 0.5399154110494316, "grad_norm": 0.8854865155301472, "learning_rate": 5.660874867537246e-06, "loss": 0.5991, "mean_token_accuracy": 0.8017793893814087, "num_tokens": 103839638.0, "step": 4085 }, { "epoch": 0.5400475812846947, "grad_norm": 0.8449088197649806, "learning_rate": 5.658799930532032e-06, "loss": 0.5792, "mean_token_accuracy": 0.8117445707321167, "num_tokens": 103905174.0, "step": 4086 }, { "epoch": 0.5401797515199577, "grad_norm": 0.9190321955310927, "learning_rate": 5.656724959721626e-06, "loss": 0.5991, "mean_token_accuracy": 0.8026492595672607, "num_tokens": 103970710.0, "step": 4087 }, { "epoch": 0.5403119217552207, "grad_norm": 0.8718108262782748, "learning_rate": 5.65464995554775e-06, "loss": 0.6222, "mean_token_accuracy": 0.7941795587539673, "num_tokens": 104036246.0, "step": 4088 }, { "epoch": 0.5404440919904837, "grad_norm": 0.9025064928914134, "learning_rate": 5.652574918452127e-06, "loss": 0.6159, "mean_token_accuracy": 0.7984372973442078, "num_tokens": 104101782.0, "step": 4089 }, { "epoch": 0.5405762622257467, "grad_norm": 0.9120768919548129, "learning_rate": 5.650499848876489e-06, "loss": 0.6266, "mean_token_accuracy": 0.7916005253791809, "num_tokens": 104167318.0, "step": 4090 }, { "epoch": 0.5407084324610097, "grad_norm": 0.8902350446462641, "learning_rate": 5.648424747262576e-06, "loss": 0.5505, "mean_token_accuracy": 0.8154071569442749, "num_tokens": 104232854.0, "step": 4091 }, { "epoch": 0.5408406026962728, "grad_norm": 0.9038971196886988, "learning_rate": 5.6463496140521334e-06, "loss": 0.6284, "mean_token_accuracy": 0.7923635840415955, "num_tokens": 104298390.0, "step": 4092 }, { "epoch": 0.5409727729315358, "grad_norm": 0.8417329612379919, "learning_rate": 5.644274449686913e-06, "loss": 0.5579, "mean_token_accuracy": 0.8144152164459229, "num_tokens": 104363926.0, "step": 4093 }, { "epoch": 0.5411049431667988, "grad_norm": 0.8902339731585297, "learning_rate": 5.642199254608675e-06, "loss": 0.656, "mean_token_accuracy": 0.7835124135017395, "num_tokens": 104429462.0, "step": 4094 }, { "epoch": 0.5412371134020618, "grad_norm": 0.908303330079715, "learning_rate": 5.640124029259184e-06, "loss": 0.6011, "mean_token_accuracy": 0.8027102947235107, "num_tokens": 104494998.0, "step": 4095 }, { "epoch": 0.5413692836373248, "grad_norm": 0.9321402569081214, "learning_rate": 5.6380487740802125e-06, "loss": 0.6334, "mean_token_accuracy": 0.7904254794120789, "num_tokens": 104560534.0, "step": 4096 }, { "epoch": 0.5415014538725879, "grad_norm": 0.9139940144164979, "learning_rate": 5.63597348951354e-06, "loss": 0.6055, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 104626070.0, "step": 4097 }, { "epoch": 0.5416336241078509, "grad_norm": 0.9680589065782672, "learning_rate": 5.633898176000949e-06, "loss": 0.5886, "mean_token_accuracy": 0.8040837645530701, "num_tokens": 104691606.0, "step": 4098 }, { "epoch": 0.5417657943431139, "grad_norm": 0.8617187970637651, "learning_rate": 5.631822833984234e-06, "loss": 0.5737, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 104757142.0, "step": 4099 }, { "epoch": 0.5418979645783769, "grad_norm": 0.8576056178763097, "learning_rate": 5.62974746390519e-06, "loss": 0.5891, "mean_token_accuracy": 0.8071053624153137, "num_tokens": 104822678.0, "step": 4100 }, { "epoch": 0.5420301348136399, "grad_norm": 0.9312431599546291, "learning_rate": 5.627672066205617e-06, "loss": 0.575, "mean_token_accuracy": 0.8091350197792053, "num_tokens": 104888214.0, "step": 4101 }, { "epoch": 0.542162305048903, "grad_norm": 0.8487387829907308, "learning_rate": 5.625596641327329e-06, "loss": 0.6146, "mean_token_accuracy": 0.7956293225288391, "num_tokens": 104953750.0, "step": 4102 }, { "epoch": 0.542294475284166, "grad_norm": 0.9283843336042882, "learning_rate": 5.623521189712138e-06, "loss": 0.6203, "mean_token_accuracy": 0.7949578762054443, "num_tokens": 105019286.0, "step": 4103 }, { "epoch": 0.542426645519429, "grad_norm": 0.9422768390465309, "learning_rate": 5.621445711801865e-06, "loss": 0.6381, "mean_token_accuracy": 0.7887772917747498, "num_tokens": 105084822.0, "step": 4104 }, { "epoch": 0.542558815754692, "grad_norm": 0.8782121830396539, "learning_rate": 5.619370208038336e-06, "loss": 0.5829, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 105150358.0, "step": 4105 }, { "epoch": 0.542690985989955, "grad_norm": 0.9239334159043724, "learning_rate": 5.617294678863381e-06, "loss": 0.5828, "mean_token_accuracy": 0.807700514793396, "num_tokens": 105215894.0, "step": 4106 }, { "epoch": 0.542823156225218, "grad_norm": 0.971560074001999, "learning_rate": 5.615219124718838e-06, "loss": 0.6487, "mean_token_accuracy": 0.7856183648109436, "num_tokens": 105281430.0, "step": 4107 }, { "epoch": 0.5429553264604811, "grad_norm": 0.9121033938087743, "learning_rate": 5.6131435460465486e-06, "loss": 0.5591, "mean_token_accuracy": 0.8169332146644592, "num_tokens": 105346966.0, "step": 4108 }, { "epoch": 0.5430874966957441, "grad_norm": 0.870636717081384, "learning_rate": 5.611067943288361e-06, "loss": 0.5573, "mean_token_accuracy": 0.8159565329551697, "num_tokens": 105412502.0, "step": 4109 }, { "epoch": 0.5432196669310071, "grad_norm": 0.9012249883767698, "learning_rate": 5.608992316886127e-06, "loss": 0.5906, "mean_token_accuracy": 0.8063880801200867, "num_tokens": 105478038.0, "step": 4110 }, { "epoch": 0.5433518371662701, "grad_norm": 0.8596809124701402, "learning_rate": 5.606916667281705e-06, "loss": 0.6328, "mean_token_accuracy": 0.7931418418884277, "num_tokens": 105543574.0, "step": 4111 }, { "epoch": 0.5434840074015331, "grad_norm": 0.8667798246097612, "learning_rate": 5.604840994916955e-06, "loss": 0.5737, "mean_token_accuracy": 0.8093944787979126, "num_tokens": 105609110.0, "step": 4112 }, { "epoch": 0.5436161776367961, "grad_norm": 0.9665792505501084, "learning_rate": 5.602765300233747e-06, "loss": 0.6215, "mean_token_accuracy": 0.7954614758491516, "num_tokens": 105674646.0, "step": 4113 }, { "epoch": 0.5437483478720592, "grad_norm": 0.8660665349537099, "learning_rate": 5.6006895836739515e-06, "loss": 0.5791, "mean_token_accuracy": 0.807456374168396, "num_tokens": 105740182.0, "step": 4114 }, { "epoch": 0.5438805181073222, "grad_norm": 0.8604545520041749, "learning_rate": 5.598613845679448e-06, "loss": 0.5996, "mean_token_accuracy": 0.7990477085113525, "num_tokens": 105805718.0, "step": 4115 }, { "epoch": 0.5440126883425852, "grad_norm": 0.8220957710740103, "learning_rate": 5.596538086692115e-06, "loss": 0.5479, "mean_token_accuracy": 0.8185508251190186, "num_tokens": 105871254.0, "step": 4116 }, { "epoch": 0.5441448585778482, "grad_norm": 0.8588306191526035, "learning_rate": 5.594462307153837e-06, "loss": 0.5861, "mean_token_accuracy": 0.8085856437683105, "num_tokens": 105936790.0, "step": 4117 }, { "epoch": 0.5442770288131112, "grad_norm": 0.8153298474042114, "learning_rate": 5.592386507506507e-06, "loss": 0.5588, "mean_token_accuracy": 0.8169942498207092, "num_tokens": 106002326.0, "step": 4118 }, { "epoch": 0.5444091990483743, "grad_norm": 0.8245006864790727, "learning_rate": 5.5903106881920184e-06, "loss": 0.5879, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 106067862.0, "step": 4119 }, { "epoch": 0.5445413692836373, "grad_norm": 0.8445595460865826, "learning_rate": 5.588234849652271e-06, "loss": 0.532, "mean_token_accuracy": 0.8225491642951965, "num_tokens": 106133398.0, "step": 4120 }, { "epoch": 0.5446735395189003, "grad_norm": 0.7808914360260383, "learning_rate": 5.586158992329166e-06, "loss": 0.54, "mean_token_accuracy": 0.8209620118141174, "num_tokens": 106198934.0, "step": 4121 }, { "epoch": 0.5448057097541633, "grad_norm": 0.8547973305029453, "learning_rate": 5.584083116664609e-06, "loss": 0.5839, "mean_token_accuracy": 0.807578444480896, "num_tokens": 106264470.0, "step": 4122 }, { "epoch": 0.5449378799894263, "grad_norm": 0.8413069474587296, "learning_rate": 5.582007223100514e-06, "loss": 0.624, "mean_token_accuracy": 0.7918447256088257, "num_tokens": 106330006.0, "step": 4123 }, { "epoch": 0.5450700502246894, "grad_norm": 0.8347317085526896, "learning_rate": 5.579931312078793e-06, "loss": 0.5692, "mean_token_accuracy": 0.8124618530273438, "num_tokens": 106395542.0, "step": 4124 }, { "epoch": 0.5452022204599524, "grad_norm": 0.8704576937517935, "learning_rate": 5.577855384041364e-06, "loss": 0.5445, "mean_token_accuracy": 0.8174368143081665, "num_tokens": 106461078.0, "step": 4125 }, { "epoch": 0.5453343906952154, "grad_norm": 0.8993462143932581, "learning_rate": 5.575779439430149e-06, "loss": 0.5985, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 106526614.0, "step": 4126 }, { "epoch": 0.5454665609304784, "grad_norm": 0.8815918685604047, "learning_rate": 5.573703478687074e-06, "loss": 0.568, "mean_token_accuracy": 0.8098217844963074, "num_tokens": 106592150.0, "step": 4127 }, { "epoch": 0.5455987311657414, "grad_norm": 0.8030115500128256, "learning_rate": 5.571627502254067e-06, "loss": 0.5489, "mean_token_accuracy": 0.8208094239234924, "num_tokens": 106657686.0, "step": 4128 }, { "epoch": 0.5457309014010044, "grad_norm": 0.9122869849939732, "learning_rate": 5.569551510573059e-06, "loss": 0.5912, "mean_token_accuracy": 0.8051214814186096, "num_tokens": 106723222.0, "step": 4129 }, { "epoch": 0.5458630716362675, "grad_norm": 0.9379865428409723, "learning_rate": 5.567475504085989e-06, "loss": 0.6184, "mean_token_accuracy": 0.7979642152786255, "num_tokens": 106788758.0, "step": 4130 }, { "epoch": 0.5459952418715305, "grad_norm": 0.9073126598853699, "learning_rate": 5.565399483234792e-06, "loss": 0.6114, "mean_token_accuracy": 0.7985899448394775, "num_tokens": 106854294.0, "step": 4131 }, { "epoch": 0.5461274121067935, "grad_norm": 0.8248319740771712, "learning_rate": 5.563323448461409e-06, "loss": 0.6072, "mean_token_accuracy": 0.7983610033988953, "num_tokens": 106919830.0, "step": 4132 }, { "epoch": 0.5462595823420565, "grad_norm": 0.8453630164973989, "learning_rate": 5.561247400207785e-06, "loss": 0.5709, "mean_token_accuracy": 0.8107221126556396, "num_tokens": 106985366.0, "step": 4133 }, { "epoch": 0.5463917525773195, "grad_norm": 0.8943044010482296, "learning_rate": 5.559171338915872e-06, "loss": 0.6516, "mean_token_accuracy": 0.784794270992279, "num_tokens": 107050902.0, "step": 4134 }, { "epoch": 0.5465239228125826, "grad_norm": 0.847897331081217, "learning_rate": 5.557095265027613e-06, "loss": 0.5971, "mean_token_accuracy": 0.8017641305923462, "num_tokens": 107116438.0, "step": 4135 }, { "epoch": 0.5466560930478456, "grad_norm": 0.8248305943722409, "learning_rate": 5.555019178984967e-06, "loss": 0.5602, "mean_token_accuracy": 0.8146288394927979, "num_tokens": 107181974.0, "step": 4136 }, { "epoch": 0.5467882632831086, "grad_norm": 0.8357307356011554, "learning_rate": 5.5529430812298844e-06, "loss": 0.5881, "mean_token_accuracy": 0.8044805526733398, "num_tokens": 107247510.0, "step": 4137 }, { "epoch": 0.5469204335183717, "grad_norm": 0.9120743775544142, "learning_rate": 5.550866972204326e-06, "loss": 0.6216, "mean_token_accuracy": 0.7943322062492371, "num_tokens": 107313046.0, "step": 4138 }, { "epoch": 0.5470526037536347, "grad_norm": 0.8009144857038297, "learning_rate": 5.548790852350252e-06, "loss": 0.5396, "mean_token_accuracy": 0.818474531173706, "num_tokens": 107378582.0, "step": 4139 }, { "epoch": 0.5471847739888978, "grad_norm": 0.8239225482949156, "learning_rate": 5.546714722109622e-06, "loss": 0.5587, "mean_token_accuracy": 0.816185474395752, "num_tokens": 107444118.0, "step": 4140 }, { "epoch": 0.5473169442241608, "grad_norm": 0.853893037143958, "learning_rate": 5.544638581924404e-06, "loss": 0.6105, "mean_token_accuracy": 0.7990935444831848, "num_tokens": 107509654.0, "step": 4141 }, { "epoch": 0.5474491144594238, "grad_norm": 0.8197764302449568, "learning_rate": 5.5425624322365655e-06, "loss": 0.5699, "mean_token_accuracy": 0.8143541812896729, "num_tokens": 107575190.0, "step": 4142 }, { "epoch": 0.5475812846946868, "grad_norm": 0.9776787078699085, "learning_rate": 5.5404862734880725e-06, "loss": 0.6222, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 107640726.0, "step": 4143 }, { "epoch": 0.5477134549299498, "grad_norm": 0.8993668914745525, "learning_rate": 5.538410106120897e-06, "loss": 0.6424, "mean_token_accuracy": 0.7870833873748779, "num_tokens": 107706262.0, "step": 4144 }, { "epoch": 0.5478456251652128, "grad_norm": 0.920476326911888, "learning_rate": 5.536333930577014e-06, "loss": 0.6294, "mean_token_accuracy": 0.7922415137290955, "num_tokens": 107771798.0, "step": 4145 }, { "epoch": 0.5479777954004759, "grad_norm": 0.823620959730054, "learning_rate": 5.534257747298393e-06, "loss": 0.5776, "mean_token_accuracy": 0.8072274327278137, "num_tokens": 107837334.0, "step": 4146 }, { "epoch": 0.5481099656357389, "grad_norm": 0.7953830385158991, "learning_rate": 5.532181556727014e-06, "loss": 0.5962, "mean_token_accuracy": 0.8068459033966064, "num_tokens": 107902870.0, "step": 4147 }, { "epoch": 0.5482421358710019, "grad_norm": 0.7677527689005865, "learning_rate": 5.5301053593048525e-06, "loss": 0.5332, "mean_token_accuracy": 0.8250519037246704, "num_tokens": 107968406.0, "step": 4148 }, { "epoch": 0.5483743061062649, "grad_norm": 0.7898083045854789, "learning_rate": 5.52802915547389e-06, "loss": 0.5417, "mean_token_accuracy": 0.822839081287384, "num_tokens": 108033942.0, "step": 4149 }, { "epoch": 0.5485064763415279, "grad_norm": 0.940575180496674, "learning_rate": 5.525952945676105e-06, "loss": 0.6293, "mean_token_accuracy": 0.7907306551933289, "num_tokens": 108099478.0, "step": 4150 }, { "epoch": 0.548638646576791, "grad_norm": 0.8541813033347084, "learning_rate": 5.523876730353479e-06, "loss": 0.6212, "mean_token_accuracy": 0.7954919934272766, "num_tokens": 108165014.0, "step": 4151 }, { "epoch": 0.548770816812054, "grad_norm": 0.8283725806222831, "learning_rate": 5.521800509947996e-06, "loss": 0.5669, "mean_token_accuracy": 0.8104627132415771, "num_tokens": 108230550.0, "step": 4152 }, { "epoch": 0.548902987047317, "grad_norm": 0.9238718973478853, "learning_rate": 5.519724284901641e-06, "loss": 0.6402, "mean_token_accuracy": 0.7895556092262268, "num_tokens": 108296086.0, "step": 4153 }, { "epoch": 0.54903515728258, "grad_norm": 0.8278569306035112, "learning_rate": 5.517648055656395e-06, "loss": 0.5378, "mean_token_accuracy": 0.8220455646514893, "num_tokens": 108361622.0, "step": 4154 }, { "epoch": 0.549167327517843, "grad_norm": 0.802451387804868, "learning_rate": 5.515571822654249e-06, "loss": 0.537, "mean_token_accuracy": 0.821847140789032, "num_tokens": 108427158.0, "step": 4155 }, { "epoch": 0.549299497753106, "grad_norm": 0.8702825978519233, "learning_rate": 5.513495586337186e-06, "loss": 0.5888, "mean_token_accuracy": 0.8052588105201721, "num_tokens": 108492694.0, "step": 4156 }, { "epoch": 0.5494316679883691, "grad_norm": 0.8574075690564847, "learning_rate": 5.511419347147193e-06, "loss": 0.5908, "mean_token_accuracy": 0.8028781414031982, "num_tokens": 108558230.0, "step": 4157 }, { "epoch": 0.5495638382236321, "grad_norm": 0.9052415778459816, "learning_rate": 5.5093431055262615e-06, "loss": 0.6229, "mean_token_accuracy": 0.79337078332901, "num_tokens": 108623766.0, "step": 4158 }, { "epoch": 0.5496960084588951, "grad_norm": 0.8855440560290689, "learning_rate": 5.507266861916376e-06, "loss": 0.5928, "mean_token_accuracy": 0.8044958114624023, "num_tokens": 108689302.0, "step": 4159 }, { "epoch": 0.5498281786941581, "grad_norm": 0.8190604060378238, "learning_rate": 5.5051906167595295e-06, "loss": 0.574, "mean_token_accuracy": 0.8115919828414917, "num_tokens": 108754838.0, "step": 4160 }, { "epoch": 0.5499603489294211, "grad_norm": 0.8323215988373917, "learning_rate": 5.5031143704977065e-06, "loss": 0.5323, "mean_token_accuracy": 0.8267915844917297, "num_tokens": 108820374.0, "step": 4161 }, { "epoch": 0.5500925191646842, "grad_norm": 0.8695180402696515, "learning_rate": 5.501038123572902e-06, "loss": 0.5964, "mean_token_accuracy": 0.8036869764328003, "num_tokens": 108885910.0, "step": 4162 }, { "epoch": 0.5502246893999472, "grad_norm": 0.9079558451652724, "learning_rate": 5.4989618764271e-06, "loss": 0.6319, "mean_token_accuracy": 0.7863966822624207, "num_tokens": 108951446.0, "step": 4163 }, { "epoch": 0.5503568596352102, "grad_norm": 0.8815654176868734, "learning_rate": 5.496885629502296e-06, "loss": 0.5747, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 109016982.0, "step": 4164 }, { "epoch": 0.5504890298704732, "grad_norm": 0.8615161762754173, "learning_rate": 5.494809383240473e-06, "loss": 0.6224, "mean_token_accuracy": 0.792714536190033, "num_tokens": 109082518.0, "step": 4165 }, { "epoch": 0.5506212001057362, "grad_norm": 0.8556935975707214, "learning_rate": 5.4927331380836255e-06, "loss": 0.5544, "mean_token_accuracy": 0.818138837814331, "num_tokens": 109148054.0, "step": 4166 }, { "epoch": 0.5507533703409992, "grad_norm": 0.9167188429355744, "learning_rate": 5.490656894473742e-06, "loss": 0.6697, "mean_token_accuracy": 0.7793614864349365, "num_tokens": 109213590.0, "step": 4167 }, { "epoch": 0.5508855405762623, "grad_norm": 0.885040666381545, "learning_rate": 5.488580652852809e-06, "loss": 0.6108, "mean_token_accuracy": 0.8001922965049744, "num_tokens": 109279126.0, "step": 4168 }, { "epoch": 0.5510177108115253, "grad_norm": 0.9206961230782044, "learning_rate": 5.4865044136628165e-06, "loss": 0.6298, "mean_token_accuracy": 0.7912037372589111, "num_tokens": 109344662.0, "step": 4169 }, { "epoch": 0.5511498810467883, "grad_norm": 1.0351927630365108, "learning_rate": 5.484428177345754e-06, "loss": 0.6098, "mean_token_accuracy": 0.7965754866600037, "num_tokens": 109410198.0, "step": 4170 }, { "epoch": 0.5512820512820513, "grad_norm": 0.8469934247975871, "learning_rate": 5.482351944343606e-06, "loss": 0.6269, "mean_token_accuracy": 0.792592465877533, "num_tokens": 109475734.0, "step": 4171 }, { "epoch": 0.5514142215173143, "grad_norm": 0.8114187612906694, "learning_rate": 5.4802757150983625e-06, "loss": 0.5725, "mean_token_accuracy": 0.8107373714447021, "num_tokens": 109541270.0, "step": 4172 }, { "epoch": 0.5515463917525774, "grad_norm": 0.8519417554758267, "learning_rate": 5.478199490052006e-06, "loss": 0.5516, "mean_token_accuracy": 0.8169942498207092, "num_tokens": 109606806.0, "step": 4173 }, { "epoch": 0.5516785619878404, "grad_norm": 0.7806144201617145, "learning_rate": 5.476123269646522e-06, "loss": 0.5147, "mean_token_accuracy": 0.8291875123977661, "num_tokens": 109672342.0, "step": 4174 }, { "epoch": 0.5518107322231034, "grad_norm": 0.8668671344539167, "learning_rate": 5.474047054323896e-06, "loss": 0.5909, "mean_token_accuracy": 0.802221953868866, "num_tokens": 109737878.0, "step": 4175 }, { "epoch": 0.5519429024583664, "grad_norm": 0.8446701877279486, "learning_rate": 5.4719708445261125e-06, "loss": 0.5561, "mean_token_accuracy": 0.8151171803474426, "num_tokens": 109803414.0, "step": 4176 }, { "epoch": 0.5520750726936294, "grad_norm": 0.8164059230055635, "learning_rate": 5.469894640695148e-06, "loss": 0.5442, "mean_token_accuracy": 0.8196801543235779, "num_tokens": 109868950.0, "step": 4177 }, { "epoch": 0.5522072429288924, "grad_norm": 0.802026264108095, "learning_rate": 5.467818443272987e-06, "loss": 0.5939, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 109934486.0, "step": 4178 }, { "epoch": 0.5523394131641555, "grad_norm": 0.88415112157129, "learning_rate": 5.465742252701609e-06, "loss": 0.587, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 110000022.0, "step": 4179 }, { "epoch": 0.5524715833994185, "grad_norm": 0.9033273283519956, "learning_rate": 5.463666069422989e-06, "loss": 0.5694, "mean_token_accuracy": 0.8119887709617615, "num_tokens": 110065558.0, "step": 4180 }, { "epoch": 0.5526037536346815, "grad_norm": 0.9035000481914658, "learning_rate": 5.461589893879104e-06, "loss": 0.6471, "mean_token_accuracy": 0.7861677408218384, "num_tokens": 110131094.0, "step": 4181 }, { "epoch": 0.5527359238699445, "grad_norm": 0.8871970382436353, "learning_rate": 5.459513726511929e-06, "loss": 0.5823, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 110196630.0, "step": 4182 }, { "epoch": 0.5528680941052075, "grad_norm": 0.8234379641334107, "learning_rate": 5.457437567763436e-06, "loss": 0.5635, "mean_token_accuracy": 0.811851441860199, "num_tokens": 110262166.0, "step": 4183 }, { "epoch": 0.5530002643404706, "grad_norm": 0.8503527117514975, "learning_rate": 5.455361418075595e-06, "loss": 0.6146, "mean_token_accuracy": 0.7961176633834839, "num_tokens": 110327702.0, "step": 4184 }, { "epoch": 0.5531324345757336, "grad_norm": 0.9213982650331617, "learning_rate": 5.453285277890379e-06, "loss": 0.5925, "mean_token_accuracy": 0.8053656220436096, "num_tokens": 110393238.0, "step": 4185 }, { "epoch": 0.5532646048109966, "grad_norm": 0.964324561041276, "learning_rate": 5.451209147649751e-06, "loss": 0.5848, "mean_token_accuracy": 0.8083414435386658, "num_tokens": 110458774.0, "step": 4186 }, { "epoch": 0.5533967750462596, "grad_norm": 0.8593368045409301, "learning_rate": 5.4491330277956745e-06, "loss": 0.5362, "mean_token_accuracy": 0.8239073157310486, "num_tokens": 110524310.0, "step": 4187 }, { "epoch": 0.5535289452815226, "grad_norm": 0.8974916449369977, "learning_rate": 5.447056918770117e-06, "loss": 0.6014, "mean_token_accuracy": 0.8002380728721619, "num_tokens": 110589846.0, "step": 4188 }, { "epoch": 0.5536611155167857, "grad_norm": 0.866171202934257, "learning_rate": 5.444980821015036e-06, "loss": 0.5601, "mean_token_accuracy": 0.8132401704788208, "num_tokens": 110655382.0, "step": 4189 }, { "epoch": 0.5537932857520487, "grad_norm": 0.891663016926969, "learning_rate": 5.4429047349723875e-06, "loss": 0.6057, "mean_token_accuracy": 0.7974758744239807, "num_tokens": 110720918.0, "step": 4190 }, { "epoch": 0.5539254559873117, "grad_norm": 0.8836119346379304, "learning_rate": 5.44082866108413e-06, "loss": 0.5801, "mean_token_accuracy": 0.8101422190666199, "num_tokens": 110786454.0, "step": 4191 }, { "epoch": 0.5540576262225747, "grad_norm": 0.8705105133858436, "learning_rate": 5.438752599792216e-06, "loss": 0.5893, "mean_token_accuracy": 0.8056555986404419, "num_tokens": 110851990.0, "step": 4192 }, { "epoch": 0.5541897964578377, "grad_norm": 0.8938741823953802, "learning_rate": 5.4366765515385945e-06, "loss": 0.5885, "mean_token_accuracy": 0.8072732090950012, "num_tokens": 110917526.0, "step": 4193 }, { "epoch": 0.5543219666931007, "grad_norm": 0.8914606058194195, "learning_rate": 5.434600516765211e-06, "loss": 0.6254, "mean_token_accuracy": 0.7928671836853027, "num_tokens": 110983062.0, "step": 4194 }, { "epoch": 0.5544541369283638, "grad_norm": 0.861750288670832, "learning_rate": 5.432524495914014e-06, "loss": 0.5979, "mean_token_accuracy": 0.8030155301094055, "num_tokens": 111048598.0, "step": 4195 }, { "epoch": 0.5545863071636268, "grad_norm": 0.8856780208860117, "learning_rate": 5.430448489426944e-06, "loss": 0.6383, "mean_token_accuracy": 0.78831946849823, "num_tokens": 111114134.0, "step": 4196 }, { "epoch": 0.5547184773988898, "grad_norm": 0.888010889180882, "learning_rate": 5.428372497745935e-06, "loss": 0.5868, "mean_token_accuracy": 0.8058692216873169, "num_tokens": 111179670.0, "step": 4197 }, { "epoch": 0.5548506476341528, "grad_norm": 0.8851099555746671, "learning_rate": 5.426296521312928e-06, "loss": 0.5898, "mean_token_accuracy": 0.803061306476593, "num_tokens": 111245206.0, "step": 4198 }, { "epoch": 0.5549828178694158, "grad_norm": 0.8654363407115694, "learning_rate": 5.424220560569854e-06, "loss": 0.5689, "mean_token_accuracy": 0.8103101253509521, "num_tokens": 111310742.0, "step": 4199 }, { "epoch": 0.5551149881046789, "grad_norm": 0.8572740858092862, "learning_rate": 5.422144615958639e-06, "loss": 0.5486, "mean_token_accuracy": 0.8203210830688477, "num_tokens": 111376278.0, "step": 4200 }, { "epoch": 0.5552471583399419, "grad_norm": 0.9169026055212226, "learning_rate": 5.42006868792121e-06, "loss": 0.6177, "mean_token_accuracy": 0.7959345579147339, "num_tokens": 111441814.0, "step": 4201 }, { "epoch": 0.5553793285752049, "grad_norm": 0.9554241866198256, "learning_rate": 5.417992776899488e-06, "loss": 0.6024, "mean_token_accuracy": 0.8012910485267639, "num_tokens": 111507350.0, "step": 4202 }, { "epoch": 0.5555114988104679, "grad_norm": 0.9859071076886474, "learning_rate": 5.415916883335392e-06, "loss": 0.6549, "mean_token_accuracy": 0.7869155406951904, "num_tokens": 111572886.0, "step": 4203 }, { "epoch": 0.5556436690457309, "grad_norm": 0.8886214696669484, "learning_rate": 5.413841007670836e-06, "loss": 0.6078, "mean_token_accuracy": 0.8014436364173889, "num_tokens": 111638422.0, "step": 4204 }, { "epoch": 0.5557758392809939, "grad_norm": 0.8599662656800574, "learning_rate": 5.411765150347731e-06, "loss": 0.5854, "mean_token_accuracy": 0.8073495030403137, "num_tokens": 111703958.0, "step": 4205 }, { "epoch": 0.555908009516257, "grad_norm": 0.9161397686507131, "learning_rate": 5.409689311807982e-06, "loss": 0.6098, "mean_token_accuracy": 0.7997344732284546, "num_tokens": 111769494.0, "step": 4206 }, { "epoch": 0.55604017975152, "grad_norm": 0.8695059974064248, "learning_rate": 5.407613492493494e-06, "loss": 0.5825, "mean_token_accuracy": 0.8058692216873169, "num_tokens": 111835030.0, "step": 4207 }, { "epoch": 0.556172349986783, "grad_norm": 0.8752828024026609, "learning_rate": 5.405537692846165e-06, "loss": 0.6181, "mean_token_accuracy": 0.7945916056632996, "num_tokens": 111900566.0, "step": 4208 }, { "epoch": 0.556304520222046, "grad_norm": 0.8954976170308039, "learning_rate": 5.403461913307889e-06, "loss": 0.5799, "mean_token_accuracy": 0.8081430792808533, "num_tokens": 111966102.0, "step": 4209 }, { "epoch": 0.556436690457309, "grad_norm": 0.8610356375710329, "learning_rate": 5.401386154320554e-06, "loss": 0.6008, "mean_token_accuracy": 0.8025271892547607, "num_tokens": 112031638.0, "step": 4210 }, { "epoch": 0.556568860692572, "grad_norm": 0.9385217506625156, "learning_rate": 5.399310416326049e-06, "loss": 0.636, "mean_token_accuracy": 0.7885789275169373, "num_tokens": 112097174.0, "step": 4211 }, { "epoch": 0.5567010309278351, "grad_norm": 0.8297446598845154, "learning_rate": 5.397234699766255e-06, "loss": 0.6045, "mean_token_accuracy": 0.8004516959190369, "num_tokens": 112162710.0, "step": 4212 }, { "epoch": 0.5568332011630981, "grad_norm": 0.8687933491035265, "learning_rate": 5.395159005083046e-06, "loss": 0.6503, "mean_token_accuracy": 0.7855573296546936, "num_tokens": 112228246.0, "step": 4213 }, { "epoch": 0.5569653713983611, "grad_norm": 0.8397391864336811, "learning_rate": 5.393083332718298e-06, "loss": 0.5522, "mean_token_accuracy": 0.8156971335411072, "num_tokens": 112293782.0, "step": 4214 }, { "epoch": 0.5570975416336241, "grad_norm": 0.8345008903396033, "learning_rate": 5.391007683113876e-06, "loss": 0.5349, "mean_token_accuracy": 0.8224422931671143, "num_tokens": 112359318.0, "step": 4215 }, { "epoch": 0.5572297118688871, "grad_norm": 0.8036116238039896, "learning_rate": 5.38893205671164e-06, "loss": 0.5565, "mean_token_accuracy": 0.8178641200065613, "num_tokens": 112424854.0, "step": 4216 }, { "epoch": 0.5573618821041502, "grad_norm": 0.9095889096148676, "learning_rate": 5.386856453953453e-06, "loss": 0.5886, "mean_token_accuracy": 0.8050146698951721, "num_tokens": 112490390.0, "step": 4217 }, { "epoch": 0.5574940523394132, "grad_norm": 0.8523217026103577, "learning_rate": 5.384780875281165e-06, "loss": 0.5862, "mean_token_accuracy": 0.8028934001922607, "num_tokens": 112555926.0, "step": 4218 }, { "epoch": 0.5576262225746762, "grad_norm": 0.8353048809071162, "learning_rate": 5.382705321136621e-06, "loss": 0.5906, "mean_token_accuracy": 0.8038548231124878, "num_tokens": 112621462.0, "step": 4219 }, { "epoch": 0.5577583928099392, "grad_norm": 0.8658993433589263, "learning_rate": 5.380629791961665e-06, "loss": 0.6247, "mean_token_accuracy": 0.7944542765617371, "num_tokens": 112686998.0, "step": 4220 }, { "epoch": 0.5578905630452022, "grad_norm": 0.843948704428029, "learning_rate": 5.378554288198137e-06, "loss": 0.557, "mean_token_accuracy": 0.8178336024284363, "num_tokens": 112752534.0, "step": 4221 }, { "epoch": 0.5580227332804653, "grad_norm": 0.8850375075092924, "learning_rate": 5.3764788102878626e-06, "loss": 0.5537, "mean_token_accuracy": 0.8188408017158508, "num_tokens": 112818070.0, "step": 4222 }, { "epoch": 0.5581549035157283, "grad_norm": 0.9154680280595344, "learning_rate": 5.3744033586726715e-06, "loss": 0.6487, "mean_token_accuracy": 0.784672200679779, "num_tokens": 112883606.0, "step": 4223 }, { "epoch": 0.5582870737509913, "grad_norm": 0.8766971553511164, "learning_rate": 5.3723279337943846e-06, "loss": 0.5672, "mean_token_accuracy": 0.8138048052787781, "num_tokens": 112949142.0, "step": 4224 }, { "epoch": 0.5584192439862543, "grad_norm": 0.8466772183159975, "learning_rate": 5.370252536094813e-06, "loss": 0.5742, "mean_token_accuracy": 0.8111494183540344, "num_tokens": 113014678.0, "step": 4225 }, { "epoch": 0.5585514142215173, "grad_norm": 0.9244042677714495, "learning_rate": 5.368177166015768e-06, "loss": 0.6378, "mean_token_accuracy": 0.792470395565033, "num_tokens": 113080214.0, "step": 4226 }, { "epoch": 0.5586835844567803, "grad_norm": 0.8523238884750444, "learning_rate": 5.366101823999051e-06, "loss": 0.567, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 113145750.0, "step": 4227 }, { "epoch": 0.5588157546920434, "grad_norm": 0.8497459147425259, "learning_rate": 5.364026510486463e-06, "loss": 0.6241, "mean_token_accuracy": 0.7951104640960693, "num_tokens": 113211286.0, "step": 4228 }, { "epoch": 0.5589479249273064, "grad_norm": 0.9104531439379071, "learning_rate": 5.361951225919788e-06, "loss": 0.6187, "mean_token_accuracy": 0.7948358058929443, "num_tokens": 113276822.0, "step": 4229 }, { "epoch": 0.5590800951625694, "grad_norm": 0.8872428328929539, "learning_rate": 5.3598759707408176e-06, "loss": 0.561, "mean_token_accuracy": 0.8141862750053406, "num_tokens": 113342358.0, "step": 4230 }, { "epoch": 0.5592122653978324, "grad_norm": 0.8141038523916168, "learning_rate": 5.357800745391329e-06, "loss": 0.5773, "mean_token_accuracy": 0.808570384979248, "num_tokens": 113407894.0, "step": 4231 }, { "epoch": 0.5593444356330954, "grad_norm": 0.8564226890557249, "learning_rate": 5.355725550313088e-06, "loss": 0.5901, "mean_token_accuracy": 0.8055182695388794, "num_tokens": 113473430.0, "step": 4232 }, { "epoch": 0.5594766058683585, "grad_norm": 0.8830038957301406, "learning_rate": 5.353650385947868e-06, "loss": 0.5728, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 113538966.0, "step": 4233 }, { "epoch": 0.5596087761036215, "grad_norm": 1.0147332234000397, "learning_rate": 5.351575252737425e-06, "loss": 0.6234, "mean_token_accuracy": 0.7934470772743225, "num_tokens": 113604502.0, "step": 4234 }, { "epoch": 0.5597409463388845, "grad_norm": 0.8332744383290164, "learning_rate": 5.349500151123511e-06, "loss": 0.5274, "mean_token_accuracy": 0.8275241255760193, "num_tokens": 113670038.0, "step": 4235 }, { "epoch": 0.5598731165741475, "grad_norm": 0.8561473102917744, "learning_rate": 5.347425081547875e-06, "loss": 0.6246, "mean_token_accuracy": 0.7937675714492798, "num_tokens": 113735574.0, "step": 4236 }, { "epoch": 0.5600052868094105, "grad_norm": 0.8622172453642214, "learning_rate": 5.345350044452251e-06, "loss": 0.6007, "mean_token_accuracy": 0.8042058348655701, "num_tokens": 113801110.0, "step": 4237 }, { "epoch": 0.5601374570446735, "grad_norm": 0.8460649908404934, "learning_rate": 5.343275040278375e-06, "loss": 0.5662, "mean_token_accuracy": 0.8139268755912781, "num_tokens": 113866646.0, "step": 4238 }, { "epoch": 0.5602696272799366, "grad_norm": 0.8518499094441778, "learning_rate": 5.34120006946797e-06, "loss": 0.5561, "mean_token_accuracy": 0.8140794634819031, "num_tokens": 113932182.0, "step": 4239 }, { "epoch": 0.5604017975151996, "grad_norm": 0.8824029523482942, "learning_rate": 5.339125132462756e-06, "loss": 0.595, "mean_token_accuracy": 0.802374541759491, "num_tokens": 113997718.0, "step": 4240 }, { "epoch": 0.5605339677504626, "grad_norm": 0.8636293000325397, "learning_rate": 5.337050229704442e-06, "loss": 0.5957, "mean_token_accuracy": 0.8029391765594482, "num_tokens": 114063254.0, "step": 4241 }, { "epoch": 0.5606661379857256, "grad_norm": 0.8859706983805181, "learning_rate": 5.334975361634731e-06, "loss": 0.5851, "mean_token_accuracy": 0.8082346320152283, "num_tokens": 114128790.0, "step": 4242 }, { "epoch": 0.5607983082209886, "grad_norm": 0.8168357098741473, "learning_rate": 5.3329005286953205e-06, "loss": 0.5389, "mean_token_accuracy": 0.8224575519561768, "num_tokens": 114194326.0, "step": 4243 }, { "epoch": 0.5609304784562517, "grad_norm": 0.9093312674121521, "learning_rate": 5.3308257313279e-06, "loss": 0.5684, "mean_token_accuracy": 0.8112868070602417, "num_tokens": 114259862.0, "step": 4244 }, { "epoch": 0.5610626486915147, "grad_norm": 0.8489080756794174, "learning_rate": 5.3287509699741455e-06, "loss": 0.5263, "mean_token_accuracy": 0.8255707621574402, "num_tokens": 114325398.0, "step": 4245 }, { "epoch": 0.5611948189267777, "grad_norm": 0.8884908214775061, "learning_rate": 5.326676245075738e-06, "loss": 0.6035, "mean_token_accuracy": 0.8005738258361816, "num_tokens": 114390934.0, "step": 4246 }, { "epoch": 0.5613269891620407, "grad_norm": 0.8284916952972259, "learning_rate": 5.324601557074338e-06, "loss": 0.55, "mean_token_accuracy": 0.8179098963737488, "num_tokens": 114456470.0, "step": 4247 }, { "epoch": 0.5614591593973037, "grad_norm": 0.8669979693964046, "learning_rate": 5.322526906411602e-06, "loss": 0.577, "mean_token_accuracy": 0.8111799359321594, "num_tokens": 114522006.0, "step": 4248 }, { "epoch": 0.5615913296325667, "grad_norm": 0.8848183103589063, "learning_rate": 5.320452293529186e-06, "loss": 0.5667, "mean_token_accuracy": 0.8130722641944885, "num_tokens": 114587542.0, "step": 4249 }, { "epoch": 0.5617234998678298, "grad_norm": 0.8641116808261334, "learning_rate": 5.318377718868727e-06, "loss": 0.6065, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 114653078.0, "step": 4250 }, { "epoch": 0.5618556701030928, "grad_norm": 0.8930159676702208, "learning_rate": 5.316303182871858e-06, "loss": 0.6383, "mean_token_accuracy": 0.792409360408783, "num_tokens": 114718614.0, "step": 4251 }, { "epoch": 0.5619878403383558, "grad_norm": 0.8106952999675732, "learning_rate": 5.314228685980207e-06, "loss": 0.5859, "mean_token_accuracy": 0.8054113984107971, "num_tokens": 114784150.0, "step": 4252 }, { "epoch": 0.5621200105736188, "grad_norm": 0.8082399695460011, "learning_rate": 5.312154228635391e-06, "loss": 0.6058, "mean_token_accuracy": 0.7995971441268921, "num_tokens": 114849686.0, "step": 4253 }, { "epoch": 0.5622521808088818, "grad_norm": 0.9135680577244771, "learning_rate": 5.310079811279018e-06, "loss": 0.6224, "mean_token_accuracy": 0.79361492395401, "num_tokens": 114915222.0, "step": 4254 }, { "epoch": 0.5623843510441449, "grad_norm": 0.8385870144084139, "learning_rate": 5.308005434352686e-06, "loss": 0.6145, "mean_token_accuracy": 0.7958124876022339, "num_tokens": 114980758.0, "step": 4255 }, { "epoch": 0.5625165212794079, "grad_norm": 0.8477229564438684, "learning_rate": 5.30593109829799e-06, "loss": 0.5838, "mean_token_accuracy": 0.8101117014884949, "num_tokens": 115046294.0, "step": 4256 }, { "epoch": 0.5626486915146709, "grad_norm": 0.9149192619134063, "learning_rate": 5.303856803556511e-06, "loss": 0.5706, "mean_token_accuracy": 0.8125991821289062, "num_tokens": 115111830.0, "step": 4257 }, { "epoch": 0.5627808617499339, "grad_norm": 0.911761961371127, "learning_rate": 5.301782550569823e-06, "loss": 0.6452, "mean_token_accuracy": 0.7872664928436279, "num_tokens": 115177366.0, "step": 4258 }, { "epoch": 0.5629130319851969, "grad_norm": 0.882626066239066, "learning_rate": 5.29970833977949e-06, "loss": 0.5776, "mean_token_accuracy": 0.808539867401123, "num_tokens": 115242902.0, "step": 4259 }, { "epoch": 0.56304520222046, "grad_norm": 0.8195303704324649, "learning_rate": 5.2976341716270715e-06, "loss": 0.5624, "mean_token_accuracy": 0.8151019215583801, "num_tokens": 115308438.0, "step": 4260 }, { "epoch": 0.563177372455723, "grad_norm": 0.864691719650516, "learning_rate": 5.295560046554113e-06, "loss": 0.6175, "mean_token_accuracy": 0.7971859574317932, "num_tokens": 115373974.0, "step": 4261 }, { "epoch": 0.563309542690986, "grad_norm": 0.8546510264028513, "learning_rate": 5.293485965002152e-06, "loss": 0.5814, "mean_token_accuracy": 0.8071206212043762, "num_tokens": 115439510.0, "step": 4262 }, { "epoch": 0.563441712926249, "grad_norm": 0.9249947524336286, "learning_rate": 5.291411927412717e-06, "loss": 0.6136, "mean_token_accuracy": 0.7995818853378296, "num_tokens": 115505046.0, "step": 4263 }, { "epoch": 0.563573883161512, "grad_norm": 0.9059169552704831, "learning_rate": 5.289337934227327e-06, "loss": 0.6038, "mean_token_accuracy": 0.802466094493866, "num_tokens": 115570582.0, "step": 4264 }, { "epoch": 0.563706053396775, "grad_norm": 0.8363968890176124, "learning_rate": 5.287263985887494e-06, "loss": 0.6028, "mean_token_accuracy": 0.8011994957923889, "num_tokens": 115636118.0, "step": 4265 }, { "epoch": 0.5638382236320381, "grad_norm": 0.877651712456255, "learning_rate": 5.2851900828347154e-06, "loss": 0.6026, "mean_token_accuracy": 0.7963771224021912, "num_tokens": 115701654.0, "step": 4266 }, { "epoch": 0.5639703938673011, "grad_norm": 0.8674621541947083, "learning_rate": 5.283116225510483e-06, "loss": 0.5644, "mean_token_accuracy": 0.8112562298774719, "num_tokens": 115767190.0, "step": 4267 }, { "epoch": 0.5641025641025641, "grad_norm": 0.90904427821127, "learning_rate": 5.28104241435628e-06, "loss": 0.5807, "mean_token_accuracy": 0.8070290684700012, "num_tokens": 115832726.0, "step": 4268 }, { "epoch": 0.5642347343378271, "grad_norm": 0.8295871903727586, "learning_rate": 5.278968649813575e-06, "loss": 0.5945, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 115898262.0, "step": 4269 }, { "epoch": 0.5643669045730901, "grad_norm": 0.8397371908444199, "learning_rate": 5.276894932323829e-06, "loss": 0.5878, "mean_token_accuracy": 0.803427517414093, "num_tokens": 115963798.0, "step": 4270 }, { "epoch": 0.5644990748083532, "grad_norm": 0.8988844646831615, "learning_rate": 5.274821262328492e-06, "loss": 0.5967, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 116029334.0, "step": 4271 }, { "epoch": 0.5646312450436162, "grad_norm": 0.84711104490997, "learning_rate": 5.2727476402690096e-06, "loss": 0.6248, "mean_token_accuracy": 0.7968502044677734, "num_tokens": 116094870.0, "step": 4272 }, { "epoch": 0.5647634152788792, "grad_norm": 0.7941850805356551, "learning_rate": 5.270674066586809e-06, "loss": 0.5493, "mean_token_accuracy": 0.8169179558753967, "num_tokens": 116160406.0, "step": 4273 }, { "epoch": 0.5648955855141422, "grad_norm": 0.7924484578852757, "learning_rate": 5.268600541723311e-06, "loss": 0.5715, "mean_token_accuracy": 0.8111494183540344, "num_tokens": 116225942.0, "step": 4274 }, { "epoch": 0.5650277557494052, "grad_norm": 0.8791099332191631, "learning_rate": 5.266527066119927e-06, "loss": 0.5947, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 116291478.0, "step": 4275 }, { "epoch": 0.5651599259846682, "grad_norm": 0.8580288684643845, "learning_rate": 5.2644536402180565e-06, "loss": 0.5847, "mean_token_accuracy": 0.806617021560669, "num_tokens": 116357014.0, "step": 4276 }, { "epoch": 0.5652920962199313, "grad_norm": 0.8371665757954282, "learning_rate": 5.262380264459088e-06, "loss": 0.6351, "mean_token_accuracy": 0.7900744676589966, "num_tokens": 116422550.0, "step": 4277 }, { "epoch": 0.5654242664551943, "grad_norm": 0.871281370581281, "learning_rate": 5.260306939284398e-06, "loss": 0.6035, "mean_token_accuracy": 0.8024966716766357, "num_tokens": 116488086.0, "step": 4278 }, { "epoch": 0.5655564366904573, "grad_norm": 0.8571842479779257, "learning_rate": 5.25823366513536e-06, "loss": 0.6323, "mean_token_accuracy": 0.7915852665901184, "num_tokens": 116553622.0, "step": 4279 }, { "epoch": 0.5656886069257203, "grad_norm": 0.8856542816634027, "learning_rate": 5.2561604424533244e-06, "loss": 0.6253, "mean_token_accuracy": 0.7963313460350037, "num_tokens": 116619158.0, "step": 4280 }, { "epoch": 0.5658207771609833, "grad_norm": 0.8995993108089616, "learning_rate": 5.25408727167964e-06, "loss": 0.5451, "mean_token_accuracy": 0.8174368143081665, "num_tokens": 116684694.0, "step": 4281 }, { "epoch": 0.5659529473962464, "grad_norm": 0.8584597629821293, "learning_rate": 5.252014153255642e-06, "loss": 0.6367, "mean_token_accuracy": 0.7915852665901184, "num_tokens": 116750230.0, "step": 4282 }, { "epoch": 0.5660851176315094, "grad_norm": 0.8248404136902519, "learning_rate": 5.2499410876226505e-06, "loss": 0.5769, "mean_token_accuracy": 0.8082041144371033, "num_tokens": 116815766.0, "step": 4283 }, { "epoch": 0.5662172878667724, "grad_norm": 0.7174464605028283, "learning_rate": 5.247868075221983e-06, "loss": 0.4757, "mean_token_accuracy": 0.8409687280654907, "num_tokens": 116881302.0, "step": 4284 }, { "epoch": 0.5663494581020354, "grad_norm": 0.9106734655684273, "learning_rate": 5.2457951164949395e-06, "loss": 0.6177, "mean_token_accuracy": 0.7970485687255859, "num_tokens": 116946838.0, "step": 4285 }, { "epoch": 0.5664816283372984, "grad_norm": 0.899937505226692, "learning_rate": 5.243722211882807e-06, "loss": 0.58, "mean_token_accuracy": 0.8104016780853271, "num_tokens": 117012374.0, "step": 4286 }, { "epoch": 0.5666137985725614, "grad_norm": 0.7754093424489206, "learning_rate": 5.241649361826862e-06, "loss": 0.5616, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 117077910.0, "step": 4287 }, { "epoch": 0.5667459688078245, "grad_norm": 0.9289933088767939, "learning_rate": 5.239576566768377e-06, "loss": 0.6092, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 117143446.0, "step": 4288 }, { "epoch": 0.5668781390430875, "grad_norm": 0.9107494467954774, "learning_rate": 5.237503827148604e-06, "loss": 0.564, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 117208982.0, "step": 4289 }, { "epoch": 0.5670103092783505, "grad_norm": 0.936350206083914, "learning_rate": 5.235431143408783e-06, "loss": 0.5624, "mean_token_accuracy": 0.8129196763038635, "num_tokens": 117274518.0, "step": 4290 }, { "epoch": 0.5671424795136135, "grad_norm": 0.9142708763065277, "learning_rate": 5.23335851599015e-06, "loss": 0.6102, "mean_token_accuracy": 0.7965297102928162, "num_tokens": 117340054.0, "step": 4291 }, { "epoch": 0.5672746497488765, "grad_norm": 0.7864348207565975, "learning_rate": 5.231285945333922e-06, "loss": 0.5718, "mean_token_accuracy": 0.8100659251213074, "num_tokens": 117405590.0, "step": 4292 }, { "epoch": 0.5674068199841396, "grad_norm": 0.8402357318656584, "learning_rate": 5.229213431881304e-06, "loss": 0.6071, "mean_token_accuracy": 0.7999175786972046, "num_tokens": 117471126.0, "step": 4293 }, { "epoch": 0.5675389902194026, "grad_norm": 0.8343736574116395, "learning_rate": 5.2271409760734935e-06, "loss": 0.5672, "mean_token_accuracy": 0.8113783597946167, "num_tokens": 117536662.0, "step": 4294 }, { "epoch": 0.5676711604546656, "grad_norm": 0.9838093059585978, "learning_rate": 5.225068578351674e-06, "loss": 0.6108, "mean_token_accuracy": 0.7994750142097473, "num_tokens": 117602198.0, "step": 4295 }, { "epoch": 0.5678033306899286, "grad_norm": 0.8741818582989004, "learning_rate": 5.222996239157012e-06, "loss": 0.5635, "mean_token_accuracy": 0.8122329115867615, "num_tokens": 117667734.0, "step": 4296 }, { "epoch": 0.5679355009251916, "grad_norm": 0.8684275859651182, "learning_rate": 5.220923958930668e-06, "loss": 0.5609, "mean_token_accuracy": 0.8152240514755249, "num_tokens": 117733270.0, "step": 4297 }, { "epoch": 0.5680676711604546, "grad_norm": 0.8819262200130568, "learning_rate": 5.218851738113787e-06, "loss": 0.5699, "mean_token_accuracy": 0.8141862750053406, "num_tokens": 117798806.0, "step": 4298 }, { "epoch": 0.5681998413957177, "grad_norm": 0.8806943261214275, "learning_rate": 5.216779577147502e-06, "loss": 0.5997, "mean_token_accuracy": 0.7999480962753296, "num_tokens": 117864342.0, "step": 4299 }, { "epoch": 0.5683320116309807, "grad_norm": 0.8044409606768258, "learning_rate": 5.214707476472929e-06, "loss": 0.5373, "mean_token_accuracy": 0.8209314942359924, "num_tokens": 117929878.0, "step": 4300 }, { "epoch": 0.5684641818662437, "grad_norm": 0.8675647187707443, "learning_rate": 5.2126354365311795e-06, "loss": 0.5668, "mean_token_accuracy": 0.8115156888961792, "num_tokens": 117995414.0, "step": 4301 }, { "epoch": 0.5685963521015067, "grad_norm": 0.9196959762864984, "learning_rate": 5.210563457763345e-06, "loss": 0.6363, "mean_token_accuracy": 0.7927451133728027, "num_tokens": 118060950.0, "step": 4302 }, { "epoch": 0.5687285223367697, "grad_norm": 0.8782828030737477, "learning_rate": 5.208491540610507e-06, "loss": 0.5724, "mean_token_accuracy": 0.8093334436416626, "num_tokens": 118126486.0, "step": 4303 }, { "epoch": 0.5688606925720328, "grad_norm": 0.7963832419400722, "learning_rate": 5.206419685513735e-06, "loss": 0.5828, "mean_token_accuracy": 0.8071053624153137, "num_tokens": 118192022.0, "step": 4304 }, { "epoch": 0.5689928628072958, "grad_norm": 0.7573365397353528, "learning_rate": 5.204347892914082e-06, "loss": 0.5131, "mean_token_accuracy": 0.8306830525398254, "num_tokens": 118257558.0, "step": 4305 }, { "epoch": 0.5691250330425588, "grad_norm": 0.874955430671001, "learning_rate": 5.202276163252589e-06, "loss": 0.6163, "mean_token_accuracy": 0.7971706986427307, "num_tokens": 118323094.0, "step": 4306 }, { "epoch": 0.5692572032778218, "grad_norm": 0.8207048527006264, "learning_rate": 5.2002044969702856e-06, "loss": 0.5466, "mean_token_accuracy": 0.817207932472229, "num_tokens": 118388630.0, "step": 4307 }, { "epoch": 0.5693893735130848, "grad_norm": 0.8350157534733269, "learning_rate": 5.198132894508186e-06, "loss": 0.5747, "mean_token_accuracy": 0.8112868070602417, "num_tokens": 118454166.0, "step": 4308 }, { "epoch": 0.5695215437483478, "grad_norm": 0.8621109554067061, "learning_rate": 5.196061356307289e-06, "loss": 0.6174, "mean_token_accuracy": 0.797887921333313, "num_tokens": 118519702.0, "step": 4309 }, { "epoch": 0.5696537139836109, "grad_norm": 0.8888026889538672, "learning_rate": 5.193989882808585e-06, "loss": 0.5799, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 118585238.0, "step": 4310 }, { "epoch": 0.5697858842188739, "grad_norm": 0.8603523003051505, "learning_rate": 5.191918474453046e-06, "loss": 0.6089, "mean_token_accuracy": 0.797826886177063, "num_tokens": 118650774.0, "step": 4311 }, { "epoch": 0.5699180544541369, "grad_norm": 0.8735990476578722, "learning_rate": 5.18984713168163e-06, "loss": 0.5455, "mean_token_accuracy": 0.8190696835517883, "num_tokens": 118716310.0, "step": 4312 }, { "epoch": 0.5700502246893999, "grad_norm": 0.8100651658094228, "learning_rate": 5.187775854935285e-06, "loss": 0.5879, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 118781846.0, "step": 4313 }, { "epoch": 0.5701823949246629, "grad_norm": 0.8368119184901791, "learning_rate": 5.185704644654943e-06, "loss": 0.5266, "mean_token_accuracy": 0.8245177865028381, "num_tokens": 118847382.0, "step": 4314 }, { "epoch": 0.570314565159926, "grad_norm": 0.9430793426586588, "learning_rate": 5.18363350128152e-06, "loss": 0.6133, "mean_token_accuracy": 0.7996276617050171, "num_tokens": 118912918.0, "step": 4315 }, { "epoch": 0.570446735395189, "grad_norm": 0.8302386391005429, "learning_rate": 5.181562425255916e-06, "loss": 0.5829, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 118978454.0, "step": 4316 }, { "epoch": 0.570578905630452, "grad_norm": 0.8517955291003189, "learning_rate": 5.179491417019026e-06, "loss": 0.5415, "mean_token_accuracy": 0.821755588054657, "num_tokens": 119043990.0, "step": 4317 }, { "epoch": 0.570711075865715, "grad_norm": 0.8855435092965293, "learning_rate": 5.177420477011722e-06, "loss": 0.5884, "mean_token_accuracy": 0.8062049746513367, "num_tokens": 119109526.0, "step": 4318 }, { "epoch": 0.570843246100978, "grad_norm": 0.7668440541596173, "learning_rate": 5.175349605674862e-06, "loss": 0.561, "mean_token_accuracy": 0.8123703002929688, "num_tokens": 119175062.0, "step": 4319 }, { "epoch": 0.570975416336241, "grad_norm": 0.8817370639159625, "learning_rate": 5.173278803449295e-06, "loss": 0.6212, "mean_token_accuracy": 0.793462336063385, "num_tokens": 119240598.0, "step": 4320 }, { "epoch": 0.5711075865715041, "grad_norm": 0.8028684064280012, "learning_rate": 5.171208070775846e-06, "loss": 0.5739, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 119306134.0, "step": 4321 }, { "epoch": 0.5712397568067671, "grad_norm": 0.9245480967058208, "learning_rate": 5.169137408095335e-06, "loss": 0.5857, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 119371670.0, "step": 4322 }, { "epoch": 0.5713719270420301, "grad_norm": 0.9289458884475247, "learning_rate": 5.167066815848564e-06, "loss": 0.6209, "mean_token_accuracy": 0.7969570159912109, "num_tokens": 119437206.0, "step": 4323 }, { "epoch": 0.5715040972772931, "grad_norm": 0.8058448505120991, "learning_rate": 5.1649962944763146e-06, "loss": 0.563, "mean_token_accuracy": 0.8136979341506958, "num_tokens": 119502742.0, "step": 4324 }, { "epoch": 0.5716362675125561, "grad_norm": 0.9828187138750021, "learning_rate": 5.1629258444193595e-06, "loss": 0.6114, "mean_token_accuracy": 0.798071026802063, "num_tokens": 119568278.0, "step": 4325 }, { "epoch": 0.5717684377478192, "grad_norm": 0.8609176069031014, "learning_rate": 5.160855466118456e-06, "loss": 0.6088, "mean_token_accuracy": 0.7995055317878723, "num_tokens": 119633814.0, "step": 4326 }, { "epoch": 0.5719006079830822, "grad_norm": 0.8605189866819907, "learning_rate": 5.158785160014342e-06, "loss": 0.6236, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 119699350.0, "step": 4327 }, { "epoch": 0.5720327782183452, "grad_norm": 0.8703567816293218, "learning_rate": 5.1567149265477434e-06, "loss": 0.6003, "mean_token_accuracy": 0.8010774254798889, "num_tokens": 119764886.0, "step": 4328 }, { "epoch": 0.5721649484536082, "grad_norm": 0.9014597302458315, "learning_rate": 5.15464476615937e-06, "loss": 0.5788, "mean_token_accuracy": 0.807456374168396, "num_tokens": 119830422.0, "step": 4329 }, { "epoch": 0.5722971186888712, "grad_norm": 0.9047892838755817, "learning_rate": 5.152574679289917e-06, "loss": 0.6578, "mean_token_accuracy": 0.7817879319190979, "num_tokens": 119895958.0, "step": 4330 }, { "epoch": 0.5724292889241342, "grad_norm": 0.8819861228326702, "learning_rate": 5.1505046663800584e-06, "loss": 0.5959, "mean_token_accuracy": 0.803458034992218, "num_tokens": 119961494.0, "step": 4331 }, { "epoch": 0.5725614591593973, "grad_norm": 0.7800171660408597, "learning_rate": 5.148434727870459e-06, "loss": 0.5387, "mean_token_accuracy": 0.8207069635391235, "num_tokens": 120026340.0, "step": 4332 }, { "epoch": 0.5726936293946603, "grad_norm": 0.9612835892325846, "learning_rate": 5.146364864201768e-06, "loss": 0.621, "mean_token_accuracy": 0.7960413694381714, "num_tokens": 120091876.0, "step": 4333 }, { "epoch": 0.5728257996299233, "grad_norm": 0.795103373144833, "learning_rate": 5.1442950758146145e-06, "loss": 0.5785, "mean_token_accuracy": 0.8103253841400146, "num_tokens": 120157412.0, "step": 4334 }, { "epoch": 0.5729579698651863, "grad_norm": 0.9019016789695327, "learning_rate": 5.14222536314961e-06, "loss": 0.6247, "mean_token_accuracy": 0.7917989492416382, "num_tokens": 120222948.0, "step": 4335 }, { "epoch": 0.5730901401004493, "grad_norm": 0.8434251336497461, "learning_rate": 5.140155726647358e-06, "loss": 0.5796, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 120288484.0, "step": 4336 }, { "epoch": 0.5732223103357124, "grad_norm": 0.8443032914333825, "learning_rate": 5.138086166748439e-06, "loss": 0.5886, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 120354020.0, "step": 4337 }, { "epoch": 0.5733544805709754, "grad_norm": 0.9468207100216928, "learning_rate": 5.136016683893417e-06, "loss": 0.6205, "mean_token_accuracy": 0.7935081124305725, "num_tokens": 120419556.0, "step": 4338 }, { "epoch": 0.5734866508062384, "grad_norm": 0.8298144537236432, "learning_rate": 5.133947278522846e-06, "loss": 0.582, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 120485092.0, "step": 4339 }, { "epoch": 0.5736188210415014, "grad_norm": 0.8204439688890628, "learning_rate": 5.1318779510772554e-06, "loss": 0.5202, "mean_token_accuracy": 0.826303243637085, "num_tokens": 120550628.0, "step": 4340 }, { "epoch": 0.5737509912767644, "grad_norm": 0.8585793362330352, "learning_rate": 5.129808701997163e-06, "loss": 0.5987, "mean_token_accuracy": 0.8005585670471191, "num_tokens": 120616164.0, "step": 4341 }, { "epoch": 0.5738831615120275, "grad_norm": 0.9776289725477295, "learning_rate": 5.127739531723069e-06, "loss": 0.6743, "mean_token_accuracy": 0.7788426280021667, "num_tokens": 120681700.0, "step": 4342 }, { "epoch": 0.5740153317472905, "grad_norm": 0.8394043402800573, "learning_rate": 5.125670440695458e-06, "loss": 0.6047, "mean_token_accuracy": 0.7997192144393921, "num_tokens": 120747236.0, "step": 4343 }, { "epoch": 0.5741475019825535, "grad_norm": 0.8451231027148212, "learning_rate": 5.123601429354792e-06, "loss": 0.6229, "mean_token_accuracy": 0.7965297102928162, "num_tokens": 120812772.0, "step": 4344 }, { "epoch": 0.5742796722178165, "grad_norm": 0.8513598483187059, "learning_rate": 5.121532498141525e-06, "loss": 0.5891, "mean_token_accuracy": 0.8065102100372314, "num_tokens": 120878308.0, "step": 4345 }, { "epoch": 0.5744118424530795, "grad_norm": 0.7968332721041819, "learning_rate": 5.119463647496087e-06, "loss": 0.5553, "mean_token_accuracy": 0.8162007331848145, "num_tokens": 120943844.0, "step": 4346 }, { "epoch": 0.5745440126883425, "grad_norm": 0.7372086879674793, "learning_rate": 5.117394877858893e-06, "loss": 0.5619, "mean_token_accuracy": 0.8142626285552979, "num_tokens": 121009380.0, "step": 4347 }, { "epoch": 0.5746761829236056, "grad_norm": 0.8305892832514947, "learning_rate": 5.115326189670341e-06, "loss": 0.6302, "mean_token_accuracy": 0.7915089726448059, "num_tokens": 121074916.0, "step": 4348 }, { "epoch": 0.5748083531588686, "grad_norm": 0.8743278178282501, "learning_rate": 5.113257583370813e-06, "loss": 0.5347, "mean_token_accuracy": 0.8227017521858215, "num_tokens": 121140452.0, "step": 4349 }, { "epoch": 0.5749405233941316, "grad_norm": 0.912120192024468, "learning_rate": 5.111189059400671e-06, "loss": 0.6305, "mean_token_accuracy": 0.7913258671760559, "num_tokens": 121205988.0, "step": 4350 }, { "epoch": 0.5750726936293946, "grad_norm": 0.840733898410121, "learning_rate": 5.109120618200257e-06, "loss": 0.5808, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 121271524.0, "step": 4351 }, { "epoch": 0.5752048638646576, "grad_norm": 0.9085108008262351, "learning_rate": 5.107052260209906e-06, "loss": 0.6384, "mean_token_accuracy": 0.7884873747825623, "num_tokens": 121337060.0, "step": 4352 }, { "epoch": 0.5753370340999207, "grad_norm": 0.8212367338877069, "learning_rate": 5.104983985869925e-06, "loss": 0.5648, "mean_token_accuracy": 0.811912477016449, "num_tokens": 121402596.0, "step": 4353 }, { "epoch": 0.5754692043351837, "grad_norm": 0.877417285119545, "learning_rate": 5.102915795620603e-06, "loss": 0.5754, "mean_token_accuracy": 0.8095775842666626, "num_tokens": 121468132.0, "step": 4354 }, { "epoch": 0.5756013745704467, "grad_norm": 0.8256141976566093, "learning_rate": 5.10084768990222e-06, "loss": 0.5546, "mean_token_accuracy": 0.8186882138252258, "num_tokens": 121533668.0, "step": 4355 }, { "epoch": 0.5757335448057097, "grad_norm": 0.8163737587170024, "learning_rate": 5.09877966915503e-06, "loss": 0.5573, "mean_token_accuracy": 0.8136521577835083, "num_tokens": 121599204.0, "step": 4356 }, { "epoch": 0.5758657150409727, "grad_norm": 0.8998025978659182, "learning_rate": 5.0967117338192705e-06, "loss": 0.5861, "mean_token_accuracy": 0.8047552108764648, "num_tokens": 121664740.0, "step": 4357 }, { "epoch": 0.5759978852762357, "grad_norm": 0.8317027485360707, "learning_rate": 5.094643884335164e-06, "loss": 0.6154, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 121730276.0, "step": 4358 }, { "epoch": 0.5761300555114988, "grad_norm": 0.8732682910586184, "learning_rate": 5.09257612114291e-06, "loss": 0.563, "mean_token_accuracy": 0.8148425221443176, "num_tokens": 121795812.0, "step": 4359 }, { "epoch": 0.5762622257467618, "grad_norm": 0.8266342771506555, "learning_rate": 5.0905084446826945e-06, "loss": 0.5983, "mean_token_accuracy": 0.803244411945343, "num_tokens": 121861348.0, "step": 4360 }, { "epoch": 0.5763943959820248, "grad_norm": 0.8017811314342664, "learning_rate": 5.088440855394682e-06, "loss": 0.5621, "mean_token_accuracy": 0.8139116168022156, "num_tokens": 121926884.0, "step": 4361 }, { "epoch": 0.5765265662172878, "grad_norm": 0.818062266310681, "learning_rate": 5.086373353719018e-06, "loss": 0.5395, "mean_token_accuracy": 0.822869598865509, "num_tokens": 121992420.0, "step": 4362 }, { "epoch": 0.5766587364525508, "grad_norm": 0.7938179847712641, "learning_rate": 5.084305940095831e-06, "loss": 0.5257, "mean_token_accuracy": 0.826181173324585, "num_tokens": 122057956.0, "step": 4363 }, { "epoch": 0.5767909066878139, "grad_norm": 0.9160727508893893, "learning_rate": 5.0822386149652295e-06, "loss": 0.5949, "mean_token_accuracy": 0.8032596707344055, "num_tokens": 122123492.0, "step": 4364 }, { "epoch": 0.5769230769230769, "grad_norm": 0.8142566009316372, "learning_rate": 5.080171378767305e-06, "loss": 0.573, "mean_token_accuracy": 0.8112409710884094, "num_tokens": 122189028.0, "step": 4365 }, { "epoch": 0.5770552471583399, "grad_norm": 0.8678104980670602, "learning_rate": 5.078104231942129e-06, "loss": 0.5732, "mean_token_accuracy": 0.8086771965026855, "num_tokens": 122254564.0, "step": 4366 }, { "epoch": 0.5771874173936029, "grad_norm": 0.839883040677487, "learning_rate": 5.076037174929752e-06, "loss": 0.5959, "mean_token_accuracy": 0.8011842370033264, "num_tokens": 122320100.0, "step": 4367 }, { "epoch": 0.5773195876288659, "grad_norm": 0.8552354769278905, "learning_rate": 5.073970208170212e-06, "loss": 0.5995, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 122385636.0, "step": 4368 }, { "epoch": 0.577451757864129, "grad_norm": 0.9361030271523529, "learning_rate": 5.071903332103518e-06, "loss": 0.6186, "mean_token_accuracy": 0.7939354181289673, "num_tokens": 122451172.0, "step": 4369 }, { "epoch": 0.577583928099392, "grad_norm": 0.8962704026086628, "learning_rate": 5.069836547169664e-06, "loss": 0.5993, "mean_token_accuracy": 0.8052282929420471, "num_tokens": 122516708.0, "step": 4370 }, { "epoch": 0.577716098334655, "grad_norm": 0.8301412851674099, "learning_rate": 5.067769853808632e-06, "loss": 0.5324, "mean_token_accuracy": 0.8238005042076111, "num_tokens": 122582244.0, "step": 4371 }, { "epoch": 0.577848268569918, "grad_norm": 0.8426557428295846, "learning_rate": 5.0657032524603726e-06, "loss": 0.6151, "mean_token_accuracy": 0.7974606156349182, "num_tokens": 122647780.0, "step": 4372 }, { "epoch": 0.577980438805181, "grad_norm": 0.7937128030242371, "learning_rate": 5.063636743564821e-06, "loss": 0.5218, "mean_token_accuracy": 0.827356219291687, "num_tokens": 122713316.0, "step": 4373 }, { "epoch": 0.578112609040444, "grad_norm": 0.8766356277935429, "learning_rate": 5.061570327561899e-06, "loss": 0.5522, "mean_token_accuracy": 0.8153156042098999, "num_tokens": 122778852.0, "step": 4374 }, { "epoch": 0.5782447792757072, "grad_norm": 0.8999188070298016, "learning_rate": 5.0595040048915e-06, "loss": 0.615, "mean_token_accuracy": 0.7972317337989807, "num_tokens": 122844388.0, "step": 4375 }, { "epoch": 0.5783769495109702, "grad_norm": 0.8443443689589931, "learning_rate": 5.057437775993502e-06, "loss": 0.5693, "mean_token_accuracy": 0.8124923706054688, "num_tokens": 122909924.0, "step": 4376 }, { "epoch": 0.5785091197462332, "grad_norm": 0.8969730715386295, "learning_rate": 5.055371641307759e-06, "loss": 0.6033, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 122975460.0, "step": 4377 }, { "epoch": 0.5786412899814962, "grad_norm": 0.8890776191587039, "learning_rate": 5.053305601274113e-06, "loss": 0.6188, "mean_token_accuracy": 0.7946984767913818, "num_tokens": 123040996.0, "step": 4378 }, { "epoch": 0.5787734602167592, "grad_norm": 0.8935253862545494, "learning_rate": 5.051239656332377e-06, "loss": 0.5865, "mean_token_accuracy": 0.8035343885421753, "num_tokens": 123106532.0, "step": 4379 }, { "epoch": 0.5789056304520223, "grad_norm": 0.9034819278226871, "learning_rate": 5.0491738069223485e-06, "loss": 0.628, "mean_token_accuracy": 0.7960566282272339, "num_tokens": 123172068.0, "step": 4380 }, { "epoch": 0.5790378006872853, "grad_norm": 0.904216088875803, "learning_rate": 5.047108053483805e-06, "loss": 0.6536, "mean_token_accuracy": 0.7839397192001343, "num_tokens": 123237604.0, "step": 4381 }, { "epoch": 0.5791699709225483, "grad_norm": 0.8861313046957753, "learning_rate": 5.0450423964565e-06, "loss": 0.6139, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 123303140.0, "step": 4382 }, { "epoch": 0.5793021411578113, "grad_norm": 0.9473207456522998, "learning_rate": 5.042976836280169e-06, "loss": 0.5867, "mean_token_accuracy": 0.8048620223999023, "num_tokens": 123368676.0, "step": 4383 }, { "epoch": 0.5794343113930743, "grad_norm": 0.8792873502162326, "learning_rate": 5.040911373394529e-06, "loss": 0.564, "mean_token_accuracy": 0.8142015337944031, "num_tokens": 123434212.0, "step": 4384 }, { "epoch": 0.5795664816283373, "grad_norm": 0.847198076737536, "learning_rate": 5.038846008239274e-06, "loss": 0.5955, "mean_token_accuracy": 0.8041753172874451, "num_tokens": 123499748.0, "step": 4385 }, { "epoch": 0.5796986518636004, "grad_norm": 0.8483195526700982, "learning_rate": 5.036780741254073e-06, "loss": 0.6154, "mean_token_accuracy": 0.7960261106491089, "num_tokens": 123565284.0, "step": 4386 }, { "epoch": 0.5798308220988634, "grad_norm": 0.8074082401095215, "learning_rate": 5.034715572878582e-06, "loss": 0.5428, "mean_token_accuracy": 0.8200616240501404, "num_tokens": 123630820.0, "step": 4387 }, { "epoch": 0.5799629923341264, "grad_norm": 0.8195523881264615, "learning_rate": 5.032650503552432e-06, "loss": 0.575, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 123696356.0, "step": 4388 }, { "epoch": 0.5800951625693894, "grad_norm": 0.9002558142189907, "learning_rate": 5.03058553371523e-06, "loss": 0.5999, "mean_token_accuracy": 0.8025424480438232, "num_tokens": 123761892.0, "step": 4389 }, { "epoch": 0.5802273328046524, "grad_norm": 0.9085457043759001, "learning_rate": 5.028520663806569e-06, "loss": 0.641, "mean_token_accuracy": 0.7901812791824341, "num_tokens": 123827428.0, "step": 4390 }, { "epoch": 0.5803595030399155, "grad_norm": 0.7998594643411346, "learning_rate": 5.0264558942660145e-06, "loss": 0.5468, "mean_token_accuracy": 0.8183677196502686, "num_tokens": 123892964.0, "step": 4391 }, { "epoch": 0.5804916732751785, "grad_norm": 0.9007675546407724, "learning_rate": 5.024391225533114e-06, "loss": 0.6391, "mean_token_accuracy": 0.789036750793457, "num_tokens": 123958500.0, "step": 4392 }, { "epoch": 0.5806238435104415, "grad_norm": 0.7943748240226375, "learning_rate": 5.02232665804739e-06, "loss": 0.5498, "mean_token_accuracy": 0.8175283670425415, "num_tokens": 124024036.0, "step": 4393 }, { "epoch": 0.5807560137457045, "grad_norm": 0.8620113160725535, "learning_rate": 5.0202621922483485e-06, "loss": 0.5423, "mean_token_accuracy": 0.8208094239234924, "num_tokens": 124089572.0, "step": 4394 }, { "epoch": 0.5808881839809675, "grad_norm": 0.8119446586227429, "learning_rate": 5.01819782857547e-06, "loss": 0.5627, "mean_token_accuracy": 0.8155597448348999, "num_tokens": 124155108.0, "step": 4395 }, { "epoch": 0.5810203542162306, "grad_norm": 0.8000446271209437, "learning_rate": 5.016133567468213e-06, "loss": 0.547, "mean_token_accuracy": 0.818413496017456, "num_tokens": 124220644.0, "step": 4396 }, { "epoch": 0.5811525244514936, "grad_norm": 0.8979406182490044, "learning_rate": 5.014069409366019e-06, "loss": 0.6142, "mean_token_accuracy": 0.7950952053070068, "num_tokens": 124286180.0, "step": 4397 }, { "epoch": 0.5812846946867566, "grad_norm": 0.879382669959191, "learning_rate": 5.012005354708301e-06, "loss": 0.5966, "mean_token_accuracy": 0.8057318925857544, "num_tokens": 124351716.0, "step": 4398 }, { "epoch": 0.5814168649220196, "grad_norm": 0.9214322922080161, "learning_rate": 5.009941403934455e-06, "loss": 0.6304, "mean_token_accuracy": 0.7919210195541382, "num_tokens": 124417252.0, "step": 4399 }, { "epoch": 0.5815490351572826, "grad_norm": 0.8169980106465302, "learning_rate": 5.007877557483852e-06, "loss": 0.5658, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 124482788.0, "step": 4400 }, { "epoch": 0.5816812053925456, "grad_norm": 0.8228443389374118, "learning_rate": 5.005813815795844e-06, "loss": 0.5035, "mean_token_accuracy": 0.8338878154754639, "num_tokens": 124548324.0, "step": 4401 }, { "epoch": 0.5818133756278087, "grad_norm": 0.8502934945107372, "learning_rate": 5.003750179309754e-06, "loss": 0.5426, "mean_token_accuracy": 0.8208704590797424, "num_tokens": 124613860.0, "step": 4402 }, { "epoch": 0.5819455458630717, "grad_norm": 0.8503902473132481, "learning_rate": 5.001686648464892e-06, "loss": 0.649, "mean_token_accuracy": 0.7851910591125488, "num_tokens": 124679396.0, "step": 4403 }, { "epoch": 0.5820777160983347, "grad_norm": 0.9116806206848667, "learning_rate": 4.999623223700538e-06, "loss": 0.571, "mean_token_accuracy": 0.8099438548088074, "num_tokens": 124744932.0, "step": 4404 }, { "epoch": 0.5822098863335977, "grad_norm": 0.9389505855762869, "learning_rate": 4.997559905455951e-06, "loss": 0.5937, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 124810468.0, "step": 4405 }, { "epoch": 0.5823420565688607, "grad_norm": 0.8654526549968466, "learning_rate": 4.9954966941703704e-06, "loss": 0.5574, "mean_token_accuracy": 0.812034547328949, "num_tokens": 124876004.0, "step": 4406 }, { "epoch": 0.5824742268041238, "grad_norm": 0.8601281941698847, "learning_rate": 4.993433590283012e-06, "loss": 0.5603, "mean_token_accuracy": 0.8157429099082947, "num_tokens": 124941540.0, "step": 4407 }, { "epoch": 0.5826063970393868, "grad_norm": 0.8511276474738744, "learning_rate": 4.991370594233067e-06, "loss": 0.5984, "mean_token_accuracy": 0.8004516959190369, "num_tokens": 125007076.0, "step": 4408 }, { "epoch": 0.5827385672746498, "grad_norm": 0.9671308265888195, "learning_rate": 4.9893077064597e-06, "loss": 0.6192, "mean_token_accuracy": 0.797796368598938, "num_tokens": 125072612.0, "step": 4409 }, { "epoch": 0.5828707375099128, "grad_norm": 0.8479972370870043, "learning_rate": 4.987244927402061e-06, "loss": 0.5324, "mean_token_accuracy": 0.8224422931671143, "num_tokens": 125138148.0, "step": 4410 }, { "epoch": 0.5830029077451758, "grad_norm": 0.8887835935396863, "learning_rate": 4.985182257499271e-06, "loss": 0.601, "mean_token_accuracy": 0.8013978600502014, "num_tokens": 125203684.0, "step": 4411 }, { "epoch": 0.5831350779804388, "grad_norm": 0.9239821408052522, "learning_rate": 4.9831196971904295e-06, "loss": 0.6146, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 125269220.0, "step": 4412 }, { "epoch": 0.5832672482157019, "grad_norm": 0.9041774146562265, "learning_rate": 4.981057246914613e-06, "loss": 0.6309, "mean_token_accuracy": 0.7920430898666382, "num_tokens": 125334756.0, "step": 4413 }, { "epoch": 0.5833994184509649, "grad_norm": 0.875538745115092, "learning_rate": 4.978994907110875e-06, "loss": 0.5931, "mean_token_accuracy": 0.802374541759491, "num_tokens": 125400292.0, "step": 4414 }, { "epoch": 0.5835315886862279, "grad_norm": 1.0150172769897292, "learning_rate": 4.976932678218242e-06, "loss": 0.625, "mean_token_accuracy": 0.7920278310775757, "num_tokens": 125465828.0, "step": 4415 }, { "epoch": 0.5836637589214909, "grad_norm": 0.8523326250851133, "learning_rate": 4.974870560675722e-06, "loss": 0.5148, "mean_token_accuracy": 0.8318123817443848, "num_tokens": 125531364.0, "step": 4416 }, { "epoch": 0.5837959291567539, "grad_norm": 0.9123477652961812, "learning_rate": 4.972808554922296e-06, "loss": 0.6149, "mean_token_accuracy": 0.7953088879585266, "num_tokens": 125596900.0, "step": 4417 }, { "epoch": 0.583928099392017, "grad_norm": 0.8615621087959499, "learning_rate": 4.97074666139692e-06, "loss": 0.5454, "mean_token_accuracy": 0.8205194473266602, "num_tokens": 125662436.0, "step": 4418 }, { "epoch": 0.58406026962728, "grad_norm": 0.8550984844014771, "learning_rate": 4.9686848805385314e-06, "loss": 0.54, "mean_token_accuracy": 0.8213435411453247, "num_tokens": 125727972.0, "step": 4419 }, { "epoch": 0.584192439862543, "grad_norm": 0.8570254867537065, "learning_rate": 4.966623212786039e-06, "loss": 0.5817, "mean_token_accuracy": 0.8056250810623169, "num_tokens": 125793508.0, "step": 4420 }, { "epoch": 0.584324610097806, "grad_norm": 0.9858975247586712, "learning_rate": 4.964561658578328e-06, "loss": 0.616, "mean_token_accuracy": 0.7988646030426025, "num_tokens": 125859044.0, "step": 4421 }, { "epoch": 0.584456780333069, "grad_norm": 0.8865308756622892, "learning_rate": 4.962500218354259e-06, "loss": 0.5882, "mean_token_accuracy": 0.8049994111061096, "num_tokens": 125924580.0, "step": 4422 }, { "epoch": 0.584588950568332, "grad_norm": 0.8681984959363523, "learning_rate": 4.960438892552674e-06, "loss": 0.6354, "mean_token_accuracy": 0.7886552214622498, "num_tokens": 125990116.0, "step": 4423 }, { "epoch": 0.5847211208035951, "grad_norm": 0.9300233086545229, "learning_rate": 4.958377681612382e-06, "loss": 0.5891, "mean_token_accuracy": 0.8053045868873596, "num_tokens": 126055652.0, "step": 4424 }, { "epoch": 0.5848532910388581, "grad_norm": 0.8311756579891798, "learning_rate": 4.956316585972172e-06, "loss": 0.6019, "mean_token_accuracy": 0.8008637428283691, "num_tokens": 126121188.0, "step": 4425 }, { "epoch": 0.5849854612741211, "grad_norm": 0.8810265152249019, "learning_rate": 4.954255606070812e-06, "loss": 0.65, "mean_token_accuracy": 0.7829477190971375, "num_tokens": 126186724.0, "step": 4426 }, { "epoch": 0.5851176315093841, "grad_norm": 0.8604635235469236, "learning_rate": 4.952194742347038e-06, "loss": 0.6118, "mean_token_accuracy": 0.79903244972229, "num_tokens": 126252260.0, "step": 4427 }, { "epoch": 0.5852498017446471, "grad_norm": 0.9114955980569134, "learning_rate": 4.9501339952395655e-06, "loss": 0.6153, "mean_token_accuracy": 0.7954156994819641, "num_tokens": 126317796.0, "step": 4428 }, { "epoch": 0.5853819719799102, "grad_norm": 0.8465162597778904, "learning_rate": 4.948073365187086e-06, "loss": 0.6218, "mean_token_accuracy": 0.7962397933006287, "num_tokens": 126383332.0, "step": 4429 }, { "epoch": 0.5855141422151732, "grad_norm": 0.9527154359980105, "learning_rate": 4.946012852628265e-06, "loss": 0.6495, "mean_token_accuracy": 0.7840617895126343, "num_tokens": 126448868.0, "step": 4430 }, { "epoch": 0.5856463124504362, "grad_norm": 0.9321272376668662, "learning_rate": 4.943952458001737e-06, "loss": 0.641, "mean_token_accuracy": 0.7880295515060425, "num_tokens": 126514404.0, "step": 4431 }, { "epoch": 0.5857784826856992, "grad_norm": 0.9169295173337528, "learning_rate": 4.9418921817461255e-06, "loss": 0.6131, "mean_token_accuracy": 0.7959803342819214, "num_tokens": 126579940.0, "step": 4432 }, { "epoch": 0.5859106529209622, "grad_norm": 0.8724029816281395, "learning_rate": 4.939832024300014e-06, "loss": 0.6127, "mean_token_accuracy": 0.7977811098098755, "num_tokens": 126645476.0, "step": 4433 }, { "epoch": 0.5860428231562252, "grad_norm": 0.8151037347849721, "learning_rate": 4.93777198610197e-06, "loss": 0.5994, "mean_token_accuracy": 0.8035191297531128, "num_tokens": 126711012.0, "step": 4434 }, { "epoch": 0.5861749933914883, "grad_norm": 0.7792990419013992, "learning_rate": 4.935712067590531e-06, "loss": 0.5699, "mean_token_accuracy": 0.8114851713180542, "num_tokens": 126776548.0, "step": 4435 }, { "epoch": 0.5863071636267513, "grad_norm": 0.937460764313787, "learning_rate": 4.933652269204212e-06, "loss": 0.5743, "mean_token_accuracy": 0.8123550415039062, "num_tokens": 126842084.0, "step": 4436 }, { "epoch": 0.5864393338620143, "grad_norm": 0.859190370253252, "learning_rate": 4.9315925913815e-06, "loss": 0.5873, "mean_token_accuracy": 0.8085856437683105, "num_tokens": 126907620.0, "step": 4437 }, { "epoch": 0.5865715040972773, "grad_norm": 0.9178906315251816, "learning_rate": 4.929533034560857e-06, "loss": 0.6175, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 126973156.0, "step": 4438 }, { "epoch": 0.5867036743325403, "grad_norm": 0.8630873865544043, "learning_rate": 4.927473599180722e-06, "loss": 0.6033, "mean_token_accuracy": 0.8000549674034119, "num_tokens": 127038692.0, "step": 4439 }, { "epoch": 0.5868358445678034, "grad_norm": 0.7812303805831107, "learning_rate": 4.9254142856795054e-06, "loss": 0.5302, "mean_token_accuracy": 0.8243498802185059, "num_tokens": 127104228.0, "step": 4440 }, { "epoch": 0.5869680148030664, "grad_norm": 0.8571671758541921, "learning_rate": 4.923355094495586e-06, "loss": 0.5537, "mean_token_accuracy": 0.8151629567146301, "num_tokens": 127169764.0, "step": 4441 }, { "epoch": 0.5871001850383294, "grad_norm": 0.8228773653009314, "learning_rate": 4.921296026067331e-06, "loss": 0.6045, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 127235300.0, "step": 4442 }, { "epoch": 0.5872323552735924, "grad_norm": 0.8792731968004327, "learning_rate": 4.9192370808330706e-06, "loss": 0.5852, "mean_token_accuracy": 0.8074105978012085, "num_tokens": 127300836.0, "step": 4443 }, { "epoch": 0.5873645255088554, "grad_norm": 0.8461541041747538, "learning_rate": 4.917178259231107e-06, "loss": 0.6358, "mean_token_accuracy": 0.7916920781135559, "num_tokens": 127366372.0, "step": 4444 }, { "epoch": 0.5874966957441184, "grad_norm": 0.7849391728605442, "learning_rate": 4.915119561699726e-06, "loss": 0.5441, "mean_token_accuracy": 0.8204736709594727, "num_tokens": 127431908.0, "step": 4445 }, { "epoch": 0.5876288659793815, "grad_norm": 0.9269552026661227, "learning_rate": 4.913060988677178e-06, "loss": 0.6019, "mean_token_accuracy": 0.8036717176437378, "num_tokens": 127497444.0, "step": 4446 }, { "epoch": 0.5877610362146445, "grad_norm": 0.8053299980984251, "learning_rate": 4.91100254060169e-06, "loss": 0.571, "mean_token_accuracy": 0.8100354075431824, "num_tokens": 127562980.0, "step": 4447 }, { "epoch": 0.5878932064499075, "grad_norm": 0.8740105274224035, "learning_rate": 4.908944217911465e-06, "loss": 0.5994, "mean_token_accuracy": 0.803030788898468, "num_tokens": 127628516.0, "step": 4448 }, { "epoch": 0.5880253766851705, "grad_norm": 0.8397193103590452, "learning_rate": 4.906886021044676e-06, "loss": 0.5917, "mean_token_accuracy": 0.8047094345092773, "num_tokens": 127694052.0, "step": 4449 }, { "epoch": 0.5881575469204335, "grad_norm": 0.854727966066262, "learning_rate": 4.9048279504394695e-06, "loss": 0.5786, "mean_token_accuracy": 0.8100964426994324, "num_tokens": 127759588.0, "step": 4450 }, { "epoch": 0.5882897171556966, "grad_norm": 0.8139681176784703, "learning_rate": 4.902770006533966e-06, "loss": 0.5714, "mean_token_accuracy": 0.8076852560043335, "num_tokens": 127825124.0, "step": 4451 }, { "epoch": 0.5884218873909596, "grad_norm": 0.8887409265240834, "learning_rate": 4.9007121897662595e-06, "loss": 0.6111, "mean_token_accuracy": 0.7961176633834839, "num_tokens": 127890660.0, "step": 4452 }, { "epoch": 0.5885540576262226, "grad_norm": 0.7850243658866305, "learning_rate": 4.8986545005744145e-06, "loss": 0.5849, "mean_token_accuracy": 0.8081888556480408, "num_tokens": 127956196.0, "step": 4453 }, { "epoch": 0.5886862278614856, "grad_norm": 0.8293998855702974, "learning_rate": 4.896596939396472e-06, "loss": 0.5548, "mean_token_accuracy": 0.8166738152503967, "num_tokens": 128021732.0, "step": 4454 }, { "epoch": 0.5888183980967486, "grad_norm": 0.8828610110731282, "learning_rate": 4.894539506670441e-06, "loss": 0.6379, "mean_token_accuracy": 0.7913106083869934, "num_tokens": 128087268.0, "step": 4455 }, { "epoch": 0.5889505683320116, "grad_norm": 0.7942464341359241, "learning_rate": 4.892482202834311e-06, "loss": 0.5395, "mean_token_accuracy": 0.821633517742157, "num_tokens": 128152804.0, "step": 4456 }, { "epoch": 0.5890827385672747, "grad_norm": 0.824624390971149, "learning_rate": 4.8904250283260335e-06, "loss": 0.594, "mean_token_accuracy": 0.8049077987670898, "num_tokens": 128218340.0, "step": 4457 }, { "epoch": 0.5892149088025377, "grad_norm": 0.8334525624528543, "learning_rate": 4.888367983583541e-06, "loss": 0.5996, "mean_token_accuracy": 0.8025577068328857, "num_tokens": 128283876.0, "step": 4458 }, { "epoch": 0.5893470790378007, "grad_norm": 0.9296644481310194, "learning_rate": 4.886311069044737e-06, "loss": 0.625, "mean_token_accuracy": 0.7966060042381287, "num_tokens": 128349412.0, "step": 4459 }, { "epoch": 0.5894792492730637, "grad_norm": 0.9439134628712348, "learning_rate": 4.8842542851474905e-06, "loss": 0.6381, "mean_token_accuracy": 0.7858167290687561, "num_tokens": 128414948.0, "step": 4460 }, { "epoch": 0.5896114195083267, "grad_norm": 0.9016049466789001, "learning_rate": 4.882197632329653e-06, "loss": 0.5711, "mean_token_accuracy": 0.8107984662055969, "num_tokens": 128480484.0, "step": 4461 }, { "epoch": 0.5897435897435898, "grad_norm": 0.835870600810901, "learning_rate": 4.8801411110290395e-06, "loss": 0.5465, "mean_token_accuracy": 0.816429615020752, "num_tokens": 128546020.0, "step": 4462 }, { "epoch": 0.5898757599788528, "grad_norm": 0.8662059403063783, "learning_rate": 4.87808472168344e-06, "loss": 0.5797, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 128611556.0, "step": 4463 }, { "epoch": 0.5900079302141158, "grad_norm": 0.8953286426102867, "learning_rate": 4.876028464730622e-06, "loss": 0.6712, "mean_token_accuracy": 0.7785679697990417, "num_tokens": 128677092.0, "step": 4464 }, { "epoch": 0.5901401004493788, "grad_norm": 0.8430833966832252, "learning_rate": 4.873972340608315e-06, "loss": 0.6266, "mean_token_accuracy": 0.7916157841682434, "num_tokens": 128742628.0, "step": 4465 }, { "epoch": 0.5902722706846418, "grad_norm": 0.9271764252424297, "learning_rate": 4.871916349754227e-06, "loss": 0.5813, "mean_token_accuracy": 0.807761549949646, "num_tokens": 128808164.0, "step": 4466 }, { "epoch": 0.5904044409199048, "grad_norm": 0.8922507586970181, "learning_rate": 4.869860492606033e-06, "loss": 0.6, "mean_token_accuracy": 0.802252471446991, "num_tokens": 128873700.0, "step": 4467 }, { "epoch": 0.5905366111551679, "grad_norm": 0.848202685643496, "learning_rate": 4.867804769601384e-06, "loss": 0.6152, "mean_token_accuracy": 0.7948663234710693, "num_tokens": 128939236.0, "step": 4468 }, { "epoch": 0.5906687813904309, "grad_norm": 0.8288460304380623, "learning_rate": 4.865749181177901e-06, "loss": 0.5759, "mean_token_accuracy": 0.8098064661026001, "num_tokens": 129004772.0, "step": 4469 }, { "epoch": 0.5908009516256939, "grad_norm": 0.7964858483676109, "learning_rate": 4.8636937277731746e-06, "loss": 0.5768, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 129070308.0, "step": 4470 }, { "epoch": 0.5909331218609569, "grad_norm": 0.8582486535651981, "learning_rate": 4.861638409824768e-06, "loss": 0.5796, "mean_token_accuracy": 0.8090282082557678, "num_tokens": 129135844.0, "step": 4471 }, { "epoch": 0.5910652920962199, "grad_norm": 0.926237039908468, "learning_rate": 4.859583227770218e-06, "loss": 0.627, "mean_token_accuracy": 0.793218195438385, "num_tokens": 129201380.0, "step": 4472 }, { "epoch": 0.591197462331483, "grad_norm": 0.8191848357629229, "learning_rate": 4.857528182047027e-06, "loss": 0.5844, "mean_token_accuracy": 0.8063575625419617, "num_tokens": 129266916.0, "step": 4473 }, { "epoch": 0.591329632566746, "grad_norm": 0.8639000593655458, "learning_rate": 4.85547327309267e-06, "loss": 0.5973, "mean_token_accuracy": 0.8035191297531128, "num_tokens": 129332452.0, "step": 4474 }, { "epoch": 0.591461802802009, "grad_norm": 0.8478370718187581, "learning_rate": 4.8534185013446e-06, "loss": 0.6521, "mean_token_accuracy": 0.7851300239562988, "num_tokens": 129397988.0, "step": 4475 }, { "epoch": 0.591593973037272, "grad_norm": 0.7850952781710967, "learning_rate": 4.851363867240229e-06, "loss": 0.5667, "mean_token_accuracy": 0.811912477016449, "num_tokens": 129463524.0, "step": 4476 }, { "epoch": 0.591726143272535, "grad_norm": 0.7846688264343331, "learning_rate": 4.849309371216949e-06, "loss": 0.5818, "mean_token_accuracy": 0.806555986404419, "num_tokens": 129529060.0, "step": 4477 }, { "epoch": 0.591858313507798, "grad_norm": 0.8994855902896685, "learning_rate": 4.847255013712119e-06, "loss": 0.5944, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 129594596.0, "step": 4478 }, { "epoch": 0.5919904837430611, "grad_norm": 0.7923512420143727, "learning_rate": 4.845200795163065e-06, "loss": 0.6048, "mean_token_accuracy": 0.8015352487564087, "num_tokens": 129660132.0, "step": 4479 }, { "epoch": 0.5921226539783241, "grad_norm": 0.8484980667085376, "learning_rate": 4.843146716007093e-06, "loss": 0.6042, "mean_token_accuracy": 0.8003143668174744, "num_tokens": 129725668.0, "step": 4480 }, { "epoch": 0.5922548242135871, "grad_norm": 0.8410666570146608, "learning_rate": 4.841092776681472e-06, "loss": 0.5917, "mean_token_accuracy": 0.8055182695388794, "num_tokens": 129791204.0, "step": 4481 }, { "epoch": 0.5923869944488501, "grad_norm": 0.8464582105459096, "learning_rate": 4.839038977623442e-06, "loss": 0.5768, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 129856740.0, "step": 4482 }, { "epoch": 0.5925191646841131, "grad_norm": 0.8390631217011097, "learning_rate": 4.83698531927021e-06, "loss": 0.5734, "mean_token_accuracy": 0.8117445707321167, "num_tokens": 129922276.0, "step": 4483 }, { "epoch": 0.5926513349193762, "grad_norm": 0.825360272143716, "learning_rate": 4.8349318020589645e-06, "loss": 0.6002, "mean_token_accuracy": 0.8009858131408691, "num_tokens": 129987812.0, "step": 4484 }, { "epoch": 0.5927835051546392, "grad_norm": 0.798539320057762, "learning_rate": 4.8328784264268515e-06, "loss": 0.5581, "mean_token_accuracy": 0.8154834508895874, "num_tokens": 130053348.0, "step": 4485 }, { "epoch": 0.5929156753899022, "grad_norm": 0.8192120794439257, "learning_rate": 4.830825192810993e-06, "loss": 0.5764, "mean_token_accuracy": 0.808753490447998, "num_tokens": 130118884.0, "step": 4486 }, { "epoch": 0.5930478456251652, "grad_norm": 0.873120933206947, "learning_rate": 4.828772101648479e-06, "loss": 0.5982, "mean_token_accuracy": 0.803396999835968, "num_tokens": 130184420.0, "step": 4487 }, { "epoch": 0.5931800158604282, "grad_norm": 0.7916716473642962, "learning_rate": 4.8267191533763725e-06, "loss": 0.5362, "mean_token_accuracy": 0.8227933049201965, "num_tokens": 130249956.0, "step": 4488 }, { "epoch": 0.5933121860956913, "grad_norm": 0.8003649585399071, "learning_rate": 4.824666348431698e-06, "loss": 0.5394, "mean_token_accuracy": 0.8212825059890747, "num_tokens": 130315492.0, "step": 4489 }, { "epoch": 0.5934443563309543, "grad_norm": 0.791105470665806, "learning_rate": 4.822613687251459e-06, "loss": 0.5266, "mean_token_accuracy": 0.8252044916152954, "num_tokens": 130381028.0, "step": 4490 }, { "epoch": 0.5935765265662173, "grad_norm": 0.8585625287204209, "learning_rate": 4.820561170272625e-06, "loss": 0.5461, "mean_token_accuracy": 0.8196954131126404, "num_tokens": 130446564.0, "step": 4491 }, { "epoch": 0.5937086968014803, "grad_norm": 0.88022632950081, "learning_rate": 4.81850879793213e-06, "loss": 0.6084, "mean_token_accuracy": 0.7995818853378296, "num_tokens": 130512100.0, "step": 4492 }, { "epoch": 0.5938408670367433, "grad_norm": 0.8463457400749788, "learning_rate": 4.8164565706668845e-06, "loss": 0.5673, "mean_token_accuracy": 0.8116225004196167, "num_tokens": 130577636.0, "step": 4493 }, { "epoch": 0.5939730372720063, "grad_norm": 0.8092464889382082, "learning_rate": 4.814404488913764e-06, "loss": 0.6002, "mean_token_accuracy": 0.7984983325004578, "num_tokens": 130643172.0, "step": 4494 }, { "epoch": 0.5941052075072694, "grad_norm": 0.8143937626019457, "learning_rate": 4.812352553109613e-06, "loss": 0.5582, "mean_token_accuracy": 0.8178488612174988, "num_tokens": 130708708.0, "step": 4495 }, { "epoch": 0.5942373777425324, "grad_norm": 0.8702089486570005, "learning_rate": 4.8103007636912465e-06, "loss": 0.5843, "mean_token_accuracy": 0.8051825165748596, "num_tokens": 130774244.0, "step": 4496 }, { "epoch": 0.5943695479777954, "grad_norm": 0.8313638318122062, "learning_rate": 4.8082491210954485e-06, "loss": 0.5385, "mean_token_accuracy": 0.8195427656173706, "num_tokens": 130839780.0, "step": 4497 }, { "epoch": 0.5945017182130584, "grad_norm": 0.8567156869402162, "learning_rate": 4.8061976257589695e-06, "loss": 0.574, "mean_token_accuracy": 0.8112714886665344, "num_tokens": 130905316.0, "step": 4498 }, { "epoch": 0.5946338884483214, "grad_norm": 0.8678760745963648, "learning_rate": 4.8041462781185285e-06, "loss": 0.5645, "mean_token_accuracy": 0.8147051930427551, "num_tokens": 130970852.0, "step": 4499 }, { "epoch": 0.5947660586835845, "grad_norm": 0.9151720695828982, "learning_rate": 4.802095078610819e-06, "loss": 0.6348, "mean_token_accuracy": 0.7899981737136841, "num_tokens": 131036388.0, "step": 4500 }, { "epoch": 0.5948982289188475, "grad_norm": 0.7702442691491195, "learning_rate": 4.800044027672495e-06, "loss": 0.5341, "mean_token_accuracy": 0.8240294456481934, "num_tokens": 131101924.0, "step": 4501 }, { "epoch": 0.5950303991541105, "grad_norm": 0.7810189581682243, "learning_rate": 4.797993125740182e-06, "loss": 0.61, "mean_token_accuracy": 0.7980862855911255, "num_tokens": 131167460.0, "step": 4502 }, { "epoch": 0.5951625693893735, "grad_norm": 0.829571129665048, "learning_rate": 4.795942373250478e-06, "loss": 0.5713, "mean_token_accuracy": 0.8104932308197021, "num_tokens": 131232996.0, "step": 4503 }, { "epoch": 0.5952947396246365, "grad_norm": 0.8458403428546558, "learning_rate": 4.793891770639943e-06, "loss": 0.6366, "mean_token_accuracy": 0.7910969257354736, "num_tokens": 131298532.0, "step": 4504 }, { "epoch": 0.5954269098598995, "grad_norm": 0.82996187999421, "learning_rate": 4.7918413183451055e-06, "loss": 0.5754, "mean_token_accuracy": 0.8124313354492188, "num_tokens": 131364068.0, "step": 4505 }, { "epoch": 0.5955590800951626, "grad_norm": 0.8565815120967103, "learning_rate": 4.789791016802468e-06, "loss": 0.617, "mean_token_accuracy": 0.7950341701507568, "num_tokens": 131429604.0, "step": 4506 }, { "epoch": 0.5956912503304256, "grad_norm": 0.8362046657387708, "learning_rate": 4.787740866448495e-06, "loss": 0.5683, "mean_token_accuracy": 0.8109968304634094, "num_tokens": 131495140.0, "step": 4507 }, { "epoch": 0.5958234205656886, "grad_norm": 0.8756060397851207, "learning_rate": 4.785690867719618e-06, "loss": 0.5449, "mean_token_accuracy": 0.8171163201332092, "num_tokens": 131560676.0, "step": 4508 }, { "epoch": 0.5959555908009516, "grad_norm": 0.8968777073357199, "learning_rate": 4.783641021052244e-06, "loss": 0.6377, "mean_token_accuracy": 0.788166880607605, "num_tokens": 131626212.0, "step": 4509 }, { "epoch": 0.5960877610362146, "grad_norm": 0.8364038893709389, "learning_rate": 4.781591326882741e-06, "loss": 0.5877, "mean_token_accuracy": 0.8044041991233826, "num_tokens": 131691748.0, "step": 4510 }, { "epoch": 0.5962199312714777, "grad_norm": 0.8139015033514301, "learning_rate": 4.779541785647442e-06, "loss": 0.5793, "mean_token_accuracy": 0.808387279510498, "num_tokens": 131757284.0, "step": 4511 }, { "epoch": 0.5963521015067407, "grad_norm": 0.7723252768960097, "learning_rate": 4.7774923977826574e-06, "loss": 0.5538, "mean_token_accuracy": 0.816215991973877, "num_tokens": 131822820.0, "step": 4512 }, { "epoch": 0.5964842717420037, "grad_norm": 0.8519859225657341, "learning_rate": 4.775443163724655e-06, "loss": 0.5525, "mean_token_accuracy": 0.8189781308174133, "num_tokens": 131888356.0, "step": 4513 }, { "epoch": 0.5966164419772667, "grad_norm": 0.8119803185957426, "learning_rate": 4.773394083909677e-06, "loss": 0.5318, "mean_token_accuracy": 0.8257080912590027, "num_tokens": 131953892.0, "step": 4514 }, { "epoch": 0.5967486122125297, "grad_norm": 0.8201423980145858, "learning_rate": 4.7713451587739265e-06, "loss": 0.6119, "mean_token_accuracy": 0.7982084155082703, "num_tokens": 132019428.0, "step": 4515 }, { "epoch": 0.5968807824477927, "grad_norm": 0.8724518745111807, "learning_rate": 4.76929638875358e-06, "loss": 0.6366, "mean_token_accuracy": 0.7909138202667236, "num_tokens": 132084964.0, "step": 4516 }, { "epoch": 0.5970129526830558, "grad_norm": 0.8820103033259015, "learning_rate": 4.7672477742847764e-06, "loss": 0.5974, "mean_token_accuracy": 0.8004974722862244, "num_tokens": 132150500.0, "step": 4517 }, { "epoch": 0.5971451229183188, "grad_norm": 0.7867407955028567, "learning_rate": 4.765199315803621e-06, "loss": 0.5748, "mean_token_accuracy": 0.808570384979248, "num_tokens": 132216036.0, "step": 4518 }, { "epoch": 0.5972772931535818, "grad_norm": 0.9118931944167646, "learning_rate": 4.76315101374619e-06, "loss": 0.5331, "mean_token_accuracy": 0.8223354816436768, "num_tokens": 132281572.0, "step": 4519 }, { "epoch": 0.5974094633888448, "grad_norm": 0.8377600632171042, "learning_rate": 4.761102868548525e-06, "loss": 0.6133, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 132347108.0, "step": 4520 }, { "epoch": 0.5975416336241078, "grad_norm": 0.8456841911146232, "learning_rate": 4.7590548806466306e-06, "loss": 0.6034, "mean_token_accuracy": 0.8018251657485962, "num_tokens": 132412644.0, "step": 4521 }, { "epoch": 0.5976738038593709, "grad_norm": 0.85143532495974, "learning_rate": 4.757007050476484e-06, "loss": 0.5612, "mean_token_accuracy": 0.8133927583694458, "num_tokens": 132478180.0, "step": 4522 }, { "epoch": 0.5978059740946339, "grad_norm": 0.8051559731037008, "learning_rate": 4.754959378474023e-06, "loss": 0.5974, "mean_token_accuracy": 0.8003906607627869, "num_tokens": 132543716.0, "step": 4523 }, { "epoch": 0.5979381443298969, "grad_norm": 0.8285299583194011, "learning_rate": 4.752911865075155e-06, "loss": 0.5804, "mean_token_accuracy": 0.807456374168396, "num_tokens": 132609252.0, "step": 4524 }, { "epoch": 0.5980703145651599, "grad_norm": 0.7602180441506602, "learning_rate": 4.750864510715753e-06, "loss": 0.5365, "mean_token_accuracy": 0.821816623210907, "num_tokens": 132674788.0, "step": 4525 }, { "epoch": 0.5982024848004229, "grad_norm": 0.923861209587193, "learning_rate": 4.748817315831656e-06, "loss": 0.6392, "mean_token_accuracy": 0.7900897264480591, "num_tokens": 132740324.0, "step": 4526 }, { "epoch": 0.598334655035686, "grad_norm": 0.9922030109169714, "learning_rate": 4.74677028085867e-06, "loss": 0.6451, "mean_token_accuracy": 0.7853283882141113, "num_tokens": 132805860.0, "step": 4527 }, { "epoch": 0.598466825270949, "grad_norm": 0.8360116904359337, "learning_rate": 4.744723406232562e-06, "loss": 0.6047, "mean_token_accuracy": 0.802008330821991, "num_tokens": 132871396.0, "step": 4528 }, { "epoch": 0.598598995506212, "grad_norm": 0.8943859953237819, "learning_rate": 4.7426766923890745e-06, "loss": 0.568, "mean_token_accuracy": 0.8138658404350281, "num_tokens": 132936932.0, "step": 4529 }, { "epoch": 0.598731165741475, "grad_norm": 0.9070273816524049, "learning_rate": 4.740630139763906e-06, "loss": 0.6484, "mean_token_accuracy": 0.784672200679779, "num_tokens": 133002468.0, "step": 4530 }, { "epoch": 0.598863335976738, "grad_norm": 0.815988655064198, "learning_rate": 4.738583748792727e-06, "loss": 0.5771, "mean_token_accuracy": 0.8105847835540771, "num_tokens": 133068004.0, "step": 4531 }, { "epoch": 0.598995506212001, "grad_norm": 0.8571689500480466, "learning_rate": 4.73653751991117e-06, "loss": 0.5733, "mean_token_accuracy": 0.8089519143104553, "num_tokens": 133133540.0, "step": 4532 }, { "epoch": 0.5991276764472641, "grad_norm": 0.8504113429189133, "learning_rate": 4.734491453554836e-06, "loss": 0.5696, "mean_token_accuracy": 0.8114241361618042, "num_tokens": 133199076.0, "step": 4533 }, { "epoch": 0.5992598466825271, "grad_norm": 0.8893664515805946, "learning_rate": 4.732445550159287e-06, "loss": 0.5746, "mean_token_accuracy": 0.8093181252479553, "num_tokens": 133264612.0, "step": 4534 }, { "epoch": 0.5993920169177901, "grad_norm": 0.8684926176452453, "learning_rate": 4.730399810160057e-06, "loss": 0.5878, "mean_token_accuracy": 0.8043584227561951, "num_tokens": 133330148.0, "step": 4535 }, { "epoch": 0.5995241871530531, "grad_norm": 0.7677210372462716, "learning_rate": 4.728354233992639e-06, "loss": 0.5487, "mean_token_accuracy": 0.8169942498207092, "num_tokens": 133395684.0, "step": 4536 }, { "epoch": 0.5996563573883161, "grad_norm": 0.8265973086032525, "learning_rate": 4.726308822092491e-06, "loss": 0.5521, "mean_token_accuracy": 0.816185474395752, "num_tokens": 133461220.0, "step": 4537 }, { "epoch": 0.5997885276235791, "grad_norm": 0.8150753266097602, "learning_rate": 4.724263574895045e-06, "loss": 0.5793, "mean_token_accuracy": 0.8092876076698303, "num_tokens": 133526756.0, "step": 4538 }, { "epoch": 0.5999206978588422, "grad_norm": 0.8785937547003272, "learning_rate": 4.722218492835686e-06, "loss": 0.5932, "mean_token_accuracy": 0.8041600584983826, "num_tokens": 133592292.0, "step": 4539 }, { "epoch": 0.6000528680941052, "grad_norm": 0.8074993471085511, "learning_rate": 4.72017357634977e-06, "loss": 0.5606, "mean_token_accuracy": 0.8146593570709229, "num_tokens": 133657828.0, "step": 4540 }, { "epoch": 0.6001850383293682, "grad_norm": 0.8687899267929302, "learning_rate": 4.71812882587262e-06, "loss": 0.5923, "mean_token_accuracy": 0.8026340007781982, "num_tokens": 133723364.0, "step": 4541 }, { "epoch": 0.6003172085646312, "grad_norm": 0.8347001018919793, "learning_rate": 4.7160842418395184e-06, "loss": 0.6441, "mean_token_accuracy": 0.7894793152809143, "num_tokens": 133788900.0, "step": 4542 }, { "epoch": 0.6004493787998942, "grad_norm": 0.9387838828286742, "learning_rate": 4.714039824685715e-06, "loss": 0.6385, "mean_token_accuracy": 0.7874496579170227, "num_tokens": 133854436.0, "step": 4543 }, { "epoch": 0.6005815490351573, "grad_norm": 0.9103282034743151, "learning_rate": 4.711995574846421e-06, "loss": 0.6552, "mean_token_accuracy": 0.7835734486579895, "num_tokens": 133919972.0, "step": 4544 }, { "epoch": 0.6007137192704203, "grad_norm": 0.7932943675933684, "learning_rate": 4.709951492756819e-06, "loss": 0.5501, "mean_token_accuracy": 0.8178030848503113, "num_tokens": 133985508.0, "step": 4545 }, { "epoch": 0.6008458895056833, "grad_norm": 0.8385101434208414, "learning_rate": 4.707907578852049e-06, "loss": 0.5839, "mean_token_accuracy": 0.806830644607544, "num_tokens": 134051044.0, "step": 4546 }, { "epoch": 0.6009780597409463, "grad_norm": 0.8892880421868411, "learning_rate": 4.705863833567216e-06, "loss": 0.6089, "mean_token_accuracy": 0.7992003560066223, "num_tokens": 134116580.0, "step": 4547 }, { "epoch": 0.6011102299762093, "grad_norm": 0.8501433207127261, "learning_rate": 4.703820257337393e-06, "loss": 0.5229, "mean_token_accuracy": 0.8245788216590881, "num_tokens": 134182116.0, "step": 4548 }, { "epoch": 0.6012424002114723, "grad_norm": 0.8505342538163952, "learning_rate": 4.701776850597616e-06, "loss": 0.6333, "mean_token_accuracy": 0.7924246191978455, "num_tokens": 134247652.0, "step": 4549 }, { "epoch": 0.6013745704467354, "grad_norm": 0.8914509791951573, "learning_rate": 4.699733613782881e-06, "loss": 0.6143, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 134313188.0, "step": 4550 }, { "epoch": 0.6015067406819984, "grad_norm": 0.806349417230305, "learning_rate": 4.6976905473281495e-06, "loss": 0.5617, "mean_token_accuracy": 0.8161091208457947, "num_tokens": 134378724.0, "step": 4551 }, { "epoch": 0.6016389109172614, "grad_norm": 0.8401161474967196, "learning_rate": 4.695647651668351e-06, "loss": 0.5828, "mean_token_accuracy": 0.8087992668151855, "num_tokens": 134444260.0, "step": 4552 }, { "epoch": 0.6017710811525244, "grad_norm": 0.7945592493825455, "learning_rate": 4.693604927238373e-06, "loss": 0.5514, "mean_token_accuracy": 0.8150408864021301, "num_tokens": 134509796.0, "step": 4553 }, { "epoch": 0.6019032513877874, "grad_norm": 0.8103441954012212, "learning_rate": 4.691562374473072e-06, "loss": 0.5836, "mean_token_accuracy": 0.8069985508918762, "num_tokens": 134575332.0, "step": 4554 }, { "epoch": 0.6020354216230505, "grad_norm": 0.8783329797685578, "learning_rate": 4.689519993807261e-06, "loss": 0.596, "mean_token_accuracy": 0.802069365978241, "num_tokens": 134640868.0, "step": 4555 }, { "epoch": 0.6021675918583135, "grad_norm": 0.869417208109793, "learning_rate": 4.68747778567572e-06, "loss": 0.6309, "mean_token_accuracy": 0.7950189113616943, "num_tokens": 134706404.0, "step": 4556 }, { "epoch": 0.6022997620935765, "grad_norm": 0.8762606213567717, "learning_rate": 4.6854357505131965e-06, "loss": 0.5401, "mean_token_accuracy": 0.8212214708328247, "num_tokens": 134771940.0, "step": 4557 }, { "epoch": 0.6024319323288395, "grad_norm": 0.8487731159858521, "learning_rate": 4.683393888754395e-06, "loss": 0.5927, "mean_token_accuracy": 0.8019777536392212, "num_tokens": 134837476.0, "step": 4558 }, { "epoch": 0.6025641025641025, "grad_norm": 0.8220427117471633, "learning_rate": 4.681352200833985e-06, "loss": 0.5714, "mean_token_accuracy": 0.8118056654930115, "num_tokens": 134903012.0, "step": 4559 }, { "epoch": 0.6026962727993656, "grad_norm": 0.7744331795642795, "learning_rate": 4.679310687186597e-06, "loss": 0.5337, "mean_token_accuracy": 0.8230527639389038, "num_tokens": 134968548.0, "step": 4560 }, { "epoch": 0.6028284430346286, "grad_norm": 0.9173488162460931, "learning_rate": 4.67726934824683e-06, "loss": 0.5849, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 135034084.0, "step": 4561 }, { "epoch": 0.6029606132698916, "grad_norm": 0.9089956974402278, "learning_rate": 4.675228184449243e-06, "loss": 0.6045, "mean_token_accuracy": 0.8000243902206421, "num_tokens": 135099620.0, "step": 4562 }, { "epoch": 0.6030927835051546, "grad_norm": 0.82085159486282, "learning_rate": 4.673187196228353e-06, "loss": 0.5799, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 135165156.0, "step": 4563 }, { "epoch": 0.6032249537404176, "grad_norm": 0.7637709222076208, "learning_rate": 4.671146384018648e-06, "loss": 0.5188, "mean_token_accuracy": 0.8266389966011047, "num_tokens": 135230692.0, "step": 4564 }, { "epoch": 0.6033571239756806, "grad_norm": 0.909062216995222, "learning_rate": 4.669105748254572e-06, "loss": 0.5508, "mean_token_accuracy": 0.8156207799911499, "num_tokens": 135296228.0, "step": 4565 }, { "epoch": 0.6034892942109437, "grad_norm": 0.9660819844809176, "learning_rate": 4.667065289370534e-06, "loss": 0.6556, "mean_token_accuracy": 0.7849926948547363, "num_tokens": 135361764.0, "step": 4566 }, { "epoch": 0.6036214644462067, "grad_norm": 0.9144616317555212, "learning_rate": 4.665025007800906e-06, "loss": 0.6065, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 135427300.0, "step": 4567 }, { "epoch": 0.6037536346814697, "grad_norm": 0.8310922247515015, "learning_rate": 4.662984903980021e-06, "loss": 0.5922, "mean_token_accuracy": 0.803152859210968, "num_tokens": 135492836.0, "step": 4568 }, { "epoch": 0.6038858049167327, "grad_norm": 0.8287284538883266, "learning_rate": 4.660944978342174e-06, "loss": 0.564, "mean_token_accuracy": 0.8140336871147156, "num_tokens": 135558372.0, "step": 4569 }, { "epoch": 0.6040179751519957, "grad_norm": 0.88872402034902, "learning_rate": 4.658905231321624e-06, "loss": 0.5818, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 135623908.0, "step": 4570 }, { "epoch": 0.6041501453872588, "grad_norm": 0.8705806846507658, "learning_rate": 4.656865663352589e-06, "loss": 0.5923, "mean_token_accuracy": 0.802252471446991, "num_tokens": 135689444.0, "step": 4571 }, { "epoch": 0.6042823156225218, "grad_norm": 0.8211059130156392, "learning_rate": 4.654826274869251e-06, "loss": 0.5883, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 135754980.0, "step": 4572 }, { "epoch": 0.6044144858577848, "grad_norm": 0.871937400530164, "learning_rate": 4.652787066305752e-06, "loss": 0.5942, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 135820516.0, "step": 4573 }, { "epoch": 0.6045466560930478, "grad_norm": 0.8792915848047045, "learning_rate": 4.6507480380961986e-06, "loss": 0.6125, "mean_token_accuracy": 0.7972317337989807, "num_tokens": 135886052.0, "step": 4574 }, { "epoch": 0.6046788263283108, "grad_norm": 0.8168127001616213, "learning_rate": 4.648709190674657e-06, "loss": 0.5875, "mean_token_accuracy": 0.8059150576591492, "num_tokens": 135951588.0, "step": 4575 }, { "epoch": 0.6048109965635738, "grad_norm": 0.9023130854063327, "learning_rate": 4.646670524475154e-06, "loss": 0.6028, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 136017124.0, "step": 4576 }, { "epoch": 0.6049431667988369, "grad_norm": 0.8921496063401233, "learning_rate": 4.644632039931682e-06, "loss": 0.6077, "mean_token_accuracy": 0.7983762621879578, "num_tokens": 136082660.0, "step": 4577 }, { "epoch": 0.6050753370340999, "grad_norm": 0.8790518157610329, "learning_rate": 4.642593737478189e-06, "loss": 0.6157, "mean_token_accuracy": 0.7967281341552734, "num_tokens": 136148196.0, "step": 4578 }, { "epoch": 0.6052075072693629, "grad_norm": 0.8809377215312783, "learning_rate": 4.640555617548586e-06, "loss": 0.6056, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 136213732.0, "step": 4579 }, { "epoch": 0.6053396775046259, "grad_norm": 0.8300780129725532, "learning_rate": 4.638517680576752e-06, "loss": 0.5675, "mean_token_accuracy": 0.8114699125289917, "num_tokens": 136279268.0, "step": 4580 }, { "epoch": 0.6054718477398889, "grad_norm": 0.8557361445200227, "learning_rate": 4.6364799269965154e-06, "loss": 0.5942, "mean_token_accuracy": 0.8048772811889648, "num_tokens": 136344804.0, "step": 4581 }, { "epoch": 0.605604017975152, "grad_norm": 0.7858882266041994, "learning_rate": 4.634442357241672e-06, "loss": 0.584, "mean_token_accuracy": 0.8053045868873596, "num_tokens": 136410340.0, "step": 4582 }, { "epoch": 0.605736188210415, "grad_norm": 0.838592533965494, "learning_rate": 4.6324049717459805e-06, "loss": 0.5809, "mean_token_accuracy": 0.8082498908042908, "num_tokens": 136475876.0, "step": 4583 }, { "epoch": 0.605868358445678, "grad_norm": 0.8729618548025755, "learning_rate": 4.630367770943157e-06, "loss": 0.5686, "mean_token_accuracy": 0.8109968304634094, "num_tokens": 136541412.0, "step": 4584 }, { "epoch": 0.606000528680941, "grad_norm": 0.8898863147501281, "learning_rate": 4.6283307552668764e-06, "loss": 0.5867, "mean_token_accuracy": 0.8058081865310669, "num_tokens": 136606948.0, "step": 4585 }, { "epoch": 0.606132698916204, "grad_norm": 0.8726692801662899, "learning_rate": 4.626293925150781e-06, "loss": 0.5904, "mean_token_accuracy": 0.8058997392654419, "num_tokens": 136672484.0, "step": 4586 }, { "epoch": 0.606264869151467, "grad_norm": 0.8344950501274696, "learning_rate": 4.6242572810284665e-06, "loss": 0.6025, "mean_token_accuracy": 0.8000549674034119, "num_tokens": 136738020.0, "step": 4587 }, { "epoch": 0.6063970393867301, "grad_norm": 0.793429810146025, "learning_rate": 4.622220823333494e-06, "loss": 0.5601, "mean_token_accuracy": 0.8145525455474854, "num_tokens": 136803556.0, "step": 4588 }, { "epoch": 0.6065292096219931, "grad_norm": 0.8508404762562759, "learning_rate": 4.6201845524993786e-06, "loss": 0.6111, "mean_token_accuracy": 0.7984525561332703, "num_tokens": 136869092.0, "step": 4589 }, { "epoch": 0.6066613798572561, "grad_norm": 0.8027812665012314, "learning_rate": 4.618148468959604e-06, "loss": 0.5419, "mean_token_accuracy": 0.8207789063453674, "num_tokens": 136934628.0, "step": 4590 }, { "epoch": 0.6067935500925191, "grad_norm": 0.8420243641083659, "learning_rate": 4.616112573147611e-06, "loss": 0.5639, "mean_token_accuracy": 0.8120803236961365, "num_tokens": 137000164.0, "step": 4591 }, { "epoch": 0.6069257203277821, "grad_norm": 1.0165534364079243, "learning_rate": 4.614076865496794e-06, "loss": 0.6269, "mean_token_accuracy": 0.7917989492416382, "num_tokens": 137065700.0, "step": 4592 }, { "epoch": 0.6070578905630452, "grad_norm": 0.8984148735203783, "learning_rate": 4.612041346440516e-06, "loss": 0.6188, "mean_token_accuracy": 0.7941032648086548, "num_tokens": 137131236.0, "step": 4593 }, { "epoch": 0.6071900607983082, "grad_norm": 0.8088592861182243, "learning_rate": 4.6100060164120965e-06, "loss": 0.5326, "mean_token_accuracy": 0.8232816457748413, "num_tokens": 137196772.0, "step": 4594 }, { "epoch": 0.6073222310335712, "grad_norm": 0.8082116105521264, "learning_rate": 4.607970875844815e-06, "loss": 0.5619, "mean_token_accuracy": 0.8146440982818604, "num_tokens": 137262308.0, "step": 4595 }, { "epoch": 0.6074544012688342, "grad_norm": 0.8982364538044119, "learning_rate": 4.6059359251719096e-06, "loss": 0.6504, "mean_token_accuracy": 0.7842448949813843, "num_tokens": 137327844.0, "step": 4596 }, { "epoch": 0.6075865715040972, "grad_norm": 0.9036706633542552, "learning_rate": 4.603901164826579e-06, "loss": 0.5743, "mean_token_accuracy": 0.8081125617027283, "num_tokens": 137393380.0, "step": 4597 }, { "epoch": 0.6077187417393602, "grad_norm": 0.8346564848420485, "learning_rate": 4.6018665952419795e-06, "loss": 0.6039, "mean_token_accuracy": 0.8003143668174744, "num_tokens": 137458916.0, "step": 4598 }, { "epoch": 0.6078509119746233, "grad_norm": 0.8829353863172753, "learning_rate": 4.599832216851232e-06, "loss": 0.5504, "mean_token_accuracy": 0.8164753913879395, "num_tokens": 137524452.0, "step": 4599 }, { "epoch": 0.6079830822098863, "grad_norm": 0.7889358604667229, "learning_rate": 4.597798030087412e-06, "loss": 0.5664, "mean_token_accuracy": 0.8114393949508667, "num_tokens": 137589988.0, "step": 4600 }, { "epoch": 0.6081152524451493, "grad_norm": 0.7337803611098503, "learning_rate": 4.595764035383552e-06, "loss": 0.5254, "mean_token_accuracy": 0.8264558911323547, "num_tokens": 137655524.0, "step": 4601 }, { "epoch": 0.6082474226804123, "grad_norm": 0.7972141188782973, "learning_rate": 4.593730233172653e-06, "loss": 0.5876, "mean_token_accuracy": 0.806464433670044, "num_tokens": 137721060.0, "step": 4602 }, { "epoch": 0.6083795929156753, "grad_norm": 0.8409812648998964, "learning_rate": 4.591696623887665e-06, "loss": 0.6084, "mean_token_accuracy": 0.7993682026863098, "num_tokens": 137786596.0, "step": 4603 }, { "epoch": 0.6085117631509384, "grad_norm": 0.8816349760065476, "learning_rate": 4.589663207961501e-06, "loss": 0.5622, "mean_token_accuracy": 0.8111036419868469, "num_tokens": 137852132.0, "step": 4604 }, { "epoch": 0.6086439333862014, "grad_norm": 0.8571645067799856, "learning_rate": 4.5876299858270325e-06, "loss": 0.5754, "mean_token_accuracy": 0.8074411153793335, "num_tokens": 137917668.0, "step": 4605 }, { "epoch": 0.6087761036214644, "grad_norm": 0.9427165426467928, "learning_rate": 4.585596957917092e-06, "loss": 0.6478, "mean_token_accuracy": 0.7845959067344666, "num_tokens": 137983204.0, "step": 4606 }, { "epoch": 0.6089082738567274, "grad_norm": 0.8177149545260577, "learning_rate": 4.583564124664469e-06, "loss": 0.558, "mean_token_accuracy": 0.8156512975692749, "num_tokens": 138048740.0, "step": 4607 }, { "epoch": 0.6090404440919904, "grad_norm": 0.8648725273797789, "learning_rate": 4.581531486501908e-06, "loss": 0.563, "mean_token_accuracy": 0.8144762516021729, "num_tokens": 138114276.0, "step": 4608 }, { "epoch": 0.6091726143272534, "grad_norm": 0.8312851808474316, "learning_rate": 4.5794990438621186e-06, "loss": 0.5825, "mean_token_accuracy": 0.8082346320152283, "num_tokens": 138179812.0, "step": 4609 }, { "epoch": 0.6093047845625165, "grad_norm": 0.8015397731114784, "learning_rate": 4.5774667971777655e-06, "loss": 0.5078, "mean_token_accuracy": 0.8339030742645264, "num_tokens": 138245348.0, "step": 4610 }, { "epoch": 0.6094369547977796, "grad_norm": 0.8110566911638964, "learning_rate": 4.575434746881467e-06, "loss": 0.5626, "mean_token_accuracy": 0.8121718764305115, "num_tokens": 138310884.0, "step": 4611 }, { "epoch": 0.6095691250330426, "grad_norm": 0.847364847537295, "learning_rate": 4.5734028934058104e-06, "loss": 0.6121, "mean_token_accuracy": 0.7949273586273193, "num_tokens": 138376420.0, "step": 4612 }, { "epoch": 0.6097012952683056, "grad_norm": 0.9409679513678505, "learning_rate": 4.57137123718333e-06, "loss": 0.6241, "mean_token_accuracy": 0.7969570159912109, "num_tokens": 138441956.0, "step": 4613 }, { "epoch": 0.6098334655035687, "grad_norm": 0.8776519767682566, "learning_rate": 4.569339778646525e-06, "loss": 0.5979, "mean_token_accuracy": 0.8027255535125732, "num_tokens": 138507492.0, "step": 4614 }, { "epoch": 0.6099656357388317, "grad_norm": 0.806929229878898, "learning_rate": 4.567308518227851e-06, "loss": 0.5834, "mean_token_accuracy": 0.8055640459060669, "num_tokens": 138573028.0, "step": 4615 }, { "epoch": 0.6100978059740947, "grad_norm": 0.7739702156769214, "learning_rate": 4.56527745635972e-06, "loss": 0.5417, "mean_token_accuracy": 0.8189628720283508, "num_tokens": 138638564.0, "step": 4616 }, { "epoch": 0.6102299762093577, "grad_norm": 0.8677160923663167, "learning_rate": 4.563246593474502e-06, "loss": 0.6054, "mean_token_accuracy": 0.8020846247673035, "num_tokens": 138704100.0, "step": 4617 }, { "epoch": 0.6103621464446207, "grad_norm": 0.7954801984071358, "learning_rate": 4.561215930004527e-06, "loss": 0.55, "mean_token_accuracy": 0.8169637322425842, "num_tokens": 138769636.0, "step": 4618 }, { "epoch": 0.6104943166798837, "grad_norm": 0.8472685482911593, "learning_rate": 4.559185466382081e-06, "loss": 0.5794, "mean_token_accuracy": 0.807792067527771, "num_tokens": 138835172.0, "step": 4619 }, { "epoch": 0.6106264869151468, "grad_norm": 0.9072467730496384, "learning_rate": 4.557155203039406e-06, "loss": 0.6431, "mean_token_accuracy": 0.7857404351234436, "num_tokens": 138900708.0, "step": 4620 }, { "epoch": 0.6107586571504098, "grad_norm": 0.8961829080888429, "learning_rate": 4.555125140408702e-06, "loss": 0.5955, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 138966244.0, "step": 4621 }, { "epoch": 0.6108908273856728, "grad_norm": 0.8215213569069227, "learning_rate": 4.553095278922128e-06, "loss": 0.5922, "mean_token_accuracy": 0.8040685057640076, "num_tokens": 139031780.0, "step": 4622 }, { "epoch": 0.6110229976209358, "grad_norm": 0.8323450276636495, "learning_rate": 4.551065619011801e-06, "loss": 0.5619, "mean_token_accuracy": 0.8104932308197021, "num_tokens": 139097316.0, "step": 4623 }, { "epoch": 0.6111551678561988, "grad_norm": 0.7754865097504826, "learning_rate": 4.549036161109789e-06, "loss": 0.5703, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 139162852.0, "step": 4624 }, { "epoch": 0.6112873380914619, "grad_norm": 0.8166360811449247, "learning_rate": 4.547006905648125e-06, "loss": 0.554, "mean_token_accuracy": 0.8185813426971436, "num_tokens": 139228388.0, "step": 4625 }, { "epoch": 0.6114195083267249, "grad_norm": 0.9682789628943366, "learning_rate": 4.544977853058794e-06, "loss": 0.5922, "mean_token_accuracy": 0.8015810251235962, "num_tokens": 139293924.0, "step": 4626 }, { "epoch": 0.6115516785619879, "grad_norm": 0.9228011964351269, "learning_rate": 4.542949003773737e-06, "loss": 0.6372, "mean_token_accuracy": 0.789219856262207, "num_tokens": 139359460.0, "step": 4627 }, { "epoch": 0.6116838487972509, "grad_norm": 0.7894826077176699, "learning_rate": 4.540920358224857e-06, "loss": 0.5588, "mean_token_accuracy": 0.8128891587257385, "num_tokens": 139424996.0, "step": 4628 }, { "epoch": 0.6118160190325139, "grad_norm": 0.8717721643290839, "learning_rate": 4.538891916844008e-06, "loss": 0.5893, "mean_token_accuracy": 0.8039158582687378, "num_tokens": 139490532.0, "step": 4629 }, { "epoch": 0.6119481892677769, "grad_norm": 0.8637773991569165, "learning_rate": 4.536863680063004e-06, "loss": 0.6134, "mean_token_accuracy": 0.7953394055366516, "num_tokens": 139556068.0, "step": 4630 }, { "epoch": 0.61208035950304, "grad_norm": 0.8716930159637672, "learning_rate": 4.534835648313614e-06, "loss": 0.623, "mean_token_accuracy": 0.7962855696678162, "num_tokens": 139621604.0, "step": 4631 }, { "epoch": 0.612212529738303, "grad_norm": 0.9374425133811997, "learning_rate": 4.5328078220275614e-06, "loss": 0.6153, "mean_token_accuracy": 0.8007721900939941, "num_tokens": 139687140.0, "step": 4632 }, { "epoch": 0.612344699973566, "grad_norm": 0.8224805428157486, "learning_rate": 4.530780201636532e-06, "loss": 0.5697, "mean_token_accuracy": 0.8133622407913208, "num_tokens": 139752676.0, "step": 4633 }, { "epoch": 0.612476870208829, "grad_norm": 0.8759804065366877, "learning_rate": 4.52875278757216e-06, "loss": 0.5761, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 139818212.0, "step": 4634 }, { "epoch": 0.612609040444092, "grad_norm": 0.8630023085842069, "learning_rate": 4.526725580266042e-06, "loss": 0.5849, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 139883748.0, "step": 4635 }, { "epoch": 0.612741210679355, "grad_norm": 0.8064649587823042, "learning_rate": 4.5246985801497304e-06, "loss": 0.6007, "mean_token_accuracy": 0.8008332252502441, "num_tokens": 139949284.0, "step": 4636 }, { "epoch": 0.6128733809146181, "grad_norm": 0.8532050333096384, "learning_rate": 4.522671787654724e-06, "loss": 0.5626, "mean_token_accuracy": 0.8114241361618042, "num_tokens": 140014820.0, "step": 4637 }, { "epoch": 0.6130055511498811, "grad_norm": 0.8798126109558387, "learning_rate": 4.52064520321249e-06, "loss": 0.595, "mean_token_accuracy": 0.8022066950798035, "num_tokens": 140080356.0, "step": 4638 }, { "epoch": 0.6131377213851441, "grad_norm": 0.7863304313997864, "learning_rate": 4.518618827254446e-06, "loss": 0.5714, "mean_token_accuracy": 0.8104321956634521, "num_tokens": 140145892.0, "step": 4639 }, { "epoch": 0.6132698916204071, "grad_norm": 0.9915322487444512, "learning_rate": 4.516592660211962e-06, "loss": 0.6394, "mean_token_accuracy": 0.7919057607650757, "num_tokens": 140211428.0, "step": 4640 }, { "epoch": 0.6134020618556701, "grad_norm": 0.8214112702018037, "learning_rate": 4.5145667025163696e-06, "loss": 0.6069, "mean_token_accuracy": 0.7996124029159546, "num_tokens": 140276964.0, "step": 4641 }, { "epoch": 0.6135342320909332, "grad_norm": 0.8529485350296979, "learning_rate": 4.512540954598953e-06, "loss": 0.5957, "mean_token_accuracy": 0.8008484840393066, "num_tokens": 140342500.0, "step": 4642 }, { "epoch": 0.6136664023261962, "grad_norm": 0.8170709512236736, "learning_rate": 4.5105154168909485e-06, "loss": 0.6056, "mean_token_accuracy": 0.8008790016174316, "num_tokens": 140408036.0, "step": 4643 }, { "epoch": 0.6137985725614592, "grad_norm": 0.9007336455504549, "learning_rate": 4.508490089823553e-06, "loss": 0.6369, "mean_token_accuracy": 0.7900286912918091, "num_tokens": 140473572.0, "step": 4644 }, { "epoch": 0.6139307427967222, "grad_norm": 0.786025632763642, "learning_rate": 4.506464973827918e-06, "loss": 0.545, "mean_token_accuracy": 0.8211299180984497, "num_tokens": 140539108.0, "step": 4645 }, { "epoch": 0.6140629130319852, "grad_norm": 0.8956093898819467, "learning_rate": 4.504440069335144e-06, "loss": 0.5763, "mean_token_accuracy": 0.807761549949646, "num_tokens": 140604644.0, "step": 4646 }, { "epoch": 0.6141950832672483, "grad_norm": 0.9099844676677717, "learning_rate": 4.502415376776296e-06, "loss": 0.6203, "mean_token_accuracy": 0.7941642999649048, "num_tokens": 140670180.0, "step": 4647 }, { "epoch": 0.6143272535025113, "grad_norm": 0.8300775252422484, "learning_rate": 4.500390896582385e-06, "loss": 0.5857, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 140735716.0, "step": 4648 }, { "epoch": 0.6144594237377743, "grad_norm": 0.7359522827448938, "learning_rate": 4.498366629184383e-06, "loss": 0.5139, "mean_token_accuracy": 0.8301184177398682, "num_tokens": 140801252.0, "step": 4649 }, { "epoch": 0.6145915939730373, "grad_norm": 0.835220652886212, "learning_rate": 4.496342575013212e-06, "loss": 0.5736, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 140866788.0, "step": 4650 }, { "epoch": 0.6147237642083003, "grad_norm": 0.7913735271236044, "learning_rate": 4.494318734499751e-06, "loss": 0.5711, "mean_token_accuracy": 0.8110883831977844, "num_tokens": 140932324.0, "step": 4651 }, { "epoch": 0.6148559344435633, "grad_norm": 0.7445486602376576, "learning_rate": 4.492295108074836e-06, "loss": 0.5303, "mean_token_accuracy": 0.822747528553009, "num_tokens": 140997860.0, "step": 4652 }, { "epoch": 0.6149881046788264, "grad_norm": 0.8584635500947958, "learning_rate": 4.490271696169251e-06, "loss": 0.5725, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 141063396.0, "step": 4653 }, { "epoch": 0.6151202749140894, "grad_norm": 0.7369585101052492, "learning_rate": 4.488248499213742e-06, "loss": 0.5342, "mean_token_accuracy": 0.8246703743934631, "num_tokens": 141128932.0, "step": 4654 }, { "epoch": 0.6152524451493524, "grad_norm": 0.8252416755149717, "learning_rate": 4.486225517639005e-06, "loss": 0.5709, "mean_token_accuracy": 0.8110120892524719, "num_tokens": 141194468.0, "step": 4655 }, { "epoch": 0.6153846153846154, "grad_norm": 0.8662808655850183, "learning_rate": 4.484202751875684e-06, "loss": 0.5896, "mean_token_accuracy": 0.8031376004219055, "num_tokens": 141260004.0, "step": 4656 }, { "epoch": 0.6155167856198784, "grad_norm": 0.9435011840584125, "learning_rate": 4.4821802023543935e-06, "loss": 0.6703, "mean_token_accuracy": 0.7778812050819397, "num_tokens": 141325540.0, "step": 4657 }, { "epoch": 0.6156489558551415, "grad_norm": 0.8898831741848042, "learning_rate": 4.4801578695056866e-06, "loss": 0.6079, "mean_token_accuracy": 0.7984830737113953, "num_tokens": 141391076.0, "step": 4658 }, { "epoch": 0.6157811260904045, "grad_norm": 0.9039228083989426, "learning_rate": 4.478135753760076e-06, "loss": 0.6127, "mean_token_accuracy": 0.8005890846252441, "num_tokens": 141456612.0, "step": 4659 }, { "epoch": 0.6159132963256675, "grad_norm": 0.9593158086262062, "learning_rate": 4.476113855548029e-06, "loss": 0.6527, "mean_token_accuracy": 0.7876175045967102, "num_tokens": 141522148.0, "step": 4660 }, { "epoch": 0.6160454665609305, "grad_norm": 0.8542069307977787, "learning_rate": 4.474092175299964e-06, "loss": 0.6196, "mean_token_accuracy": 0.7966365814208984, "num_tokens": 141587684.0, "step": 4661 }, { "epoch": 0.6161776367961935, "grad_norm": 0.8207325263870285, "learning_rate": 4.472070713446255e-06, "loss": 0.5505, "mean_token_accuracy": 0.818260908126831, "num_tokens": 141653220.0, "step": 4662 }, { "epoch": 0.6163098070314565, "grad_norm": 0.8558630797551381, "learning_rate": 4.470049470417231e-06, "loss": 0.5793, "mean_token_accuracy": 0.806739091873169, "num_tokens": 141718756.0, "step": 4663 }, { "epoch": 0.6164419772667196, "grad_norm": 0.823179173182554, "learning_rate": 4.468028446643171e-06, "loss": 0.5562, "mean_token_accuracy": 0.8154681921005249, "num_tokens": 141784292.0, "step": 4664 }, { "epoch": 0.6165741475019826, "grad_norm": 0.847262684845638, "learning_rate": 4.466007642554307e-06, "loss": 0.6032, "mean_token_accuracy": 0.7991850972175598, "num_tokens": 141849828.0, "step": 4665 }, { "epoch": 0.6167063177372456, "grad_norm": 0.844460127970561, "learning_rate": 4.4639870585808285e-06, "loss": 0.5689, "mean_token_accuracy": 0.812004029750824, "num_tokens": 141915364.0, "step": 4666 }, { "epoch": 0.6168384879725086, "grad_norm": 0.8682396758389066, "learning_rate": 4.4619666951528765e-06, "loss": 0.6092, "mean_token_accuracy": 0.7993071675300598, "num_tokens": 141980900.0, "step": 4667 }, { "epoch": 0.6169706582077716, "grad_norm": 0.7839640793614593, "learning_rate": 4.45994655270054e-06, "loss": 0.5576, "mean_token_accuracy": 0.816612720489502, "num_tokens": 142046436.0, "step": 4668 }, { "epoch": 0.6171028284430347, "grad_norm": 0.8096423356503544, "learning_rate": 4.4579266316538695e-06, "loss": 0.5959, "mean_token_accuracy": 0.8024966716766357, "num_tokens": 142111972.0, "step": 4669 }, { "epoch": 0.6172349986782977, "grad_norm": 0.8167152048170555, "learning_rate": 4.455906932442862e-06, "loss": 0.5501, "mean_token_accuracy": 0.8181846141815186, "num_tokens": 142177508.0, "step": 4670 }, { "epoch": 0.6173671689135607, "grad_norm": 0.8662400790184801, "learning_rate": 4.453887455497467e-06, "loss": 0.5975, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 142243044.0, "step": 4671 }, { "epoch": 0.6174993391488237, "grad_norm": 0.8803533156660828, "learning_rate": 4.451868201247594e-06, "loss": 0.5715, "mean_token_accuracy": 0.8085551261901855, "num_tokens": 142308580.0, "step": 4672 }, { "epoch": 0.6176315093840867, "grad_norm": 0.8185017836953261, "learning_rate": 4.449849170123097e-06, "loss": 0.5944, "mean_token_accuracy": 0.8029086589813232, "num_tokens": 142374116.0, "step": 4673 }, { "epoch": 0.6177636796193497, "grad_norm": 0.8712016264108585, "learning_rate": 4.4478303625537866e-06, "loss": 0.5569, "mean_token_accuracy": 0.8145220279693604, "num_tokens": 142439652.0, "step": 4674 }, { "epoch": 0.6178958498546128, "grad_norm": 0.8118970784374424, "learning_rate": 4.445811778969423e-06, "loss": 0.5707, "mean_token_accuracy": 0.8118972182273865, "num_tokens": 142505188.0, "step": 4675 }, { "epoch": 0.6180280200898758, "grad_norm": 0.8458542255379631, "learning_rate": 4.443793419799723e-06, "loss": 0.611, "mean_token_accuracy": 0.7986509799957275, "num_tokens": 142570724.0, "step": 4676 }, { "epoch": 0.6181601903251388, "grad_norm": 1.0306890787736454, "learning_rate": 4.4417752854743524e-06, "loss": 0.6222, "mean_token_accuracy": 0.7951409816741943, "num_tokens": 142636260.0, "step": 4677 }, { "epoch": 0.6182923605604018, "grad_norm": 0.8373671540466436, "learning_rate": 4.439757376422926e-06, "loss": 0.536, "mean_token_accuracy": 0.8213282823562622, "num_tokens": 142701796.0, "step": 4678 }, { "epoch": 0.6184245307956648, "grad_norm": 0.9692551607454762, "learning_rate": 4.4377396930750215e-06, "loss": 0.6153, "mean_token_accuracy": 0.7987120151519775, "num_tokens": 142767332.0, "step": 4679 }, { "epoch": 0.6185567010309279, "grad_norm": 0.83618162497167, "learning_rate": 4.435722235860158e-06, "loss": 0.6074, "mean_token_accuracy": 0.8003601431846619, "num_tokens": 142832868.0, "step": 4680 }, { "epoch": 0.6186888712661909, "grad_norm": 0.8065376656667248, "learning_rate": 4.433705005207809e-06, "loss": 0.5392, "mean_token_accuracy": 0.8239989280700684, "num_tokens": 142898404.0, "step": 4681 }, { "epoch": 0.6188210415014539, "grad_norm": 0.8326402445961173, "learning_rate": 4.4316880015473995e-06, "loss": 0.5749, "mean_token_accuracy": 0.812187135219574, "num_tokens": 142963940.0, "step": 4682 }, { "epoch": 0.6189532117367169, "grad_norm": 0.9308005721191731, "learning_rate": 4.429671225308313e-06, "loss": 0.6109, "mean_token_accuracy": 0.8017641305923462, "num_tokens": 143029476.0, "step": 4683 }, { "epoch": 0.6190853819719799, "grad_norm": 0.7985770152082309, "learning_rate": 4.427654676919875e-06, "loss": 0.5595, "mean_token_accuracy": 0.8132706880569458, "num_tokens": 143095012.0, "step": 4684 }, { "epoch": 0.619217552207243, "grad_norm": 0.8615904755436972, "learning_rate": 4.425638356811365e-06, "loss": 0.5904, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 143160548.0, "step": 4685 }, { "epoch": 0.619349722442506, "grad_norm": 0.8707455710373317, "learning_rate": 4.42362226541202e-06, "loss": 0.5848, "mean_token_accuracy": 0.8040227293968201, "num_tokens": 143226084.0, "step": 4686 }, { "epoch": 0.619481892677769, "grad_norm": 0.7771060415320437, "learning_rate": 4.421606403151022e-06, "loss": 0.5501, "mean_token_accuracy": 0.816429615020752, "num_tokens": 143291620.0, "step": 4687 }, { "epoch": 0.619614062913032, "grad_norm": 0.9183098298287047, "learning_rate": 4.419590770457504e-06, "loss": 0.6164, "mean_token_accuracy": 0.7944237589836121, "num_tokens": 143357156.0, "step": 4688 }, { "epoch": 0.619746233148295, "grad_norm": 0.9131933555972311, "learning_rate": 4.417575367760555e-06, "loss": 0.6045, "mean_token_accuracy": 0.8005585670471191, "num_tokens": 143422692.0, "step": 4689 }, { "epoch": 0.619878403383558, "grad_norm": 0.9245525881431137, "learning_rate": 4.41556019548921e-06, "loss": 0.6055, "mean_token_accuracy": 0.7995971441268921, "num_tokens": 143488228.0, "step": 4690 }, { "epoch": 0.6200105736188211, "grad_norm": 0.9776274950405762, "learning_rate": 4.413545254072458e-06, "loss": 0.6248, "mean_token_accuracy": 0.7903797030448914, "num_tokens": 143553764.0, "step": 4691 }, { "epoch": 0.6201427438540841, "grad_norm": 0.8881439174674439, "learning_rate": 4.411530543939237e-06, "loss": 0.5975, "mean_token_accuracy": 0.803335964679718, "num_tokens": 143619300.0, "step": 4692 }, { "epoch": 0.6202749140893471, "grad_norm": 0.8551324918680194, "learning_rate": 4.40951606551844e-06, "loss": 0.594, "mean_token_accuracy": 0.806708574295044, "num_tokens": 143684836.0, "step": 4693 }, { "epoch": 0.6204070843246101, "grad_norm": 0.8781096949877216, "learning_rate": 4.4075018192389045e-06, "loss": 0.5899, "mean_token_accuracy": 0.8044041991233826, "num_tokens": 143750372.0, "step": 4694 }, { "epoch": 0.6205392545598731, "grad_norm": 0.9024461600134324, "learning_rate": 4.405487805529418e-06, "loss": 0.6035, "mean_token_accuracy": 0.7980862855911255, "num_tokens": 143815908.0, "step": 4695 }, { "epoch": 0.6206714247951362, "grad_norm": 0.8942509765431735, "learning_rate": 4.40347402481873e-06, "loss": 0.6136, "mean_token_accuracy": 0.7967586517333984, "num_tokens": 143881444.0, "step": 4696 }, { "epoch": 0.6208035950303992, "grad_norm": 0.915467367773241, "learning_rate": 4.401460477535529e-06, "loss": 0.5704, "mean_token_accuracy": 0.8118972182273865, "num_tokens": 143946980.0, "step": 4697 }, { "epoch": 0.6209357652656622, "grad_norm": 0.8614194439905921, "learning_rate": 4.3994471641084526e-06, "loss": 0.5767, "mean_token_accuracy": 0.8086771965026855, "num_tokens": 144012516.0, "step": 4698 }, { "epoch": 0.6210679355009252, "grad_norm": 0.8506152504051844, "learning_rate": 4.397434084966099e-06, "loss": 0.568, "mean_token_accuracy": 0.8119582533836365, "num_tokens": 144078052.0, "step": 4699 }, { "epoch": 0.6212001057361882, "grad_norm": 1.0069167317241736, "learning_rate": 4.39542124053701e-06, "loss": 0.6779, "mean_token_accuracy": 0.7758668065071106, "num_tokens": 144143588.0, "step": 4700 }, { "epoch": 0.6213322759714512, "grad_norm": 0.8417565793745787, "learning_rate": 4.393408631249674e-06, "loss": 0.5975, "mean_token_accuracy": 0.8002380728721619, "num_tokens": 144209124.0, "step": 4701 }, { "epoch": 0.6214644462067143, "grad_norm": 0.8795393119151936, "learning_rate": 4.391396257532538e-06, "loss": 0.6061, "mean_token_accuracy": 0.7969145774841309, "num_tokens": 144269545.0, "step": 4702 }, { "epoch": 0.6215966164419773, "grad_norm": 0.7901256695299089, "learning_rate": 4.389384119813996e-06, "loss": 0.5986, "mean_token_accuracy": 0.802221953868866, "num_tokens": 144335081.0, "step": 4703 }, { "epoch": 0.6217287866772403, "grad_norm": 0.880597567566107, "learning_rate": 4.387372218522381e-06, "loss": 0.6092, "mean_token_accuracy": 0.798696756362915, "num_tokens": 144400617.0, "step": 4704 }, { "epoch": 0.6218609569125033, "grad_norm": 0.9389958005559172, "learning_rate": 4.385360554085994e-06, "loss": 0.5575, "mean_token_accuracy": 0.8168874382972717, "num_tokens": 144466153.0, "step": 4705 }, { "epoch": 0.6219931271477663, "grad_norm": 0.9708973209274677, "learning_rate": 4.383349126933074e-06, "loss": 0.6487, "mean_token_accuracy": 0.7819405198097229, "num_tokens": 144531689.0, "step": 4706 }, { "epoch": 0.6221252973830294, "grad_norm": 0.8457743955640508, "learning_rate": 4.381337937491808e-06, "loss": 0.5356, "mean_token_accuracy": 0.8208857178688049, "num_tokens": 144597225.0, "step": 4707 }, { "epoch": 0.6222574676182924, "grad_norm": 0.8276460867968112, "learning_rate": 4.379326986190342e-06, "loss": 0.6125, "mean_token_accuracy": 0.7966976165771484, "num_tokens": 144662761.0, "step": 4708 }, { "epoch": 0.6223896378535554, "grad_norm": 0.8266665814258825, "learning_rate": 4.377316273456761e-06, "loss": 0.5822, "mean_token_accuracy": 0.8095165491104126, "num_tokens": 144728297.0, "step": 4709 }, { "epoch": 0.6225218080888184, "grad_norm": 0.8427926061162214, "learning_rate": 4.375305799719106e-06, "loss": 0.5643, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 144793833.0, "step": 4710 }, { "epoch": 0.6226539783240814, "grad_norm": 0.8922802224897954, "learning_rate": 4.373295565405365e-06, "loss": 0.5884, "mean_token_accuracy": 0.8045721054077148, "num_tokens": 144859369.0, "step": 4711 }, { "epoch": 0.6227861485593444, "grad_norm": 0.830325871744666, "learning_rate": 4.3712855709434735e-06, "loss": 0.5242, "mean_token_accuracy": 0.8262879848480225, "num_tokens": 144924905.0, "step": 4712 }, { "epoch": 0.6229183187946075, "grad_norm": 0.8681044890204567, "learning_rate": 4.3692758167613185e-06, "loss": 0.6051, "mean_token_accuracy": 0.7995055317878723, "num_tokens": 144990441.0, "step": 4713 }, { "epoch": 0.6230504890298705, "grad_norm": 0.9279732971402861, "learning_rate": 4.3672663032867325e-06, "loss": 0.6062, "mean_token_accuracy": 0.799062967300415, "num_tokens": 145055977.0, "step": 4714 }, { "epoch": 0.6231826592651335, "grad_norm": 0.9123417409282283, "learning_rate": 4.365257030947502e-06, "loss": 0.6105, "mean_token_accuracy": 0.798879861831665, "num_tokens": 145121513.0, "step": 4715 }, { "epoch": 0.6233148295003965, "grad_norm": 0.8584636116475127, "learning_rate": 4.363248000171359e-06, "loss": 0.5746, "mean_token_accuracy": 0.8115462064743042, "num_tokens": 145187049.0, "step": 4716 }, { "epoch": 0.6234469997356595, "grad_norm": 0.9634985168677465, "learning_rate": 4.36123921138598e-06, "loss": 0.6602, "mean_token_accuracy": 0.7834818959236145, "num_tokens": 145252585.0, "step": 4717 }, { "epoch": 0.6235791699709226, "grad_norm": 0.9047535599001658, "learning_rate": 4.359230665019e-06, "loss": 0.6061, "mean_token_accuracy": 0.799001932144165, "num_tokens": 145318121.0, "step": 4718 }, { "epoch": 0.6237113402061856, "grad_norm": 0.8622086968372614, "learning_rate": 4.357222361497993e-06, "loss": 0.5818, "mean_token_accuracy": 0.8109052777290344, "num_tokens": 145383657.0, "step": 4719 }, { "epoch": 0.6238435104414486, "grad_norm": 0.8364929332752704, "learning_rate": 4.355214301250486e-06, "loss": 0.6176, "mean_token_accuracy": 0.7956445813179016, "num_tokens": 145449193.0, "step": 4720 }, { "epoch": 0.6239756806767116, "grad_norm": 0.8800112370650232, "learning_rate": 4.3532064847039525e-06, "loss": 0.5832, "mean_token_accuracy": 0.8043736815452576, "num_tokens": 145514729.0, "step": 4721 }, { "epoch": 0.6241078509119746, "grad_norm": 0.8241116656781033, "learning_rate": 4.351198912285815e-06, "loss": 0.547, "mean_token_accuracy": 0.8192833662033081, "num_tokens": 145580265.0, "step": 4722 }, { "epoch": 0.6242400211472376, "grad_norm": 0.8812511063485869, "learning_rate": 4.349191584423443e-06, "loss": 0.6289, "mean_token_accuracy": 0.7931418418884277, "num_tokens": 145645801.0, "step": 4723 }, { "epoch": 0.6243721913825007, "grad_norm": 0.816602363868981, "learning_rate": 4.347184501544159e-06, "loss": 0.542, "mean_token_accuracy": 0.8202295303344727, "num_tokens": 145711337.0, "step": 4724 }, { "epoch": 0.6245043616177637, "grad_norm": 0.8445657992678054, "learning_rate": 4.345177664075223e-06, "loss": 0.5587, "mean_token_accuracy": 0.8145067691802979, "num_tokens": 145776873.0, "step": 4725 }, { "epoch": 0.6246365318530267, "grad_norm": 0.806985393795572, "learning_rate": 4.343171072443852e-06, "loss": 0.544, "mean_token_accuracy": 0.8214350938796997, "num_tokens": 145842409.0, "step": 4726 }, { "epoch": 0.6247687020882897, "grad_norm": 0.7678556551546073, "learning_rate": 4.341164727077206e-06, "loss": 0.5729, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 145907945.0, "step": 4727 }, { "epoch": 0.6249008723235527, "grad_norm": 0.8716831950820124, "learning_rate": 4.339158628402394e-06, "loss": 0.5773, "mean_token_accuracy": 0.8091655373573303, "num_tokens": 145973481.0, "step": 4728 }, { "epoch": 0.6250330425588158, "grad_norm": 0.8075384751563434, "learning_rate": 4.337152776846476e-06, "loss": 0.528, "mean_token_accuracy": 0.8243193626403809, "num_tokens": 146039017.0, "step": 4729 }, { "epoch": 0.6251652127940788, "grad_norm": 0.8138184550717408, "learning_rate": 4.3351471728364525e-06, "loss": 0.5721, "mean_token_accuracy": 0.8090892434120178, "num_tokens": 146104553.0, "step": 4730 }, { "epoch": 0.6252973830293418, "grad_norm": 0.8513425116370146, "learning_rate": 4.333141816799276e-06, "loss": 0.593, "mean_token_accuracy": 0.8032596707344055, "num_tokens": 146170089.0, "step": 4731 }, { "epoch": 0.6254295532646048, "grad_norm": 0.976698928505754, "learning_rate": 4.3311367091618426e-06, "loss": 0.6289, "mean_token_accuracy": 0.7899981737136841, "num_tokens": 146235625.0, "step": 4732 }, { "epoch": 0.6255617234998678, "grad_norm": 0.9171659481418084, "learning_rate": 4.329131850350999e-06, "loss": 0.5684, "mean_token_accuracy": 0.8146135807037354, "num_tokens": 146301161.0, "step": 4733 }, { "epoch": 0.6256938937351308, "grad_norm": 0.837696596575251, "learning_rate": 4.327127240793541e-06, "loss": 0.5678, "mean_token_accuracy": 0.812782347202301, "num_tokens": 146366697.0, "step": 4734 }, { "epoch": 0.6258260639703939, "grad_norm": 0.8020211902916298, "learning_rate": 4.325122880916206e-06, "loss": 0.5362, "mean_token_accuracy": 0.8230222463607788, "num_tokens": 146432233.0, "step": 4735 }, { "epoch": 0.6259582342056569, "grad_norm": 0.8657804857063617, "learning_rate": 4.323118771145677e-06, "loss": 0.6029, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 146497769.0, "step": 4736 }, { "epoch": 0.6260904044409199, "grad_norm": 0.8286104168964823, "learning_rate": 4.321114911908592e-06, "loss": 0.597, "mean_token_accuracy": 0.8041905760765076, "num_tokens": 146563305.0, "step": 4737 }, { "epoch": 0.6262225746761829, "grad_norm": 0.9676552953117135, "learning_rate": 4.31911130363153e-06, "loss": 0.6225, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 146628841.0, "step": 4738 }, { "epoch": 0.6263547449114459, "grad_norm": 0.8085126532192716, "learning_rate": 4.3171079467410144e-06, "loss": 0.5353, "mean_token_accuracy": 0.8223965167999268, "num_tokens": 146694377.0, "step": 4739 }, { "epoch": 0.626486915146709, "grad_norm": 0.8941662167059061, "learning_rate": 4.31510484166352e-06, "loss": 0.5794, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 146759913.0, "step": 4740 }, { "epoch": 0.626619085381972, "grad_norm": 0.833181587434127, "learning_rate": 4.313101988825467e-06, "loss": 0.5993, "mean_token_accuracy": 0.8019472360610962, "num_tokens": 146825449.0, "step": 4741 }, { "epoch": 0.626751255617235, "grad_norm": 0.9491458467445169, "learning_rate": 4.311099388653221e-06, "loss": 0.6305, "mean_token_accuracy": 0.7893267273902893, "num_tokens": 146890985.0, "step": 4742 }, { "epoch": 0.626883425852498, "grad_norm": 0.8907009156636942, "learning_rate": 4.3090970415730895e-06, "loss": 0.6091, "mean_token_accuracy": 0.8006043434143066, "num_tokens": 146956521.0, "step": 4743 }, { "epoch": 0.627015596087761, "grad_norm": 0.8534072604880992, "learning_rate": 4.307094948011335e-06, "loss": 0.621, "mean_token_accuracy": 0.7957667112350464, "num_tokens": 147022057.0, "step": 4744 }, { "epoch": 0.627147766323024, "grad_norm": 0.8406851153870656, "learning_rate": 4.305093108394161e-06, "loss": 0.5984, "mean_token_accuracy": 0.8016878366470337, "num_tokens": 147087593.0, "step": 4745 }, { "epoch": 0.6272799365582871, "grad_norm": 0.799506367120811, "learning_rate": 4.303091523147715e-06, "loss": 0.5854, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 147153129.0, "step": 4746 }, { "epoch": 0.6274121067935501, "grad_norm": 0.7863727520571749, "learning_rate": 4.301090192698094e-06, "loss": 0.5621, "mean_token_accuracy": 0.8129807114601135, "num_tokens": 147218665.0, "step": 4747 }, { "epoch": 0.6275442770288131, "grad_norm": 0.8090861915741046, "learning_rate": 4.299089117471341e-06, "loss": 0.567, "mean_token_accuracy": 0.811942994594574, "num_tokens": 147284201.0, "step": 4748 }, { "epoch": 0.6276764472640761, "grad_norm": 0.7900245427795113, "learning_rate": 4.297088297893441e-06, "loss": 0.552, "mean_token_accuracy": 0.816521167755127, "num_tokens": 147349737.0, "step": 4749 }, { "epoch": 0.6278086174993391, "grad_norm": 0.8191361735381022, "learning_rate": 4.2950877343903275e-06, "loss": 0.5516, "mean_token_accuracy": 0.8153156042098999, "num_tokens": 147415273.0, "step": 4750 }, { "epoch": 0.6279407877346022, "grad_norm": 0.8211037092046062, "learning_rate": 4.293087427387878e-06, "loss": 0.5662, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 147480809.0, "step": 4751 }, { "epoch": 0.6280729579698652, "grad_norm": 0.8716013337316459, "learning_rate": 4.291087377311916e-06, "loss": 0.599, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 147546345.0, "step": 4752 }, { "epoch": 0.6282051282051282, "grad_norm": 0.8277942849318017, "learning_rate": 4.289087584588214e-06, "loss": 0.5749, "mean_token_accuracy": 0.8101269602775574, "num_tokens": 147611881.0, "step": 4753 }, { "epoch": 0.6283372984403912, "grad_norm": 0.854125461589121, "learning_rate": 4.2870880496424814e-06, "loss": 0.5988, "mean_token_accuracy": 0.8010469079017639, "num_tokens": 147677417.0, "step": 4754 }, { "epoch": 0.6284694686756542, "grad_norm": 0.867730803290388, "learning_rate": 4.28508877290038e-06, "loss": 0.5999, "mean_token_accuracy": 0.8003754019737244, "num_tokens": 147742953.0, "step": 4755 }, { "epoch": 0.6286016389109172, "grad_norm": 0.8667975378016485, "learning_rate": 4.283089754787511e-06, "loss": 0.5886, "mean_token_accuracy": 0.8075326681137085, "num_tokens": 147808489.0, "step": 4756 }, { "epoch": 0.6287338091461803, "grad_norm": 0.7966989613292065, "learning_rate": 4.2810909957294275e-06, "loss": 0.5839, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 147874025.0, "step": 4757 }, { "epoch": 0.6288659793814433, "grad_norm": 0.7703023669580598, "learning_rate": 4.279092496151623e-06, "loss": 0.5155, "mean_token_accuracy": 0.8266695141792297, "num_tokens": 147939561.0, "step": 4758 }, { "epoch": 0.6289981496167063, "grad_norm": 0.7892669906312875, "learning_rate": 4.277094256479533e-06, "loss": 0.5581, "mean_token_accuracy": 0.8141710162162781, "num_tokens": 148005097.0, "step": 4759 }, { "epoch": 0.6291303198519693, "grad_norm": 0.8770974050895329, "learning_rate": 4.275096277138545e-06, "loss": 0.5714, "mean_token_accuracy": 0.8123855590820312, "num_tokens": 148070633.0, "step": 4760 }, { "epoch": 0.6292624900872323, "grad_norm": 0.8144846216716939, "learning_rate": 4.2730985585539865e-06, "loss": 0.5762, "mean_token_accuracy": 0.8116225004196167, "num_tokens": 148136169.0, "step": 4761 }, { "epoch": 0.6293946603224954, "grad_norm": 0.8192340737449565, "learning_rate": 4.271101101151126e-06, "loss": 0.6064, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 148201705.0, "step": 4762 }, { "epoch": 0.6295268305577584, "grad_norm": 0.8132973452578413, "learning_rate": 4.269103905355185e-06, "loss": 0.5836, "mean_token_accuracy": 0.806617021560669, "num_tokens": 148267241.0, "step": 4763 }, { "epoch": 0.6296590007930214, "grad_norm": 0.8553100359936023, "learning_rate": 4.267106971591325e-06, "loss": 0.5851, "mean_token_accuracy": 0.8026034832000732, "num_tokens": 148332777.0, "step": 4764 }, { "epoch": 0.6297911710282844, "grad_norm": 0.8653910273559422, "learning_rate": 4.265110300284647e-06, "loss": 0.5901, "mean_token_accuracy": 0.8035191297531128, "num_tokens": 148398313.0, "step": 4765 }, { "epoch": 0.6299233412635474, "grad_norm": 0.863161725504361, "learning_rate": 4.2631138918602056e-06, "loss": 0.5772, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 148463849.0, "step": 4766 }, { "epoch": 0.6300555114988104, "grad_norm": 0.8729908238109169, "learning_rate": 4.2611177467429906e-06, "loss": 0.6124, "mean_token_accuracy": 0.7962855696678162, "num_tokens": 148529385.0, "step": 4767 }, { "epoch": 0.6301876817340735, "grad_norm": 0.7727167880066343, "learning_rate": 4.259121865357942e-06, "loss": 0.5323, "mean_token_accuracy": 0.822747528553009, "num_tokens": 148594921.0, "step": 4768 }, { "epoch": 0.6303198519693365, "grad_norm": 0.843348466845674, "learning_rate": 4.257126248129942e-06, "loss": 0.5181, "mean_token_accuracy": 0.8255860209465027, "num_tokens": 148660457.0, "step": 4769 }, { "epoch": 0.6304520222045995, "grad_norm": 0.8178877990848973, "learning_rate": 4.2551308954838134e-06, "loss": 0.5354, "mean_token_accuracy": 0.8226101994514465, "num_tokens": 148725993.0, "step": 4770 }, { "epoch": 0.6305841924398625, "grad_norm": 0.8171843182859857, "learning_rate": 4.253135807844327e-06, "loss": 0.6075, "mean_token_accuracy": 0.7983304858207703, "num_tokens": 148791529.0, "step": 4771 }, { "epoch": 0.6307163626751255, "grad_norm": 0.858879852730464, "learning_rate": 4.251140985636192e-06, "loss": 0.5968, "mean_token_accuracy": 0.8020541071891785, "num_tokens": 148857065.0, "step": 4772 }, { "epoch": 0.6308485329103886, "grad_norm": 0.8716165835943345, "learning_rate": 4.249146429284069e-06, "loss": 0.6072, "mean_token_accuracy": 0.7998107671737671, "num_tokens": 148922601.0, "step": 4773 }, { "epoch": 0.6309807031456516, "grad_norm": 0.8439039138515032, "learning_rate": 4.247152139212554e-06, "loss": 0.6, "mean_token_accuracy": 0.8041142821311951, "num_tokens": 148988137.0, "step": 4774 }, { "epoch": 0.6311128733809146, "grad_norm": 0.8754269804803504, "learning_rate": 4.24515811584619e-06, "loss": 0.5977, "mean_token_accuracy": 0.8007874488830566, "num_tokens": 149053673.0, "step": 4775 }, { "epoch": 0.6312450436161776, "grad_norm": 0.853117993876722, "learning_rate": 4.243164359609466e-06, "loss": 0.5965, "mean_token_accuracy": 0.8014588952064514, "num_tokens": 149119209.0, "step": 4776 }, { "epoch": 0.6313772138514406, "grad_norm": 0.8958727101770682, "learning_rate": 4.24117087092681e-06, "loss": 0.5946, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 149184745.0, "step": 4777 }, { "epoch": 0.6315093840867037, "grad_norm": 0.8891418042167625, "learning_rate": 4.23917765022259e-06, "loss": 0.6378, "mean_token_accuracy": 0.7884415984153748, "num_tokens": 149250281.0, "step": 4778 }, { "epoch": 0.6316415543219667, "grad_norm": 0.8965321088737641, "learning_rate": 4.237184697921126e-06, "loss": 0.5807, "mean_token_accuracy": 0.807578444480896, "num_tokens": 149315817.0, "step": 4779 }, { "epoch": 0.6317737245572297, "grad_norm": 0.838579528571683, "learning_rate": 4.235192014446674e-06, "loss": 0.5241, "mean_token_accuracy": 0.8275088667869568, "num_tokens": 149381353.0, "step": 4780 }, { "epoch": 0.6319058947924927, "grad_norm": 0.8081514745688869, "learning_rate": 4.233199600223434e-06, "loss": 0.5658, "mean_token_accuracy": 0.8137437701225281, "num_tokens": 149446889.0, "step": 4781 }, { "epoch": 0.6320380650277557, "grad_norm": 0.8402561586529913, "learning_rate": 4.231207455675551e-06, "loss": 0.5943, "mean_token_accuracy": 0.79878830909729, "num_tokens": 149512425.0, "step": 4782 }, { "epoch": 0.6321702352630187, "grad_norm": 0.8309056602785163, "learning_rate": 4.2292155812271105e-06, "loss": 0.5542, "mean_token_accuracy": 0.8196496367454529, "num_tokens": 149577961.0, "step": 4783 }, { "epoch": 0.6323024054982818, "grad_norm": 0.8699218696912983, "learning_rate": 4.227223977302141e-06, "loss": 0.5908, "mean_token_accuracy": 0.8047704696655273, "num_tokens": 149643497.0, "step": 4784 }, { "epoch": 0.6324345757335448, "grad_norm": 0.8596000132096309, "learning_rate": 4.225232644324612e-06, "loss": 0.558, "mean_token_accuracy": 0.8138353228569031, "num_tokens": 149709033.0, "step": 4785 }, { "epoch": 0.6325667459688078, "grad_norm": 0.9201852428667681, "learning_rate": 4.223241582718438e-06, "loss": 0.651, "mean_token_accuracy": 0.7825204730033875, "num_tokens": 149774569.0, "step": 4786 }, { "epoch": 0.6326989162040708, "grad_norm": 0.9894131918542021, "learning_rate": 4.2212507929074755e-06, "loss": 0.6005, "mean_token_accuracy": 0.8029850125312805, "num_tokens": 149840105.0, "step": 4787 }, { "epoch": 0.6328310864393338, "grad_norm": 0.8565643208750174, "learning_rate": 4.2192602753155185e-06, "loss": 0.5718, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 149905641.0, "step": 4788 }, { "epoch": 0.6329632566745969, "grad_norm": 0.8536514772539413, "learning_rate": 4.2172700303663114e-06, "loss": 0.5906, "mean_token_accuracy": 0.8044652938842773, "num_tokens": 149971177.0, "step": 4789 }, { "epoch": 0.6330954269098599, "grad_norm": 0.9097336137743022, "learning_rate": 4.2152800584835326e-06, "loss": 0.608, "mean_token_accuracy": 0.797796368598938, "num_tokens": 150036713.0, "step": 4790 }, { "epoch": 0.6332275971451229, "grad_norm": 0.8617749239330077, "learning_rate": 4.213290360090805e-06, "loss": 0.5992, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 150102249.0, "step": 4791 }, { "epoch": 0.6333597673803859, "grad_norm": 0.8209022384274004, "learning_rate": 4.211300935611696e-06, "loss": 0.5521, "mean_token_accuracy": 0.8170400261878967, "num_tokens": 150167785.0, "step": 4792 }, { "epoch": 0.6334919376156489, "grad_norm": 0.9688862889814952, "learning_rate": 4.209311785469713e-06, "loss": 0.6449, "mean_token_accuracy": 0.7886399626731873, "num_tokens": 150233321.0, "step": 4793 }, { "epoch": 0.6336241078509119, "grad_norm": 0.8394985938928423, "learning_rate": 4.2073229100883005e-06, "loss": 0.6043, "mean_token_accuracy": 0.8016420602798462, "num_tokens": 150298857.0, "step": 4794 }, { "epoch": 0.633756278086175, "grad_norm": 0.7231815623275132, "learning_rate": 4.2053343098908525e-06, "loss": 0.4905, "mean_token_accuracy": 0.8382065892219543, "num_tokens": 150364393.0, "step": 4795 }, { "epoch": 0.633888448321438, "grad_norm": 0.8607687999489468, "learning_rate": 4.2033459853007e-06, "loss": 0.5785, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 150429929.0, "step": 4796 }, { "epoch": 0.634020618556701, "grad_norm": 0.7740210724740724, "learning_rate": 4.201357936741114e-06, "loss": 0.547, "mean_token_accuracy": 0.8189933896064758, "num_tokens": 150495465.0, "step": 4797 }, { "epoch": 0.634152788791964, "grad_norm": 0.8088234675729372, "learning_rate": 4.199370164635312e-06, "loss": 0.5453, "mean_token_accuracy": 0.8208704590797424, "num_tokens": 150561001.0, "step": 4798 }, { "epoch": 0.634284959027227, "grad_norm": 0.8658882792866841, "learning_rate": 4.197382669406445e-06, "loss": 0.5876, "mean_token_accuracy": 0.8083414435386658, "num_tokens": 150626537.0, "step": 4799 }, { "epoch": 0.63441712926249, "grad_norm": 0.8146333671885847, "learning_rate": 4.195395451477611e-06, "loss": 0.5876, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 150692073.0, "step": 4800 }, { "epoch": 0.6345492994977531, "grad_norm": 0.8143096111313416, "learning_rate": 4.1934085112718455e-06, "loss": 0.5772, "mean_token_accuracy": 0.8077462911605835, "num_tokens": 150757609.0, "step": 4801 }, { "epoch": 0.6346814697330161, "grad_norm": 0.8433378885064928, "learning_rate": 4.191421849212132e-06, "loss": 0.5647, "mean_token_accuracy": 0.8145220279693604, "num_tokens": 150823145.0, "step": 4802 }, { "epoch": 0.6348136399682791, "grad_norm": 0.8397353623085935, "learning_rate": 4.1894354657213835e-06, "loss": 0.5504, "mean_token_accuracy": 0.8177573084831238, "num_tokens": 150888681.0, "step": 4803 }, { "epoch": 0.6349458102035421, "grad_norm": 0.8539889290554779, "learning_rate": 4.18744936122246e-06, "loss": 0.5727, "mean_token_accuracy": 0.813087522983551, "num_tokens": 150954217.0, "step": 4804 }, { "epoch": 0.6350779804388051, "grad_norm": 0.9042238145748912, "learning_rate": 4.185463536138165e-06, "loss": 0.6109, "mean_token_accuracy": 0.8003143668174744, "num_tokens": 151019753.0, "step": 4805 }, { "epoch": 0.6352101506740682, "grad_norm": 0.8585102187988873, "learning_rate": 4.1834779908912394e-06, "loss": 0.6081, "mean_token_accuracy": 0.8013978600502014, "num_tokens": 151085289.0, "step": 4806 }, { "epoch": 0.6353423209093312, "grad_norm": 0.948124477827563, "learning_rate": 4.1814927259043604e-06, "loss": 0.6124, "mean_token_accuracy": 0.7995513081550598, "num_tokens": 151150825.0, "step": 4807 }, { "epoch": 0.6354744911445942, "grad_norm": 0.8335741211121231, "learning_rate": 4.179507741600153e-06, "loss": 0.6061, "mean_token_accuracy": 0.7987730503082275, "num_tokens": 151216361.0, "step": 4808 }, { "epoch": 0.6356066613798572, "grad_norm": 0.8098247143466359, "learning_rate": 4.177523038401178e-06, "loss": 0.5503, "mean_token_accuracy": 0.8173757791519165, "num_tokens": 151281897.0, "step": 4809 }, { "epoch": 0.6357388316151202, "grad_norm": 0.831705705281307, "learning_rate": 4.175538616729933e-06, "loss": 0.5868, "mean_token_accuracy": 0.803427517414093, "num_tokens": 151347433.0, "step": 4810 }, { "epoch": 0.6358710018503833, "grad_norm": 0.8596708631025807, "learning_rate": 4.173554477008868e-06, "loss": 0.588, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 151412969.0, "step": 4811 }, { "epoch": 0.6360031720856463, "grad_norm": 0.8397878971801164, "learning_rate": 4.171570619660359e-06, "loss": 0.5636, "mean_token_accuracy": 0.8135453462600708, "num_tokens": 151478505.0, "step": 4812 }, { "epoch": 0.6361353423209093, "grad_norm": 0.8949668363707413, "learning_rate": 4.169587045106728e-06, "loss": 0.6053, "mean_token_accuracy": 0.7994902729988098, "num_tokens": 151544041.0, "step": 4813 }, { "epoch": 0.6362675125561723, "grad_norm": 0.7829569455704629, "learning_rate": 4.167603753770239e-06, "loss": 0.5852, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 151609577.0, "step": 4814 }, { "epoch": 0.6363996827914353, "grad_norm": 0.8545231066840032, "learning_rate": 4.165620746073094e-06, "loss": 0.5768, "mean_token_accuracy": 0.8082498908042908, "num_tokens": 151675113.0, "step": 4815 }, { "epoch": 0.6365318530266983, "grad_norm": 0.8479262877174739, "learning_rate": 4.163638022437432e-06, "loss": 0.6081, "mean_token_accuracy": 0.8007416725158691, "num_tokens": 151740649.0, "step": 4816 }, { "epoch": 0.6366640232619614, "grad_norm": 0.8930670252714619, "learning_rate": 4.161655583285332e-06, "loss": 0.6282, "mean_token_accuracy": 0.7919362783432007, "num_tokens": 151806185.0, "step": 4817 }, { "epoch": 0.6367961934972244, "grad_norm": 0.7998250984167609, "learning_rate": 4.159673429038816e-06, "loss": 0.5674, "mean_token_accuracy": 0.8135606050491333, "num_tokens": 151871721.0, "step": 4818 }, { "epoch": 0.6369283637324874, "grad_norm": 0.9061063186558513, "learning_rate": 4.157691560119845e-06, "loss": 0.6311, "mean_token_accuracy": 0.7915394902229309, "num_tokens": 151937257.0, "step": 4819 }, { "epoch": 0.6370605339677504, "grad_norm": 0.8771075371882611, "learning_rate": 4.155709976950313e-06, "loss": 0.5988, "mean_token_accuracy": 0.8027865886688232, "num_tokens": 152002793.0, "step": 4820 }, { "epoch": 0.6371927042030134, "grad_norm": 0.8295147026961726, "learning_rate": 4.153728679952062e-06, "loss": 0.5753, "mean_token_accuracy": 0.806769609451294, "num_tokens": 152068329.0, "step": 4821 }, { "epoch": 0.6373248744382765, "grad_norm": 0.8355889692058093, "learning_rate": 4.151747669546866e-06, "loss": 0.5555, "mean_token_accuracy": 0.8162922859191895, "num_tokens": 152133865.0, "step": 4822 }, { "epoch": 0.6374570446735395, "grad_norm": 0.7845994813087335, "learning_rate": 4.149766946156442e-06, "loss": 0.5631, "mean_token_accuracy": 0.812156617641449, "num_tokens": 152199401.0, "step": 4823 }, { "epoch": 0.6375892149088025, "grad_norm": 0.8391347679217754, "learning_rate": 4.1477865102024425e-06, "loss": 0.5572, "mean_token_accuracy": 0.812843382358551, "num_tokens": 152264937.0, "step": 4824 }, { "epoch": 0.6377213851440655, "grad_norm": 0.8370076679268412, "learning_rate": 4.145806362106466e-06, "loss": 0.588, "mean_token_accuracy": 0.8054113984107971, "num_tokens": 152330473.0, "step": 4825 }, { "epoch": 0.6378535553793285, "grad_norm": 0.8770148261652588, "learning_rate": 4.143826502290039e-06, "loss": 0.6393, "mean_token_accuracy": 0.7906238436698914, "num_tokens": 152396009.0, "step": 4826 }, { "epoch": 0.6379857256145915, "grad_norm": 0.8181154455218829, "learning_rate": 4.141846931174637e-06, "loss": 0.5616, "mean_token_accuracy": 0.8124618530273438, "num_tokens": 152461545.0, "step": 4827 }, { "epoch": 0.6381178958498546, "grad_norm": 0.878079188727156, "learning_rate": 4.1398676491816655e-06, "loss": 0.6425, "mean_token_accuracy": 0.7866408228874207, "num_tokens": 152527081.0, "step": 4828 }, { "epoch": 0.6382500660851176, "grad_norm": 0.8533488483876467, "learning_rate": 4.137888656732474e-06, "loss": 0.5993, "mean_token_accuracy": 0.8015505075454712, "num_tokens": 152592617.0, "step": 4829 }, { "epoch": 0.6383822363203806, "grad_norm": 0.8619840757588619, "learning_rate": 4.135909954248349e-06, "loss": 0.6126, "mean_token_accuracy": 0.8022066950798035, "num_tokens": 152658153.0, "step": 4830 }, { "epoch": 0.6385144065556436, "grad_norm": 0.8843759630206006, "learning_rate": 4.133931542150516e-06, "loss": 0.6223, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 152723689.0, "step": 4831 }, { "epoch": 0.6386465767909066, "grad_norm": 0.8485719072818441, "learning_rate": 4.1319534208601344e-06, "loss": 0.5781, "mean_token_accuracy": 0.8062354922294617, "num_tokens": 152789225.0, "step": 4832 }, { "epoch": 0.6387787470261697, "grad_norm": 0.867537079434261, "learning_rate": 4.1299755907983066e-06, "loss": 0.6221, "mean_token_accuracy": 0.7942711710929871, "num_tokens": 152854761.0, "step": 4833 }, { "epoch": 0.6389109172614327, "grad_norm": 0.8268245955780407, "learning_rate": 4.127998052386072e-06, "loss": 0.617, "mean_token_accuracy": 0.7953546643257141, "num_tokens": 152920297.0, "step": 4834 }, { "epoch": 0.6390430874966957, "grad_norm": 0.8832332371118565, "learning_rate": 4.126020806044408e-06, "loss": 0.6036, "mean_token_accuracy": 0.7979031801223755, "num_tokens": 152985833.0, "step": 4835 }, { "epoch": 0.6391752577319587, "grad_norm": 0.851990069180944, "learning_rate": 4.124043852194225e-06, "loss": 0.5989, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 153051369.0, "step": 4836 }, { "epoch": 0.6393074279672217, "grad_norm": 0.8677093401283291, "learning_rate": 4.122067191256379e-06, "loss": 0.5643, "mean_token_accuracy": 0.811973512172699, "num_tokens": 153116905.0, "step": 4837 }, { "epoch": 0.6394395982024847, "grad_norm": 0.8490748196335701, "learning_rate": 4.12009082365166e-06, "loss": 0.6271, "mean_token_accuracy": 0.7930960655212402, "num_tokens": 153182441.0, "step": 4838 }, { "epoch": 0.6395717684377478, "grad_norm": 0.8598886132452179, "learning_rate": 4.118114749800792e-06, "loss": 0.5832, "mean_token_accuracy": 0.8071358799934387, "num_tokens": 153247977.0, "step": 4839 }, { "epoch": 0.6397039386730108, "grad_norm": 0.8415823185531551, "learning_rate": 4.116138970124444e-06, "loss": 0.5711, "mean_token_accuracy": 0.8110731244087219, "num_tokens": 153313513.0, "step": 4840 }, { "epoch": 0.6398361089082738, "grad_norm": 0.7919334938994584, "learning_rate": 4.114163485043218e-06, "loss": 0.6026, "mean_token_accuracy": 0.8025424480438232, "num_tokens": 153379049.0, "step": 4841 }, { "epoch": 0.6399682791435368, "grad_norm": 0.8027513369252101, "learning_rate": 4.112188294977648e-06, "loss": 0.582, "mean_token_accuracy": 0.8087077140808105, "num_tokens": 153444585.0, "step": 4842 }, { "epoch": 0.6401004493787998, "grad_norm": 0.8223974221721431, "learning_rate": 4.1102134003482165e-06, "loss": 0.5971, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 153510121.0, "step": 4843 }, { "epoch": 0.6402326196140629, "grad_norm": 0.841053341676517, "learning_rate": 4.108238801575336e-06, "loss": 0.5917, "mean_token_accuracy": 0.8016420602798462, "num_tokens": 153575657.0, "step": 4844 }, { "epoch": 0.6403647898493259, "grad_norm": 0.7810776217678779, "learning_rate": 4.106264499079357e-06, "loss": 0.5948, "mean_token_accuracy": 0.8033817410469055, "num_tokens": 153641193.0, "step": 4845 }, { "epoch": 0.6404969600845889, "grad_norm": 0.8018235390058217, "learning_rate": 4.104290493280565e-06, "loss": 0.5416, "mean_token_accuracy": 0.8235716223716736, "num_tokens": 153706729.0, "step": 4846 }, { "epoch": 0.640629130319852, "grad_norm": 0.8001181833384132, "learning_rate": 4.102316784599188e-06, "loss": 0.5616, "mean_token_accuracy": 0.8144457340240479, "num_tokens": 153772265.0, "step": 4847 }, { "epoch": 0.640761300555115, "grad_norm": 0.8230969255581506, "learning_rate": 4.100343373455385e-06, "loss": 0.5664, "mean_token_accuracy": 0.8126907348632812, "num_tokens": 153837801.0, "step": 4848 }, { "epoch": 0.6408934707903781, "grad_norm": 0.8472976437621506, "learning_rate": 4.098370260269255e-06, "loss": 0.5961, "mean_token_accuracy": 0.8025882244110107, "num_tokens": 153903337.0, "step": 4849 }, { "epoch": 0.6410256410256411, "grad_norm": 0.813034020733219, "learning_rate": 4.0963974454608344e-06, "loss": 0.5905, "mean_token_accuracy": 0.8043126463890076, "num_tokens": 153968873.0, "step": 4850 }, { "epoch": 0.6411578112609041, "grad_norm": 0.841393717827924, "learning_rate": 4.094424929450092e-06, "loss": 0.5993, "mean_token_accuracy": 0.803030788898468, "num_tokens": 154034409.0, "step": 4851 }, { "epoch": 0.6412899814961671, "grad_norm": 0.8712147285823801, "learning_rate": 4.092452712656935e-06, "loss": 0.6418, "mean_token_accuracy": 0.7870070934295654, "num_tokens": 154099945.0, "step": 4852 }, { "epoch": 0.6414221517314301, "grad_norm": 0.827657634942308, "learning_rate": 4.09048079550121e-06, "loss": 0.577, "mean_token_accuracy": 0.8085856437683105, "num_tokens": 154165481.0, "step": 4853 }, { "epoch": 0.6415543219666932, "grad_norm": 0.8332020943948377, "learning_rate": 4.088509178402693e-06, "loss": 0.5706, "mean_token_accuracy": 0.8100201487541199, "num_tokens": 154231017.0, "step": 4854 }, { "epoch": 0.6416864922019562, "grad_norm": 0.8619049582079314, "learning_rate": 4.086537861781101e-06, "loss": 0.6114, "mean_token_accuracy": 0.7987730503082275, "num_tokens": 154296553.0, "step": 4855 }, { "epoch": 0.6418186624372192, "grad_norm": 0.888669961673454, "learning_rate": 4.08456684605609e-06, "loss": 0.609, "mean_token_accuracy": 0.798696756362915, "num_tokens": 154362089.0, "step": 4856 }, { "epoch": 0.6419508326724822, "grad_norm": 0.8075442596597524, "learning_rate": 4.082596131647244e-06, "loss": 0.578, "mean_token_accuracy": 0.807669997215271, "num_tokens": 154427625.0, "step": 4857 }, { "epoch": 0.6420830029077452, "grad_norm": 0.8256600364980383, "learning_rate": 4.080625718974087e-06, "loss": 0.5925, "mean_token_accuracy": 0.8045568466186523, "num_tokens": 154493161.0, "step": 4858 }, { "epoch": 0.6422151731430082, "grad_norm": 0.8580961911989925, "learning_rate": 4.0786556084560835e-06, "loss": 0.6218, "mean_token_accuracy": 0.7950646877288818, "num_tokens": 154558697.0, "step": 4859 }, { "epoch": 0.6423473433782713, "grad_norm": 0.8401857540272143, "learning_rate": 4.076685800512623e-06, "loss": 0.5797, "mean_token_accuracy": 0.8078073263168335, "num_tokens": 154624233.0, "step": 4860 }, { "epoch": 0.6424795136135343, "grad_norm": 0.7896778173756221, "learning_rate": 4.07471629556304e-06, "loss": 0.5306, "mean_token_accuracy": 0.821664035320282, "num_tokens": 154689769.0, "step": 4861 }, { "epoch": 0.6426116838487973, "grad_norm": 0.8699531662676139, "learning_rate": 4.072747094026598e-06, "loss": 0.5627, "mean_token_accuracy": 0.812034547328949, "num_tokens": 154755305.0, "step": 4862 }, { "epoch": 0.6427438540840603, "grad_norm": 0.7910815519299859, "learning_rate": 4.070778196322502e-06, "loss": 0.5804, "mean_token_accuracy": 0.8056861162185669, "num_tokens": 154820841.0, "step": 4863 }, { "epoch": 0.6428760243193233, "grad_norm": 0.8090992056156427, "learning_rate": 4.068809602869889e-06, "loss": 0.6127, "mean_token_accuracy": 0.7975521683692932, "num_tokens": 154886377.0, "step": 4864 }, { "epoch": 0.6430081945545864, "grad_norm": 0.7949716217002372, "learning_rate": 4.066841314087829e-06, "loss": 0.5478, "mean_token_accuracy": 0.8178336024284363, "num_tokens": 154951913.0, "step": 4865 }, { "epoch": 0.6431403647898494, "grad_norm": 0.8136960147405441, "learning_rate": 4.0648733303953326e-06, "loss": 0.5593, "mean_token_accuracy": 0.816277027130127, "num_tokens": 155017449.0, "step": 4866 }, { "epoch": 0.6432725350251124, "grad_norm": 0.7784657389186053, "learning_rate": 4.06290565221134e-06, "loss": 0.4988, "mean_token_accuracy": 0.8337962627410889, "num_tokens": 155082985.0, "step": 4867 }, { "epoch": 0.6434047052603754, "grad_norm": 0.8197744071722689, "learning_rate": 4.060938279954728e-06, "loss": 0.571, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 155148521.0, "step": 4868 }, { "epoch": 0.6435368754956384, "grad_norm": 0.817963317610075, "learning_rate": 4.058971214044314e-06, "loss": 0.5453, "mean_token_accuracy": 0.818474531173706, "num_tokens": 155214057.0, "step": 4869 }, { "epoch": 0.6436690457309014, "grad_norm": 0.8756169410552519, "learning_rate": 4.057004454898841e-06, "loss": 0.5861, "mean_token_accuracy": 0.8098370432853699, "num_tokens": 155279593.0, "step": 4870 }, { "epoch": 0.6438012159661645, "grad_norm": 0.8528166409061811, "learning_rate": 4.055038002936988e-06, "loss": 0.6132, "mean_token_accuracy": 0.7964228987693787, "num_tokens": 155345129.0, "step": 4871 }, { "epoch": 0.6439333862014275, "grad_norm": 0.7625581191836102, "learning_rate": 4.0530718585773796e-06, "loss": 0.5761, "mean_token_accuracy": 0.8101422190666199, "num_tokens": 155410665.0, "step": 4872 }, { "epoch": 0.6440655564366905, "grad_norm": 0.9374729157136334, "learning_rate": 4.0511060222385635e-06, "loss": 0.6377, "mean_token_accuracy": 0.7890520095825195, "num_tokens": 155476201.0, "step": 4873 }, { "epoch": 0.6441977266719535, "grad_norm": 0.8094088260036917, "learning_rate": 4.049140494339022e-06, "loss": 0.5318, "mean_token_accuracy": 0.8245025277137756, "num_tokens": 155541737.0, "step": 4874 }, { "epoch": 0.6443298969072165, "grad_norm": 0.7815954095351171, "learning_rate": 4.047175275297178e-06, "loss": 0.5532, "mean_token_accuracy": 0.8144304752349854, "num_tokens": 155607273.0, "step": 4875 }, { "epoch": 0.6444620671424796, "grad_norm": 0.8690283210357143, "learning_rate": 4.045210365531387e-06, "loss": 0.5326, "mean_token_accuracy": 0.8136942386627197, "num_tokens": 155666839.0, "step": 4876 }, { "epoch": 0.6445942373777426, "grad_norm": 0.8207357599069702, "learning_rate": 4.043245765459934e-06, "loss": 0.5955, "mean_token_accuracy": 0.8045721054077148, "num_tokens": 155732375.0, "step": 4877 }, { "epoch": 0.6447264076130056, "grad_norm": 0.8652270583870654, "learning_rate": 4.0412814755010424e-06, "loss": 0.5815, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 155797911.0, "step": 4878 }, { "epoch": 0.6448585778482686, "grad_norm": 0.8383045966519026, "learning_rate": 4.039317496072869e-06, "loss": 0.5667, "mean_token_accuracy": 0.812995970249176, "num_tokens": 155863447.0, "step": 4879 }, { "epoch": 0.6449907480835316, "grad_norm": 0.8662969462341293, "learning_rate": 4.037353827593504e-06, "loss": 0.5932, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 155928983.0, "step": 4880 }, { "epoch": 0.6451229183187946, "grad_norm": 0.8022014399806274, "learning_rate": 4.03539047048097e-06, "loss": 0.5419, "mean_token_accuracy": 0.818199872970581, "num_tokens": 155994519.0, "step": 4881 }, { "epoch": 0.6452550885540577, "grad_norm": 0.7545171165123893, "learning_rate": 4.033427425153228e-06, "loss": 0.5172, "mean_token_accuracy": 0.8291875123977661, "num_tokens": 156060055.0, "step": 4882 }, { "epoch": 0.6453872587893207, "grad_norm": 0.884916193723701, "learning_rate": 4.031464692028168e-06, "loss": 0.5606, "mean_token_accuracy": 0.8142321109771729, "num_tokens": 156125591.0, "step": 4883 }, { "epoch": 0.6455194290245837, "grad_norm": 0.8154330273660816, "learning_rate": 4.029502271523611e-06, "loss": 0.5437, "mean_token_accuracy": 0.8191155195236206, "num_tokens": 156191127.0, "step": 4884 }, { "epoch": 0.6456515992598467, "grad_norm": 0.8736062907453133, "learning_rate": 4.027540164057321e-06, "loss": 0.6043, "mean_token_accuracy": 0.7985135912895203, "num_tokens": 156256663.0, "step": 4885 }, { "epoch": 0.6457837694951097, "grad_norm": 0.7982015124753378, "learning_rate": 4.025578370046987e-06, "loss": 0.5649, "mean_token_accuracy": 0.8121718764305115, "num_tokens": 156322199.0, "step": 4886 }, { "epoch": 0.6459159397303728, "grad_norm": 0.8534145403910395, "learning_rate": 4.023616889910233e-06, "loss": 0.6268, "mean_token_accuracy": 0.7930045127868652, "num_tokens": 156387735.0, "step": 4887 }, { "epoch": 0.6460481099656358, "grad_norm": 0.8192609839363861, "learning_rate": 4.02165572406462e-06, "loss": 0.5191, "mean_token_accuracy": 0.8306220173835754, "num_tokens": 156453271.0, "step": 4888 }, { "epoch": 0.6461802802008988, "grad_norm": 0.9031634109885165, "learning_rate": 4.019694872927637e-06, "loss": 0.6043, "mean_token_accuracy": 0.8011537194252014, "num_tokens": 156518807.0, "step": 4889 }, { "epoch": 0.6463124504361618, "grad_norm": 0.8564498091820094, "learning_rate": 4.017734336916709e-06, "loss": 0.579, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 156584343.0, "step": 4890 }, { "epoch": 0.6464446206714248, "grad_norm": 0.8835405299617558, "learning_rate": 4.015774116449194e-06, "loss": 0.6018, "mean_token_accuracy": 0.8004211783409119, "num_tokens": 156649879.0, "step": 4891 }, { "epoch": 0.6465767909066878, "grad_norm": 0.8681178023338509, "learning_rate": 4.013814211942381e-06, "loss": 0.5436, "mean_token_accuracy": 0.8220760822296143, "num_tokens": 156715415.0, "step": 4892 }, { "epoch": 0.6467089611419509, "grad_norm": 0.8219734357379994, "learning_rate": 4.011854623813493e-06, "loss": 0.5644, "mean_token_accuracy": 0.8131332993507385, "num_tokens": 156780951.0, "step": 4893 }, { "epoch": 0.6468411313772139, "grad_norm": 0.818022806467171, "learning_rate": 4.009895352479684e-06, "loss": 0.537, "mean_token_accuracy": 0.8222744464874268, "num_tokens": 156846487.0, "step": 4894 }, { "epoch": 0.6469733016124769, "grad_norm": 0.8105894635245635, "learning_rate": 4.0079363983580445e-06, "loss": 0.5414, "mean_token_accuracy": 0.8213740587234497, "num_tokens": 156912023.0, "step": 4895 }, { "epoch": 0.6471054718477399, "grad_norm": 0.8645606997655342, "learning_rate": 4.0059777618655935e-06, "loss": 0.5933, "mean_token_accuracy": 0.8042210936546326, "num_tokens": 156977559.0, "step": 4896 }, { "epoch": 0.6472376420830029, "grad_norm": 0.8750202849405168, "learning_rate": 4.0040194434192835e-06, "loss": 0.642, "mean_token_accuracy": 0.7884415984153748, "num_tokens": 157043095.0, "step": 4897 }, { "epoch": 0.647369812318266, "grad_norm": 0.8393464941672951, "learning_rate": 4.002061443436e-06, "loss": 0.6105, "mean_token_accuracy": 0.7969722747802734, "num_tokens": 157108631.0, "step": 4898 }, { "epoch": 0.647501982553529, "grad_norm": 0.8097520536345765, "learning_rate": 4.000103762332564e-06, "loss": 0.557, "mean_token_accuracy": 0.8151935338973999, "num_tokens": 157174167.0, "step": 4899 }, { "epoch": 0.647634152788792, "grad_norm": 0.8869676946949517, "learning_rate": 3.9981464005257166e-06, "loss": 0.6006, "mean_token_accuracy": 0.8002533316612244, "num_tokens": 157239703.0, "step": 4900 }, { "epoch": 0.647766323024055, "grad_norm": 0.859817224342072, "learning_rate": 3.9961893584321466e-06, "loss": 0.5585, "mean_token_accuracy": 0.8155597448348999, "num_tokens": 157305239.0, "step": 4901 }, { "epoch": 0.647898493259318, "grad_norm": 0.8105737935875921, "learning_rate": 3.994232636468464e-06, "loss": 0.5409, "mean_token_accuracy": 0.8214198350906372, "num_tokens": 157370775.0, "step": 4902 }, { "epoch": 0.648030663494581, "grad_norm": 0.8302746344486238, "learning_rate": 3.992276235051216e-06, "loss": 0.6, "mean_token_accuracy": 0.7995360493659973, "num_tokens": 157436311.0, "step": 4903 }, { "epoch": 0.6481628337298441, "grad_norm": 0.8644601115890106, "learning_rate": 3.9903201545968785e-06, "loss": 0.6045, "mean_token_accuracy": 0.8006501197814941, "num_tokens": 157501847.0, "step": 4904 }, { "epoch": 0.6482950039651071, "grad_norm": 0.8850573701015051, "learning_rate": 3.988364395521859e-06, "loss": 0.5749, "mean_token_accuracy": 0.8097606897354126, "num_tokens": 157567383.0, "step": 4905 }, { "epoch": 0.6484271742003701, "grad_norm": 0.8247415801359983, "learning_rate": 3.986408958242501e-06, "loss": 0.5517, "mean_token_accuracy": 0.8207026124000549, "num_tokens": 157632919.0, "step": 4906 }, { "epoch": 0.6485593444356331, "grad_norm": 0.8635028992638605, "learning_rate": 3.984453843175072e-06, "loss": 0.5833, "mean_token_accuracy": 0.8070290684700012, "num_tokens": 157698455.0, "step": 4907 }, { "epoch": 0.6486915146708961, "grad_norm": 0.8279038623419221, "learning_rate": 3.982499050735779e-06, "loss": 0.5756, "mean_token_accuracy": 0.808814525604248, "num_tokens": 157763991.0, "step": 4908 }, { "epoch": 0.6488236849061592, "grad_norm": 0.7583070716259754, "learning_rate": 3.980544581340754e-06, "loss": 0.5632, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 157829527.0, "step": 4909 }, { "epoch": 0.6489558551414222, "grad_norm": 0.8720725426789018, "learning_rate": 3.978590435406063e-06, "loss": 0.647, "mean_token_accuracy": 0.7845043540000916, "num_tokens": 157895063.0, "step": 4910 }, { "epoch": 0.6490880253766852, "grad_norm": 0.8060652049613817, "learning_rate": 3.976636613347705e-06, "loss": 0.5813, "mean_token_accuracy": 0.8052435517311096, "num_tokens": 157960599.0, "step": 4911 }, { "epoch": 0.6492201956119482, "grad_norm": 0.8834762619189899, "learning_rate": 3.974683115581607e-06, "loss": 0.5916, "mean_token_accuracy": 0.803427517414093, "num_tokens": 158026135.0, "step": 4912 }, { "epoch": 0.6493523658472112, "grad_norm": 0.8842278541888299, "learning_rate": 3.972729942523625e-06, "loss": 0.6486, "mean_token_accuracy": 0.7871291637420654, "num_tokens": 158091671.0, "step": 4913 }, { "epoch": 0.6494845360824743, "grad_norm": 0.8490199345077527, "learning_rate": 3.970777094589553e-06, "loss": 0.5728, "mean_token_accuracy": 0.8112714886665344, "num_tokens": 158157207.0, "step": 4914 }, { "epoch": 0.6496167063177373, "grad_norm": 0.8525856805536498, "learning_rate": 3.9688245721951104e-06, "loss": 0.5664, "mean_token_accuracy": 0.8137285113334656, "num_tokens": 158222743.0, "step": 4915 }, { "epoch": 0.6497488765530003, "grad_norm": 0.8227630937498718, "learning_rate": 3.966872375755944e-06, "loss": 0.5662, "mean_token_accuracy": 0.812065064907074, "num_tokens": 158288279.0, "step": 4916 }, { "epoch": 0.6498810467882633, "grad_norm": 0.8629323851221361, "learning_rate": 3.9649205056876435e-06, "loss": 0.6381, "mean_token_accuracy": 0.7890520095825195, "num_tokens": 158353815.0, "step": 4917 }, { "epoch": 0.6500132170235263, "grad_norm": 0.7928259453734627, "learning_rate": 3.962968962405717e-06, "loss": 0.5463, "mean_token_accuracy": 0.821938693523407, "num_tokens": 158419351.0, "step": 4918 }, { "epoch": 0.6501453872587893, "grad_norm": 0.8299451084545723, "learning_rate": 3.961017746325604e-06, "loss": 0.5176, "mean_token_accuracy": 0.8278445601463318, "num_tokens": 158484887.0, "step": 4919 }, { "epoch": 0.6502775574940524, "grad_norm": 0.8411023089645803, "learning_rate": 3.959066857862685e-06, "loss": 0.5813, "mean_token_accuracy": 0.8062812685966492, "num_tokens": 158550423.0, "step": 4920 }, { "epoch": 0.6504097277293154, "grad_norm": 0.8567381457548826, "learning_rate": 3.957116297432258e-06, "loss": 0.5788, "mean_token_accuracy": 0.808631420135498, "num_tokens": 158615959.0, "step": 4921 }, { "epoch": 0.6505418979645784, "grad_norm": 0.7918701309750484, "learning_rate": 3.955166065449558e-06, "loss": 0.5064, "mean_token_accuracy": 0.8309882879257202, "num_tokens": 158681495.0, "step": 4922 }, { "epoch": 0.6506740681998414, "grad_norm": 0.8491799109313454, "learning_rate": 3.95321616232975e-06, "loss": 0.5313, "mean_token_accuracy": 0.8244872689247131, "num_tokens": 158747031.0, "step": 4923 }, { "epoch": 0.6508062384351044, "grad_norm": 0.7827496539264898, "learning_rate": 3.951266588487924e-06, "loss": 0.547, "mean_token_accuracy": 0.821847140789032, "num_tokens": 158812567.0, "step": 4924 }, { "epoch": 0.6509384086703675, "grad_norm": 0.8354483469088823, "learning_rate": 3.949317344339109e-06, "loss": 0.5836, "mean_token_accuracy": 0.8078989386558533, "num_tokens": 158878103.0, "step": 4925 }, { "epoch": 0.6510705789056305, "grad_norm": 0.831678803124961, "learning_rate": 3.947368430298253e-06, "loss": 0.5719, "mean_token_accuracy": 0.8092265725135803, "num_tokens": 158943639.0, "step": 4926 }, { "epoch": 0.6512027491408935, "grad_norm": 0.8328217272915814, "learning_rate": 3.945419846780243e-06, "loss": 0.5516, "mean_token_accuracy": 0.8145525455474854, "num_tokens": 159009175.0, "step": 4927 }, { "epoch": 0.6513349193761565, "grad_norm": 0.8762957784100526, "learning_rate": 3.94347159419989e-06, "loss": 0.5952, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 159074711.0, "step": 4928 }, { "epoch": 0.6514670896114195, "grad_norm": 0.8223595634929183, "learning_rate": 3.941523672971936e-06, "loss": 0.6035, "mean_token_accuracy": 0.8011842370033264, "num_tokens": 159140247.0, "step": 4929 }, { "epoch": 0.6515992598466825, "grad_norm": 0.8742396869725367, "learning_rate": 3.939576083511054e-06, "loss": 0.5993, "mean_token_accuracy": 0.8009705543518066, "num_tokens": 159205783.0, "step": 4930 }, { "epoch": 0.6517314300819456, "grad_norm": 0.7569326842668825, "learning_rate": 3.937628826231845e-06, "loss": 0.5171, "mean_token_accuracy": 0.8303321003913879, "num_tokens": 159271319.0, "step": 4931 }, { "epoch": 0.6518636003172086, "grad_norm": 0.8145572104231095, "learning_rate": 3.935681901548838e-06, "loss": 0.5278, "mean_token_accuracy": 0.8235563635826111, "num_tokens": 159336855.0, "step": 4932 }, { "epoch": 0.6519957705524716, "grad_norm": 0.8854547224900117, "learning_rate": 3.933735309876494e-06, "loss": 0.6229, "mean_token_accuracy": 0.7917989492416382, "num_tokens": 159402391.0, "step": 4933 }, { "epoch": 0.6521279407877346, "grad_norm": 0.8274189095551144, "learning_rate": 3.931789051629203e-06, "loss": 0.5358, "mean_token_accuracy": 0.8230222463607788, "num_tokens": 159467927.0, "step": 4934 }, { "epoch": 0.6522601110229976, "grad_norm": 0.8344069688187277, "learning_rate": 3.929843127221277e-06, "loss": 0.5883, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 159533463.0, "step": 4935 }, { "epoch": 0.6523922812582607, "grad_norm": 0.8573899764194367, "learning_rate": 3.927897537066969e-06, "loss": 0.5617, "mean_token_accuracy": 0.8138658404350281, "num_tokens": 159598999.0, "step": 4936 }, { "epoch": 0.6525244514935237, "grad_norm": 0.8230713458101084, "learning_rate": 3.925952281580453e-06, "loss": 0.5915, "mean_token_accuracy": 0.8051214814186096, "num_tokens": 159664535.0, "step": 4937 }, { "epoch": 0.6526566217287867, "grad_norm": 0.9149426999479284, "learning_rate": 3.924007361175832e-06, "loss": 0.6325, "mean_token_accuracy": 0.7904865145683289, "num_tokens": 159730071.0, "step": 4938 }, { "epoch": 0.6527887919640497, "grad_norm": 0.7575300916854785, "learning_rate": 3.922062776267138e-06, "loss": 0.5306, "mean_token_accuracy": 0.8241972923278809, "num_tokens": 159795607.0, "step": 4939 }, { "epoch": 0.6529209621993127, "grad_norm": 0.8100776897649318, "learning_rate": 3.920118527268337e-06, "loss": 0.5807, "mean_token_accuracy": 0.8056708574295044, "num_tokens": 159861143.0, "step": 4940 }, { "epoch": 0.6530531324345757, "grad_norm": 0.8204177580770237, "learning_rate": 3.918174614593314e-06, "loss": 0.5543, "mean_token_accuracy": 0.8155902624130249, "num_tokens": 159926679.0, "step": 4941 }, { "epoch": 0.6531853026698388, "grad_norm": 0.7755778051559157, "learning_rate": 3.916231038655889e-06, "loss": 0.5236, "mean_token_accuracy": 0.8244109153747559, "num_tokens": 159992215.0, "step": 4942 }, { "epoch": 0.6533174729051018, "grad_norm": 0.8042781270497726, "learning_rate": 3.91428779986981e-06, "loss": 0.6002, "mean_token_accuracy": 0.802008330821991, "num_tokens": 160057751.0, "step": 4943 }, { "epoch": 0.6534496431403648, "grad_norm": 0.891992687381938, "learning_rate": 3.912344898648751e-06, "loss": 0.5963, "mean_token_accuracy": 0.8029391765594482, "num_tokens": 160123287.0, "step": 4944 }, { "epoch": 0.6535818133756278, "grad_norm": 0.8116618768190778, "learning_rate": 3.9104023354063145e-06, "loss": 0.5728, "mean_token_accuracy": 0.8073800802230835, "num_tokens": 160188823.0, "step": 4945 }, { "epoch": 0.6537139836108908, "grad_norm": 0.8010638697166655, "learning_rate": 3.908460110556034e-06, "loss": 0.5586, "mean_token_accuracy": 0.816307544708252, "num_tokens": 160254359.0, "step": 4946 }, { "epoch": 0.6538461538461539, "grad_norm": 0.8533872618752971, "learning_rate": 3.9065182245113655e-06, "loss": 0.5952, "mean_token_accuracy": 0.8029544353485107, "num_tokens": 160319895.0, "step": 4947 }, { "epoch": 0.6539783240814169, "grad_norm": 0.914705622980637, "learning_rate": 3.904576677685698e-06, "loss": 0.6117, "mean_token_accuracy": 0.7981778979301453, "num_tokens": 160385431.0, "step": 4948 }, { "epoch": 0.6541104943166799, "grad_norm": 0.9057617629059798, "learning_rate": 3.902635470492345e-06, "loss": 0.6022, "mean_token_accuracy": 0.7984678149223328, "num_tokens": 160450967.0, "step": 4949 }, { "epoch": 0.6542426645519429, "grad_norm": 0.7951749879307348, "learning_rate": 3.900694603344552e-06, "loss": 0.5469, "mean_token_accuracy": 0.8203973770141602, "num_tokens": 160516503.0, "step": 4950 }, { "epoch": 0.6543748347872059, "grad_norm": 0.8389772017624061, "learning_rate": 3.898754076655483e-06, "loss": 0.5722, "mean_token_accuracy": 0.8093028664588928, "num_tokens": 160582039.0, "step": 4951 }, { "epoch": 0.654507005022469, "grad_norm": 0.8075780323357188, "learning_rate": 3.896813890838242e-06, "loss": 0.5612, "mean_token_accuracy": 0.8150103688240051, "num_tokens": 160647575.0, "step": 4952 }, { "epoch": 0.654639175257732, "grad_norm": 0.82267788512695, "learning_rate": 3.894874046305853e-06, "loss": 0.5684, "mean_token_accuracy": 0.8124160766601562, "num_tokens": 160713111.0, "step": 4953 }, { "epoch": 0.654771345492995, "grad_norm": 0.903777355096979, "learning_rate": 3.8929345434712655e-06, "loss": 0.6143, "mean_token_accuracy": 0.7984220385551453, "num_tokens": 160778647.0, "step": 4954 }, { "epoch": 0.654903515728258, "grad_norm": 0.8585232247820991, "learning_rate": 3.890995382747359e-06, "loss": 0.605, "mean_token_accuracy": 0.7961634993553162, "num_tokens": 160844183.0, "step": 4955 }, { "epoch": 0.655035685963521, "grad_norm": 0.8257658870946202, "learning_rate": 3.889056564546944e-06, "loss": 0.6641, "mean_token_accuracy": 0.7834360599517822, "num_tokens": 160909719.0, "step": 4956 }, { "epoch": 0.655167856198784, "grad_norm": 0.8712346033041847, "learning_rate": 3.887118089282752e-06, "loss": 0.6046, "mean_token_accuracy": 0.79891037940979, "num_tokens": 160975255.0, "step": 4957 }, { "epoch": 0.6553000264340471, "grad_norm": 0.8001997138725402, "learning_rate": 3.885179957367442e-06, "loss": 0.5327, "mean_token_accuracy": 0.8238157629966736, "num_tokens": 161040791.0, "step": 4958 }, { "epoch": 0.6554321966693101, "grad_norm": 0.8179044671484705, "learning_rate": 3.883242169213607e-06, "loss": 0.591, "mean_token_accuracy": 0.8058387041091919, "num_tokens": 161106327.0, "step": 4959 }, { "epoch": 0.6555643669045731, "grad_norm": 0.8797553280036379, "learning_rate": 3.881304725233758e-06, "loss": 0.6074, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 161171863.0, "step": 4960 }, { "epoch": 0.6556965371398361, "grad_norm": 0.8032104478655445, "learning_rate": 3.879367625840336e-06, "loss": 0.578, "mean_token_accuracy": 0.8073037266731262, "num_tokens": 161237399.0, "step": 4961 }, { "epoch": 0.6558287073750991, "grad_norm": 0.8370591894700096, "learning_rate": 3.877430871445712e-06, "loss": 0.5633, "mean_token_accuracy": 0.8116072416305542, "num_tokens": 161302935.0, "step": 4962 }, { "epoch": 0.6559608776103621, "grad_norm": 0.8675926360405939, "learning_rate": 3.875494462462176e-06, "loss": 0.6133, "mean_token_accuracy": 0.7965602278709412, "num_tokens": 161368471.0, "step": 4963 }, { "epoch": 0.6560930478456252, "grad_norm": 0.7748972597227443, "learning_rate": 3.873558399301953e-06, "loss": 0.543, "mean_token_accuracy": 0.8204584121704102, "num_tokens": 161434007.0, "step": 4964 }, { "epoch": 0.6562252180808882, "grad_norm": 0.9343639460964237, "learning_rate": 3.871622682377191e-06, "loss": 0.6208, "mean_token_accuracy": 0.7981168627738953, "num_tokens": 161499543.0, "step": 4965 }, { "epoch": 0.6563573883161512, "grad_norm": 0.8708109440501017, "learning_rate": 3.86968731209996e-06, "loss": 0.6352, "mean_token_accuracy": 0.7881211042404175, "num_tokens": 161565079.0, "step": 4966 }, { "epoch": 0.6564895585514142, "grad_norm": 0.8256353027366791, "learning_rate": 3.867752288882263e-06, "loss": 0.636, "mean_token_accuracy": 0.7892961502075195, "num_tokens": 161630615.0, "step": 4967 }, { "epoch": 0.6566217287866772, "grad_norm": 0.8553540252470614, "learning_rate": 3.865817613136021e-06, "loss": 0.589, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 161696151.0, "step": 4968 }, { "epoch": 0.6567538990219403, "grad_norm": 0.8105615327973247, "learning_rate": 3.863883285273093e-06, "loss": 0.6157, "mean_token_accuracy": 0.7983915209770203, "num_tokens": 161761687.0, "step": 4969 }, { "epoch": 0.6568860692572033, "grad_norm": 0.8953650841076062, "learning_rate": 3.861949305705255e-06, "loss": 0.6716, "mean_token_accuracy": 0.7757447361946106, "num_tokens": 161827223.0, "step": 4970 }, { "epoch": 0.6570182394924663, "grad_norm": 0.832798393613056, "learning_rate": 3.8600156748442075e-06, "loss": 0.5407, "mean_token_accuracy": 0.8197259306907654, "num_tokens": 161892759.0, "step": 4971 }, { "epoch": 0.6571504097277293, "grad_norm": 0.8614213818922009, "learning_rate": 3.858082393101583e-06, "loss": 0.5881, "mean_token_accuracy": 0.8021456599235535, "num_tokens": 161958295.0, "step": 4972 }, { "epoch": 0.6572825799629923, "grad_norm": 0.7841079131298239, "learning_rate": 3.856149460888937e-06, "loss": 0.5431, "mean_token_accuracy": 0.8172537088394165, "num_tokens": 162023831.0, "step": 4973 }, { "epoch": 0.6574147501982553, "grad_norm": 0.8111325634805607, "learning_rate": 3.854216878617748e-06, "loss": 0.5743, "mean_token_accuracy": 0.8096386194229126, "num_tokens": 162089367.0, "step": 4974 }, { "epoch": 0.6575469204335184, "grad_norm": 0.8202808961430328, "learning_rate": 3.8522846466994255e-06, "loss": 0.543, "mean_token_accuracy": 0.8193291425704956, "num_tokens": 162154903.0, "step": 4975 }, { "epoch": 0.6576790906687814, "grad_norm": 0.9642206449149251, "learning_rate": 3.8503527655452996e-06, "loss": 0.6378, "mean_token_accuracy": 0.7879990339279175, "num_tokens": 162220439.0, "step": 4976 }, { "epoch": 0.6578112609040444, "grad_norm": 0.7443323845613552, "learning_rate": 3.848421235566627e-06, "loss": 0.504, "mean_token_accuracy": 0.8349713087081909, "num_tokens": 162285975.0, "step": 4977 }, { "epoch": 0.6579434311393074, "grad_norm": 0.8556553220331682, "learning_rate": 3.846490057174592e-06, "loss": 0.587, "mean_token_accuracy": 0.8059913516044617, "num_tokens": 162351511.0, "step": 4978 }, { "epoch": 0.6580756013745704, "grad_norm": 0.7839035557474808, "learning_rate": 3.844559230780301e-06, "loss": 0.5887, "mean_token_accuracy": 0.8062507510185242, "num_tokens": 162417047.0, "step": 4979 }, { "epoch": 0.6582077716098335, "grad_norm": 0.7556480252743042, "learning_rate": 3.842628756794786e-06, "loss": 0.5662, "mean_token_accuracy": 0.8115156888961792, "num_tokens": 162482583.0, "step": 4980 }, { "epoch": 0.6583399418450965, "grad_norm": 0.8621848527767011, "learning_rate": 3.840698635629006e-06, "loss": 0.595, "mean_token_accuracy": 0.8015352487564087, "num_tokens": 162548119.0, "step": 4981 }, { "epoch": 0.6584721120803595, "grad_norm": 0.8457967335713367, "learning_rate": 3.838768867693843e-06, "loss": 0.6259, "mean_token_accuracy": 0.7916310429573059, "num_tokens": 162613655.0, "step": 4982 }, { "epoch": 0.6586042823156225, "grad_norm": 0.8816555842781328, "learning_rate": 3.836839453400103e-06, "loss": 0.5654, "mean_token_accuracy": 0.8127976059913635, "num_tokens": 162679191.0, "step": 4983 }, { "epoch": 0.6587364525508855, "grad_norm": 0.8970295559497471, "learning_rate": 3.83491039315852e-06, "loss": 0.5844, "mean_token_accuracy": 0.8066017627716064, "num_tokens": 162744727.0, "step": 4984 }, { "epoch": 0.6588686227861486, "grad_norm": 0.839039709309914, "learning_rate": 3.832981687379749e-06, "loss": 0.5472, "mean_token_accuracy": 0.8181540966033936, "num_tokens": 162810263.0, "step": 4985 }, { "epoch": 0.6590007930214116, "grad_norm": 0.7754031688503507, "learning_rate": 3.831053336474373e-06, "loss": 0.5502, "mean_token_accuracy": 0.8185813426971436, "num_tokens": 162875799.0, "step": 4986 }, { "epoch": 0.6591329632566746, "grad_norm": 0.8755965043428734, "learning_rate": 3.829125340852894e-06, "loss": 0.5991, "mean_token_accuracy": 0.7999328374862671, "num_tokens": 162941335.0, "step": 4987 }, { "epoch": 0.6592651334919376, "grad_norm": 0.9219707552483789, "learning_rate": 3.827197700925747e-06, "loss": 0.593, "mean_token_accuracy": 0.8059608340263367, "num_tokens": 163006871.0, "step": 4988 }, { "epoch": 0.6593973037272006, "grad_norm": 0.8760131132177404, "learning_rate": 3.825270417103283e-06, "loss": 0.5599, "mean_token_accuracy": 0.8138658404350281, "num_tokens": 163072407.0, "step": 4989 }, { "epoch": 0.6595294739624636, "grad_norm": 0.7958720719608428, "learning_rate": 3.82334348979578e-06, "loss": 0.5586, "mean_token_accuracy": 0.8133164644241333, "num_tokens": 163137943.0, "step": 4990 }, { "epoch": 0.6596616441977267, "grad_norm": 0.820409627994777, "learning_rate": 3.821416919413444e-06, "loss": 0.5781, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 163203479.0, "step": 4991 }, { "epoch": 0.6597938144329897, "grad_norm": 0.8327913398675447, "learning_rate": 3.8194907063664e-06, "loss": 0.5565, "mean_token_accuracy": 0.8136826753616333, "num_tokens": 163269015.0, "step": 4992 }, { "epoch": 0.6599259846682527, "grad_norm": 0.8452543533864727, "learning_rate": 3.817564851064696e-06, "loss": 0.5594, "mean_token_accuracy": 0.816277027130127, "num_tokens": 163334551.0, "step": 4993 }, { "epoch": 0.6600581549035157, "grad_norm": 0.8509282064904714, "learning_rate": 3.81563935391831e-06, "loss": 0.5719, "mean_token_accuracy": 0.8094860315322876, "num_tokens": 163400087.0, "step": 4994 }, { "epoch": 0.6601903251387787, "grad_norm": 0.8721438450816411, "learning_rate": 3.8137142153371386e-06, "loss": 0.534, "mean_token_accuracy": 0.822778046131134, "num_tokens": 163465623.0, "step": 4995 }, { "epoch": 0.6603224953740418, "grad_norm": 0.9520739669371887, "learning_rate": 3.8117894357310025e-06, "loss": 0.6283, "mean_token_accuracy": 0.7904559969902039, "num_tokens": 163531159.0, "step": 4996 }, { "epoch": 0.6604546656093048, "grad_norm": 0.8948117096548935, "learning_rate": 3.8098650155096483e-06, "loss": 0.5782, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 163596695.0, "step": 4997 }, { "epoch": 0.6605868358445678, "grad_norm": 0.818649569609967, "learning_rate": 3.8079409550827462e-06, "loss": 0.5687, "mean_token_accuracy": 0.8115462064743042, "num_tokens": 163662231.0, "step": 4998 }, { "epoch": 0.6607190060798308, "grad_norm": 0.8130888257269367, "learning_rate": 3.8060172548598873e-06, "loss": 0.5692, "mean_token_accuracy": 0.8119277358055115, "num_tokens": 163727767.0, "step": 4999 }, { "epoch": 0.6608511763150938, "grad_norm": 0.8495567261967699, "learning_rate": 3.8040939152505864e-06, "loss": 0.6341, "mean_token_accuracy": 0.7884110808372498, "num_tokens": 163793303.0, "step": 5000 }, { "epoch": 0.6609833465503568, "grad_norm": 0.8351094415824869, "learning_rate": 3.8021709366642824e-06, "loss": 0.5437, "mean_token_accuracy": 0.8169332146644592, "num_tokens": 163858839.0, "step": 5001 }, { "epoch": 0.6611155167856199, "grad_norm": 0.8671164487258679, "learning_rate": 3.800248319510338e-06, "loss": 0.5577, "mean_token_accuracy": 0.8144457340240479, "num_tokens": 163924375.0, "step": 5002 }, { "epoch": 0.6612476870208829, "grad_norm": 0.9273861160795973, "learning_rate": 3.798326064198038e-06, "loss": 0.5577, "mean_token_accuracy": 0.812843382358551, "num_tokens": 163989911.0, "step": 5003 }, { "epoch": 0.6613798572561459, "grad_norm": 0.94456837079981, "learning_rate": 3.796404171136592e-06, "loss": 0.589, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 164055447.0, "step": 5004 }, { "epoch": 0.6615120274914089, "grad_norm": 0.9336614624224515, "learning_rate": 3.794482640735129e-06, "loss": 0.6113, "mean_token_accuracy": 0.7997649908065796, "num_tokens": 164120983.0, "step": 5005 }, { "epoch": 0.6616441977266719, "grad_norm": 0.8625727955275193, "learning_rate": 3.7925614734027004e-06, "loss": 0.6037, "mean_token_accuracy": 0.7981168627738953, "num_tokens": 164186519.0, "step": 5006 }, { "epoch": 0.661776367961935, "grad_norm": 0.7707212704369252, "learning_rate": 3.7906406695482876e-06, "loss": 0.5119, "mean_token_accuracy": 0.8286381363868713, "num_tokens": 164252055.0, "step": 5007 }, { "epoch": 0.661908538197198, "grad_norm": 1.0299111156984784, "learning_rate": 3.788720229580788e-06, "loss": 0.605, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 164317591.0, "step": 5008 }, { "epoch": 0.662040708432461, "grad_norm": 0.9919698415950008, "learning_rate": 3.7868001539090204e-06, "loss": 0.6383, "mean_token_accuracy": 0.7893725037574768, "num_tokens": 164383127.0, "step": 5009 }, { "epoch": 0.662172878667724, "grad_norm": 0.9013610613598915, "learning_rate": 3.784880442941734e-06, "loss": 0.5874, "mean_token_accuracy": 0.8046331405639648, "num_tokens": 164448663.0, "step": 5010 }, { "epoch": 0.662305048902987, "grad_norm": 0.8574150900060457, "learning_rate": 3.7829610970875907e-06, "loss": 0.5162, "mean_token_accuracy": 0.8289891481399536, "num_tokens": 164514199.0, "step": 5011 }, { "epoch": 0.66243721913825, "grad_norm": 0.872813891422083, "learning_rate": 3.7810421167551805e-06, "loss": 0.589, "mean_token_accuracy": 0.806678056716919, "num_tokens": 164579735.0, "step": 5012 }, { "epoch": 0.6625693893735131, "grad_norm": 0.8652825606356284, "learning_rate": 3.7791235023530136e-06, "loss": 0.594, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 164645271.0, "step": 5013 }, { "epoch": 0.6627015596087761, "grad_norm": 0.8646036463272312, "learning_rate": 3.7772052542895254e-06, "loss": 0.57, "mean_token_accuracy": 0.8124771118164062, "num_tokens": 164710807.0, "step": 5014 }, { "epoch": 0.6628337298440391, "grad_norm": 0.9750946981878397, "learning_rate": 3.775287372973069e-06, "loss": 0.6577, "mean_token_accuracy": 0.7793614864349365, "num_tokens": 164776343.0, "step": 5015 }, { "epoch": 0.6629659000793021, "grad_norm": 0.9257759605925797, "learning_rate": 3.7733698588119182e-06, "loss": 0.6392, "mean_token_accuracy": 0.7874496579170227, "num_tokens": 164841879.0, "step": 5016 }, { "epoch": 0.6630980703145651, "grad_norm": 0.8747308532715742, "learning_rate": 3.7714527122142784e-06, "loss": 0.5964, "mean_token_accuracy": 0.803366482257843, "num_tokens": 164907415.0, "step": 5017 }, { "epoch": 0.6632302405498282, "grad_norm": 0.8399308337422321, "learning_rate": 3.7695359335882664e-06, "loss": 0.6245, "mean_token_accuracy": 0.792500913143158, "num_tokens": 164972951.0, "step": 5018 }, { "epoch": 0.6633624107850912, "grad_norm": 0.7743145332925642, "learning_rate": 3.7676195233419243e-06, "loss": 0.5481, "mean_token_accuracy": 0.817207932472229, "num_tokens": 165038487.0, "step": 5019 }, { "epoch": 0.6634945810203542, "grad_norm": 0.8625532510952753, "learning_rate": 3.765703481883215e-06, "loss": 0.6007, "mean_token_accuracy": 0.8029391765594482, "num_tokens": 165104023.0, "step": 5020 }, { "epoch": 0.6636267512556172, "grad_norm": 0.876922612017035, "learning_rate": 3.7637878096200277e-06, "loss": 0.6104, "mean_token_accuracy": 0.8011689782142639, "num_tokens": 165169559.0, "step": 5021 }, { "epoch": 0.6637589214908802, "grad_norm": 0.8650923930856216, "learning_rate": 3.7618725069601634e-06, "loss": 0.6139, "mean_token_accuracy": 0.7951868176460266, "num_tokens": 165235095.0, "step": 5022 }, { "epoch": 0.6638910917261432, "grad_norm": 0.7969632768134779, "learning_rate": 3.759957574311355e-06, "loss": 0.5563, "mean_token_accuracy": 0.816338062286377, "num_tokens": 165300631.0, "step": 5023 }, { "epoch": 0.6640232619614063, "grad_norm": 0.8901209012346769, "learning_rate": 3.7580430120812483e-06, "loss": 0.6001, "mean_token_accuracy": 0.8002533316612244, "num_tokens": 165366167.0, "step": 5024 }, { "epoch": 0.6641554321966693, "grad_norm": 0.8189646870164605, "learning_rate": 3.7561288206774136e-06, "loss": 0.5663, "mean_token_accuracy": 0.8125076293945312, "num_tokens": 165431703.0, "step": 5025 }, { "epoch": 0.6642876024319323, "grad_norm": 0.8704745235832411, "learning_rate": 3.7542150005073462e-06, "loss": 0.6092, "mean_token_accuracy": 0.79872727394104, "num_tokens": 165497239.0, "step": 5026 }, { "epoch": 0.6644197726671953, "grad_norm": 0.7898333188169798, "learning_rate": 3.7523015519784555e-06, "loss": 0.5331, "mean_token_accuracy": 0.8223812580108643, "num_tokens": 165562775.0, "step": 5027 }, { "epoch": 0.6645519429024583, "grad_norm": 0.968460626359598, "learning_rate": 3.7503884754980748e-06, "loss": 0.6507, "mean_token_accuracy": 0.7869307994842529, "num_tokens": 165628311.0, "step": 5028 }, { "epoch": 0.6646841131377214, "grad_norm": 0.8668995717636637, "learning_rate": 3.7484757714734563e-06, "loss": 0.5699, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 165693847.0, "step": 5029 }, { "epoch": 0.6648162833729844, "grad_norm": 0.8409270068518331, "learning_rate": 3.746563440311779e-06, "loss": 0.5833, "mean_token_accuracy": 0.8070443272590637, "num_tokens": 165759383.0, "step": 5030 }, { "epoch": 0.6649484536082474, "grad_norm": 0.8056950860469473, "learning_rate": 3.7446514824201363e-06, "loss": 0.5366, "mean_token_accuracy": 0.8220760822296143, "num_tokens": 165824919.0, "step": 5031 }, { "epoch": 0.6650806238435104, "grad_norm": 0.8642970167913898, "learning_rate": 3.7427398982055417e-06, "loss": 0.6275, "mean_token_accuracy": 0.7922720313072205, "num_tokens": 165890455.0, "step": 5032 }, { "epoch": 0.6652127940787734, "grad_norm": 0.8287685660268708, "learning_rate": 3.7408286880749354e-06, "loss": 0.5433, "mean_token_accuracy": 0.817482590675354, "num_tokens": 165955991.0, "step": 5033 }, { "epoch": 0.6653449643140364, "grad_norm": 0.8151866739315845, "learning_rate": 3.7389178524351745e-06, "loss": 0.5792, "mean_token_accuracy": 0.8082651495933533, "num_tokens": 166021527.0, "step": 5034 }, { "epoch": 0.6654771345492995, "grad_norm": 0.7785183874931286, "learning_rate": 3.7370073916930307e-06, "loss": 0.5635, "mean_token_accuracy": 0.8111951947212219, "num_tokens": 166087063.0, "step": 5035 }, { "epoch": 0.6656093047845625, "grad_norm": 0.8963917755847187, "learning_rate": 3.7350973062552053e-06, "loss": 0.5766, "mean_token_accuracy": 0.8080210089683533, "num_tokens": 166152599.0, "step": 5036 }, { "epoch": 0.6657414750198255, "grad_norm": 0.8754298968366429, "learning_rate": 3.733187596528317e-06, "loss": 0.5787, "mean_token_accuracy": 0.808722972869873, "num_tokens": 166218135.0, "step": 5037 }, { "epoch": 0.6658736452550885, "grad_norm": 0.750408887017762, "learning_rate": 3.7312782629188993e-06, "loss": 0.5741, "mean_token_accuracy": 0.8102795481681824, "num_tokens": 166283671.0, "step": 5038 }, { "epoch": 0.6660058154903515, "grad_norm": 0.8622569620555028, "learning_rate": 3.7293693058334103e-06, "loss": 0.6484, "mean_token_accuracy": 0.7873122692108154, "num_tokens": 166349207.0, "step": 5039 }, { "epoch": 0.6661379857256146, "grad_norm": 0.7778227618314544, "learning_rate": 3.7274607256782303e-06, "loss": 0.5524, "mean_token_accuracy": 0.8177725672721863, "num_tokens": 166414743.0, "step": 5040 }, { "epoch": 0.6662701559608776, "grad_norm": 0.8273723371134382, "learning_rate": 3.7255525228596524e-06, "loss": 0.597, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 166480279.0, "step": 5041 }, { "epoch": 0.6664023261961406, "grad_norm": 0.8813518220751297, "learning_rate": 3.7236446977838947e-06, "loss": 0.5982, "mean_token_accuracy": 0.8034427762031555, "num_tokens": 166545815.0, "step": 5042 }, { "epoch": 0.6665344964314036, "grad_norm": 0.8560327555657503, "learning_rate": 3.7217372508570937e-06, "loss": 0.6124, "mean_token_accuracy": 0.7963466048240662, "num_tokens": 166611351.0, "step": 5043 }, { "epoch": 0.6666666666666666, "grad_norm": 0.8307853585909124, "learning_rate": 3.7198301824853045e-06, "loss": 0.596, "mean_token_accuracy": 0.8042516112327576, "num_tokens": 166676887.0, "step": 5044 }, { "epoch": 0.6667988369019296, "grad_norm": 0.8305452233344572, "learning_rate": 3.717923493074501e-06, "loss": 0.5493, "mean_token_accuracy": 0.8188408017158508, "num_tokens": 166742423.0, "step": 5045 }, { "epoch": 0.6669310071371927, "grad_norm": 0.8048809290585754, "learning_rate": 3.7160171830305803e-06, "loss": 0.5829, "mean_token_accuracy": 0.808356761932373, "num_tokens": 166807959.0, "step": 5046 }, { "epoch": 0.6670631773724557, "grad_norm": 0.8507762781959782, "learning_rate": 3.7141112527593548e-06, "loss": 0.544, "mean_token_accuracy": 0.8179098963737488, "num_tokens": 166873495.0, "step": 5047 }, { "epoch": 0.6671953476077187, "grad_norm": 0.7949298713317772, "learning_rate": 3.7122057026665547e-06, "loss": 0.5667, "mean_token_accuracy": 0.8139116168022156, "num_tokens": 166939031.0, "step": 5048 }, { "epoch": 0.6673275178429817, "grad_norm": 0.8024969192234909, "learning_rate": 3.7103005331578357e-06, "loss": 0.5119, "mean_token_accuracy": 0.8310493230819702, "num_tokens": 167004567.0, "step": 5049 }, { "epoch": 0.6674596880782447, "grad_norm": 0.7649686175409838, "learning_rate": 3.7083957446387663e-06, "loss": 0.579, "mean_token_accuracy": 0.807822585105896, "num_tokens": 167070103.0, "step": 5050 }, { "epoch": 0.6675918583135078, "grad_norm": 0.8564524351862851, "learning_rate": 3.7064913375148358e-06, "loss": 0.5686, "mean_token_accuracy": 0.8133469820022583, "num_tokens": 167135639.0, "step": 5051 }, { "epoch": 0.6677240285487708, "grad_norm": 0.8073044301245379, "learning_rate": 3.7045873121914568e-06, "loss": 0.6015, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 167201175.0, "step": 5052 }, { "epoch": 0.6678561987840338, "grad_norm": 0.8009777238531911, "learning_rate": 3.702683669073952e-06, "loss": 0.5866, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 167266711.0, "step": 5053 }, { "epoch": 0.6679883690192968, "grad_norm": 0.794257025200546, "learning_rate": 3.700780408567567e-06, "loss": 0.5184, "mean_token_accuracy": 0.8267152905464172, "num_tokens": 167332247.0, "step": 5054 }, { "epoch": 0.6681205392545598, "grad_norm": 0.8778259391722767, "learning_rate": 3.6988775310774707e-06, "loss": 0.653, "mean_token_accuracy": 0.7840770483016968, "num_tokens": 167397783.0, "step": 5055 }, { "epoch": 0.6682527094898228, "grad_norm": 0.9261750521554167, "learning_rate": 3.696975037008742e-06, "loss": 0.5722, "mean_token_accuracy": 0.8121413588523865, "num_tokens": 167463319.0, "step": 5056 }, { "epoch": 0.6683848797250859, "grad_norm": 0.8281998807859576, "learning_rate": 3.6950729267663834e-06, "loss": 0.5987, "mean_token_accuracy": 0.8011079430580139, "num_tokens": 167528855.0, "step": 5057 }, { "epoch": 0.6685170499603489, "grad_norm": 0.8067769319578225, "learning_rate": 3.6931712007553143e-06, "loss": 0.5859, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 167594391.0, "step": 5058 }, { "epoch": 0.6686492201956119, "grad_norm": 0.8289341066574213, "learning_rate": 3.6912698593803743e-06, "loss": 0.5696, "mean_token_accuracy": 0.8112257122993469, "num_tokens": 167659927.0, "step": 5059 }, { "epoch": 0.6687813904308749, "grad_norm": 0.8412704407848246, "learning_rate": 3.6893689030463163e-06, "loss": 0.6011, "mean_token_accuracy": 0.8027255535125732, "num_tokens": 167725463.0, "step": 5060 }, { "epoch": 0.6689135606661379, "grad_norm": 0.8625128755902325, "learning_rate": 3.687468332157815e-06, "loss": 0.6133, "mean_token_accuracy": 0.7973690629005432, "num_tokens": 167790999.0, "step": 5061 }, { "epoch": 0.669045730901401, "grad_norm": 0.7927136935119842, "learning_rate": 3.6855681471194648e-06, "loss": 0.5451, "mean_token_accuracy": 0.8199395537376404, "num_tokens": 167856535.0, "step": 5062 }, { "epoch": 0.669177901136664, "grad_norm": 0.7879867524608914, "learning_rate": 3.6836683483357714e-06, "loss": 0.5544, "mean_token_accuracy": 0.8177573084831238, "num_tokens": 167922071.0, "step": 5063 }, { "epoch": 0.669310071371927, "grad_norm": 0.8144093106102667, "learning_rate": 3.6817689362111653e-06, "loss": 0.5777, "mean_token_accuracy": 0.8084025382995605, "num_tokens": 167987607.0, "step": 5064 }, { "epoch": 0.66944224160719, "grad_norm": 0.7964260694372173, "learning_rate": 3.6798699111499912e-06, "loss": 0.5979, "mean_token_accuracy": 0.7996429204940796, "num_tokens": 168053143.0, "step": 5065 }, { "epoch": 0.669574411842453, "grad_norm": 0.7852901330539018, "learning_rate": 3.677971273556512e-06, "loss": 0.5833, "mean_token_accuracy": 0.8070138096809387, "num_tokens": 168118679.0, "step": 5066 }, { "epoch": 0.669706582077716, "grad_norm": 0.821636843523618, "learning_rate": 3.676073023834905e-06, "loss": 0.5766, "mean_token_accuracy": 0.8094097375869751, "num_tokens": 168184215.0, "step": 5067 }, { "epoch": 0.6698387523129791, "grad_norm": 0.8248866017506038, "learning_rate": 3.674175162389273e-06, "loss": 0.5998, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 168249751.0, "step": 5068 }, { "epoch": 0.6699709225482421, "grad_norm": 0.8786646798810924, "learning_rate": 3.6722776896236266e-06, "loss": 0.6112, "mean_token_accuracy": 0.7980557680130005, "num_tokens": 168315287.0, "step": 5069 }, { "epoch": 0.6701030927835051, "grad_norm": 0.8118508757710355, "learning_rate": 3.6703806059419e-06, "loss": 0.5951, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 168380823.0, "step": 5070 }, { "epoch": 0.6702352630187681, "grad_norm": 0.8962250103224734, "learning_rate": 3.6684839117479444e-06, "loss": 0.6048, "mean_token_accuracy": 0.7967433929443359, "num_tokens": 168446359.0, "step": 5071 }, { "epoch": 0.6703674332540311, "grad_norm": 0.8004795805865534, "learning_rate": 3.666587607445523e-06, "loss": 0.5719, "mean_token_accuracy": 0.8093792200088501, "num_tokens": 168511895.0, "step": 5072 }, { "epoch": 0.6704996034892942, "grad_norm": 0.8409522233578767, "learning_rate": 3.6646916934383227e-06, "loss": 0.5837, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 168577431.0, "step": 5073 }, { "epoch": 0.6706317737245572, "grad_norm": 0.8315391239804136, "learning_rate": 3.6627961701299397e-06, "loss": 0.5475, "mean_token_accuracy": 0.8178488612174988, "num_tokens": 168642967.0, "step": 5074 }, { "epoch": 0.6707639439598202, "grad_norm": 0.809127971923302, "learning_rate": 3.6609010379238963e-06, "loss": 0.6028, "mean_token_accuracy": 0.8000091314315796, "num_tokens": 168708503.0, "step": 5075 }, { "epoch": 0.6708961141950832, "grad_norm": 0.7886082160674938, "learning_rate": 3.659006297223625e-06, "loss": 0.5424, "mean_token_accuracy": 0.8207026124000549, "num_tokens": 168774039.0, "step": 5076 }, { "epoch": 0.6710282844303462, "grad_norm": 0.7904001164204738, "learning_rate": 3.6571119484324737e-06, "loss": 0.5414, "mean_token_accuracy": 0.8174062967300415, "num_tokens": 168839575.0, "step": 5077 }, { "epoch": 0.6711604546656093, "grad_norm": 0.7483149563885213, "learning_rate": 3.6552179919537124e-06, "loss": 0.5323, "mean_token_accuracy": 0.8221065998077393, "num_tokens": 168905111.0, "step": 5078 }, { "epoch": 0.6712926249008723, "grad_norm": 0.8158221003981955, "learning_rate": 3.6533244281905256e-06, "loss": 0.5347, "mean_token_accuracy": 0.82621169090271, "num_tokens": 168970647.0, "step": 5079 }, { "epoch": 0.6714247951361353, "grad_norm": 0.9312985507461801, "learning_rate": 3.651431257546011e-06, "loss": 0.6537, "mean_token_accuracy": 0.783802330493927, "num_tokens": 169036183.0, "step": 5080 }, { "epoch": 0.6715569653713983, "grad_norm": 0.8745642637947537, "learning_rate": 3.649538480423187e-06, "loss": 0.5828, "mean_token_accuracy": 0.8055945634841919, "num_tokens": 169101719.0, "step": 5081 }, { "epoch": 0.6716891356066613, "grad_norm": 0.792045863717396, "learning_rate": 3.647646097224985e-06, "loss": 0.5344, "mean_token_accuracy": 0.8232511281967163, "num_tokens": 169167255.0, "step": 5082 }, { "epoch": 0.6718213058419243, "grad_norm": 0.7940358841095847, "learning_rate": 3.6457541083542536e-06, "loss": 0.4968, "mean_token_accuracy": 0.8340556621551514, "num_tokens": 169232791.0, "step": 5083 }, { "epoch": 0.6719534760771875, "grad_norm": 0.8801432488430612, "learning_rate": 3.6438625142137608e-06, "loss": 0.569, "mean_token_accuracy": 0.8137590289115906, "num_tokens": 169298327.0, "step": 5084 }, { "epoch": 0.6720856463124505, "grad_norm": 0.8363011279682196, "learning_rate": 3.6419713152061854e-06, "loss": 0.5933, "mean_token_accuracy": 0.8042210936546326, "num_tokens": 169363863.0, "step": 5085 }, { "epoch": 0.6722178165477135, "grad_norm": 0.8912046969009146, "learning_rate": 3.6400805117341207e-06, "loss": 0.6548, "mean_token_accuracy": 0.783924400806427, "num_tokens": 169429399.0, "step": 5086 }, { "epoch": 0.6723499867829765, "grad_norm": 0.7925620739596253, "learning_rate": 3.638190104200087e-06, "loss": 0.5868, "mean_token_accuracy": 0.802008330821991, "num_tokens": 169494935.0, "step": 5087 }, { "epoch": 0.6724821570182395, "grad_norm": 0.862670464602778, "learning_rate": 3.636300093006506e-06, "loss": 0.594, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 169560471.0, "step": 5088 }, { "epoch": 0.6726143272535026, "grad_norm": 0.8691156639262799, "learning_rate": 3.6344104785557256e-06, "loss": 0.6122, "mean_token_accuracy": 0.7997497320175171, "num_tokens": 169626007.0, "step": 5089 }, { "epoch": 0.6727464974887656, "grad_norm": 0.796188313780028, "learning_rate": 3.6325212612500004e-06, "loss": 0.5688, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 169691543.0, "step": 5090 }, { "epoch": 0.6728786677240286, "grad_norm": 0.8189888035343836, "learning_rate": 3.630632441491512e-06, "loss": 0.568, "mean_token_accuracy": 0.8097149133682251, "num_tokens": 169757079.0, "step": 5091 }, { "epoch": 0.6730108379592916, "grad_norm": 0.8372614994997248, "learning_rate": 3.6287440196823467e-06, "loss": 0.5615, "mean_token_accuracy": 0.8162312507629395, "num_tokens": 169822615.0, "step": 5092 }, { "epoch": 0.6731430081945546, "grad_norm": 0.8306720113879278, "learning_rate": 3.6268559962245086e-06, "loss": 0.572, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 169888151.0, "step": 5093 }, { "epoch": 0.6732751784298177, "grad_norm": 0.8524152471246548, "learning_rate": 3.6249683715199213e-06, "loss": 0.5994, "mean_token_accuracy": 0.7999328374862671, "num_tokens": 169953687.0, "step": 5094 }, { "epoch": 0.6734073486650807, "grad_norm": 0.8741948798026821, "learning_rate": 3.6230811459704217e-06, "loss": 0.6105, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 170019223.0, "step": 5095 }, { "epoch": 0.6735395189003437, "grad_norm": 0.8113660270226255, "learning_rate": 3.6211943199777555e-06, "loss": 0.5758, "mean_token_accuracy": 0.8097454309463501, "num_tokens": 170084759.0, "step": 5096 }, { "epoch": 0.6736716891356067, "grad_norm": 0.738702333716414, "learning_rate": 3.6193078939435955e-06, "loss": 0.5176, "mean_token_accuracy": 0.8266847729682922, "num_tokens": 170150295.0, "step": 5097 }, { "epoch": 0.6738038593708697, "grad_norm": 0.8709728623480243, "learning_rate": 3.617421868269516e-06, "loss": 0.6506, "mean_token_accuracy": 0.7844127416610718, "num_tokens": 170215831.0, "step": 5098 }, { "epoch": 0.6739360296061327, "grad_norm": 0.8106243775219025, "learning_rate": 3.615536243357016e-06, "loss": 0.6074, "mean_token_accuracy": 0.7973843216896057, "num_tokens": 170281367.0, "step": 5099 }, { "epoch": 0.6740681998413958, "grad_norm": 0.8031793783565678, "learning_rate": 3.6136510196075057e-06, "loss": 0.5668, "mean_token_accuracy": 0.8135606050491333, "num_tokens": 170346903.0, "step": 5100 }, { "epoch": 0.6742003700766588, "grad_norm": 0.8489510127983717, "learning_rate": 3.6117661974223107e-06, "loss": 0.6041, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 170412439.0, "step": 5101 }, { "epoch": 0.6743325403119218, "grad_norm": 0.8330532567371717, "learning_rate": 3.609881777202665e-06, "loss": 0.5683, "mean_token_accuracy": 0.8142778873443604, "num_tokens": 170477975.0, "step": 5102 }, { "epoch": 0.6744647105471848, "grad_norm": 0.9408645001734066, "learning_rate": 3.6079977593497294e-06, "loss": 0.6016, "mean_token_accuracy": 0.7973538041114807, "num_tokens": 170543511.0, "step": 5103 }, { "epoch": 0.6745968807824478, "grad_norm": 0.8613165053809936, "learning_rate": 3.6061141442645662e-06, "loss": 0.6033, "mean_token_accuracy": 0.7989256381988525, "num_tokens": 170609047.0, "step": 5104 }, { "epoch": 0.6747290510177109, "grad_norm": 0.7884570353831277, "learning_rate": 3.6042309323481637e-06, "loss": 0.5769, "mean_token_accuracy": 0.8077157735824585, "num_tokens": 170674583.0, "step": 5105 }, { "epoch": 0.6748612212529739, "grad_norm": 0.83679213537192, "learning_rate": 3.60234812400141e-06, "loss": 0.5853, "mean_token_accuracy": 0.8067543506622314, "num_tokens": 170740119.0, "step": 5106 }, { "epoch": 0.6749933914882369, "grad_norm": 0.8773875111332975, "learning_rate": 3.6004657196251246e-06, "loss": 0.5838, "mean_token_accuracy": 0.8046331405639648, "num_tokens": 170805655.0, "step": 5107 }, { "epoch": 0.6751255617234999, "grad_norm": 0.7751478201250503, "learning_rate": 3.598583719620027e-06, "loss": 0.5515, "mean_token_accuracy": 0.8178183436393738, "num_tokens": 170871191.0, "step": 5108 }, { "epoch": 0.6752577319587629, "grad_norm": 0.849225636071374, "learning_rate": 3.596702124386756e-06, "loss": 0.5859, "mean_token_accuracy": 0.8025424480438232, "num_tokens": 170936727.0, "step": 5109 }, { "epoch": 0.675389902194026, "grad_norm": 0.9094207274431237, "learning_rate": 3.5948209343258665e-06, "loss": 0.5675, "mean_token_accuracy": 0.8108442425727844, "num_tokens": 171002263.0, "step": 5110 }, { "epoch": 0.675522072429289, "grad_norm": 0.8310127123576205, "learning_rate": 3.5929401498378212e-06, "loss": 0.5815, "mean_token_accuracy": 0.8069985508918762, "num_tokens": 171067799.0, "step": 5111 }, { "epoch": 0.675654242664552, "grad_norm": 0.855258204261132, "learning_rate": 3.591059771323002e-06, "loss": 0.572, "mean_token_accuracy": 0.8090129494667053, "num_tokens": 171133335.0, "step": 5112 }, { "epoch": 0.675786412899815, "grad_norm": 0.8055423928083232, "learning_rate": 3.5891797991817012e-06, "loss": 0.6012, "mean_token_accuracy": 0.8033207058906555, "num_tokens": 171198871.0, "step": 5113 }, { "epoch": 0.675918583135078, "grad_norm": 0.8216179050217444, "learning_rate": 3.5873002338141272e-06, "loss": 0.5483, "mean_token_accuracy": 0.8207026124000549, "num_tokens": 171264407.0, "step": 5114 }, { "epoch": 0.676050753370341, "grad_norm": 0.9256971675292467, "learning_rate": 3.5854210756203965e-06, "loss": 0.5971, "mean_token_accuracy": 0.8039005994796753, "num_tokens": 171329943.0, "step": 5115 }, { "epoch": 0.6761829236056041, "grad_norm": 0.7635387010427479, "learning_rate": 3.5835423250005486e-06, "loss": 0.54, "mean_token_accuracy": 0.8207636475563049, "num_tokens": 171395479.0, "step": 5116 }, { "epoch": 0.6763150938408671, "grad_norm": 0.7562146824835658, "learning_rate": 3.581663982354525e-06, "loss": 0.502, "mean_token_accuracy": 0.8333384394645691, "num_tokens": 171461015.0, "step": 5117 }, { "epoch": 0.6764472640761301, "grad_norm": 0.8130715829818601, "learning_rate": 3.5797860480821887e-06, "loss": 0.6029, "mean_token_accuracy": 0.7985746264457703, "num_tokens": 171526551.0, "step": 5118 }, { "epoch": 0.6765794343113931, "grad_norm": 0.8273698386038113, "learning_rate": 3.57790852258331e-06, "loss": 0.5699, "mean_token_accuracy": 0.8080210089683533, "num_tokens": 171592087.0, "step": 5119 }, { "epoch": 0.6767116045466561, "grad_norm": 0.8533918427433395, "learning_rate": 3.5760314062575776e-06, "loss": 0.6047, "mean_token_accuracy": 0.7993834614753723, "num_tokens": 171657623.0, "step": 5120 }, { "epoch": 0.6768437747819191, "grad_norm": 0.9008972460370794, "learning_rate": 3.5741546995045895e-06, "loss": 0.6258, "mean_token_accuracy": 0.7929129600524902, "num_tokens": 171723159.0, "step": 5121 }, { "epoch": 0.6769759450171822, "grad_norm": 0.8205529969124657, "learning_rate": 3.5722784027238564e-06, "loss": 0.5759, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 171788695.0, "step": 5122 }, { "epoch": 0.6771081152524452, "grad_norm": 0.8678053307007905, "learning_rate": 3.5704025163148027e-06, "loss": 0.5536, "mean_token_accuracy": 0.8180472254753113, "num_tokens": 171854231.0, "step": 5123 }, { "epoch": 0.6772402854877082, "grad_norm": 0.8004632931467279, "learning_rate": 3.568527040676768e-06, "loss": 0.5647, "mean_token_accuracy": 0.8099743723869324, "num_tokens": 171919767.0, "step": 5124 }, { "epoch": 0.6773724557229712, "grad_norm": 0.833411889455058, "learning_rate": 3.566651976208998e-06, "loss": 0.5906, "mean_token_accuracy": 0.8028628826141357, "num_tokens": 171985303.0, "step": 5125 }, { "epoch": 0.6775046259582342, "grad_norm": 0.7880493262802999, "learning_rate": 3.564777323310659e-06, "loss": 0.556, "mean_token_accuracy": 0.8153918981552124, "num_tokens": 172050839.0, "step": 5126 }, { "epoch": 0.6776367961934973, "grad_norm": 0.7161599125573913, "learning_rate": 3.562903082380822e-06, "loss": 0.4544, "mean_token_accuracy": 0.8497588634490967, "num_tokens": 172116375.0, "step": 5127 }, { "epoch": 0.6777689664287603, "grad_norm": 0.7791029837489275, "learning_rate": 3.5610292538184752e-06, "loss": 0.5549, "mean_token_accuracy": 0.8155292272567749, "num_tokens": 172181911.0, "step": 5128 }, { "epoch": 0.6779011366640233, "grad_norm": 0.8476927985035551, "learning_rate": 3.559155838022519e-06, "loss": 0.5866, "mean_token_accuracy": 0.8082804083824158, "num_tokens": 172247447.0, "step": 5129 }, { "epoch": 0.6780333068992863, "grad_norm": 0.8400065555926849, "learning_rate": 3.557282835391761e-06, "loss": 0.5816, "mean_token_accuracy": 0.8070595860481262, "num_tokens": 172312983.0, "step": 5130 }, { "epoch": 0.6781654771345493, "grad_norm": 0.7932837813705365, "learning_rate": 3.5554102463249273e-06, "loss": 0.5864, "mean_token_accuracy": 0.8026492595672607, "num_tokens": 172378519.0, "step": 5131 }, { "epoch": 0.6782976473698124, "grad_norm": 0.7636403108523012, "learning_rate": 3.5535380712206513e-06, "loss": 0.5536, "mean_token_accuracy": 0.8163533210754395, "num_tokens": 172444055.0, "step": 5132 }, { "epoch": 0.6784298176050754, "grad_norm": 0.8511765987341285, "learning_rate": 3.551666310477483e-06, "loss": 0.5809, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 172509591.0, "step": 5133 }, { "epoch": 0.6785619878403384, "grad_norm": 0.7678982026880762, "learning_rate": 3.5497949644938768e-06, "loss": 0.5664, "mean_token_accuracy": 0.8152393102645874, "num_tokens": 172575127.0, "step": 5134 }, { "epoch": 0.6786941580756014, "grad_norm": 0.8232840427786446, "learning_rate": 3.5479240336682054e-06, "loss": 0.5669, "mean_token_accuracy": 0.8112104535102844, "num_tokens": 172640663.0, "step": 5135 }, { "epoch": 0.6788263283108644, "grad_norm": 0.9049121163930456, "learning_rate": 3.546053518398751e-06, "loss": 0.5595, "mean_token_accuracy": 0.8155902624130249, "num_tokens": 172706199.0, "step": 5136 }, { "epoch": 0.6789584985461274, "grad_norm": 0.7659504106698806, "learning_rate": 3.544183419083709e-06, "loss": 0.5385, "mean_token_accuracy": 0.8208399415016174, "num_tokens": 172771735.0, "step": 5137 }, { "epoch": 0.6790906687813905, "grad_norm": 0.9229280656563668, "learning_rate": 3.542313736121178e-06, "loss": 0.658, "mean_token_accuracy": 0.7835429310798645, "num_tokens": 172837271.0, "step": 5138 }, { "epoch": 0.6792228390166535, "grad_norm": 0.8545418128235046, "learning_rate": 3.5404444699091834e-06, "loss": 0.5647, "mean_token_accuracy": 0.8141557574272156, "num_tokens": 172902807.0, "step": 5139 }, { "epoch": 0.6793550092519165, "grad_norm": 0.835988732133873, "learning_rate": 3.5385756208456467e-06, "loss": 0.6116, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 172968343.0, "step": 5140 }, { "epoch": 0.6794871794871795, "grad_norm": 0.8941888465507006, "learning_rate": 3.536707189328408e-06, "loss": 0.6137, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 173033879.0, "step": 5141 }, { "epoch": 0.6796193497224425, "grad_norm": 0.8455028502453955, "learning_rate": 3.534839175755218e-06, "loss": 0.5952, "mean_token_accuracy": 0.8032291531562805, "num_tokens": 173099415.0, "step": 5142 }, { "epoch": 0.6797515199577056, "grad_norm": 0.8492436447119529, "learning_rate": 3.532971580523741e-06, "loss": 0.5913, "mean_token_accuracy": 0.8039769530296326, "num_tokens": 173164951.0, "step": 5143 }, { "epoch": 0.6798836901929686, "grad_norm": 0.8491351324039025, "learning_rate": 3.5311044040315425e-06, "loss": 0.6157, "mean_token_accuracy": 0.7963313460350037, "num_tokens": 173230487.0, "step": 5144 }, { "epoch": 0.6800158604282316, "grad_norm": 0.7836071469756016, "learning_rate": 3.5292376466761114e-06, "loss": 0.5579, "mean_token_accuracy": 0.8129807114601135, "num_tokens": 173296023.0, "step": 5145 }, { "epoch": 0.6801480306634946, "grad_norm": 0.8209603755308007, "learning_rate": 3.5273713088548367e-06, "loss": 0.5942, "mean_token_accuracy": 0.803427517414093, "num_tokens": 173361559.0, "step": 5146 }, { "epoch": 0.6802802008987576, "grad_norm": 0.8750463946276656, "learning_rate": 3.525505390965025e-06, "loss": 0.6045, "mean_token_accuracy": 0.7983915209770203, "num_tokens": 173427095.0, "step": 5147 }, { "epoch": 0.6804123711340206, "grad_norm": 0.850668020072863, "learning_rate": 3.523639893403894e-06, "loss": 0.5641, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 173492631.0, "step": 5148 }, { "epoch": 0.6805445413692837, "grad_norm": 0.8612779418651118, "learning_rate": 3.521774816568563e-06, "loss": 0.5963, "mean_token_accuracy": 0.8039158582687378, "num_tokens": 173558167.0, "step": 5149 }, { "epoch": 0.6806767116045467, "grad_norm": 0.903205498732406, "learning_rate": 3.5199101608560716e-06, "loss": 0.6293, "mean_token_accuracy": 0.7910206317901611, "num_tokens": 173623703.0, "step": 5150 }, { "epoch": 0.6808088818398097, "grad_norm": 0.8641900558876596, "learning_rate": 3.518045926663366e-06, "loss": 0.586, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 173689239.0, "step": 5151 }, { "epoch": 0.6809410520750727, "grad_norm": 0.8118268959160745, "learning_rate": 3.516182114387304e-06, "loss": 0.5662, "mean_token_accuracy": 0.812004029750824, "num_tokens": 173754775.0, "step": 5152 }, { "epoch": 0.6810732223103357, "grad_norm": 0.7380897309756093, "learning_rate": 3.514318724424649e-06, "loss": 0.5252, "mean_token_accuracy": 0.8259522914886475, "num_tokens": 173820311.0, "step": 5153 }, { "epoch": 0.6812053925455988, "grad_norm": 0.8855178080719539, "learning_rate": 3.5124557571720796e-06, "loss": 0.5902, "mean_token_accuracy": 0.8052130341529846, "num_tokens": 173885847.0, "step": 5154 }, { "epoch": 0.6813375627808618, "grad_norm": 0.9245090244938281, "learning_rate": 3.5105932130261832e-06, "loss": 0.6081, "mean_token_accuracy": 0.797857403755188, "num_tokens": 173951383.0, "step": 5155 }, { "epoch": 0.6814697330161248, "grad_norm": 0.9344954292560167, "learning_rate": 3.508731092383456e-06, "loss": 0.636, "mean_token_accuracy": 0.7882736921310425, "num_tokens": 174016919.0, "step": 5156 }, { "epoch": 0.6816019032513878, "grad_norm": 0.7993058629106943, "learning_rate": 3.506869395640302e-06, "loss": 0.5528, "mean_token_accuracy": 0.8171163201332092, "num_tokens": 174082455.0, "step": 5157 }, { "epoch": 0.6817340734866508, "grad_norm": 0.7561275869357169, "learning_rate": 3.5050081231930446e-06, "loss": 0.513, "mean_token_accuracy": 0.8292332887649536, "num_tokens": 174147991.0, "step": 5158 }, { "epoch": 0.6818662437219138, "grad_norm": 0.8445628882731411, "learning_rate": 3.5031472754379025e-06, "loss": 0.5707, "mean_token_accuracy": 0.8101117014884949, "num_tokens": 174213527.0, "step": 5159 }, { "epoch": 0.6819984139571769, "grad_norm": 0.8709460448578463, "learning_rate": 3.501286852771013e-06, "loss": 0.6246, "mean_token_accuracy": 0.7918752431869507, "num_tokens": 174279063.0, "step": 5160 }, { "epoch": 0.6821305841924399, "grad_norm": 0.8853042512784853, "learning_rate": 3.499426855588425e-06, "loss": 0.6232, "mean_token_accuracy": 0.7965754866600037, "num_tokens": 174344599.0, "step": 5161 }, { "epoch": 0.6822627544277029, "grad_norm": 0.9249399297332651, "learning_rate": 3.497567284286091e-06, "loss": 0.5809, "mean_token_accuracy": 0.8079599738121033, "num_tokens": 174410135.0, "step": 5162 }, { "epoch": 0.6823949246629659, "grad_norm": 0.803227642244651, "learning_rate": 3.4957081392598704e-06, "loss": 0.5255, "mean_token_accuracy": 0.8264100551605225, "num_tokens": 174475671.0, "step": 5163 }, { "epoch": 0.6825270948982289, "grad_norm": 0.8583249060004873, "learning_rate": 3.493849420905544e-06, "loss": 0.6116, "mean_token_accuracy": 0.8006348609924316, "num_tokens": 174541207.0, "step": 5164 }, { "epoch": 0.682659265133492, "grad_norm": 0.7952525006001215, "learning_rate": 3.491991129618789e-06, "loss": 0.5616, "mean_token_accuracy": 0.8144609928131104, "num_tokens": 174606743.0, "step": 5165 }, { "epoch": 0.682791435368755, "grad_norm": 0.882893642126634, "learning_rate": 3.4901332657952e-06, "loss": 0.5991, "mean_token_accuracy": 0.8013978600502014, "num_tokens": 174672279.0, "step": 5166 }, { "epoch": 0.682923605604018, "grad_norm": 0.8200684139205355, "learning_rate": 3.4882758298302744e-06, "loss": 0.5606, "mean_token_accuracy": 0.8151629567146301, "num_tokens": 174737815.0, "step": 5167 }, { "epoch": 0.683055775839281, "grad_norm": 0.8953731780871067, "learning_rate": 3.4864188221194224e-06, "loss": 0.5895, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 174803351.0, "step": 5168 }, { "epoch": 0.683187946074544, "grad_norm": 0.7775358809034661, "learning_rate": 3.484562243057963e-06, "loss": 0.549, "mean_token_accuracy": 0.8179404139518738, "num_tokens": 174868887.0, "step": 5169 }, { "epoch": 0.683320116309807, "grad_norm": 0.8166487661440622, "learning_rate": 3.4827060930411214e-06, "loss": 0.577, "mean_token_accuracy": 0.8080667853355408, "num_tokens": 174934423.0, "step": 5170 }, { "epoch": 0.6834522865450701, "grad_norm": 0.7757619260641773, "learning_rate": 3.4808503724640376e-06, "loss": 0.5425, "mean_token_accuracy": 0.8193596601486206, "num_tokens": 174999959.0, "step": 5171 }, { "epoch": 0.6835844567803331, "grad_norm": 0.8361985940563234, "learning_rate": 3.478995081721751e-06, "loss": 0.5405, "mean_token_accuracy": 0.8230527639389038, "num_tokens": 175065495.0, "step": 5172 }, { "epoch": 0.6837166270155961, "grad_norm": 0.844258185131038, "learning_rate": 3.4771402212092153e-06, "loss": 0.5678, "mean_token_accuracy": 0.8099591135978699, "num_tokens": 175131031.0, "step": 5173 }, { "epoch": 0.6838487972508591, "grad_norm": 0.8372301867323615, "learning_rate": 3.475285791321293e-06, "loss": 0.5879, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 175196567.0, "step": 5174 }, { "epoch": 0.6839809674861221, "grad_norm": 0.8162360276243419, "learning_rate": 3.4734317924527537e-06, "loss": 0.6031, "mean_token_accuracy": 0.8009858131408691, "num_tokens": 175262103.0, "step": 5175 }, { "epoch": 0.6841131377213852, "grad_norm": 0.8703276684815082, "learning_rate": 3.471578224998271e-06, "loss": 0.5951, "mean_token_accuracy": 0.8022982478141785, "num_tokens": 175327639.0, "step": 5176 }, { "epoch": 0.6842453079566482, "grad_norm": 0.7970071793231109, "learning_rate": 3.4697250893524376e-06, "loss": 0.5627, "mean_token_accuracy": 0.8150561451911926, "num_tokens": 175393175.0, "step": 5177 }, { "epoch": 0.6843774781919112, "grad_norm": 0.8966886399488153, "learning_rate": 3.467872385909741e-06, "loss": 0.6245, "mean_token_accuracy": 0.7946678996086121, "num_tokens": 175458711.0, "step": 5178 }, { "epoch": 0.6845096484271742, "grad_norm": 0.8857902426510985, "learning_rate": 3.4660201150645877e-06, "loss": 0.5878, "mean_token_accuracy": 0.8056861162185669, "num_tokens": 175524247.0, "step": 5179 }, { "epoch": 0.6846418186624372, "grad_norm": 0.8203061288508369, "learning_rate": 3.464168277211283e-06, "loss": 0.5807, "mean_token_accuracy": 0.8090434670448303, "num_tokens": 175589783.0, "step": 5180 }, { "epoch": 0.6847739888977002, "grad_norm": 0.8037660763640666, "learning_rate": 3.462316872744049e-06, "loss": 0.592, "mean_token_accuracy": 0.8019777536392212, "num_tokens": 175655319.0, "step": 5181 }, { "epoch": 0.6849061591329633, "grad_norm": 0.9081921307349898, "learning_rate": 3.4604659020570074e-06, "loss": 0.6235, "mean_token_accuracy": 0.7940574884414673, "num_tokens": 175720855.0, "step": 5182 }, { "epoch": 0.6850383293682263, "grad_norm": 0.7640599560660812, "learning_rate": 3.458615365544194e-06, "loss": 0.5539, "mean_token_accuracy": 0.816582202911377, "num_tokens": 175786391.0, "step": 5183 }, { "epoch": 0.6851704996034893, "grad_norm": 0.8462759410278633, "learning_rate": 3.4567652635995473e-06, "loss": 0.5724, "mean_token_accuracy": 0.8121413588523865, "num_tokens": 175851927.0, "step": 5184 }, { "epoch": 0.6853026698387523, "grad_norm": 0.8184241275415898, "learning_rate": 3.454915596616919e-06, "loss": 0.5978, "mean_token_accuracy": 0.8012605309486389, "num_tokens": 175917463.0, "step": 5185 }, { "epoch": 0.6854348400740153, "grad_norm": 0.8398830070944543, "learning_rate": 3.4530663649900587e-06, "loss": 0.5696, "mean_token_accuracy": 0.8116682767868042, "num_tokens": 175982999.0, "step": 5186 }, { "epoch": 0.6855670103092784, "grad_norm": 0.8346834436928271, "learning_rate": 3.451217569112633e-06, "loss": 0.6012, "mean_token_accuracy": 0.8008637428283691, "num_tokens": 176048535.0, "step": 5187 }, { "epoch": 0.6856991805445414, "grad_norm": 0.8733179602914387, "learning_rate": 3.4493692093782116e-06, "loss": 0.6306, "mean_token_accuracy": 0.7937217950820923, "num_tokens": 176114071.0, "step": 5188 }, { "epoch": 0.6858313507798044, "grad_norm": 0.8492653823835947, "learning_rate": 3.447521286180272e-06, "loss": 0.6004, "mean_token_accuracy": 0.7997039556503296, "num_tokens": 176179607.0, "step": 5189 }, { "epoch": 0.6859635210150674, "grad_norm": 0.8527635639040537, "learning_rate": 3.445673799912199e-06, "loss": 0.5719, "mean_token_accuracy": 0.8121718764305115, "num_tokens": 176245143.0, "step": 5190 }, { "epoch": 0.6860956912503304, "grad_norm": 0.76164594005373, "learning_rate": 3.4438267509672805e-06, "loss": 0.5344, "mean_token_accuracy": 0.8246398568153381, "num_tokens": 176310679.0, "step": 5191 }, { "epoch": 0.6862278614855934, "grad_norm": 0.8986230969193013, "learning_rate": 3.4419801397387176e-06, "loss": 0.5896, "mean_token_accuracy": 0.8029544353485107, "num_tokens": 176376215.0, "step": 5192 }, { "epoch": 0.6863600317208565, "grad_norm": 0.7949072158782211, "learning_rate": 3.440133966619615e-06, "loss": 0.5417, "mean_token_accuracy": 0.8200463652610779, "num_tokens": 176441751.0, "step": 5193 }, { "epoch": 0.6864922019561195, "grad_norm": 0.8299742541242329, "learning_rate": 3.4382882320029844e-06, "loss": 0.5346, "mean_token_accuracy": 0.8232663869857788, "num_tokens": 176507287.0, "step": 5194 }, { "epoch": 0.6866243721913825, "grad_norm": 0.8031501081773779, "learning_rate": 3.436442936281743e-06, "loss": 0.5321, "mean_token_accuracy": 0.8256165385246277, "num_tokens": 176572823.0, "step": 5195 }, { "epoch": 0.6867565424266455, "grad_norm": 0.8766300311909566, "learning_rate": 3.4345980798487165e-06, "loss": 0.5977, "mean_token_accuracy": 0.8028781414031982, "num_tokens": 176638359.0, "step": 5196 }, { "epoch": 0.6868887126619085, "grad_norm": 0.8896318918303304, "learning_rate": 3.4327536630966356e-06, "loss": 0.5621, "mean_token_accuracy": 0.8109205365180969, "num_tokens": 176703895.0, "step": 5197 }, { "epoch": 0.6870208828971716, "grad_norm": 0.7997466745597513, "learning_rate": 3.4309096864181414e-06, "loss": 0.5488, "mean_token_accuracy": 0.8167806267738342, "num_tokens": 176769431.0, "step": 5198 }, { "epoch": 0.6871530531324346, "grad_norm": 0.8251113997184156, "learning_rate": 3.4290661502057716e-06, "loss": 0.5878, "mean_token_accuracy": 0.8053656220436096, "num_tokens": 176834967.0, "step": 5199 }, { "epoch": 0.6872852233676976, "grad_norm": 0.8659649107063393, "learning_rate": 3.4272230548519834e-06, "loss": 0.5993, "mean_token_accuracy": 0.8017030954360962, "num_tokens": 176900503.0, "step": 5200 }, { "epoch": 0.6874173936029606, "grad_norm": 0.8255857984860826, "learning_rate": 3.42538040074913e-06, "loss": 0.588, "mean_token_accuracy": 0.8037480115890503, "num_tokens": 176966039.0, "step": 5201 }, { "epoch": 0.6875495638382236, "grad_norm": 0.8405796084685548, "learning_rate": 3.4235381882894735e-06, "loss": 0.5772, "mean_token_accuracy": 0.8109357953071594, "num_tokens": 177031575.0, "step": 5202 }, { "epoch": 0.6876817340734867, "grad_norm": 0.8839079431822809, "learning_rate": 3.4216964178651834e-06, "loss": 0.5912, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 177097111.0, "step": 5203 }, { "epoch": 0.6878139043087497, "grad_norm": 0.7707480867705162, "learning_rate": 3.419855089868336e-06, "loss": 0.5431, "mean_token_accuracy": 0.8196343779563904, "num_tokens": 177162647.0, "step": 5204 }, { "epoch": 0.6879460745440127, "grad_norm": 0.9798108032189649, "learning_rate": 3.418014204690907e-06, "loss": 0.6099, "mean_token_accuracy": 0.7983457446098328, "num_tokens": 177228183.0, "step": 5205 }, { "epoch": 0.6880782447792757, "grad_norm": 0.8779293042185041, "learning_rate": 3.416173762724786e-06, "loss": 0.6414, "mean_token_accuracy": 0.788288950920105, "num_tokens": 177293719.0, "step": 5206 }, { "epoch": 0.6882104150145387, "grad_norm": 0.8183332388274118, "learning_rate": 3.4143337643617637e-06, "loss": 0.5886, "mean_token_accuracy": 0.8039463758468628, "num_tokens": 177359255.0, "step": 5207 }, { "epoch": 0.6883425852498017, "grad_norm": 0.8345699213691068, "learning_rate": 3.412494209993537e-06, "loss": 0.61, "mean_token_accuracy": 0.7976285219192505, "num_tokens": 177424791.0, "step": 5208 }, { "epoch": 0.6884747554850648, "grad_norm": 0.857810831928575, "learning_rate": 3.410655100011711e-06, "loss": 0.6135, "mean_token_accuracy": 0.7958124876022339, "num_tokens": 177490327.0, "step": 5209 }, { "epoch": 0.6886069257203278, "grad_norm": 0.7830213013699053, "learning_rate": 3.4088164348077903e-06, "loss": 0.5602, "mean_token_accuracy": 0.8153003454208374, "num_tokens": 177555863.0, "step": 5210 }, { "epoch": 0.6887390959555908, "grad_norm": 0.8388346460106363, "learning_rate": 3.4069782147731913e-06, "loss": 0.5986, "mean_token_accuracy": 0.8030155301094055, "num_tokens": 177621399.0, "step": 5211 }, { "epoch": 0.6888712661908538, "grad_norm": 0.8610017216459181, "learning_rate": 3.405140440299228e-06, "loss": 0.5839, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 177686935.0, "step": 5212 }, { "epoch": 0.6890034364261168, "grad_norm": 0.8564890642448262, "learning_rate": 3.4033031117771303e-06, "loss": 0.5961, "mean_token_accuracy": 0.8032291531562805, "num_tokens": 177752471.0, "step": 5213 }, { "epoch": 0.6891356066613799, "grad_norm": 0.808602394816967, "learning_rate": 3.401466229598023e-06, "loss": 0.5836, "mean_token_accuracy": 0.806861162185669, "num_tokens": 177818007.0, "step": 5214 }, { "epoch": 0.6892677768966429, "grad_norm": 0.800384329374746, "learning_rate": 3.399629794152941e-06, "loss": 0.5547, "mean_token_accuracy": 0.8170400261878967, "num_tokens": 177883543.0, "step": 5215 }, { "epoch": 0.6893999471319059, "grad_norm": 0.8306988930331096, "learning_rate": 3.397793805832823e-06, "loss": 0.6132, "mean_token_accuracy": 0.7978116273880005, "num_tokens": 177949079.0, "step": 5216 }, { "epoch": 0.6895321173671689, "grad_norm": 0.8309370088847656, "learning_rate": 3.395958265028515e-06, "loss": 0.5629, "mean_token_accuracy": 0.8117598295211792, "num_tokens": 178014615.0, "step": 5217 }, { "epoch": 0.6896642876024319, "grad_norm": 0.7902885755690737, "learning_rate": 3.3941231721307603e-06, "loss": 0.5321, "mean_token_accuracy": 0.8251129388809204, "num_tokens": 178080151.0, "step": 5218 }, { "epoch": 0.6897964578376949, "grad_norm": 0.832077390727301, "learning_rate": 3.3922885275302177e-06, "loss": 0.5713, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 178145687.0, "step": 5219 }, { "epoch": 0.689928628072958, "grad_norm": 0.8414379805152907, "learning_rate": 3.39045433161744e-06, "loss": 0.5823, "mean_token_accuracy": 0.8070595860481262, "num_tokens": 178211223.0, "step": 5220 }, { "epoch": 0.690060798308221, "grad_norm": 0.8872832695837085, "learning_rate": 3.3886205847828903e-06, "loss": 0.5796, "mean_token_accuracy": 0.8082651495933533, "num_tokens": 178276759.0, "step": 5221 }, { "epoch": 0.690192968543484, "grad_norm": 0.8455874009219133, "learning_rate": 3.386787287416936e-06, "loss": 0.5699, "mean_token_accuracy": 0.8103406429290771, "num_tokens": 178342295.0, "step": 5222 }, { "epoch": 0.690325138778747, "grad_norm": 0.9328063223101866, "learning_rate": 3.3849544399098495e-06, "loss": 0.6201, "mean_token_accuracy": 0.7971096038818359, "num_tokens": 178407831.0, "step": 5223 }, { "epoch": 0.69045730901401, "grad_norm": 0.8783435401554882, "learning_rate": 3.383122042651802e-06, "loss": 0.6011, "mean_token_accuracy": 0.8025271892547607, "num_tokens": 178473367.0, "step": 5224 }, { "epoch": 0.690589479249273, "grad_norm": 0.804229964703639, "learning_rate": 3.3812900960328736e-06, "loss": 0.5665, "mean_token_accuracy": 0.811973512172699, "num_tokens": 178538903.0, "step": 5225 }, { "epoch": 0.6907216494845361, "grad_norm": 0.9077700857884655, "learning_rate": 3.3794586004430496e-06, "loss": 0.6123, "mean_token_accuracy": 0.7976590394973755, "num_tokens": 178604439.0, "step": 5226 }, { "epoch": 0.6908538197197991, "grad_norm": 0.9095071446214025, "learning_rate": 3.377627556272217e-06, "loss": 0.6367, "mean_token_accuracy": 0.7880295515060425, "num_tokens": 178669975.0, "step": 5227 }, { "epoch": 0.6909859899550621, "grad_norm": 0.9570037660794197, "learning_rate": 3.375796963910164e-06, "loss": 0.6118, "mean_token_accuracy": 0.7980252504348755, "num_tokens": 178735511.0, "step": 5228 }, { "epoch": 0.6911181601903251, "grad_norm": 0.8623226862228008, "learning_rate": 3.3739668237465874e-06, "loss": 0.5967, "mean_token_accuracy": 0.802313506603241, "num_tokens": 178801047.0, "step": 5229 }, { "epoch": 0.6912503304255881, "grad_norm": 0.8462241242369869, "learning_rate": 3.3721371361710866e-06, "loss": 0.5877, "mean_token_accuracy": 0.806464433670044, "num_tokens": 178866583.0, "step": 5230 }, { "epoch": 0.6913825006608512, "grad_norm": 0.8482755533789877, "learning_rate": 3.37030790157316e-06, "loss": 0.6199, "mean_token_accuracy": 0.7945916056632996, "num_tokens": 178932119.0, "step": 5231 }, { "epoch": 0.6915146708961142, "grad_norm": 0.7487135730021498, "learning_rate": 3.3684791203422196e-06, "loss": 0.5238, "mean_token_accuracy": 0.8248839974403381, "num_tokens": 178997655.0, "step": 5232 }, { "epoch": 0.6916468411313772, "grad_norm": 0.8850362475459155, "learning_rate": 3.3666507928675706e-06, "loss": 0.5755, "mean_token_accuracy": 0.8099133372306824, "num_tokens": 179063191.0, "step": 5233 }, { "epoch": 0.6917790113666402, "grad_norm": 0.9135884400802535, "learning_rate": 3.3648229195384264e-06, "loss": 0.6062, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 179128727.0, "step": 5234 }, { "epoch": 0.6919111816019032, "grad_norm": 0.8325989652382045, "learning_rate": 3.3629955007439025e-06, "loss": 0.5835, "mean_token_accuracy": 0.8085856437683105, "num_tokens": 179194263.0, "step": 5235 }, { "epoch": 0.6920433518371663, "grad_norm": 0.8010635457239221, "learning_rate": 3.361168536873022e-06, "loss": 0.5428, "mean_token_accuracy": 0.8189628720283508, "num_tokens": 179259799.0, "step": 5236 }, { "epoch": 0.6921755220724293, "grad_norm": 0.8175074927510034, "learning_rate": 3.359342028314699e-06, "loss": 0.5909, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 179325335.0, "step": 5237 }, { "epoch": 0.6923076923076923, "grad_norm": 0.769132982045477, "learning_rate": 3.35751597545777e-06, "loss": 0.5295, "mean_token_accuracy": 0.8255097270011902, "num_tokens": 179390871.0, "step": 5238 }, { "epoch": 0.6924398625429553, "grad_norm": 0.7643056886354039, "learning_rate": 3.3556903786909556e-06, "loss": 0.5567, "mean_token_accuracy": 0.8133317232131958, "num_tokens": 179456407.0, "step": 5239 }, { "epoch": 0.6925720327782183, "grad_norm": 0.8557975996751807, "learning_rate": 3.353865238402891e-06, "loss": 0.5835, "mean_token_accuracy": 0.8019319772720337, "num_tokens": 179521943.0, "step": 5240 }, { "epoch": 0.6927042030134813, "grad_norm": 0.8767735889297972, "learning_rate": 3.3520405549821057e-06, "loss": 0.6127, "mean_token_accuracy": 0.7992156147956848, "num_tokens": 179587479.0, "step": 5241 }, { "epoch": 0.6928363732487444, "grad_norm": 0.8698373740393047, "learning_rate": 3.3502163288170442e-06, "loss": 0.6142, "mean_token_accuracy": 0.797613263130188, "num_tokens": 179653015.0, "step": 5242 }, { "epoch": 0.6929685434840074, "grad_norm": 0.8537896384685018, "learning_rate": 3.34839256029604e-06, "loss": 0.5985, "mean_token_accuracy": 0.8028323650360107, "num_tokens": 179718551.0, "step": 5243 }, { "epoch": 0.6931007137192704, "grad_norm": 0.7935223706964538, "learning_rate": 3.3465692498073368e-06, "loss": 0.5149, "mean_token_accuracy": 0.8288212418556213, "num_tokens": 179784087.0, "step": 5244 }, { "epoch": 0.6932328839545334, "grad_norm": 0.9659823977675455, "learning_rate": 3.3447463977390803e-06, "loss": 0.6122, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 179849623.0, "step": 5245 }, { "epoch": 0.6933650541897964, "grad_norm": 0.8137976418722412, "learning_rate": 3.3429240044793186e-06, "loss": 0.5748, "mean_token_accuracy": 0.8079599738121033, "num_tokens": 179915159.0, "step": 5246 }, { "epoch": 0.6934972244250595, "grad_norm": 0.7942416623898801, "learning_rate": 3.3411020704159992e-06, "loss": 0.5569, "mean_token_accuracy": 0.8125991821289062, "num_tokens": 179980695.0, "step": 5247 }, { "epoch": 0.6936293946603225, "grad_norm": 0.8257788957333705, "learning_rate": 3.3392805959369734e-06, "loss": 0.5882, "mean_token_accuracy": 0.8047552108764648, "num_tokens": 180046231.0, "step": 5248 }, { "epoch": 0.6937615648955855, "grad_norm": 0.8529143003198807, "learning_rate": 3.337459581429998e-06, "loss": 0.6223, "mean_token_accuracy": 0.7932639718055725, "num_tokens": 180111767.0, "step": 5249 }, { "epoch": 0.6938937351308485, "grad_norm": 0.8652241691856596, "learning_rate": 3.3356390272827245e-06, "loss": 0.6328, "mean_token_accuracy": 0.7917531728744507, "num_tokens": 180177303.0, "step": 5250 }, { "epoch": 0.6940259053661115, "grad_norm": 0.8172660577537811, "learning_rate": 3.333818933882716e-06, "loss": 0.5683, "mean_token_accuracy": 0.8101422190666199, "num_tokens": 180242839.0, "step": 5251 }, { "epoch": 0.6941580756013745, "grad_norm": 0.8711318438742528, "learning_rate": 3.331999301617428e-06, "loss": 0.6068, "mean_token_accuracy": 0.7991240620613098, "num_tokens": 180308375.0, "step": 5252 }, { "epoch": 0.6942902458366376, "grad_norm": 0.8687281910541251, "learning_rate": 3.3301801308742253e-06, "loss": 0.5985, "mean_token_accuracy": 0.8014131188392639, "num_tokens": 180373911.0, "step": 5253 }, { "epoch": 0.6944224160719006, "grad_norm": 0.7976527997387458, "learning_rate": 3.3283614220403694e-06, "loss": 0.5728, "mean_token_accuracy": 0.8093944787979126, "num_tokens": 180439447.0, "step": 5254 }, { "epoch": 0.6945545863071636, "grad_norm": 0.8598442850712195, "learning_rate": 3.3265431755030277e-06, "loss": 0.6092, "mean_token_accuracy": 0.7981931567192078, "num_tokens": 180504983.0, "step": 5255 }, { "epoch": 0.6946867565424266, "grad_norm": 0.8550910753628476, "learning_rate": 3.3247253916492643e-06, "loss": 0.5563, "mean_token_accuracy": 0.8155139684677124, "num_tokens": 180570519.0, "step": 5256 }, { "epoch": 0.6948189267776896, "grad_norm": 0.8021199697615204, "learning_rate": 3.322908070866048e-06, "loss": 0.5866, "mean_token_accuracy": 0.8062965273857117, "num_tokens": 180636055.0, "step": 5257 }, { "epoch": 0.6949510970129527, "grad_norm": 0.838894496832081, "learning_rate": 3.3210912135402495e-06, "loss": 0.5993, "mean_token_accuracy": 0.7989866733551025, "num_tokens": 180701591.0, "step": 5258 }, { "epoch": 0.6950832672482157, "grad_norm": 0.8654614598005032, "learning_rate": 3.3192748200586412e-06, "loss": 0.6243, "mean_token_accuracy": 0.7956751585006714, "num_tokens": 180767127.0, "step": 5259 }, { "epoch": 0.6952154374834787, "grad_norm": 0.8407308916543178, "learning_rate": 3.3174588908078888e-06, "loss": 0.5252, "mean_token_accuracy": 0.8261659145355225, "num_tokens": 180832663.0, "step": 5260 }, { "epoch": 0.6953476077187417, "grad_norm": 0.8112340901581223, "learning_rate": 3.315643426174575e-06, "loss": 0.5864, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 180898199.0, "step": 5261 }, { "epoch": 0.6954797779540047, "grad_norm": 0.8770996281598051, "learning_rate": 3.3138284265451677e-06, "loss": 0.5993, "mean_token_accuracy": 0.802466094493866, "num_tokens": 180963735.0, "step": 5262 }, { "epoch": 0.6956119481892677, "grad_norm": 0.8132470177823973, "learning_rate": 3.3120138923060434e-06, "loss": 0.5801, "mean_token_accuracy": 0.8070138096809387, "num_tokens": 181029271.0, "step": 5263 }, { "epoch": 0.6957441184245308, "grad_norm": 0.7785979866359217, "learning_rate": 3.310199823843481e-06, "loss": 0.5195, "mean_token_accuracy": 0.8323897123336792, "num_tokens": 181092157.0, "step": 5264 }, { "epoch": 0.6958762886597938, "grad_norm": 0.856770109782001, "learning_rate": 3.3083862215436573e-06, "loss": 0.6007, "mean_token_accuracy": 0.802466094493866, "num_tokens": 181157693.0, "step": 5265 }, { "epoch": 0.6960084588950568, "grad_norm": 0.777899494220536, "learning_rate": 3.3065730857926474e-06, "loss": 0.5658, "mean_token_accuracy": 0.8104932308197021, "num_tokens": 181223229.0, "step": 5266 }, { "epoch": 0.6961406291303198, "grad_norm": 0.9158727348488851, "learning_rate": 3.3047604169764326e-06, "loss": 0.6151, "mean_token_accuracy": 0.7972164750099182, "num_tokens": 181288765.0, "step": 5267 }, { "epoch": 0.6962727993655828, "grad_norm": 0.7675738306495486, "learning_rate": 3.302948215480894e-06, "loss": 0.5528, "mean_token_accuracy": 0.8167043328285217, "num_tokens": 181354301.0, "step": 5268 }, { "epoch": 0.6964049696008459, "grad_norm": 0.8148921376576147, "learning_rate": 3.3011364816918045e-06, "loss": 0.5671, "mean_token_accuracy": 0.8117598295211792, "num_tokens": 181419837.0, "step": 5269 }, { "epoch": 0.6965371398361089, "grad_norm": 0.8131272010170134, "learning_rate": 3.299325215994854e-06, "loss": 0.6416, "mean_token_accuracy": 0.7861829996109009, "num_tokens": 181485373.0, "step": 5270 }, { "epoch": 0.6966693100713719, "grad_norm": 0.8303409153043165, "learning_rate": 3.297514418775615e-06, "loss": 0.5723, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 181550909.0, "step": 5271 }, { "epoch": 0.6968014803066349, "grad_norm": 0.8348041724227745, "learning_rate": 3.2957040904195747e-06, "loss": 0.5688, "mean_token_accuracy": 0.8118056654930115, "num_tokens": 181616445.0, "step": 5272 }, { "epoch": 0.6969336505418979, "grad_norm": 0.9034960059970573, "learning_rate": 3.293894231312107e-06, "loss": 0.6327, "mean_token_accuracy": 0.7895861268043518, "num_tokens": 181681981.0, "step": 5273 }, { "epoch": 0.697065820777161, "grad_norm": 0.8838407080144419, "learning_rate": 3.292084841838501e-06, "loss": 0.6366, "mean_token_accuracy": 0.7877853512763977, "num_tokens": 181747517.0, "step": 5274 }, { "epoch": 0.697197991012424, "grad_norm": 0.8918876628400774, "learning_rate": 3.290275922383932e-06, "loss": 0.6591, "mean_token_accuracy": 0.7830393314361572, "num_tokens": 181813053.0, "step": 5275 }, { "epoch": 0.697330161247687, "grad_norm": 0.7611508751463488, "learning_rate": 3.288467473333484e-06, "loss": 0.5344, "mean_token_accuracy": 0.8208552002906799, "num_tokens": 181878589.0, "step": 5276 }, { "epoch": 0.69746233148295, "grad_norm": 0.8707173110447177, "learning_rate": 3.2866594950721375e-06, "loss": 0.6126, "mean_token_accuracy": 0.7983457446098328, "num_tokens": 181944125.0, "step": 5277 }, { "epoch": 0.697594501718213, "grad_norm": 0.8614684569976971, "learning_rate": 3.2848519879847746e-06, "loss": 0.5972, "mean_token_accuracy": 0.802008330821991, "num_tokens": 182009661.0, "step": 5278 }, { "epoch": 0.697726671953476, "grad_norm": 0.8389267439011512, "learning_rate": 3.2830449524561715e-06, "loss": 0.5437, "mean_token_accuracy": 0.8207789063453674, "num_tokens": 182075197.0, "step": 5279 }, { "epoch": 0.6978588421887391, "grad_norm": 0.7976842209457586, "learning_rate": 3.281238388871014e-06, "loss": 0.544, "mean_token_accuracy": 0.8199242949485779, "num_tokens": 182140733.0, "step": 5280 }, { "epoch": 0.6979910124240021, "grad_norm": 0.7990458016152452, "learning_rate": 3.2794322976138774e-06, "loss": 0.5616, "mean_token_accuracy": 0.813179075717926, "num_tokens": 182206269.0, "step": 5281 }, { "epoch": 0.6981231826592651, "grad_norm": 0.783104326404792, "learning_rate": 3.2776266790692433e-06, "loss": 0.5384, "mean_token_accuracy": 0.8227322697639465, "num_tokens": 182271805.0, "step": 5282 }, { "epoch": 0.6982553528945281, "grad_norm": 0.7942565251234158, "learning_rate": 3.2758215336214894e-06, "loss": 0.5487, "mean_token_accuracy": 0.8177267909049988, "num_tokens": 182337341.0, "step": 5283 }, { "epoch": 0.6983875231297911, "grad_norm": 0.7862546073886032, "learning_rate": 3.2740168616548945e-06, "loss": 0.5207, "mean_token_accuracy": 0.8250671625137329, "num_tokens": 182402877.0, "step": 5284 }, { "epoch": 0.6985196933650542, "grad_norm": 0.7875463731435361, "learning_rate": 3.2722126635536335e-06, "loss": 0.5956, "mean_token_accuracy": 0.8026645183563232, "num_tokens": 182468413.0, "step": 5285 }, { "epoch": 0.6986518636003172, "grad_norm": 0.8264902347602655, "learning_rate": 3.270408939701783e-06, "loss": 0.5932, "mean_token_accuracy": 0.8029850125312805, "num_tokens": 182533949.0, "step": 5286 }, { "epoch": 0.6987840338355802, "grad_norm": 0.8085004231596183, "learning_rate": 3.268605690483321e-06, "loss": 0.5474, "mean_token_accuracy": 0.8193291425704956, "num_tokens": 182599485.0, "step": 5287 }, { "epoch": 0.6989162040708432, "grad_norm": 0.8018720036937564, "learning_rate": 3.2668029162821172e-06, "loss": 0.5852, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 182665021.0, "step": 5288 }, { "epoch": 0.6990483743061062, "grad_norm": 0.802436527693144, "learning_rate": 3.265000617481947e-06, "loss": 0.5434, "mean_token_accuracy": 0.8195580244064331, "num_tokens": 182730557.0, "step": 5289 }, { "epoch": 0.6991805445413692, "grad_norm": 0.8388322095799747, "learning_rate": 3.2631987944664822e-06, "loss": 0.575, "mean_token_accuracy": 0.8076547384262085, "num_tokens": 182796093.0, "step": 5290 }, { "epoch": 0.6993127147766323, "grad_norm": 0.755880418972267, "learning_rate": 3.2613974476192937e-06, "loss": 0.5605, "mean_token_accuracy": 0.8141252398490906, "num_tokens": 182861629.0, "step": 5291 }, { "epoch": 0.6994448850118953, "grad_norm": 0.8445979057791584, "learning_rate": 3.259596577323847e-06, "loss": 0.6128, "mean_token_accuracy": 0.7954309582710266, "num_tokens": 182927165.0, "step": 5292 }, { "epoch": 0.6995770552471583, "grad_norm": 0.8641054760607599, "learning_rate": 3.2577961839635163e-06, "loss": 0.61, "mean_token_accuracy": 0.7975063920021057, "num_tokens": 182992701.0, "step": 5293 }, { "epoch": 0.6997092254824213, "grad_norm": 0.852207085364796, "learning_rate": 3.255996267921563e-06, "loss": 0.5777, "mean_token_accuracy": 0.806861162185669, "num_tokens": 183058237.0, "step": 5294 }, { "epoch": 0.6998413957176843, "grad_norm": 0.8248970453330569, "learning_rate": 3.254196829581152e-06, "loss": 0.5815, "mean_token_accuracy": 0.8042058348655701, "num_tokens": 183123773.0, "step": 5295 }, { "epoch": 0.6999735659529474, "grad_norm": 0.8307454976822476, "learning_rate": 3.2523978693253477e-06, "loss": 0.5922, "mean_token_accuracy": 0.8031681180000305, "num_tokens": 183189309.0, "step": 5296 }, { "epoch": 0.7001057361882104, "grad_norm": 0.800545666531027, "learning_rate": 3.2505993875371123e-06, "loss": 0.5811, "mean_token_accuracy": 0.8044194579124451, "num_tokens": 183254845.0, "step": 5297 }, { "epoch": 0.7002379064234734, "grad_norm": 0.8431721117738507, "learning_rate": 3.2488013845992988e-06, "loss": 0.6233, "mean_token_accuracy": 0.79330974817276, "num_tokens": 183320381.0, "step": 5298 }, { "epoch": 0.7003700766587364, "grad_norm": 0.7612841157701584, "learning_rate": 3.247003860894673e-06, "loss": 0.526, "mean_token_accuracy": 0.8235105872154236, "num_tokens": 183385917.0, "step": 5299 }, { "epoch": 0.7005022468939994, "grad_norm": 0.8487412531493027, "learning_rate": 3.2452068168058837e-06, "loss": 0.6137, "mean_token_accuracy": 0.7960108518600464, "num_tokens": 183451453.0, "step": 5300 }, { "epoch": 0.7006344171292624, "grad_norm": 0.8815163959380712, "learning_rate": 3.243410252715489e-06, "loss": 0.6423, "mean_token_accuracy": 0.7882736921310425, "num_tokens": 183516989.0, "step": 5301 }, { "epoch": 0.7007665873645255, "grad_norm": 0.8247850444331674, "learning_rate": 3.2416141690059327e-06, "loss": 0.5975, "mean_token_accuracy": 0.79872727394104, "num_tokens": 183582525.0, "step": 5302 }, { "epoch": 0.7008987575997885, "grad_norm": 0.8197691990167584, "learning_rate": 3.239818566059572e-06, "loss": 0.6016, "mean_token_accuracy": 0.7986814975738525, "num_tokens": 183648061.0, "step": 5303 }, { "epoch": 0.7010309278350515, "grad_norm": 0.7865655129957245, "learning_rate": 3.238023444258648e-06, "loss": 0.5562, "mean_token_accuracy": 0.8154376745223999, "num_tokens": 183713597.0, "step": 5304 }, { "epoch": 0.7011630980703145, "grad_norm": 0.8630266253360112, "learning_rate": 3.2362288039853064e-06, "loss": 0.5887, "mean_token_accuracy": 0.8057624101638794, "num_tokens": 183779133.0, "step": 5305 }, { "epoch": 0.7012952683055775, "grad_norm": 0.8136972307451582, "learning_rate": 3.2344346456215897e-06, "loss": 0.5694, "mean_token_accuracy": 0.8095928430557251, "num_tokens": 183844669.0, "step": 5306 }, { "epoch": 0.7014274385408406, "grad_norm": 0.8352962548746512, "learning_rate": 3.232640969549433e-06, "loss": 0.6223, "mean_token_accuracy": 0.7951562404632568, "num_tokens": 183910205.0, "step": 5307 }, { "epoch": 0.7015596087761036, "grad_norm": 0.8113629104874441, "learning_rate": 3.230847776150676e-06, "loss": 0.5641, "mean_token_accuracy": 0.8127670884132385, "num_tokens": 183975741.0, "step": 5308 }, { "epoch": 0.7016917790113666, "grad_norm": 0.8207730978688772, "learning_rate": 3.2290550658070497e-06, "loss": 0.6181, "mean_token_accuracy": 0.793401300907135, "num_tokens": 184041277.0, "step": 5309 }, { "epoch": 0.7018239492466296, "grad_norm": 0.7789367791830333, "learning_rate": 3.2272628389001886e-06, "loss": 0.6197, "mean_token_accuracy": 0.7917531728744507, "num_tokens": 184106813.0, "step": 5310 }, { "epoch": 0.7019561194818926, "grad_norm": 0.8493881738637138, "learning_rate": 3.2254710958116144e-06, "loss": 0.5919, "mean_token_accuracy": 0.8023592829704285, "num_tokens": 184172349.0, "step": 5311 }, { "epoch": 0.7020882897171556, "grad_norm": 1.0426467676252478, "learning_rate": 3.223679836922757e-06, "loss": 0.6102, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 184237885.0, "step": 5312 }, { "epoch": 0.7022204599524187, "grad_norm": 0.7810258001854099, "learning_rate": 3.221889062614937e-06, "loss": 0.5519, "mean_token_accuracy": 0.8131027817726135, "num_tokens": 184303421.0, "step": 5313 }, { "epoch": 0.7023526301876817, "grad_norm": 0.8115443516008561, "learning_rate": 3.22009877326937e-06, "loss": 0.5599, "mean_token_accuracy": 0.8145983219146729, "num_tokens": 184368957.0, "step": 5314 }, { "epoch": 0.7024848004229447, "grad_norm": 0.8669820564090445, "learning_rate": 3.218308969267174e-06, "loss": 0.6149, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 184434493.0, "step": 5315 }, { "epoch": 0.7026169706582077, "grad_norm": 0.8128035317960869, "learning_rate": 3.216519650989362e-06, "loss": 0.5836, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 184500029.0, "step": 5316 }, { "epoch": 0.7027491408934707, "grad_norm": 0.8547100179220224, "learning_rate": 3.2147308188168393e-06, "loss": 0.6196, "mean_token_accuracy": 0.7927451133728027, "num_tokens": 184565565.0, "step": 5317 }, { "epoch": 0.7028813111287338, "grad_norm": 0.7831888587320488, "learning_rate": 3.2129424731304117e-06, "loss": 0.5183, "mean_token_accuracy": 0.827203631401062, "num_tokens": 184631101.0, "step": 5318 }, { "epoch": 0.7030134813639968, "grad_norm": 0.8343188564253615, "learning_rate": 3.211154614310781e-06, "loss": 0.586, "mean_token_accuracy": 0.807822585105896, "num_tokens": 184696637.0, "step": 5319 }, { "epoch": 0.7031456515992599, "grad_norm": 0.7869344827766714, "learning_rate": 3.2093672427385468e-06, "loss": 0.579, "mean_token_accuracy": 0.8066322803497314, "num_tokens": 184762173.0, "step": 5320 }, { "epoch": 0.7032778218345229, "grad_norm": 0.7444006904164882, "learning_rate": 3.207580358794201e-06, "loss": 0.5406, "mean_token_accuracy": 0.8204584121704102, "num_tokens": 184827709.0, "step": 5321 }, { "epoch": 0.7034099920697859, "grad_norm": 0.8391406591283483, "learning_rate": 3.2057939628581344e-06, "loss": 0.5968, "mean_token_accuracy": 0.8016573190689087, "num_tokens": 184893245.0, "step": 5322 }, { "epoch": 0.703542162305049, "grad_norm": 0.7826276172555956, "learning_rate": 3.2040080553106327e-06, "loss": 0.5592, "mean_token_accuracy": 0.8178488612174988, "num_tokens": 184958781.0, "step": 5323 }, { "epoch": 0.703674332540312, "grad_norm": 0.8384345763010819, "learning_rate": 3.20222263653188e-06, "loss": 0.5809, "mean_token_accuracy": 0.8109815716743469, "num_tokens": 185024317.0, "step": 5324 }, { "epoch": 0.703806502775575, "grad_norm": 0.7808628653735765, "learning_rate": 3.200437706901955e-06, "loss": 0.5866, "mean_token_accuracy": 0.8053961396217346, "num_tokens": 185089853.0, "step": 5325 }, { "epoch": 0.703938673010838, "grad_norm": 0.8318477188505613, "learning_rate": 3.198653266800829e-06, "loss": 0.5888, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 185155389.0, "step": 5326 }, { "epoch": 0.704070843246101, "grad_norm": 0.81102383483299, "learning_rate": 3.196869316608374e-06, "loss": 0.6038, "mean_token_accuracy": 0.799062967300415, "num_tokens": 185220925.0, "step": 5327 }, { "epoch": 0.704203013481364, "grad_norm": 0.8053650798491032, "learning_rate": 3.195085856704356e-06, "loss": 0.5642, "mean_token_accuracy": 0.8158039450645447, "num_tokens": 185286461.0, "step": 5328 }, { "epoch": 0.7043351837166271, "grad_norm": 0.8413842541820432, "learning_rate": 3.1933028874684367e-06, "loss": 0.5931, "mean_token_accuracy": 0.802252471446991, "num_tokens": 185351997.0, "step": 5329 }, { "epoch": 0.7044673539518901, "grad_norm": 0.7636752315171673, "learning_rate": 3.1915204092801698e-06, "loss": 0.5406, "mean_token_accuracy": 0.8193749189376831, "num_tokens": 185417533.0, "step": 5330 }, { "epoch": 0.7045995241871531, "grad_norm": 0.7964607928378498, "learning_rate": 3.189738422519013e-06, "loss": 0.5648, "mean_token_accuracy": 0.8137285113334656, "num_tokens": 185483069.0, "step": 5331 }, { "epoch": 0.7047316944224161, "grad_norm": 0.8479330475260176, "learning_rate": 3.1879569275643102e-06, "loss": 0.6018, "mean_token_accuracy": 0.8021456599235535, "num_tokens": 185548605.0, "step": 5332 }, { "epoch": 0.7048638646576791, "grad_norm": 0.9339380459224953, "learning_rate": 3.186175924795307e-06, "loss": 0.6113, "mean_token_accuracy": 0.8004059195518494, "num_tokens": 185614141.0, "step": 5333 }, { "epoch": 0.7049960348929422, "grad_norm": 0.8085451125912857, "learning_rate": 3.1843954145911367e-06, "loss": 0.6102, "mean_token_accuracy": 0.7973385453224182, "num_tokens": 185679677.0, "step": 5334 }, { "epoch": 0.7051282051282052, "grad_norm": 0.7422669678170687, "learning_rate": 3.1826153973308397e-06, "loss": 0.5593, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 185745213.0, "step": 5335 }, { "epoch": 0.7052603753634682, "grad_norm": 0.8576103695859731, "learning_rate": 3.1808358733933398e-06, "loss": 0.5936, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 185810749.0, "step": 5336 }, { "epoch": 0.7053925455987312, "grad_norm": 0.8964321211330658, "learning_rate": 3.1790568431574613e-06, "loss": 0.6383, "mean_token_accuracy": 0.7898761034011841, "num_tokens": 185876285.0, "step": 5337 }, { "epoch": 0.7055247158339942, "grad_norm": 0.7983088231578396, "learning_rate": 3.177278307001923e-06, "loss": 0.4972, "mean_token_accuracy": 0.8366194367408752, "num_tokens": 185941821.0, "step": 5338 }, { "epoch": 0.7056568860692572, "grad_norm": 0.8280936406727692, "learning_rate": 3.17550026530534e-06, "loss": 0.6162, "mean_token_accuracy": 0.7901812791824341, "num_tokens": 186007357.0, "step": 5339 }, { "epoch": 0.7057890563045203, "grad_norm": 0.8072795310500528, "learning_rate": 3.173722718446217e-06, "loss": 0.5697, "mean_token_accuracy": 0.8109663128852844, "num_tokens": 186072893.0, "step": 5340 }, { "epoch": 0.7059212265397833, "grad_norm": 0.8785721543189867, "learning_rate": 3.171945666802958e-06, "loss": 0.5994, "mean_token_accuracy": 0.7985288500785828, "num_tokens": 186138429.0, "step": 5341 }, { "epoch": 0.7060533967750463, "grad_norm": 0.8640367476886059, "learning_rate": 3.1701691107538603e-06, "loss": 0.5948, "mean_token_accuracy": 0.803305447101593, "num_tokens": 186203965.0, "step": 5342 }, { "epoch": 0.7061855670103093, "grad_norm": 0.8826441992412413, "learning_rate": 3.1683930506771156e-06, "loss": 0.5852, "mean_token_accuracy": 0.8049841523170471, "num_tokens": 186269501.0, "step": 5343 }, { "epoch": 0.7063177372455723, "grad_norm": 0.7818209418104739, "learning_rate": 3.1666174869508114e-06, "loss": 0.5656, "mean_token_accuracy": 0.8105084896087646, "num_tokens": 186335037.0, "step": 5344 }, { "epoch": 0.7064499074808354, "grad_norm": 0.7910700578742839, "learning_rate": 3.1648424199529248e-06, "loss": 0.5756, "mean_token_accuracy": 0.8075021505355835, "num_tokens": 186400573.0, "step": 5345 }, { "epoch": 0.7065820777160984, "grad_norm": 0.7809588841907381, "learning_rate": 3.163067850061333e-06, "loss": 0.5714, "mean_token_accuracy": 0.8075326681137085, "num_tokens": 186466109.0, "step": 5346 }, { "epoch": 0.7067142479513614, "grad_norm": 0.8621071121613144, "learning_rate": 3.1612937776538043e-06, "loss": 0.6349, "mean_token_accuracy": 0.7903034090995789, "num_tokens": 186531645.0, "step": 5347 }, { "epoch": 0.7068464181866244, "grad_norm": 0.8360428891853133, "learning_rate": 3.159520203108003e-06, "loss": 0.5924, "mean_token_accuracy": 0.8022677302360535, "num_tokens": 186597181.0, "step": 5348 }, { "epoch": 0.7069785884218874, "grad_norm": 0.7868344132657259, "learning_rate": 3.1577471268014837e-06, "loss": 0.5539, "mean_token_accuracy": 0.8166890740394592, "num_tokens": 186662717.0, "step": 5349 }, { "epoch": 0.7071107586571505, "grad_norm": 0.8120925416765706, "learning_rate": 3.1559745491116976e-06, "loss": 0.5402, "mean_token_accuracy": 0.8204431533813477, "num_tokens": 186728253.0, "step": 5350 }, { "epoch": 0.7072429288924135, "grad_norm": 0.7840932773262401, "learning_rate": 3.1542024704159902e-06, "loss": 0.5577, "mean_token_accuracy": 0.8136979341506958, "num_tokens": 186793789.0, "step": 5351 }, { "epoch": 0.7073750991276765, "grad_norm": 0.8232615157719445, "learning_rate": 3.1524308910916024e-06, "loss": 0.6195, "mean_token_accuracy": 0.7936912178993225, "num_tokens": 186859325.0, "step": 5352 }, { "epoch": 0.7075072693629395, "grad_norm": 0.8373153537564482, "learning_rate": 3.1506598115156607e-06, "loss": 0.5551, "mean_token_accuracy": 0.8141710162162781, "num_tokens": 186924861.0, "step": 5353 }, { "epoch": 0.7076394395982025, "grad_norm": 0.8625092214479737, "learning_rate": 3.1488892320651978e-06, "loss": 0.5956, "mean_token_accuracy": 0.8037480115890503, "num_tokens": 186990397.0, "step": 5354 }, { "epoch": 0.7077716098334655, "grad_norm": 0.8398375225062938, "learning_rate": 3.1471191531171275e-06, "loss": 0.6036, "mean_token_accuracy": 0.8011994957923889, "num_tokens": 187055933.0, "step": 5355 }, { "epoch": 0.7079037800687286, "grad_norm": 0.7761634771390717, "learning_rate": 3.1453495750482653e-06, "loss": 0.526, "mean_token_accuracy": 0.8240141868591309, "num_tokens": 187121469.0, "step": 5356 }, { "epoch": 0.7080359503039916, "grad_norm": 0.8888972690386235, "learning_rate": 3.143580498235317e-06, "loss": 0.5656, "mean_token_accuracy": 0.8142778873443604, "num_tokens": 187187005.0, "step": 5357 }, { "epoch": 0.7081681205392546, "grad_norm": 0.7790041593175501, "learning_rate": 3.141811923054884e-06, "loss": 0.5542, "mean_token_accuracy": 0.8169484734535217, "num_tokens": 187252541.0, "step": 5358 }, { "epoch": 0.7083002907745176, "grad_norm": 0.7282448469804906, "learning_rate": 3.140043849883455e-06, "loss": 0.5677, "mean_token_accuracy": 0.8115156888961792, "num_tokens": 187318077.0, "step": 5359 }, { "epoch": 0.7084324610097806, "grad_norm": 0.7918602861190897, "learning_rate": 3.1382762790974187e-06, "loss": 0.544, "mean_token_accuracy": 0.817574143409729, "num_tokens": 187383613.0, "step": 5360 }, { "epoch": 0.7085646312450437, "grad_norm": 0.7810095785232933, "learning_rate": 3.1365092110730524e-06, "loss": 0.5511, "mean_token_accuracy": 0.8171010613441467, "num_tokens": 187449149.0, "step": 5361 }, { "epoch": 0.7086968014803067, "grad_norm": 0.8277869735837485, "learning_rate": 3.1347426461865316e-06, "loss": 0.6058, "mean_token_accuracy": 0.7993241548538208, "num_tokens": 187513669.0, "step": 5362 }, { "epoch": 0.7088289717155697, "grad_norm": 0.8512904311702495, "learning_rate": 3.132976584813916e-06, "loss": 0.538, "mean_token_accuracy": 0.821847140789032, "num_tokens": 187579205.0, "step": 5363 }, { "epoch": 0.7089611419508327, "grad_norm": 0.8726906153278909, "learning_rate": 3.131211027331166e-06, "loss": 0.5969, "mean_token_accuracy": 0.8013368248939514, "num_tokens": 187644741.0, "step": 5364 }, { "epoch": 0.7090933121860957, "grad_norm": 0.8547471062792207, "learning_rate": 3.1294459741141315e-06, "loss": 0.6023, "mean_token_accuracy": 0.7991240620613098, "num_tokens": 187710277.0, "step": 5365 }, { "epoch": 0.7092254824213587, "grad_norm": 0.8437605221211579, "learning_rate": 3.127681425538556e-06, "loss": 0.5664, "mean_token_accuracy": 0.8125534057617188, "num_tokens": 187775813.0, "step": 5366 }, { "epoch": 0.7093576526566218, "grad_norm": 0.8820619509220154, "learning_rate": 3.1259173819800746e-06, "loss": 0.5803, "mean_token_accuracy": 0.8061591982841492, "num_tokens": 187841349.0, "step": 5367 }, { "epoch": 0.7094898228918848, "grad_norm": 0.9313290414440135, "learning_rate": 3.124153843814215e-06, "loss": 0.6049, "mean_token_accuracy": 0.8018099069595337, "num_tokens": 187906885.0, "step": 5368 }, { "epoch": 0.7096219931271478, "grad_norm": 0.7981160444762617, "learning_rate": 3.122390811416398e-06, "loss": 0.5675, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 187972421.0, "step": 5369 }, { "epoch": 0.7097541633624108, "grad_norm": 0.841046251098619, "learning_rate": 3.1206282851619364e-06, "loss": 0.5843, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 188037957.0, "step": 5370 }, { "epoch": 0.7098863335976738, "grad_norm": 0.816036043316569, "learning_rate": 3.118866265426036e-06, "loss": 0.5872, "mean_token_accuracy": 0.8068764209747314, "num_tokens": 188103493.0, "step": 5371 }, { "epoch": 0.7100185038329369, "grad_norm": 0.8453357069478791, "learning_rate": 3.11710475258379e-06, "loss": 0.5575, "mean_token_accuracy": 0.8140947222709656, "num_tokens": 188169029.0, "step": 5372 }, { "epoch": 0.7101506740681999, "grad_norm": 0.884424603925073, "learning_rate": 3.1153437470101945e-06, "loss": 0.5793, "mean_token_accuracy": 0.8058539628982544, "num_tokens": 188234565.0, "step": 5373 }, { "epoch": 0.7102828443034629, "grad_norm": 0.9217152633865379, "learning_rate": 3.113583249080126e-06, "loss": 0.6637, "mean_token_accuracy": 0.7802466154098511, "num_tokens": 188300101.0, "step": 5374 }, { "epoch": 0.7104150145387259, "grad_norm": 0.8755605966210478, "learning_rate": 3.111823259168358e-06, "loss": 0.5371, "mean_token_accuracy": 0.8201227188110352, "num_tokens": 188365637.0, "step": 5375 }, { "epoch": 0.7105471847739889, "grad_norm": 0.8061696705685192, "learning_rate": 3.110063777649558e-06, "loss": 0.5966, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 188431173.0, "step": 5376 }, { "epoch": 0.710679355009252, "grad_norm": 0.853377364484542, "learning_rate": 3.1083048048982823e-06, "loss": 0.5607, "mean_token_accuracy": 0.8107068538665771, "num_tokens": 188496709.0, "step": 5377 }, { "epoch": 0.710811525244515, "grad_norm": 0.7321677698126827, "learning_rate": 3.106546341288977e-06, "loss": 0.5315, "mean_token_accuracy": 0.8235716223716736, "num_tokens": 188562245.0, "step": 5378 }, { "epoch": 0.710943695479778, "grad_norm": 0.8471056391691498, "learning_rate": 3.1047883871959838e-06, "loss": 0.5689, "mean_token_accuracy": 0.8116682767868042, "num_tokens": 188627781.0, "step": 5379 }, { "epoch": 0.711075865715041, "grad_norm": 0.8124952868437275, "learning_rate": 3.1030309429935345e-06, "loss": 0.5641, "mean_token_accuracy": 0.8133622407913208, "num_tokens": 188693317.0, "step": 5380 }, { "epoch": 0.711208035950304, "grad_norm": 0.8589483663080786, "learning_rate": 3.101274009055754e-06, "loss": 0.6233, "mean_token_accuracy": 0.7925772070884705, "num_tokens": 188758853.0, "step": 5381 }, { "epoch": 0.711340206185567, "grad_norm": 0.8636852732849029, "learning_rate": 3.0995175857566536e-06, "loss": 0.5885, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 188824389.0, "step": 5382 }, { "epoch": 0.71147237642083, "grad_norm": 0.8626033099591522, "learning_rate": 3.097761673470141e-06, "loss": 0.6036, "mean_token_accuracy": 0.8003754019737244, "num_tokens": 188889925.0, "step": 5383 }, { "epoch": 0.7116045466560931, "grad_norm": 0.8529005027076093, "learning_rate": 3.0960062725700125e-06, "loss": 0.5888, "mean_token_accuracy": 0.8019777536392212, "num_tokens": 188955461.0, "step": 5384 }, { "epoch": 0.7117367168913561, "grad_norm": 0.9036213192095492, "learning_rate": 3.094251383429957e-06, "loss": 0.6398, "mean_token_accuracy": 0.7878769636154175, "num_tokens": 189020997.0, "step": 5385 }, { "epoch": 0.7118688871266191, "grad_norm": 0.7756667920429116, "learning_rate": 3.0924970064235554e-06, "loss": 0.5486, "mean_token_accuracy": 0.8181235790252686, "num_tokens": 189086533.0, "step": 5386 }, { "epoch": 0.7120010573618821, "grad_norm": 0.8523884453582976, "learning_rate": 3.0907431419242743e-06, "loss": 0.6164, "mean_token_accuracy": 0.7957667112350464, "num_tokens": 189152069.0, "step": 5387 }, { "epoch": 0.7121332275971451, "grad_norm": 0.868955964775857, "learning_rate": 3.0889897903054765e-06, "loss": 0.5905, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 189217605.0, "step": 5388 }, { "epoch": 0.7122653978324082, "grad_norm": 0.7871528210504051, "learning_rate": 3.0872369519404143e-06, "loss": 0.5458, "mean_token_accuracy": 0.8190696835517883, "num_tokens": 189283141.0, "step": 5389 }, { "epoch": 0.7123975680676712, "grad_norm": 0.7652278581038692, "learning_rate": 3.0854846272022323e-06, "loss": 0.5367, "mean_token_accuracy": 0.8200158476829529, "num_tokens": 189348677.0, "step": 5390 }, { "epoch": 0.7125297383029342, "grad_norm": 0.8336455195428548, "learning_rate": 3.083732816463959e-06, "loss": 0.5929, "mean_token_accuracy": 0.802435576915741, "num_tokens": 189414213.0, "step": 5391 }, { "epoch": 0.7126619085381972, "grad_norm": 0.795891635672386, "learning_rate": 3.0819815200985216e-06, "loss": 0.5593, "mean_token_accuracy": 0.8154224157333374, "num_tokens": 189479749.0, "step": 5392 }, { "epoch": 0.7127940787734602, "grad_norm": 0.7925275677536749, "learning_rate": 3.0802307384787338e-06, "loss": 0.564, "mean_token_accuracy": 0.8136979341506958, "num_tokens": 189545285.0, "step": 5393 }, { "epoch": 0.7129262490087233, "grad_norm": 0.796806496949463, "learning_rate": 3.0784804719773026e-06, "loss": 0.6021, "mean_token_accuracy": 0.7982847094535828, "num_tokens": 189610821.0, "step": 5394 }, { "epoch": 0.7130584192439863, "grad_norm": 0.8977249065339525, "learning_rate": 3.0767307209668176e-06, "loss": 0.6124, "mean_token_accuracy": 0.7971859574317932, "num_tokens": 189676357.0, "step": 5395 }, { "epoch": 0.7131905894792493, "grad_norm": 0.7918570054958026, "learning_rate": 3.0749814858197715e-06, "loss": 0.5316, "mean_token_accuracy": 0.8234342336654663, "num_tokens": 189741893.0, "step": 5396 }, { "epoch": 0.7133227597145123, "grad_norm": 0.930312470513881, "learning_rate": 3.073232766908535e-06, "loss": 0.6314, "mean_token_accuracy": 0.7916310429573059, "num_tokens": 189807429.0, "step": 5397 }, { "epoch": 0.7134549299497753, "grad_norm": 0.8826625222319114, "learning_rate": 3.0714845646053737e-06, "loss": 0.5875, "mean_token_accuracy": 0.8048620223999023, "num_tokens": 189872965.0, "step": 5398 }, { "epoch": 0.7135871001850383, "grad_norm": 0.8098373781963543, "learning_rate": 3.069736879282446e-06, "loss": 0.586, "mean_token_accuracy": 0.8048315048217773, "num_tokens": 189938501.0, "step": 5399 }, { "epoch": 0.7137192704203014, "grad_norm": 0.7909825599860894, "learning_rate": 3.0679897113117985e-06, "loss": 0.5697, "mean_token_accuracy": 0.8107068538665771, "num_tokens": 190004037.0, "step": 5400 }, { "epoch": 0.7138514406555644, "grad_norm": 0.9027790874430766, "learning_rate": 3.0662430610653628e-06, "loss": 0.5803, "mean_token_accuracy": 0.8058234453201294, "num_tokens": 190069573.0, "step": 5401 }, { "epoch": 0.7139836108908274, "grad_norm": 0.789769485394909, "learning_rate": 3.064496928914965e-06, "loss": 0.55, "mean_token_accuracy": 0.816429615020752, "num_tokens": 190135109.0, "step": 5402 }, { "epoch": 0.7141157811260904, "grad_norm": 0.7451456302036338, "learning_rate": 3.0627513152323227e-06, "loss": 0.5148, "mean_token_accuracy": 0.8277987837791443, "num_tokens": 190200645.0, "step": 5403 }, { "epoch": 0.7142479513613534, "grad_norm": 0.7522991795771474, "learning_rate": 3.061006220389039e-06, "loss": 0.5429, "mean_token_accuracy": 0.8196191191673279, "num_tokens": 190266181.0, "step": 5404 }, { "epoch": 0.7143801215966165, "grad_norm": 0.8273730137632636, "learning_rate": 3.0592616447566103e-06, "loss": 0.576, "mean_token_accuracy": 0.8071816563606262, "num_tokens": 190331717.0, "step": 5405 }, { "epoch": 0.7145122918318795, "grad_norm": 0.8465658147069588, "learning_rate": 3.057517588706417e-06, "loss": 0.606, "mean_token_accuracy": 0.7994139790534973, "num_tokens": 190397253.0, "step": 5406 }, { "epoch": 0.7146444620671425, "grad_norm": 0.877612746705085, "learning_rate": 3.0557740526097345e-06, "loss": 0.6441, "mean_token_accuracy": 0.7859083414077759, "num_tokens": 190462789.0, "step": 5407 }, { "epoch": 0.7147766323024055, "grad_norm": 0.8478059290383626, "learning_rate": 3.0540310368377213e-06, "loss": 0.6072, "mean_token_accuracy": 0.7951868176460266, "num_tokens": 190528325.0, "step": 5408 }, { "epoch": 0.7149088025376685, "grad_norm": 0.7621409216495232, "learning_rate": 3.0522885417614354e-06, "loss": 0.589, "mean_token_accuracy": 0.8040532469749451, "num_tokens": 190593861.0, "step": 5409 }, { "epoch": 0.7150409727729315, "grad_norm": 0.836160258346611, "learning_rate": 3.050546567751813e-06, "loss": 0.5982, "mean_token_accuracy": 0.7959761023521423, "num_tokens": 190655401.0, "step": 5410 }, { "epoch": 0.7151731430081946, "grad_norm": 0.7684705846412905, "learning_rate": 3.048805115179686e-06, "loss": 0.525, "mean_token_accuracy": 0.8245177865028381, "num_tokens": 190720937.0, "step": 5411 }, { "epoch": 0.7153053132434576, "grad_norm": 0.8817784893617384, "learning_rate": 3.0470641844157726e-06, "loss": 0.6326, "mean_token_accuracy": 0.7920278310775757, "num_tokens": 190786473.0, "step": 5412 }, { "epoch": 0.7154374834787206, "grad_norm": 0.8656261860252942, "learning_rate": 3.045323775830682e-06, "loss": 0.6202, "mean_token_accuracy": 0.7958430051803589, "num_tokens": 190852009.0, "step": 5413 }, { "epoch": 0.7155696537139836, "grad_norm": 0.8296765956016714, "learning_rate": 3.043583889794906e-06, "loss": 0.5835, "mean_token_accuracy": 0.8081888556480408, "num_tokens": 190917545.0, "step": 5414 }, { "epoch": 0.7157018239492466, "grad_norm": 0.807712840946216, "learning_rate": 3.041844526678838e-06, "loss": 0.5278, "mean_token_accuracy": 0.8264406323432922, "num_tokens": 190983081.0, "step": 5415 }, { "epoch": 0.7158339941845097, "grad_norm": 0.8179859271840554, "learning_rate": 3.0401056868527445e-06, "loss": 0.557, "mean_token_accuracy": 0.8140642046928406, "num_tokens": 191048617.0, "step": 5416 }, { "epoch": 0.7159661644197727, "grad_norm": 0.8478101972153582, "learning_rate": 3.038367370686792e-06, "loss": 0.6286, "mean_token_accuracy": 0.7924246191978455, "num_tokens": 191114153.0, "step": 5417 }, { "epoch": 0.7160983346550357, "grad_norm": 0.7991682631411334, "learning_rate": 3.036629578551031e-06, "loss": 0.5633, "mean_token_accuracy": 0.8094402551651001, "num_tokens": 191179689.0, "step": 5418 }, { "epoch": 0.7162305048902987, "grad_norm": 0.8300658512019299, "learning_rate": 3.0348923108154037e-06, "loss": 0.5841, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 191245225.0, "step": 5419 }, { "epoch": 0.7163626751255617, "grad_norm": 0.8639929642799369, "learning_rate": 3.0331555678497336e-06, "loss": 0.6041, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 191310761.0, "step": 5420 }, { "epoch": 0.7164948453608248, "grad_norm": 0.8853700327890464, "learning_rate": 3.031419350023739e-06, "loss": 0.6261, "mean_token_accuracy": 0.7939659357070923, "num_tokens": 191376297.0, "step": 5421 }, { "epoch": 0.7166270155960878, "grad_norm": 0.8166119659049418, "learning_rate": 3.029683657707024e-06, "loss": 0.5636, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 191441833.0, "step": 5422 }, { "epoch": 0.7167591858313508, "grad_norm": 0.7690950568890899, "learning_rate": 3.0279484912690842e-06, "loss": 0.5381, "mean_token_accuracy": 0.821999728679657, "num_tokens": 191507369.0, "step": 5423 }, { "epoch": 0.7168913560666138, "grad_norm": 0.8076490584442741, "learning_rate": 3.026213851079296e-06, "loss": 0.5662, "mean_token_accuracy": 0.8144762516021729, "num_tokens": 191572905.0, "step": 5424 }, { "epoch": 0.7170235263018768, "grad_norm": 0.8401843769616235, "learning_rate": 3.0244797375069303e-06, "loss": 0.5418, "mean_token_accuracy": 0.8190391659736633, "num_tokens": 191638441.0, "step": 5425 }, { "epoch": 0.7171556965371398, "grad_norm": 0.8151287974710575, "learning_rate": 3.0227461509211443e-06, "loss": 0.5918, "mean_token_accuracy": 0.8016725778579712, "num_tokens": 191703977.0, "step": 5426 }, { "epoch": 0.7172878667724029, "grad_norm": 0.8122101870871313, "learning_rate": 3.0210130916909787e-06, "loss": 0.5704, "mean_token_accuracy": 0.8104016780853271, "num_tokens": 191769513.0, "step": 5427 }, { "epoch": 0.7174200370076659, "grad_norm": 0.8398183606162913, "learning_rate": 3.0192805601853713e-06, "loss": 0.6128, "mean_token_accuracy": 0.7970943450927734, "num_tokens": 191835049.0, "step": 5428 }, { "epoch": 0.7175522072429289, "grad_norm": 0.7878514018153342, "learning_rate": 3.0175485567731372e-06, "loss": 0.5526, "mean_token_accuracy": 0.816246509552002, "num_tokens": 191900585.0, "step": 5429 }, { "epoch": 0.7176843774781919, "grad_norm": 0.8279086494723522, "learning_rate": 3.015817081822986e-06, "loss": 0.6211, "mean_token_accuracy": 0.7970180511474609, "num_tokens": 191966121.0, "step": 5430 }, { "epoch": 0.7178165477134549, "grad_norm": 0.7558359237818414, "learning_rate": 3.014086135703511e-06, "loss": 0.4956, "mean_token_accuracy": 0.8368178606033325, "num_tokens": 192031657.0, "step": 5431 }, { "epoch": 0.717948717948718, "grad_norm": 0.8328934950469075, "learning_rate": 3.0123557187831987e-06, "loss": 0.5934, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 192097193.0, "step": 5432 }, { "epoch": 0.718080888183981, "grad_norm": 0.8318825945235999, "learning_rate": 3.0106258314304116e-06, "loss": 0.6376, "mean_token_accuracy": 0.7896013855934143, "num_tokens": 192162729.0, "step": 5433 }, { "epoch": 0.718213058419244, "grad_norm": 0.8532175703915357, "learning_rate": 3.0088964740134143e-06, "loss": 0.6011, "mean_token_accuracy": 0.799062967300415, "num_tokens": 192228265.0, "step": 5434 }, { "epoch": 0.718345228654507, "grad_norm": 0.8730115551114802, "learning_rate": 3.0071676469003447e-06, "loss": 0.6158, "mean_token_accuracy": 0.79349285364151, "num_tokens": 192293801.0, "step": 5435 }, { "epoch": 0.71847739888977, "grad_norm": 0.7421519812881983, "learning_rate": 3.0054393504592366e-06, "loss": 0.5327, "mean_token_accuracy": 0.821603000164032, "num_tokens": 192359337.0, "step": 5436 }, { "epoch": 0.718609569125033, "grad_norm": 0.7914175811507131, "learning_rate": 3.0037115850580083e-06, "loss": 0.5268, "mean_token_accuracy": 0.8250671625137329, "num_tokens": 192424873.0, "step": 5437 }, { "epoch": 0.7187417393602961, "grad_norm": 0.846071481960163, "learning_rate": 3.001984351064467e-06, "loss": 0.5673, "mean_token_accuracy": 0.8099438548088074, "num_tokens": 192490409.0, "step": 5438 }, { "epoch": 0.7188739095955591, "grad_norm": 0.7463624765220054, "learning_rate": 3.0002576488463003e-06, "loss": 0.5097, "mean_token_accuracy": 0.8293401002883911, "num_tokens": 192555945.0, "step": 5439 }, { "epoch": 0.7190060798308221, "grad_norm": 0.7835927490726089, "learning_rate": 2.9985314787710895e-06, "loss": 0.5703, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 192621481.0, "step": 5440 }, { "epoch": 0.7191382500660851, "grad_norm": 0.7567277018020261, "learning_rate": 2.996805841206302e-06, "loss": 0.4969, "mean_token_accuracy": 0.8371230363845825, "num_tokens": 192687017.0, "step": 5441 }, { "epoch": 0.7192704203013481, "grad_norm": 0.8493108182207835, "learning_rate": 2.9950807365192858e-06, "loss": 0.543, "mean_token_accuracy": 0.8192070722579956, "num_tokens": 192752553.0, "step": 5442 }, { "epoch": 0.7194025905366112, "grad_norm": 0.8363819581925149, "learning_rate": 2.9933561650772817e-06, "loss": 0.5927, "mean_token_accuracy": 0.8039616942405701, "num_tokens": 192818089.0, "step": 5443 }, { "epoch": 0.7195347607718742, "grad_norm": 0.8435751609237532, "learning_rate": 2.991632127247416e-06, "loss": 0.6487, "mean_token_accuracy": 0.7832987308502197, "num_tokens": 192883625.0, "step": 5444 }, { "epoch": 0.7196669310071372, "grad_norm": 0.8410864606126462, "learning_rate": 2.9899086233967003e-06, "loss": 0.5803, "mean_token_accuracy": 0.807792067527771, "num_tokens": 192949161.0, "step": 5445 }, { "epoch": 0.7197991012424002, "grad_norm": 0.8045561139550892, "learning_rate": 2.9881856538920286e-06, "loss": 0.5634, "mean_token_accuracy": 0.8132706880569458, "num_tokens": 193014697.0, "step": 5446 }, { "epoch": 0.7199312714776632, "grad_norm": 0.8062963475643644, "learning_rate": 2.9864632191001915e-06, "loss": 0.5233, "mean_token_accuracy": 0.8243803977966309, "num_tokens": 193080233.0, "step": 5447 }, { "epoch": 0.7200634417129262, "grad_norm": 0.84537791166157, "learning_rate": 2.984741319387854e-06, "loss": 0.5818, "mean_token_accuracy": 0.8062812685966492, "num_tokens": 193145769.0, "step": 5448 }, { "epoch": 0.7201956119481893, "grad_norm": 0.7610367915522307, "learning_rate": 2.983019955121576e-06, "loss": 0.5575, "mean_token_accuracy": 0.8130112290382385, "num_tokens": 193211305.0, "step": 5449 }, { "epoch": 0.7203277821834523, "grad_norm": 0.8437581235596313, "learning_rate": 2.9812991266677984e-06, "loss": 0.625, "mean_token_accuracy": 0.7950341701507568, "num_tokens": 193276841.0, "step": 5450 }, { "epoch": 0.7204599524187153, "grad_norm": 0.7963157920882843, "learning_rate": 2.9795788343928512e-06, "loss": 0.5661, "mean_token_accuracy": 0.8128891587257385, "num_tokens": 193342377.0, "step": 5451 }, { "epoch": 0.7205921226539783, "grad_norm": 0.7750613826389281, "learning_rate": 2.977859078662946e-06, "loss": 0.5275, "mean_token_accuracy": 0.8238157629966736, "num_tokens": 193407913.0, "step": 5452 }, { "epoch": 0.7207242928892413, "grad_norm": 0.8417841751623636, "learning_rate": 2.976139859844185e-06, "loss": 0.5949, "mean_token_accuracy": 0.8012452721595764, "num_tokens": 193473449.0, "step": 5453 }, { "epoch": 0.7208564631245044, "grad_norm": 0.8280855788993061, "learning_rate": 2.974421178302553e-06, "loss": 0.5712, "mean_token_accuracy": 0.8095165491104126, "num_tokens": 193538985.0, "step": 5454 }, { "epoch": 0.7209886333597674, "grad_norm": 0.8786653968718129, "learning_rate": 2.972703034403923e-06, "loss": 0.6243, "mean_token_accuracy": 0.7955988049507141, "num_tokens": 193604521.0, "step": 5455 }, { "epoch": 0.7211208035950304, "grad_norm": 0.7655082174100808, "learning_rate": 2.9709854285140487e-06, "loss": 0.5186, "mean_token_accuracy": 0.8280429840087891, "num_tokens": 193670057.0, "step": 5456 }, { "epoch": 0.7212529738302934, "grad_norm": 0.8059859189960397, "learning_rate": 2.9692683609985774e-06, "loss": 0.5849, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 193735593.0, "step": 5457 }, { "epoch": 0.7213851440655564, "grad_norm": 0.8227218264371662, "learning_rate": 2.967551832223032e-06, "loss": 0.5776, "mean_token_accuracy": 0.8075326681137085, "num_tokens": 193801129.0, "step": 5458 }, { "epoch": 0.7215173143008194, "grad_norm": 0.8448864489785901, "learning_rate": 2.965835842552829e-06, "loss": 0.5663, "mean_token_accuracy": 0.8123703002929688, "num_tokens": 193866665.0, "step": 5459 }, { "epoch": 0.7216494845360825, "grad_norm": 0.8297806494333214, "learning_rate": 2.9641203923532665e-06, "loss": 0.5434, "mean_token_accuracy": 0.8196343779563904, "num_tokens": 193932201.0, "step": 5460 }, { "epoch": 0.7217816547713455, "grad_norm": 0.857581479159413, "learning_rate": 2.9624054819895256e-06, "loss": 0.5637, "mean_token_accuracy": 0.8143541812896729, "num_tokens": 193997737.0, "step": 5461 }, { "epoch": 0.7219138250066085, "grad_norm": 0.8522352093076266, "learning_rate": 2.9606911118266757e-06, "loss": 0.5713, "mean_token_accuracy": 0.8117751479148865, "num_tokens": 194063273.0, "step": 5462 }, { "epoch": 0.7220459952418715, "grad_norm": 0.8354258026945549, "learning_rate": 2.958977282229672e-06, "loss": 0.5989, "mean_token_accuracy": 0.8019319772720337, "num_tokens": 194128809.0, "step": 5463 }, { "epoch": 0.7221781654771345, "grad_norm": 0.8705119245707238, "learning_rate": 2.9572639935633533e-06, "loss": 0.573, "mean_token_accuracy": 0.808509349822998, "num_tokens": 194194345.0, "step": 5464 }, { "epoch": 0.7223103357123976, "grad_norm": 0.9310814265276997, "learning_rate": 2.955551246192439e-06, "loss": 0.6157, "mean_token_accuracy": 0.7988951206207275, "num_tokens": 194259881.0, "step": 5465 }, { "epoch": 0.7224425059476606, "grad_norm": 0.784765347751049, "learning_rate": 2.953839040481543e-06, "loss": 0.5623, "mean_token_accuracy": 0.8167195916175842, "num_tokens": 194325417.0, "step": 5466 }, { "epoch": 0.7225746761829236, "grad_norm": 0.9115134698919436, "learning_rate": 2.952127376795154e-06, "loss": 0.615, "mean_token_accuracy": 0.7958887815475464, "num_tokens": 194390953.0, "step": 5467 }, { "epoch": 0.7227068464181866, "grad_norm": 0.7735214307011355, "learning_rate": 2.950416255497652e-06, "loss": 0.5362, "mean_token_accuracy": 0.821786105632782, "num_tokens": 194456489.0, "step": 5468 }, { "epoch": 0.7228390166534496, "grad_norm": 0.757003080824117, "learning_rate": 2.948705676953295e-06, "loss": 0.5357, "mean_token_accuracy": 0.8206415772438049, "num_tokens": 194522025.0, "step": 5469 }, { "epoch": 0.7229711868887126, "grad_norm": 0.8749391385686787, "learning_rate": 2.946995641526236e-06, "loss": 0.6007, "mean_token_accuracy": 0.8009247779846191, "num_tokens": 194587561.0, "step": 5470 }, { "epoch": 0.7231033571239757, "grad_norm": 0.872103220147921, "learning_rate": 2.9452861495804984e-06, "loss": 0.6013, "mean_token_accuracy": 0.8023898005485535, "num_tokens": 194653097.0, "step": 5471 }, { "epoch": 0.7232355273592387, "grad_norm": 0.8359343724484011, "learning_rate": 2.9435772014800024e-06, "loss": 0.5466, "mean_token_accuracy": 0.8191002607345581, "num_tokens": 194718633.0, "step": 5472 }, { "epoch": 0.7233676975945017, "grad_norm": 0.8849245152774009, "learning_rate": 2.9418687975885455e-06, "loss": 0.5969, "mean_token_accuracy": 0.8028171062469482, "num_tokens": 194784169.0, "step": 5473 }, { "epoch": 0.7234998678297647, "grad_norm": 0.8661681412302036, "learning_rate": 2.9401609382698137e-06, "loss": 0.6052, "mean_token_accuracy": 0.7998870611190796, "num_tokens": 194849705.0, "step": 5474 }, { "epoch": 0.7236320380650277, "grad_norm": 0.8528553727577467, "learning_rate": 2.9384536238873694e-06, "loss": 0.6261, "mean_token_accuracy": 0.7928519248962402, "num_tokens": 194915241.0, "step": 5475 }, { "epoch": 0.7237642083002908, "grad_norm": 0.8282744983337703, "learning_rate": 2.9367468548046696e-06, "loss": 0.6095, "mean_token_accuracy": 0.7956293225288391, "num_tokens": 194980777.0, "step": 5476 }, { "epoch": 0.7238963785355538, "grad_norm": 0.8470648046445529, "learning_rate": 2.935040631385047e-06, "loss": 0.5729, "mean_token_accuracy": 0.8081430792808533, "num_tokens": 195046313.0, "step": 5477 }, { "epoch": 0.7240285487708168, "grad_norm": 0.825005915170052, "learning_rate": 2.933334953991721e-06, "loss": 0.5835, "mean_token_accuracy": 0.8066933155059814, "num_tokens": 195111849.0, "step": 5478 }, { "epoch": 0.7241607190060798, "grad_norm": 0.910626136559436, "learning_rate": 2.931629822987797e-06, "loss": 0.566, "mean_token_accuracy": 0.8138963580131531, "num_tokens": 195177385.0, "step": 5479 }, { "epoch": 0.7242928892413428, "grad_norm": 0.8349267006279545, "learning_rate": 2.929925238736258e-06, "loss": 0.5622, "mean_token_accuracy": 0.8142168521881104, "num_tokens": 195242921.0, "step": 5480 }, { "epoch": 0.7244250594766058, "grad_norm": 0.848975077605387, "learning_rate": 2.9282212015999766e-06, "loss": 0.6033, "mean_token_accuracy": 0.7994597554206848, "num_tokens": 195308457.0, "step": 5481 }, { "epoch": 0.7245572297118689, "grad_norm": 0.8501871801916764, "learning_rate": 2.9265177119417065e-06, "loss": 0.6164, "mean_token_accuracy": 0.7935691475868225, "num_tokens": 195373993.0, "step": 5482 }, { "epoch": 0.7246893999471319, "grad_norm": 0.8601022440084639, "learning_rate": 2.924814770124087e-06, "loss": 0.6082, "mean_token_accuracy": 0.7971554398536682, "num_tokens": 195439529.0, "step": 5483 }, { "epoch": 0.7248215701823949, "grad_norm": 0.8566940881979904, "learning_rate": 2.9231123765096348e-06, "loss": 0.5728, "mean_token_accuracy": 0.8087077140808105, "num_tokens": 195505065.0, "step": 5484 }, { "epoch": 0.7249537404176579, "grad_norm": 0.7932233310700447, "learning_rate": 2.9214105314607566e-06, "loss": 0.5767, "mean_token_accuracy": 0.807517409324646, "num_tokens": 195570601.0, "step": 5485 }, { "epoch": 0.7250859106529209, "grad_norm": 0.8442319321060676, "learning_rate": 2.919709235339739e-06, "loss": 0.6418, "mean_token_accuracy": 0.7852215766906738, "num_tokens": 195636137.0, "step": 5486 }, { "epoch": 0.725218080888184, "grad_norm": 0.9035944815791314, "learning_rate": 2.918008488508754e-06, "loss": 0.5877, "mean_token_accuracy": 0.8058081865310669, "num_tokens": 195701673.0, "step": 5487 }, { "epoch": 0.725350251123447, "grad_norm": 0.8289858163847236, "learning_rate": 2.9163082913298503e-06, "loss": 0.5173, "mean_token_accuracy": 0.8289433717727661, "num_tokens": 195767209.0, "step": 5488 }, { "epoch": 0.72548242135871, "grad_norm": 0.795749546254593, "learning_rate": 2.914608644164971e-06, "loss": 0.5635, "mean_token_accuracy": 0.812298595905304, "num_tokens": 195832704.0, "step": 5489 }, { "epoch": 0.725614591593973, "grad_norm": 0.9226486786840575, "learning_rate": 2.9129095473759315e-06, "loss": 0.603, "mean_token_accuracy": 0.7993834614753723, "num_tokens": 195898240.0, "step": 5490 }, { "epoch": 0.725746761829236, "grad_norm": 0.8408183379948904, "learning_rate": 2.9112110013244345e-06, "loss": 0.549, "mean_token_accuracy": 0.8149188160896301, "num_tokens": 195963776.0, "step": 5491 }, { "epoch": 0.725878932064499, "grad_norm": 0.9083969648654344, "learning_rate": 2.9095130063720645e-06, "loss": 0.5989, "mean_token_accuracy": 0.8011079430580139, "num_tokens": 196029312.0, "step": 5492 }, { "epoch": 0.7260111022997621, "grad_norm": 0.8507941264228468, "learning_rate": 2.9078155628802924e-06, "loss": 0.5519, "mean_token_accuracy": 0.8161091208457947, "num_tokens": 196094848.0, "step": 5493 }, { "epoch": 0.7261432725350251, "grad_norm": 0.7951524508469963, "learning_rate": 2.9061186712104624e-06, "loss": 0.5681, "mean_token_accuracy": 0.8110273480415344, "num_tokens": 196160384.0, "step": 5494 }, { "epoch": 0.7262754427702881, "grad_norm": 0.8042421845796699, "learning_rate": 2.9044223317238145e-06, "loss": 0.6066, "mean_token_accuracy": 0.7988035678863525, "num_tokens": 196225920.0, "step": 5495 }, { "epoch": 0.7264076130055511, "grad_norm": 0.841248913682327, "learning_rate": 2.902726544781459e-06, "loss": 0.6097, "mean_token_accuracy": 0.7967281341552734, "num_tokens": 196291456.0, "step": 5496 }, { "epoch": 0.7265397832408141, "grad_norm": 0.8623225298402409, "learning_rate": 2.9010313107443943e-06, "loss": 0.6029, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 196356992.0, "step": 5497 }, { "epoch": 0.7266719534760772, "grad_norm": 0.8144146239420443, "learning_rate": 2.8993366299735027e-06, "loss": 0.5902, "mean_token_accuracy": 0.8035038709640503, "num_tokens": 196422528.0, "step": 5498 }, { "epoch": 0.7268041237113402, "grad_norm": 0.8307552037807752, "learning_rate": 2.897642502829543e-06, "loss": 0.5904, "mean_token_accuracy": 0.803458034992218, "num_tokens": 196488064.0, "step": 5499 }, { "epoch": 0.7269362939466032, "grad_norm": 0.8206179783402182, "learning_rate": 2.895948929673162e-06, "loss": 0.5769, "mean_token_accuracy": 0.8056403398513794, "num_tokens": 196553600.0, "step": 5500 }, { "epoch": 0.7270684641818662, "grad_norm": 0.8191099043086519, "learning_rate": 2.8942559108648845e-06, "loss": 0.5348, "mean_token_accuracy": 0.8205652832984924, "num_tokens": 196619136.0, "step": 5501 }, { "epoch": 0.7272006344171292, "grad_norm": 0.8270435296796407, "learning_rate": 2.892563446765121e-06, "loss": 0.5998, "mean_token_accuracy": 0.8031070828437805, "num_tokens": 196684672.0, "step": 5502 }, { "epoch": 0.7273328046523923, "grad_norm": 0.9270414759486388, "learning_rate": 2.890871537734159e-06, "loss": 0.6383, "mean_token_accuracy": 0.7859693765640259, "num_tokens": 196750208.0, "step": 5503 }, { "epoch": 0.7274649748876553, "grad_norm": 0.7754082248765737, "learning_rate": 2.889180184132172e-06, "loss": 0.5679, "mean_token_accuracy": 0.812873899936676, "num_tokens": 196815744.0, "step": 5504 }, { "epoch": 0.7275971451229183, "grad_norm": 0.8709256404774492, "learning_rate": 2.8874893863192134e-06, "loss": 0.6198, "mean_token_accuracy": 0.7940574884414673, "num_tokens": 196881280.0, "step": 5505 }, { "epoch": 0.7277293153581813, "grad_norm": 0.8751576788540847, "learning_rate": 2.8857991446552213e-06, "loss": 0.6189, "mean_token_accuracy": 0.7940269708633423, "num_tokens": 196946816.0, "step": 5506 }, { "epoch": 0.7278614855934443, "grad_norm": 0.7705594253657568, "learning_rate": 2.8841094595000075e-06, "loss": 0.5344, "mean_token_accuracy": 0.821664035320282, "num_tokens": 197012352.0, "step": 5507 }, { "epoch": 0.7279936558287073, "grad_norm": 0.8616527961812496, "learning_rate": 2.882420331213277e-06, "loss": 0.5751, "mean_token_accuracy": 0.8075631856918335, "num_tokens": 197077888.0, "step": 5508 }, { "epoch": 0.7281258260639704, "grad_norm": 0.847937722591521, "learning_rate": 2.880731760154605e-06, "loss": 0.5911, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 197143424.0, "step": 5509 }, { "epoch": 0.7282579962992334, "grad_norm": 0.7388522319138318, "learning_rate": 2.8790437466834548e-06, "loss": 0.4959, "mean_token_accuracy": 0.8334299921989441, "num_tokens": 197208960.0, "step": 5510 }, { "epoch": 0.7283901665344964, "grad_norm": 0.874112277591049, "learning_rate": 2.8773562911591695e-06, "loss": 0.5849, "mean_token_accuracy": 0.8051977753639221, "num_tokens": 197274496.0, "step": 5511 }, { "epoch": 0.7285223367697594, "grad_norm": 0.8472067205225051, "learning_rate": 2.8756693939409745e-06, "loss": 0.6333, "mean_token_accuracy": 0.7897081971168518, "num_tokens": 197340032.0, "step": 5512 }, { "epoch": 0.7286545070050224, "grad_norm": 0.8219204975678687, "learning_rate": 2.873983055387972e-06, "loss": 0.5748, "mean_token_accuracy": 0.8100659251213074, "num_tokens": 197405568.0, "step": 5513 }, { "epoch": 0.7287866772402855, "grad_norm": 0.893849698271372, "learning_rate": 2.8722972758591493e-06, "loss": 0.6159, "mean_token_accuracy": 0.7964839339256287, "num_tokens": 197471104.0, "step": 5514 }, { "epoch": 0.7289188474755485, "grad_norm": 0.7833431946012684, "learning_rate": 2.870612055713375e-06, "loss": 0.5732, "mean_token_accuracy": 0.8099896311759949, "num_tokens": 197536640.0, "step": 5515 }, { "epoch": 0.7290510177108115, "grad_norm": 0.8100351164628206, "learning_rate": 2.868927395309397e-06, "loss": 0.5818, "mean_token_accuracy": 0.8061897158622742, "num_tokens": 197602176.0, "step": 5516 }, { "epoch": 0.7291831879460745, "grad_norm": 0.8763019495118993, "learning_rate": 2.8672432950058427e-06, "loss": 0.6376, "mean_token_accuracy": 0.7885026335716248, "num_tokens": 197667712.0, "step": 5517 }, { "epoch": 0.7293153581813375, "grad_norm": 0.8897918683359164, "learning_rate": 2.8655597551612225e-06, "loss": 0.6199, "mean_token_accuracy": 0.7942711710929871, "num_tokens": 197733248.0, "step": 5518 }, { "epoch": 0.7294475284166005, "grad_norm": 0.8284619215148081, "learning_rate": 2.8638767761339276e-06, "loss": 0.5987, "mean_token_accuracy": 0.8008332252502441, "num_tokens": 197798784.0, "step": 5519 }, { "epoch": 0.7295796986518636, "grad_norm": 0.850791133357161, "learning_rate": 2.862194358282229e-06, "loss": 0.5873, "mean_token_accuracy": 0.8048620223999023, "num_tokens": 197864320.0, "step": 5520 }, { "epoch": 0.7297118688871266, "grad_norm": 0.7450316362592889, "learning_rate": 2.8605125019642787e-06, "loss": 0.5099, "mean_token_accuracy": 0.8303167819976807, "num_tokens": 197929856.0, "step": 5521 }, { "epoch": 0.7298440391223896, "grad_norm": 0.8342651529525521, "learning_rate": 2.8588312075381076e-06, "loss": 0.5568, "mean_token_accuracy": 0.8146135807037354, "num_tokens": 197995392.0, "step": 5522 }, { "epoch": 0.7299762093576526, "grad_norm": 0.7890224852105553, "learning_rate": 2.8571504753616284e-06, "loss": 0.5392, "mean_token_accuracy": 0.8181540966033936, "num_tokens": 198060928.0, "step": 5523 }, { "epoch": 0.7301083795929156, "grad_norm": 0.7581997357962819, "learning_rate": 2.8554703057926346e-06, "loss": 0.5248, "mean_token_accuracy": 0.8244566917419434, "num_tokens": 198126464.0, "step": 5524 }, { "epoch": 0.7302405498281787, "grad_norm": 0.8719553461490172, "learning_rate": 2.8537906991887998e-06, "loss": 0.6294, "mean_token_accuracy": 0.7898455858230591, "num_tokens": 198192000.0, "step": 5525 }, { "epoch": 0.7303727200634417, "grad_norm": 0.8554468641562586, "learning_rate": 2.8521116559076735e-06, "loss": 0.5933, "mean_token_accuracy": 0.8046941757202148, "num_tokens": 198257536.0, "step": 5526 }, { "epoch": 0.7305048902987047, "grad_norm": 0.7343959401603378, "learning_rate": 2.850433176306695e-06, "loss": 0.524, "mean_token_accuracy": 0.827020525932312, "num_tokens": 198323072.0, "step": 5527 }, { "epoch": 0.7306370605339677, "grad_norm": 0.7374142556082205, "learning_rate": 2.8487552607431724e-06, "loss": 0.494, "mean_token_accuracy": 0.8346660733222961, "num_tokens": 198388608.0, "step": 5528 }, { "epoch": 0.7307692307692307, "grad_norm": 0.8154331408493187, "learning_rate": 2.8470779095743028e-06, "loss": 0.6037, "mean_token_accuracy": 0.8006806373596191, "num_tokens": 198454144.0, "step": 5529 }, { "epoch": 0.7309014010044937, "grad_norm": 0.7723348094545817, "learning_rate": 2.8454011231571537e-06, "loss": 0.5286, "mean_token_accuracy": 0.8251434564590454, "num_tokens": 198519680.0, "step": 5530 }, { "epoch": 0.7310335712397568, "grad_norm": 0.8053788080844512, "learning_rate": 2.843724901848685e-06, "loss": 0.6503, "mean_token_accuracy": 0.7867018580436707, "num_tokens": 198585216.0, "step": 5531 }, { "epoch": 0.7311657414750198, "grad_norm": 0.9053365203391704, "learning_rate": 2.8420492460057236e-06, "loss": 0.6481, "mean_token_accuracy": 0.7867781519889832, "num_tokens": 198650752.0, "step": 5532 }, { "epoch": 0.7312979117102828, "grad_norm": 0.8336053599361084, "learning_rate": 2.840374155984984e-06, "loss": 0.5503, "mean_token_accuracy": 0.818535566329956, "num_tokens": 198716288.0, "step": 5533 }, { "epoch": 0.7314300819455458, "grad_norm": 0.7921492186007887, "learning_rate": 2.8386996321430575e-06, "loss": 0.583, "mean_token_accuracy": 0.8057929277420044, "num_tokens": 198781824.0, "step": 5534 }, { "epoch": 0.7315622521808088, "grad_norm": 0.838804902028901, "learning_rate": 2.837025674836418e-06, "loss": 0.607, "mean_token_accuracy": 0.7975369095802307, "num_tokens": 198847360.0, "step": 5535 }, { "epoch": 0.7316944224160719, "grad_norm": 0.7725637697546623, "learning_rate": 2.83535228442141e-06, "loss": 0.5831, "mean_token_accuracy": 0.8053198456764221, "num_tokens": 198912896.0, "step": 5536 }, { "epoch": 0.7318265926513349, "grad_norm": 0.8117866819983476, "learning_rate": 2.833679461254268e-06, "loss": 0.5818, "mean_token_accuracy": 0.8094097375869751, "num_tokens": 198978432.0, "step": 5537 }, { "epoch": 0.7319587628865979, "grad_norm": 0.8796973759914218, "learning_rate": 2.832007205691099e-06, "loss": 0.6576, "mean_token_accuracy": 0.7811622619628906, "num_tokens": 199043968.0, "step": 5538 }, { "epoch": 0.7320909331218609, "grad_norm": 0.7555023841699766, "learning_rate": 2.8303355180878933e-06, "loss": 0.5497, "mean_token_accuracy": 0.817482590675354, "num_tokens": 199109504.0, "step": 5539 }, { "epoch": 0.7322231033571239, "grad_norm": 0.8267318631552975, "learning_rate": 2.828664398800518e-06, "loss": 0.5512, "mean_token_accuracy": 0.816368579864502, "num_tokens": 199175040.0, "step": 5540 }, { "epoch": 0.732355273592387, "grad_norm": 0.8939372297490488, "learning_rate": 2.8269938481847183e-06, "loss": 0.5773, "mean_token_accuracy": 0.808356761932373, "num_tokens": 199240576.0, "step": 5541 }, { "epoch": 0.73248744382765, "grad_norm": 0.8626342327217169, "learning_rate": 2.825323866596118e-06, "loss": 0.5774, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 199306112.0, "step": 5542 }, { "epoch": 0.732619614062913, "grad_norm": 0.7827543167071318, "learning_rate": 2.8236544543902246e-06, "loss": 0.5285, "mean_token_accuracy": 0.8221218585968018, "num_tokens": 199371648.0, "step": 5543 }, { "epoch": 0.732751784298176, "grad_norm": 0.8198728365541706, "learning_rate": 2.821985611922421e-06, "loss": 0.6328, "mean_token_accuracy": 0.7883042097091675, "num_tokens": 199437184.0, "step": 5544 }, { "epoch": 0.732883954533439, "grad_norm": 0.7663073922549255, "learning_rate": 2.8203173395479666e-06, "loss": 0.5952, "mean_token_accuracy": 0.8018251657485962, "num_tokens": 199502720.0, "step": 5545 }, { "epoch": 0.733016124768702, "grad_norm": 0.8610522667369183, "learning_rate": 2.818649637622003e-06, "loss": 0.5819, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 199568256.0, "step": 5546 }, { "epoch": 0.7331482950039651, "grad_norm": 0.8293107429823853, "learning_rate": 2.816982506499548e-06, "loss": 0.6051, "mean_token_accuracy": 0.7984830737113953, "num_tokens": 199633792.0, "step": 5547 }, { "epoch": 0.7332804652392281, "grad_norm": 0.8795073629540809, "learning_rate": 2.8153159465355023e-06, "loss": 0.5852, "mean_token_accuracy": 0.8099285960197449, "num_tokens": 199699328.0, "step": 5548 }, { "epoch": 0.7334126354744911, "grad_norm": 0.8506222788218464, "learning_rate": 2.8136499580846366e-06, "loss": 0.5913, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 199764864.0, "step": 5549 }, { "epoch": 0.7335448057097541, "grad_norm": 0.9006512892768893, "learning_rate": 2.81198454150161e-06, "loss": 0.6265, "mean_token_accuracy": 0.7917836904525757, "num_tokens": 199830400.0, "step": 5550 }, { "epoch": 0.7336769759450171, "grad_norm": 0.8183939932575525, "learning_rate": 2.810319697140952e-06, "loss": 0.5619, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 199895936.0, "step": 5551 }, { "epoch": 0.7338091461802801, "grad_norm": 0.8515651485106521, "learning_rate": 2.808655425357073e-06, "loss": 0.5833, "mean_token_accuracy": 0.8046941757202148, "num_tokens": 199961472.0, "step": 5552 }, { "epoch": 0.7339413164155432, "grad_norm": 0.8110384876076194, "learning_rate": 2.8069917265042635e-06, "loss": 0.5206, "mean_token_accuracy": 0.8293554186820984, "num_tokens": 200027008.0, "step": 5553 }, { "epoch": 0.7340734866508062, "grad_norm": 0.8171472044612944, "learning_rate": 2.805328600936692e-06, "loss": 0.5709, "mean_token_accuracy": 0.8088908791542053, "num_tokens": 200092544.0, "step": 5554 }, { "epoch": 0.7342056568860692, "grad_norm": 0.8607833720505265, "learning_rate": 2.803666049008398e-06, "loss": 0.584, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 200158080.0, "step": 5555 }, { "epoch": 0.7343378271213322, "grad_norm": 0.9108822615983245, "learning_rate": 2.8020040710733065e-06, "loss": 0.5969, "mean_token_accuracy": 0.802466094493866, "num_tokens": 200223616.0, "step": 5556 }, { "epoch": 0.7344699973565953, "grad_norm": 0.9453262809722449, "learning_rate": 2.8003426674852186e-06, "loss": 0.6377, "mean_token_accuracy": 0.7868391871452332, "num_tokens": 200289152.0, "step": 5557 }, { "epoch": 0.7346021675918584, "grad_norm": 0.8076535962488922, "learning_rate": 2.7986818385978134e-06, "loss": 0.5515, "mean_token_accuracy": 0.8169637322425842, "num_tokens": 200354688.0, "step": 5558 }, { "epoch": 0.7347343378271214, "grad_norm": 0.9103356993329641, "learning_rate": 2.797021584764644e-06, "loss": 0.6425, "mean_token_accuracy": 0.7860304117202759, "num_tokens": 200420224.0, "step": 5559 }, { "epoch": 0.7348665080623844, "grad_norm": 0.8176308778231278, "learning_rate": 2.7953619063391444e-06, "loss": 0.5854, "mean_token_accuracy": 0.8019624948501587, "num_tokens": 200485760.0, "step": 5560 }, { "epoch": 0.7349986782976474, "grad_norm": 0.8375344036850441, "learning_rate": 2.793702803674627e-06, "loss": 0.5936, "mean_token_accuracy": 0.8028476238250732, "num_tokens": 200551296.0, "step": 5561 }, { "epoch": 0.7351308485329104, "grad_norm": 0.8166374868151322, "learning_rate": 2.792044277124279e-06, "loss": 0.5219, "mean_token_accuracy": 0.8276767134666443, "num_tokens": 200616832.0, "step": 5562 }, { "epoch": 0.7352630187681735, "grad_norm": 0.7866891530707762, "learning_rate": 2.790386327041168e-06, "loss": 0.5619, "mean_token_accuracy": 0.8134385347366333, "num_tokens": 200682368.0, "step": 5563 }, { "epoch": 0.7353951890034365, "grad_norm": 0.8063550204033757, "learning_rate": 2.7887289537782335e-06, "loss": 0.5772, "mean_token_accuracy": 0.8060523867607117, "num_tokens": 200747904.0, "step": 5564 }, { "epoch": 0.7355273592386995, "grad_norm": 0.8351221526629065, "learning_rate": 2.7870721576882974e-06, "loss": 0.5855, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 200813440.0, "step": 5565 }, { "epoch": 0.7356595294739625, "grad_norm": 0.8511290556205755, "learning_rate": 2.7854159391240577e-06, "loss": 0.59, "mean_token_accuracy": 0.8017336130142212, "num_tokens": 200878976.0, "step": 5566 }, { "epoch": 0.7357916997092255, "grad_norm": 0.8668952044806567, "learning_rate": 2.7837602984380885e-06, "loss": 0.5964, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 200944512.0, "step": 5567 }, { "epoch": 0.7359238699444886, "grad_norm": 0.8097394819262775, "learning_rate": 2.7821052359828386e-06, "loss": 0.591, "mean_token_accuracy": 0.802435576915741, "num_tokens": 201010048.0, "step": 5568 }, { "epoch": 0.7360560401797516, "grad_norm": 0.8655691708021561, "learning_rate": 2.7804507521106415e-06, "loss": 0.5569, "mean_token_accuracy": 0.8146898746490479, "num_tokens": 201075584.0, "step": 5569 }, { "epoch": 0.7361882104150146, "grad_norm": 0.8013827259536346, "learning_rate": 2.778796847173697e-06, "loss": 0.5708, "mean_token_accuracy": 0.8103101253509521, "num_tokens": 201141120.0, "step": 5570 }, { "epoch": 0.7363203806502776, "grad_norm": 0.8514779355649329, "learning_rate": 2.77714352152409e-06, "loss": 0.5895, "mean_token_accuracy": 0.8026034832000732, "num_tokens": 201206656.0, "step": 5571 }, { "epoch": 0.7364525508855406, "grad_norm": 0.8057407623471993, "learning_rate": 2.775490775513777e-06, "loss": 0.5935, "mean_token_accuracy": 0.8048315048217773, "num_tokens": 201272192.0, "step": 5572 }, { "epoch": 0.7365847211208036, "grad_norm": 0.7911276180235943, "learning_rate": 2.7738386094945967e-06, "loss": 0.5642, "mean_token_accuracy": 0.8140489459037781, "num_tokens": 201337728.0, "step": 5573 }, { "epoch": 0.7367168913560667, "grad_norm": 0.8044292379666514, "learning_rate": 2.772187023818257e-06, "loss": 0.552, "mean_token_accuracy": 0.8179556727409363, "num_tokens": 201403264.0, "step": 5574 }, { "epoch": 0.7368490615913297, "grad_norm": 0.8419346432415009, "learning_rate": 2.7705360188363463e-06, "loss": 0.5982, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 201468800.0, "step": 5575 }, { "epoch": 0.7369812318265927, "grad_norm": 0.7520717592055883, "learning_rate": 2.768885594900331e-06, "loss": 0.5227, "mean_token_accuracy": 0.8244109153747559, "num_tokens": 201534336.0, "step": 5576 }, { "epoch": 0.7371134020618557, "grad_norm": 0.8156039438733221, "learning_rate": 2.7672357523615522e-06, "loss": 0.5675, "mean_token_accuracy": 0.8120192885398865, "num_tokens": 201599872.0, "step": 5577 }, { "epoch": 0.7372455722971187, "grad_norm": 0.7954466254037429, "learning_rate": 2.765586491571224e-06, "loss": 0.5135, "mean_token_accuracy": 0.8311103582382202, "num_tokens": 201665408.0, "step": 5578 }, { "epoch": 0.7373777425323818, "grad_norm": 0.8243675877822474, "learning_rate": 2.763937812880442e-06, "loss": 0.5703, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 201730944.0, "step": 5579 }, { "epoch": 0.7375099127676448, "grad_norm": 0.8340174459865708, "learning_rate": 2.7622897166401754e-06, "loss": 0.6001, "mean_token_accuracy": 0.8041447997093201, "num_tokens": 201796480.0, "step": 5580 }, { "epoch": 0.7376420830029078, "grad_norm": 0.8479450367929272, "learning_rate": 2.760642203201266e-06, "loss": 0.5812, "mean_token_accuracy": 0.8074105978012085, "num_tokens": 201862016.0, "step": 5581 }, { "epoch": 0.7377742532381708, "grad_norm": 0.913031289263559, "learning_rate": 2.758995272914441e-06, "loss": 0.632, "mean_token_accuracy": 0.7913563847541809, "num_tokens": 201927552.0, "step": 5582 }, { "epoch": 0.7379064234734338, "grad_norm": 0.7960914543219411, "learning_rate": 2.757348926130293e-06, "loss": 0.6096, "mean_token_accuracy": 0.7958582639694214, "num_tokens": 201993088.0, "step": 5583 }, { "epoch": 0.7380385937086968, "grad_norm": 0.7505862437594079, "learning_rate": 2.7557031631992957e-06, "loss": 0.5284, "mean_token_accuracy": 0.8243041038513184, "num_tokens": 202058624.0, "step": 5584 }, { "epoch": 0.7381707639439599, "grad_norm": 0.8548292162608784, "learning_rate": 2.7540579844717987e-06, "loss": 0.5717, "mean_token_accuracy": 0.8099743723869324, "num_tokens": 202124160.0, "step": 5585 }, { "epoch": 0.7383029341792229, "grad_norm": 0.8709674860975708, "learning_rate": 2.7524133902980266e-06, "loss": 0.6096, "mean_token_accuracy": 0.7966212630271912, "num_tokens": 202189696.0, "step": 5586 }, { "epoch": 0.7384351044144859, "grad_norm": 0.7233841735023282, "learning_rate": 2.7507693810280755e-06, "loss": 0.5245, "mean_token_accuracy": 0.8230069875717163, "num_tokens": 202255232.0, "step": 5587 }, { "epoch": 0.7385672746497489, "grad_norm": 0.8775700541150372, "learning_rate": 2.7491259570119268e-06, "loss": 0.6514, "mean_token_accuracy": 0.782688319683075, "num_tokens": 202320768.0, "step": 5588 }, { "epoch": 0.7386994448850119, "grad_norm": 0.78536536795424, "learning_rate": 2.7474831185994256e-06, "loss": 0.5513, "mean_token_accuracy": 0.8189628720283508, "num_tokens": 202386304.0, "step": 5589 }, { "epoch": 0.738831615120275, "grad_norm": 0.8712445838129974, "learning_rate": 2.745840866140302e-06, "loss": 0.6341, "mean_token_accuracy": 0.7906848788261414, "num_tokens": 202451840.0, "step": 5590 }, { "epoch": 0.738963785355538, "grad_norm": 0.8401064914197429, "learning_rate": 2.744199199984152e-06, "loss": 0.6018, "mean_token_accuracy": 0.7998107671737671, "num_tokens": 202517376.0, "step": 5591 }, { "epoch": 0.739095955590801, "grad_norm": 0.8373663147040347, "learning_rate": 2.7425581204804586e-06, "loss": 0.582, "mean_token_accuracy": 0.806861162185669, "num_tokens": 202582912.0, "step": 5592 }, { "epoch": 0.739228125826064, "grad_norm": 0.8517988504190944, "learning_rate": 2.7409176279785682e-06, "loss": 0.5872, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 202648448.0, "step": 5593 }, { "epoch": 0.739360296061327, "grad_norm": 0.8280066528043499, "learning_rate": 2.7392777228277096e-06, "loss": 0.5538, "mean_token_accuracy": 0.8203973770141602, "num_tokens": 202713984.0, "step": 5594 }, { "epoch": 0.73949246629659, "grad_norm": 0.8202481449846518, "learning_rate": 2.737638405376984e-06, "loss": 0.6207, "mean_token_accuracy": 0.7964839339256287, "num_tokens": 202779520.0, "step": 5595 }, { "epoch": 0.7396246365318531, "grad_norm": 0.8354420307637789, "learning_rate": 2.7359996759753694e-06, "loss": 0.5658, "mean_token_accuracy": 0.8118667006492615, "num_tokens": 202845056.0, "step": 5596 }, { "epoch": 0.7397568067671161, "grad_norm": 0.7765667671954831, "learning_rate": 2.7343615349717145e-06, "loss": 0.552, "mean_token_accuracy": 0.8160328269004822, "num_tokens": 202910592.0, "step": 5597 }, { "epoch": 0.7398889770023791, "grad_norm": 0.880770739332514, "learning_rate": 2.7327239827147455e-06, "loss": 0.5741, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 202976128.0, "step": 5598 }, { "epoch": 0.7400211472376421, "grad_norm": 0.7729326120445795, "learning_rate": 2.731087019553067e-06, "loss": 0.5374, "mean_token_accuracy": 0.821908175945282, "num_tokens": 203041664.0, "step": 5599 }, { "epoch": 0.7401533174729051, "grad_norm": 0.8878892851772251, "learning_rate": 2.7294506458351473e-06, "loss": 0.5856, "mean_token_accuracy": 0.802313506603241, "num_tokens": 203107200.0, "step": 5600 }, { "epoch": 0.7402854877081682, "grad_norm": 0.748413546005177, "learning_rate": 2.7278148619093436e-06, "loss": 0.5629, "mean_token_accuracy": 0.813026487827301, "num_tokens": 203172736.0, "step": 5601 }, { "epoch": 0.7404176579434312, "grad_norm": 0.8210427762842539, "learning_rate": 2.7261796681238745e-06, "loss": 0.5427, "mean_token_accuracy": 0.8213435411453247, "num_tokens": 203238272.0, "step": 5602 }, { "epoch": 0.7405498281786942, "grad_norm": 0.7890170489008074, "learning_rate": 2.7245450648268413e-06, "loss": 0.54, "mean_token_accuracy": 0.8229764103889465, "num_tokens": 203303808.0, "step": 5603 }, { "epoch": 0.7406819984139572, "grad_norm": 0.8154382659230385, "learning_rate": 2.7229110523662165e-06, "loss": 0.5581, "mean_token_accuracy": 0.8134690523147583, "num_tokens": 203369344.0, "step": 5604 }, { "epoch": 0.7408141686492202, "grad_norm": 0.7859893514412671, "learning_rate": 2.7212776310898486e-06, "loss": 0.5726, "mean_token_accuracy": 0.8095928430557251, "num_tokens": 203434880.0, "step": 5605 }, { "epoch": 0.7409463388844832, "grad_norm": 0.7899864361337311, "learning_rate": 2.7196448013454564e-06, "loss": 0.5875, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 203500416.0, "step": 5606 }, { "epoch": 0.7410785091197463, "grad_norm": 0.7196996532474934, "learning_rate": 2.7180125634806366e-06, "loss": 0.5046, "mean_token_accuracy": 0.8304694294929504, "num_tokens": 203565952.0, "step": 5607 }, { "epoch": 0.7412106793550093, "grad_norm": 0.7944844580249376, "learning_rate": 2.7163809178428576e-06, "loss": 0.604, "mean_token_accuracy": 0.7973538041114807, "num_tokens": 203631488.0, "step": 5608 }, { "epoch": 0.7413428495902723, "grad_norm": 0.7662031378125868, "learning_rate": 2.7147498647794644e-06, "loss": 0.5506, "mean_token_accuracy": 0.8181082606315613, "num_tokens": 203697024.0, "step": 5609 }, { "epoch": 0.7414750198255353, "grad_norm": 0.8381580946925701, "learning_rate": 2.7131194046376714e-06, "loss": 0.5736, "mean_token_accuracy": 0.8115309476852417, "num_tokens": 203762560.0, "step": 5610 }, { "epoch": 0.7416071900607983, "grad_norm": 0.8199458586997718, "learning_rate": 2.7114895377645733e-06, "loss": 0.5488, "mean_token_accuracy": 0.816643238067627, "num_tokens": 203828096.0, "step": 5611 }, { "epoch": 0.7417393602960614, "grad_norm": 0.8095488460861137, "learning_rate": 2.709860264507132e-06, "loss": 0.6507, "mean_token_accuracy": 0.7848705649375916, "num_tokens": 203893632.0, "step": 5612 }, { "epoch": 0.7418715305313244, "grad_norm": 0.9478031336905595, "learning_rate": 2.7082315852121867e-06, "loss": 0.5757, "mean_token_accuracy": 0.8085246086120605, "num_tokens": 203959168.0, "step": 5613 }, { "epoch": 0.7420037007665874, "grad_norm": 0.867256075146635, "learning_rate": 2.706603500226448e-06, "loss": 0.5872, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 204024704.0, "step": 5614 }, { "epoch": 0.7421358710018504, "grad_norm": 0.7821952863915067, "learning_rate": 2.704976009896504e-06, "loss": 0.5966, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 204090240.0, "step": 5615 }, { "epoch": 0.7422680412371134, "grad_norm": 0.8071755160973567, "learning_rate": 2.7033491145688094e-06, "loss": 0.587, "mean_token_accuracy": 0.806433916091919, "num_tokens": 204155776.0, "step": 5616 }, { "epoch": 0.7424002114723764, "grad_norm": 0.8334243682882838, "learning_rate": 2.7017228145897e-06, "loss": 0.6097, "mean_token_accuracy": 0.798879861831665, "num_tokens": 204221312.0, "step": 5617 }, { "epoch": 0.7425323817076395, "grad_norm": 0.8213386571807365, "learning_rate": 2.70009711030538e-06, "loss": 0.568, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 204286848.0, "step": 5618 }, { "epoch": 0.7426645519429025, "grad_norm": 0.7484224685408981, "learning_rate": 2.6984720020619265e-06, "loss": 0.5159, "mean_token_accuracy": 0.8268373608589172, "num_tokens": 204352384.0, "step": 5619 }, { "epoch": 0.7427967221781655, "grad_norm": 0.8282468070464063, "learning_rate": 2.6968474902052914e-06, "loss": 0.5673, "mean_token_accuracy": 0.8093487024307251, "num_tokens": 204417920.0, "step": 5620 }, { "epoch": 0.7429288924134285, "grad_norm": 0.8102366250604753, "learning_rate": 2.695223575081301e-06, "loss": 0.568, "mean_token_accuracy": 0.8099133372306824, "num_tokens": 204483456.0, "step": 5621 }, { "epoch": 0.7430610626486915, "grad_norm": 0.8618413521699193, "learning_rate": 2.6936002570356524e-06, "loss": 0.6034, "mean_token_accuracy": 0.8013673424720764, "num_tokens": 204548992.0, "step": 5622 }, { "epoch": 0.7431932328839546, "grad_norm": 0.8617402881769499, "learning_rate": 2.6919775364139133e-06, "loss": 0.5582, "mean_token_accuracy": 0.8167806267738342, "num_tokens": 204614528.0, "step": 5623 }, { "epoch": 0.7433254031192176, "grad_norm": 0.7999020334973447, "learning_rate": 2.690355413561533e-06, "loss": 0.5646, "mean_token_accuracy": 0.812995970249176, "num_tokens": 204680064.0, "step": 5624 }, { "epoch": 0.7434575733544806, "grad_norm": 0.8551366386764157, "learning_rate": 2.688733888823821e-06, "loss": 0.6157, "mean_token_accuracy": 0.7994597554206848, "num_tokens": 204745600.0, "step": 5625 }, { "epoch": 0.7435897435897436, "grad_norm": 1.6065269334250396, "learning_rate": 2.68711296254597e-06, "loss": 0.5971, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 204811136.0, "step": 5626 }, { "epoch": 0.7437219138250066, "grad_norm": 0.7748979308639116, "learning_rate": 2.6854926350730397e-06, "loss": 0.552, "mean_token_accuracy": 0.8152850866317749, "num_tokens": 204876672.0, "step": 5627 }, { "epoch": 0.7438540840602696, "grad_norm": 0.8164127714540916, "learning_rate": 2.683872906749967e-06, "loss": 0.5676, "mean_token_accuracy": 0.8127212524414062, "num_tokens": 204942208.0, "step": 5628 }, { "epoch": 0.7439862542955327, "grad_norm": 0.8396557942511191, "learning_rate": 2.682253777921553e-06, "loss": 0.5814, "mean_token_accuracy": 0.806739091873169, "num_tokens": 205007744.0, "step": 5629 }, { "epoch": 0.7441184245307957, "grad_norm": 0.8697819403639571, "learning_rate": 2.680635248932482e-06, "loss": 0.5884, "mean_token_accuracy": 0.803427517414093, "num_tokens": 205073280.0, "step": 5630 }, { "epoch": 0.7442505947660587, "grad_norm": 0.8843418308952146, "learning_rate": 2.679017320127301e-06, "loss": 0.5532, "mean_token_accuracy": 0.8141710162162781, "num_tokens": 205138816.0, "step": 5631 }, { "epoch": 0.7443827650013217, "grad_norm": 0.8581822992835844, "learning_rate": 2.6773999918504347e-06, "loss": 0.5741, "mean_token_accuracy": 0.807822585105896, "num_tokens": 205204352.0, "step": 5632 }, { "epoch": 0.7445149352365847, "grad_norm": 0.8560662164117199, "learning_rate": 2.6757832644461786e-06, "loss": 0.5572, "mean_token_accuracy": 0.8127976059913635, "num_tokens": 205269888.0, "step": 5633 }, { "epoch": 0.7446471054718478, "grad_norm": 0.8095617114831232, "learning_rate": 2.674167138258701e-06, "loss": 0.5367, "mean_token_accuracy": 0.8215113878250122, "num_tokens": 205335424.0, "step": 5634 }, { "epoch": 0.7447792757071108, "grad_norm": 0.889905916618709, "learning_rate": 2.6725516136320396e-06, "loss": 0.6051, "mean_token_accuracy": 0.79872727394104, "num_tokens": 205400960.0, "step": 5635 }, { "epoch": 0.7449114459423738, "grad_norm": 0.7646975541984291, "learning_rate": 2.6709366909101064e-06, "loss": 0.5525, "mean_token_accuracy": 0.8135606050491333, "num_tokens": 205466496.0, "step": 5636 }, { "epoch": 0.7450436161776368, "grad_norm": 0.8162562276025601, "learning_rate": 2.669322370436687e-06, "loss": 0.5102, "mean_token_accuracy": 0.8288212418556213, "num_tokens": 205532032.0, "step": 5637 }, { "epoch": 0.7451757864128998, "grad_norm": 0.8263960587379398, "learning_rate": 2.667708652555433e-06, "loss": 0.5532, "mean_token_accuracy": 0.8173147439956665, "num_tokens": 205597568.0, "step": 5638 }, { "epoch": 0.7453079566481629, "grad_norm": 0.7970208020683861, "learning_rate": 2.666095537609873e-06, "loss": 0.6002, "mean_token_accuracy": 0.7991088032722473, "num_tokens": 205663104.0, "step": 5639 }, { "epoch": 0.7454401268834259, "grad_norm": 0.8502862019329394, "learning_rate": 2.664483025943405e-06, "loss": 0.6016, "mean_token_accuracy": 0.8004669547080994, "num_tokens": 205728640.0, "step": 5640 }, { "epoch": 0.7455722971186889, "grad_norm": 0.8083511948875461, "learning_rate": 2.6628711178993018e-06, "loss": 0.5536, "mean_token_accuracy": 0.8153613805770874, "num_tokens": 205794176.0, "step": 5641 }, { "epoch": 0.7457044673539519, "grad_norm": 0.8200454240617056, "learning_rate": 2.661259813820699e-06, "loss": 0.5626, "mean_token_accuracy": 0.8139879107475281, "num_tokens": 205859712.0, "step": 5642 }, { "epoch": 0.7458366375892149, "grad_norm": 0.8487593802674895, "learning_rate": 2.6596491140506157e-06, "loss": 0.5831, "mean_token_accuracy": 0.8040227293968201, "num_tokens": 205925248.0, "step": 5643 }, { "epoch": 0.7459688078244779, "grad_norm": 0.7871441322794742, "learning_rate": 2.658039018931933e-06, "loss": 0.5759, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 205990784.0, "step": 5644 }, { "epoch": 0.746100978059741, "grad_norm": 0.8466484839401686, "learning_rate": 2.656429528807407e-06, "loss": 0.6026, "mean_token_accuracy": 0.8017793893814087, "num_tokens": 206056320.0, "step": 5645 }, { "epoch": 0.746233148295004, "grad_norm": 0.875459798427381, "learning_rate": 2.6548206440196655e-06, "loss": 0.5967, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 206121856.0, "step": 5646 }, { "epoch": 0.746365318530267, "grad_norm": 0.8720265442506682, "learning_rate": 2.653212364911206e-06, "loss": 0.6081, "mean_token_accuracy": 0.8001465201377869, "num_tokens": 206187392.0, "step": 5647 }, { "epoch": 0.74649748876553, "grad_norm": 0.7695708197495456, "learning_rate": 2.6516046918243955e-06, "loss": 0.5251, "mean_token_accuracy": 0.8247466683387756, "num_tokens": 206252928.0, "step": 5648 }, { "epoch": 0.746629659000793, "grad_norm": 0.781111481226148, "learning_rate": 2.6499976251014786e-06, "loss": 0.5727, "mean_token_accuracy": 0.8079447150230408, "num_tokens": 206318464.0, "step": 5649 }, { "epoch": 0.746761829236056, "grad_norm": 0.8213522692977068, "learning_rate": 2.648391165084562e-06, "loss": 0.5728, "mean_token_accuracy": 0.808814525604248, "num_tokens": 206384000.0, "step": 5650 }, { "epoch": 0.7468939994713191, "grad_norm": 0.813628976679746, "learning_rate": 2.6467853121156305e-06, "loss": 0.5437, "mean_token_accuracy": 0.8180930018424988, "num_tokens": 206449536.0, "step": 5651 }, { "epoch": 0.7470261697065821, "grad_norm": 0.8512711467782358, "learning_rate": 2.6451800665365323e-06, "loss": 0.5822, "mean_token_accuracy": 0.8097149133682251, "num_tokens": 206515072.0, "step": 5652 }, { "epoch": 0.7471583399418451, "grad_norm": 0.820731867719377, "learning_rate": 2.643575428688996e-06, "loss": 0.5796, "mean_token_accuracy": 0.8086466789245605, "num_tokens": 206580608.0, "step": 5653 }, { "epoch": 0.7472905101771081, "grad_norm": 0.84220324945443, "learning_rate": 2.6419713989146123e-06, "loss": 0.5782, "mean_token_accuracy": 0.8052588105201721, "num_tokens": 206646144.0, "step": 5654 }, { "epoch": 0.7474226804123711, "grad_norm": 0.8136348697263598, "learning_rate": 2.6403679775548463e-06, "loss": 0.6034, "mean_token_accuracy": 0.7994292378425598, "num_tokens": 206711680.0, "step": 5655 }, { "epoch": 0.7475548506476342, "grad_norm": 0.810023270981081, "learning_rate": 2.638765164951035e-06, "loss": 0.531, "mean_token_accuracy": 0.8236479163169861, "num_tokens": 206777216.0, "step": 5656 }, { "epoch": 0.7476870208828972, "grad_norm": 0.9356668547570224, "learning_rate": 2.6371629614443795e-06, "loss": 0.6337, "mean_token_accuracy": 0.789067268371582, "num_tokens": 206842752.0, "step": 5657 }, { "epoch": 0.7478191911181602, "grad_norm": 0.8394690143106354, "learning_rate": 2.6355613673759576e-06, "loss": 0.5773, "mean_token_accuracy": 0.8082804083824158, "num_tokens": 206908288.0, "step": 5658 }, { "epoch": 0.7479513613534232, "grad_norm": 0.9158483404776228, "learning_rate": 2.633960383086716e-06, "loss": 0.6469, "mean_token_accuracy": 0.7856031060218811, "num_tokens": 206973824.0, "step": 5659 }, { "epoch": 0.7480835315886862, "grad_norm": 0.7948510793206659, "learning_rate": 2.6323600089174715e-06, "loss": 0.5468, "mean_token_accuracy": 0.8163533210754395, "num_tokens": 207039360.0, "step": 5660 }, { "epoch": 0.7482157018239493, "grad_norm": 0.8620354330125672, "learning_rate": 2.6307602452089067e-06, "loss": 0.6159, "mean_token_accuracy": 0.7958887815475464, "num_tokens": 207104896.0, "step": 5661 }, { "epoch": 0.7483478720592123, "grad_norm": 0.758021124080798, "learning_rate": 2.6291610923015828e-06, "loss": 0.5424, "mean_token_accuracy": 0.8188865780830383, "num_tokens": 207170432.0, "step": 5662 }, { "epoch": 0.7484800422944753, "grad_norm": 0.8513438723958443, "learning_rate": 2.6275625505359215e-06, "loss": 0.5422, "mean_token_accuracy": 0.8194664716720581, "num_tokens": 207235968.0, "step": 5663 }, { "epoch": 0.7486122125297383, "grad_norm": 0.8772578725845022, "learning_rate": 2.625964620252224e-06, "loss": 0.5931, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 207301504.0, "step": 5664 }, { "epoch": 0.7487443827650013, "grad_norm": 0.818577199130588, "learning_rate": 2.624367301790649e-06, "loss": 0.5883, "mean_token_accuracy": 0.8053351044654846, "num_tokens": 207367040.0, "step": 5665 }, { "epoch": 0.7488765530002643, "grad_norm": 0.8010681294859685, "learning_rate": 2.6227705954912396e-06, "loss": 0.5662, "mean_token_accuracy": 0.8123092651367188, "num_tokens": 207432576.0, "step": 5666 }, { "epoch": 0.7490087232355274, "grad_norm": 0.8687173024156646, "learning_rate": 2.621174501693897e-06, "loss": 0.5912, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 207498112.0, "step": 5667 }, { "epoch": 0.7491408934707904, "grad_norm": 0.8665056579284747, "learning_rate": 2.6195790207383976e-06, "loss": 0.6138, "mean_token_accuracy": 0.8008027076721191, "num_tokens": 207563648.0, "step": 5668 }, { "epoch": 0.7492730637060534, "grad_norm": 0.8397567498555935, "learning_rate": 2.617984152964386e-06, "loss": 0.5787, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 207629184.0, "step": 5669 }, { "epoch": 0.7494052339413164, "grad_norm": 0.8932218486256771, "learning_rate": 2.616389898711377e-06, "loss": 0.5529, "mean_token_accuracy": 0.8147357106208801, "num_tokens": 207694720.0, "step": 5670 }, { "epoch": 0.7495374041765794, "grad_norm": 0.8897692598091832, "learning_rate": 2.6147962583187505e-06, "loss": 0.5934, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 207760256.0, "step": 5671 }, { "epoch": 0.7496695744118425, "grad_norm": 0.7999703623191684, "learning_rate": 2.613203232125765e-06, "loss": 0.6025, "mean_token_accuracy": 0.798818826675415, "num_tokens": 207825792.0, "step": 5672 }, { "epoch": 0.7498017446471055, "grad_norm": 0.813636914298697, "learning_rate": 2.6116108204715373e-06, "loss": 0.6088, "mean_token_accuracy": 0.7985746264457703, "num_tokens": 207891328.0, "step": 5673 }, { "epoch": 0.7499339148823685, "grad_norm": 0.8197475462203532, "learning_rate": 2.6100190236950618e-06, "loss": 0.6149, "mean_token_accuracy": 0.7959192991256714, "num_tokens": 207956864.0, "step": 5674 }, { "epoch": 0.7500660851176315, "grad_norm": 0.8252985360205983, "learning_rate": 2.608427842135198e-06, "loss": 0.5418, "mean_token_accuracy": 0.8202905654907227, "num_tokens": 208022400.0, "step": 5675 }, { "epoch": 0.7501982553528945, "grad_norm": 0.9201768393607581, "learning_rate": 2.6068372761306738e-06, "loss": 0.61, "mean_token_accuracy": 0.7972317337989807, "num_tokens": 208087936.0, "step": 5676 }, { "epoch": 0.7503304255881575, "grad_norm": 0.7991791248085414, "learning_rate": 2.6052473260200895e-06, "loss": 0.5544, "mean_token_accuracy": 0.8149340748786926, "num_tokens": 208153472.0, "step": 5677 }, { "epoch": 0.7504625958234206, "grad_norm": 0.8322398042966809, "learning_rate": 2.6036579921419096e-06, "loss": 0.5732, "mean_token_accuracy": 0.8120192885398865, "num_tokens": 208219008.0, "step": 5678 }, { "epoch": 0.7505947660586836, "grad_norm": 0.8594251066613122, "learning_rate": 2.602069274834475e-06, "loss": 0.5978, "mean_token_accuracy": 0.8019777536392212, "num_tokens": 208284544.0, "step": 5679 }, { "epoch": 0.7507269362939466, "grad_norm": 0.8191627949402255, "learning_rate": 2.6004811744359852e-06, "loss": 0.5749, "mean_token_accuracy": 0.8097606897354126, "num_tokens": 208350080.0, "step": 5680 }, { "epoch": 0.7508591065292096, "grad_norm": 0.8581063460193238, "learning_rate": 2.5988936912845157e-06, "loss": 0.6119, "mean_token_accuracy": 0.7961024045944214, "num_tokens": 208415616.0, "step": 5681 }, { "epoch": 0.7509912767644726, "grad_norm": 0.7656293475369151, "learning_rate": 2.597306825718009e-06, "loss": 0.502, "mean_token_accuracy": 0.8320870399475098, "num_tokens": 208481152.0, "step": 5682 }, { "epoch": 0.7511234469997357, "grad_norm": 0.9430208561363086, "learning_rate": 2.5957205780742766e-06, "loss": 0.6281, "mean_token_accuracy": 0.793645441532135, "num_tokens": 208546688.0, "step": 5683 }, { "epoch": 0.7512556172349987, "grad_norm": 0.8078753010551386, "learning_rate": 2.594134948690993e-06, "loss": 0.5651, "mean_token_accuracy": 0.8122634887695312, "num_tokens": 208612224.0, "step": 5684 }, { "epoch": 0.7513877874702617, "grad_norm": 0.7748434175640585, "learning_rate": 2.592549937905712e-06, "loss": 0.55, "mean_token_accuracy": 0.8162007331848145, "num_tokens": 208677760.0, "step": 5685 }, { "epoch": 0.7515199577055247, "grad_norm": 0.920740766128451, "learning_rate": 2.5909655460558436e-06, "loss": 0.5976, "mean_token_accuracy": 0.8010926842689514, "num_tokens": 208743296.0, "step": 5686 }, { "epoch": 0.7516521279407877, "grad_norm": 0.8448800791451748, "learning_rate": 2.5893817734786752e-06, "loss": 0.6261, "mean_token_accuracy": 0.7928519248962402, "num_tokens": 208808832.0, "step": 5687 }, { "epoch": 0.7517842981760507, "grad_norm": 0.8671315240917993, "learning_rate": 2.587798620511358e-06, "loss": 0.5805, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 208874368.0, "step": 5688 }, { "epoch": 0.7519164684113138, "grad_norm": 0.8362935761214403, "learning_rate": 2.586216087490913e-06, "loss": 0.5935, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 208939904.0, "step": 5689 }, { "epoch": 0.7520486386465768, "grad_norm": 0.7820741268957315, "learning_rate": 2.5846341747542247e-06, "loss": 0.5194, "mean_token_accuracy": 0.8274020552635193, "num_tokens": 209005440.0, "step": 5690 }, { "epoch": 0.7521808088818398, "grad_norm": 0.853227698031727, "learning_rate": 2.5830528826380554e-06, "loss": 0.5755, "mean_token_accuracy": 0.8091502785682678, "num_tokens": 209070976.0, "step": 5691 }, { "epoch": 0.7523129791171028, "grad_norm": 0.8155035940306645, "learning_rate": 2.581472211479024e-06, "loss": 0.5463, "mean_token_accuracy": 0.8189933896064758, "num_tokens": 209136512.0, "step": 5692 }, { "epoch": 0.7524451493523658, "grad_norm": 0.8448407375295478, "learning_rate": 2.579892161613624e-06, "loss": 0.5355, "mean_token_accuracy": 0.8214961290359497, "num_tokens": 209202048.0, "step": 5693 }, { "epoch": 0.7525773195876289, "grad_norm": 0.784234240548336, "learning_rate": 2.5783127333782174e-06, "loss": 0.5551, "mean_token_accuracy": 0.8143846988677979, "num_tokens": 209267584.0, "step": 5694 }, { "epoch": 0.7527094898228919, "grad_norm": 0.7777100748651822, "learning_rate": 2.576733927109028e-06, "loss": 0.5046, "mean_token_accuracy": 0.8336130976676941, "num_tokens": 209333120.0, "step": 5695 }, { "epoch": 0.7528416600581549, "grad_norm": 0.9311783033328249, "learning_rate": 2.575155743142152e-06, "loss": 0.6618, "mean_token_accuracy": 0.7809943556785583, "num_tokens": 209398656.0, "step": 5696 }, { "epoch": 0.7529738302934179, "grad_norm": 0.8195946450202402, "learning_rate": 2.573578181813552e-06, "loss": 0.5631, "mean_token_accuracy": 0.8123855590820312, "num_tokens": 209464192.0, "step": 5697 }, { "epoch": 0.7531060005286809, "grad_norm": 0.8317305610156661, "learning_rate": 2.5720012434590593e-06, "loss": 0.5851, "mean_token_accuracy": 0.8059760928153992, "num_tokens": 209529728.0, "step": 5698 }, { "epoch": 0.753238170763944, "grad_norm": 0.8005083762197013, "learning_rate": 2.570424928414368e-06, "loss": 0.5391, "mean_token_accuracy": 0.8194664716720581, "num_tokens": 209595264.0, "step": 5699 }, { "epoch": 0.753370340999207, "grad_norm": 0.820763871880238, "learning_rate": 2.5688492370150442e-06, "loss": 0.6054, "mean_token_accuracy": 0.7980862855911255, "num_tokens": 209660800.0, "step": 5700 }, { "epoch": 0.75350251123447, "grad_norm": 0.8350495080348659, "learning_rate": 2.5672741695965197e-06, "loss": 0.5965, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 209726336.0, "step": 5701 }, { "epoch": 0.753634681469733, "grad_norm": 0.8558419475420961, "learning_rate": 2.565699726494096e-06, "loss": 0.5718, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 209791872.0, "step": 5702 }, { "epoch": 0.753766851704996, "grad_norm": 0.8527687398313675, "learning_rate": 2.5641259080429324e-06, "loss": 0.6147, "mean_token_accuracy": 0.7985746264457703, "num_tokens": 209857408.0, "step": 5703 }, { "epoch": 0.753899021940259, "grad_norm": 0.8481906039362467, "learning_rate": 2.5625527145780704e-06, "loss": 0.6379, "mean_token_accuracy": 0.7877395749092102, "num_tokens": 209922944.0, "step": 5704 }, { "epoch": 0.7540311921755221, "grad_norm": 0.8570262093283293, "learning_rate": 2.560980146434404e-06, "loss": 0.5824, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 209988480.0, "step": 5705 }, { "epoch": 0.7541633624107851, "grad_norm": 0.7801668662344892, "learning_rate": 2.5594082039467027e-06, "loss": 0.568, "mean_token_accuracy": 0.8139116168022156, "num_tokens": 210054016.0, "step": 5706 }, { "epoch": 0.7542955326460481, "grad_norm": 0.8838193698949773, "learning_rate": 2.5578368874495996e-06, "loss": 0.5874, "mean_token_accuracy": 0.8062660098075867, "num_tokens": 210119552.0, "step": 5707 }, { "epoch": 0.7544277028813111, "grad_norm": 0.7735147198289988, "learning_rate": 2.556266197277596e-06, "loss": 0.5438, "mean_token_accuracy": 0.8179556727409363, "num_tokens": 210185088.0, "step": 5708 }, { "epoch": 0.7545598731165741, "grad_norm": 0.8253159090757203, "learning_rate": 2.5546961337650555e-06, "loss": 0.5539, "mean_token_accuracy": 0.8152087926864624, "num_tokens": 210250624.0, "step": 5709 }, { "epoch": 0.7546920433518371, "grad_norm": 0.7691742513135201, "learning_rate": 2.5531266972462176e-06, "loss": 0.5793, "mean_token_accuracy": 0.8108595013618469, "num_tokens": 210316160.0, "step": 5710 }, { "epoch": 0.7548242135871002, "grad_norm": 0.8245675689141889, "learning_rate": 2.5515578880551782e-06, "loss": 0.57, "mean_token_accuracy": 0.8102948069572449, "num_tokens": 210381696.0, "step": 5711 }, { "epoch": 0.7549563838223632, "grad_norm": 0.8374702293971308, "learning_rate": 2.5499897065259065e-06, "loss": 0.5998, "mean_token_accuracy": 0.8014588952064514, "num_tokens": 210447232.0, "step": 5712 }, { "epoch": 0.7550885540576262, "grad_norm": 0.8619496080040986, "learning_rate": 2.548422152992234e-06, "loss": 0.5909, "mean_token_accuracy": 0.8011079430580139, "num_tokens": 210512768.0, "step": 5713 }, { "epoch": 0.7552207242928892, "grad_norm": 0.8259840586505472, "learning_rate": 2.54685522778786e-06, "loss": 0.5864, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 210578304.0, "step": 5714 }, { "epoch": 0.7553528945281522, "grad_norm": 0.8068566561315587, "learning_rate": 2.5452889312463504e-06, "loss": 0.5359, "mean_token_accuracy": 0.8216487765312195, "num_tokens": 210643840.0, "step": 5715 }, { "epoch": 0.7554850647634153, "grad_norm": 0.7993526396801635, "learning_rate": 2.5437232637011376e-06, "loss": 0.5663, "mean_token_accuracy": 0.812965452671051, "num_tokens": 210709376.0, "step": 5716 }, { "epoch": 0.7556172349986783, "grad_norm": 0.8315302467236179, "learning_rate": 2.5421582254855214e-06, "loss": 0.5834, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 210774912.0, "step": 5717 }, { "epoch": 0.7557494052339413, "grad_norm": 0.8187141421006529, "learning_rate": 2.5405938169326624e-06, "loss": 0.5832, "mean_token_accuracy": 0.8040227293968201, "num_tokens": 210840448.0, "step": 5718 }, { "epoch": 0.7558815754692043, "grad_norm": 0.8648460560961242, "learning_rate": 2.5390300383755906e-06, "loss": 0.611, "mean_token_accuracy": 0.7989866733551025, "num_tokens": 210905984.0, "step": 5719 }, { "epoch": 0.7560137457044673, "grad_norm": 0.7602959728511898, "learning_rate": 2.537466890147204e-06, "loss": 0.5457, "mean_token_accuracy": 0.8201684951782227, "num_tokens": 210971520.0, "step": 5720 }, { "epoch": 0.7561459159397304, "grad_norm": 0.7837918324319006, "learning_rate": 2.5359043725802645e-06, "loss": 0.5695, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 211037056.0, "step": 5721 }, { "epoch": 0.7562780861749934, "grad_norm": 0.8147647277397491, "learning_rate": 2.5343424860073952e-06, "loss": 0.5625, "mean_token_accuracy": 0.8134537935256958, "num_tokens": 211102592.0, "step": 5722 }, { "epoch": 0.7564102564102564, "grad_norm": 0.8588325144584258, "learning_rate": 2.532781230761095e-06, "loss": 0.598, "mean_token_accuracy": 0.7991545796394348, "num_tokens": 211168128.0, "step": 5723 }, { "epoch": 0.7565424266455194, "grad_norm": 0.8176470511758632, "learning_rate": 2.531220607173718e-06, "loss": 0.5398, "mean_token_accuracy": 0.8224575519561768, "num_tokens": 211233664.0, "step": 5724 }, { "epoch": 0.7566745968807824, "grad_norm": 0.8604341576826839, "learning_rate": 2.5296606155774915e-06, "loss": 0.5874, "mean_token_accuracy": 0.8062202334403992, "num_tokens": 211299200.0, "step": 5725 }, { "epoch": 0.7568067671160454, "grad_norm": 0.7844396206434515, "learning_rate": 2.5281012563045015e-06, "loss": 0.5211, "mean_token_accuracy": 0.8267763257026672, "num_tokens": 211364736.0, "step": 5726 }, { "epoch": 0.7569389373513085, "grad_norm": 0.8373322491375443, "learning_rate": 2.526542529686708e-06, "loss": 0.593, "mean_token_accuracy": 0.8016878366470337, "num_tokens": 211430272.0, "step": 5727 }, { "epoch": 0.7570711075865715, "grad_norm": 0.798812277038364, "learning_rate": 2.5249844360559273e-06, "loss": 0.5495, "mean_token_accuracy": 0.8178488612174988, "num_tokens": 211495808.0, "step": 5728 }, { "epoch": 0.7572032778218345, "grad_norm": 0.818109960564222, "learning_rate": 2.5234269757438457e-06, "loss": 0.5714, "mean_token_accuracy": 0.8100811839103699, "num_tokens": 211561344.0, "step": 5729 }, { "epoch": 0.7573354480570975, "grad_norm": 0.826498385424818, "learning_rate": 2.521870149082015e-06, "loss": 0.5935, "mean_token_accuracy": 0.803030788898468, "num_tokens": 211626880.0, "step": 5730 }, { "epoch": 0.7574676182923605, "grad_norm": 0.8106924171624444, "learning_rate": 2.520313956401852e-06, "loss": 0.559, "mean_token_accuracy": 0.8128891587257385, "num_tokens": 211692416.0, "step": 5731 }, { "epoch": 0.7575997885276236, "grad_norm": 0.8472880591455135, "learning_rate": 2.5187583980346347e-06, "loss": 0.5906, "mean_token_accuracy": 0.8018099069595337, "num_tokens": 211757952.0, "step": 5732 }, { "epoch": 0.7577319587628866, "grad_norm": 0.8052354131067048, "learning_rate": 2.5172034743115103e-06, "loss": 0.6099, "mean_token_accuracy": 0.797948956489563, "num_tokens": 211823488.0, "step": 5733 }, { "epoch": 0.7578641289981496, "grad_norm": 0.820511521886538, "learning_rate": 2.5156491855634913e-06, "loss": 0.5518, "mean_token_accuracy": 0.8182761669158936, "num_tokens": 211889024.0, "step": 5734 }, { "epoch": 0.7579962992334126, "grad_norm": 0.8711923482834261, "learning_rate": 2.514095532121451e-06, "loss": 0.4887, "mean_token_accuracy": 0.8375961184501648, "num_tokens": 211954560.0, "step": 5735 }, { "epoch": 0.7581284694686756, "grad_norm": 0.7956477368362559, "learning_rate": 2.512542514316132e-06, "loss": 0.5544, "mean_token_accuracy": 0.8174368143081665, "num_tokens": 212020096.0, "step": 5736 }, { "epoch": 0.7582606397039386, "grad_norm": 0.7584283666189996, "learning_rate": 2.5109901324781374e-06, "loss": 0.5157, "mean_token_accuracy": 0.8287296891212463, "num_tokens": 212085632.0, "step": 5737 }, { "epoch": 0.7583928099392017, "grad_norm": 0.7906200499078296, "learning_rate": 2.5094383869379367e-06, "loss": 0.5399, "mean_token_accuracy": 0.8201990127563477, "num_tokens": 212151168.0, "step": 5738 }, { "epoch": 0.7585249801744647, "grad_norm": 0.9061913798782689, "learning_rate": 2.507887278025865e-06, "loss": 0.6232, "mean_token_accuracy": 0.7940117120742798, "num_tokens": 212216704.0, "step": 5739 }, { "epoch": 0.7586571504097277, "grad_norm": 0.9706491775989295, "learning_rate": 2.5063368060721227e-06, "loss": 0.6075, "mean_token_accuracy": 0.7976895570755005, "num_tokens": 212282240.0, "step": 5740 }, { "epoch": 0.7587893206449907, "grad_norm": 0.8634258215048309, "learning_rate": 2.50478697140677e-06, "loss": 0.5972, "mean_token_accuracy": 0.8019167184829712, "num_tokens": 212347776.0, "step": 5741 }, { "epoch": 0.7589214908802537, "grad_norm": 0.8704568161364427, "learning_rate": 2.5032377743597358e-06, "loss": 0.6228, "mean_token_accuracy": 0.7940117120742798, "num_tokens": 212413312.0, "step": 5742 }, { "epoch": 0.7590536611155168, "grad_norm": 0.8093672148219969, "learning_rate": 2.501689215260811e-06, "loss": 0.5952, "mean_token_accuracy": 0.8033817410469055, "num_tokens": 212478848.0, "step": 5743 }, { "epoch": 0.7591858313507798, "grad_norm": 0.8040957084589093, "learning_rate": 2.500141294439653e-06, "loss": 0.5743, "mean_token_accuracy": 0.8058539628982544, "num_tokens": 212544384.0, "step": 5744 }, { "epoch": 0.7593180015860428, "grad_norm": 0.7725463864820956, "learning_rate": 2.4985940122257787e-06, "loss": 0.5648, "mean_token_accuracy": 0.8121108412742615, "num_tokens": 212609920.0, "step": 5745 }, { "epoch": 0.7594501718213058, "grad_norm": 0.8053876611722302, "learning_rate": 2.4970473689485777e-06, "loss": 0.5615, "mean_token_accuracy": 0.8126754760742188, "num_tokens": 212675456.0, "step": 5746 }, { "epoch": 0.7595823420565688, "grad_norm": 0.8684556575587742, "learning_rate": 2.4955013649372927e-06, "loss": 0.605, "mean_token_accuracy": 0.8001617789268494, "num_tokens": 212740992.0, "step": 5747 }, { "epoch": 0.7597145122918318, "grad_norm": 0.8702827998209604, "learning_rate": 2.4939560005210373e-06, "loss": 0.6319, "mean_token_accuracy": 0.7899676561355591, "num_tokens": 212806528.0, "step": 5748 }, { "epoch": 0.7598466825270949, "grad_norm": 0.8537326227780826, "learning_rate": 2.4924112760287884e-06, "loss": 0.5995, "mean_token_accuracy": 0.8005738258361816, "num_tokens": 212872064.0, "step": 5749 }, { "epoch": 0.7599788527623579, "grad_norm": 0.8523231136927509, "learning_rate": 2.4908671917893863e-06, "loss": 0.5895, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 212937600.0, "step": 5750 }, { "epoch": 0.7601110229976209, "grad_norm": 0.8533224941156982, "learning_rate": 2.4893237481315304e-06, "loss": 0.5709, "mean_token_accuracy": 0.8080820441246033, "num_tokens": 213003136.0, "step": 5751 }, { "epoch": 0.7602431932328839, "grad_norm": 0.823297248552569, "learning_rate": 2.487780945383791e-06, "loss": 0.6172, "mean_token_accuracy": 0.7950646877288818, "num_tokens": 213068672.0, "step": 5752 }, { "epoch": 0.7603753634681469, "grad_norm": 0.8248711445778804, "learning_rate": 2.486238783874598e-06, "loss": 0.5933, "mean_token_accuracy": 0.8002075552940369, "num_tokens": 213134208.0, "step": 5753 }, { "epoch": 0.76050753370341, "grad_norm": 0.7990725934677444, "learning_rate": 2.484697263932245e-06, "loss": 0.5734, "mean_token_accuracy": 0.8098675608634949, "num_tokens": 213199744.0, "step": 5754 }, { "epoch": 0.760639703938673, "grad_norm": 0.7823295342995589, "learning_rate": 2.4831563858848905e-06, "loss": 0.5867, "mean_token_accuracy": 0.8062812685966492, "num_tokens": 213265280.0, "step": 5755 }, { "epoch": 0.760771874173936, "grad_norm": 0.8469042805860202, "learning_rate": 2.481616150060553e-06, "loss": 0.5731, "mean_token_accuracy": 0.8100354075431824, "num_tokens": 213330816.0, "step": 5756 }, { "epoch": 0.760904044409199, "grad_norm": 0.8879537085564982, "learning_rate": 2.4800765567871196e-06, "loss": 0.6416, "mean_token_accuracy": 0.7898150682449341, "num_tokens": 213396352.0, "step": 5757 }, { "epoch": 0.761036214644462, "grad_norm": 0.8581228559219067, "learning_rate": 2.478537606392333e-06, "loss": 0.6047, "mean_token_accuracy": 0.7999938726425171, "num_tokens": 213461888.0, "step": 5758 }, { "epoch": 0.761168384879725, "grad_norm": 0.8298041394541962, "learning_rate": 2.4769992992038087e-06, "loss": 0.557, "mean_token_accuracy": 0.812217652797699, "num_tokens": 213527424.0, "step": 5759 }, { "epoch": 0.7613005551149881, "grad_norm": 0.8115140445024177, "learning_rate": 2.4754616355490164e-06, "loss": 0.5594, "mean_token_accuracy": 0.816307544708252, "num_tokens": 213592960.0, "step": 5760 }, { "epoch": 0.7614327253502511, "grad_norm": 0.8005405474962068, "learning_rate": 2.4739246157552944e-06, "loss": 0.5687, "mean_token_accuracy": 0.8100201487541199, "num_tokens": 213658496.0, "step": 5761 }, { "epoch": 0.7615648955855141, "grad_norm": 0.8249807439555643, "learning_rate": 2.4723882401498418e-06, "loss": 0.6028, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 213724032.0, "step": 5762 }, { "epoch": 0.7616970658207771, "grad_norm": 0.7774893916689385, "learning_rate": 2.4708525090597225e-06, "loss": 0.5734, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 213789568.0, "step": 5763 }, { "epoch": 0.7618292360560401, "grad_norm": 0.7676476179458654, "learning_rate": 2.469317422811857e-06, "loss": 0.5417, "mean_token_accuracy": 0.8212367296218872, "num_tokens": 213855104.0, "step": 5764 }, { "epoch": 0.7619614062913032, "grad_norm": 0.8165190617235439, "learning_rate": 2.4677829817330387e-06, "loss": 0.6038, "mean_token_accuracy": 0.7981931567192078, "num_tokens": 213920640.0, "step": 5765 }, { "epoch": 0.7620935765265662, "grad_norm": 0.9029065906595721, "learning_rate": 2.4662491861499144e-06, "loss": 0.5978, "mean_token_accuracy": 0.8018404245376587, "num_tokens": 213986176.0, "step": 5766 }, { "epoch": 0.7622257467618292, "grad_norm": 0.8017716529192696, "learning_rate": 2.464716036388998e-06, "loss": 0.5252, "mean_token_accuracy": 0.8244872689247131, "num_tokens": 214051712.0, "step": 5767 }, { "epoch": 0.7623579169970922, "grad_norm": 0.8155358831354691, "learning_rate": 2.4631835327766652e-06, "loss": 0.5748, "mean_token_accuracy": 0.8076547384262085, "num_tokens": 214117248.0, "step": 5768 }, { "epoch": 0.7624900872323552, "grad_norm": 0.8230795632701865, "learning_rate": 2.461651675639155e-06, "loss": 0.6081, "mean_token_accuracy": 0.797918438911438, "num_tokens": 214182784.0, "step": 5769 }, { "epoch": 0.7626222574676182, "grad_norm": 0.7856353967075153, "learning_rate": 2.4601204653025665e-06, "loss": 0.555, "mean_token_accuracy": 0.8160938620567322, "num_tokens": 214248320.0, "step": 5770 }, { "epoch": 0.7627544277028813, "grad_norm": 0.7707218227293612, "learning_rate": 2.4585899020928624e-06, "loss": 0.5643, "mean_token_accuracy": 0.8136063814163208, "num_tokens": 214313856.0, "step": 5771 }, { "epoch": 0.7628865979381443, "grad_norm": 0.7944601701093504, "learning_rate": 2.457059986335868e-06, "loss": 0.588, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 214379392.0, "step": 5772 }, { "epoch": 0.7630187681734073, "grad_norm": 0.824973099852325, "learning_rate": 2.455530718357273e-06, "loss": 0.5754, "mean_token_accuracy": 0.8093028664588928, "num_tokens": 214444928.0, "step": 5773 }, { "epoch": 0.7631509384086703, "grad_norm": 0.8437844494701054, "learning_rate": 2.454002098482621e-06, "loss": 0.5388, "mean_token_accuracy": 0.8214656114578247, "num_tokens": 214510464.0, "step": 5774 }, { "epoch": 0.7632831086439333, "grad_norm": 0.8332076031729021, "learning_rate": 2.452474127037327e-06, "loss": 0.5724, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 214576000.0, "step": 5775 }, { "epoch": 0.7634152788791964, "grad_norm": 0.8231780782105215, "learning_rate": 2.4509468043466657e-06, "loss": 0.5559, "mean_token_accuracy": 0.8174673318862915, "num_tokens": 214641536.0, "step": 5776 }, { "epoch": 0.7635474491144594, "grad_norm": 0.8235406356789128, "learning_rate": 2.4494201307357665e-06, "loss": 0.5336, "mean_token_accuracy": 0.8203210830688477, "num_tokens": 214707072.0, "step": 5777 }, { "epoch": 0.7636796193497224, "grad_norm": 0.8070599531075852, "learning_rate": 2.4478941065296327e-06, "loss": 0.5557, "mean_token_accuracy": 0.8138658404350281, "num_tokens": 214772608.0, "step": 5778 }, { "epoch": 0.7638117895849854, "grad_norm": 0.819041295250835, "learning_rate": 2.446368732053119e-06, "loss": 0.6, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 214838144.0, "step": 5779 }, { "epoch": 0.7639439598202484, "grad_norm": 0.7988245443416646, "learning_rate": 2.444844007630947e-06, "loss": 0.5828, "mean_token_accuracy": 0.8041447997093201, "num_tokens": 214903680.0, "step": 5780 }, { "epoch": 0.7640761300555114, "grad_norm": 0.9232837867738496, "learning_rate": 2.443319933587699e-06, "loss": 0.628, "mean_token_accuracy": 0.792378842830658, "num_tokens": 214969216.0, "step": 5781 }, { "epoch": 0.7642083002907745, "grad_norm": 0.882269239847354, "learning_rate": 2.4417965102478197e-06, "loss": 0.641, "mean_token_accuracy": 0.787861704826355, "num_tokens": 215034752.0, "step": 5782 }, { "epoch": 0.7643404705260375, "grad_norm": 0.8601849553189408, "learning_rate": 2.4402737379356094e-06, "loss": 0.5778, "mean_token_accuracy": 0.8079752326011658, "num_tokens": 215100288.0, "step": 5783 }, { "epoch": 0.7644726407613005, "grad_norm": 0.803537915270914, "learning_rate": 2.4387516169752412e-06, "loss": 0.5472, "mean_token_accuracy": 0.8186424374580383, "num_tokens": 215165824.0, "step": 5784 }, { "epoch": 0.7646048109965635, "grad_norm": 0.7846763903061944, "learning_rate": 2.4372301476907375e-06, "loss": 0.5547, "mean_token_accuracy": 0.8146746158599854, "num_tokens": 215231360.0, "step": 5785 }, { "epoch": 0.7647369812318265, "grad_norm": 0.7524762725608235, "learning_rate": 2.4357093304059914e-06, "loss": 0.5234, "mean_token_accuracy": 0.8234647512435913, "num_tokens": 215296896.0, "step": 5786 }, { "epoch": 0.7648691514670896, "grad_norm": 0.8761926723194247, "learning_rate": 2.4341891654447477e-06, "loss": 0.6251, "mean_token_accuracy": 0.7895098328590393, "num_tokens": 215362432.0, "step": 5787 }, { "epoch": 0.7650013217023526, "grad_norm": 0.8380117510385423, "learning_rate": 2.4326696531306243e-06, "loss": 0.5886, "mean_token_accuracy": 0.8042516112327576, "num_tokens": 215427968.0, "step": 5788 }, { "epoch": 0.7651334919376156, "grad_norm": 0.9152035255465744, "learning_rate": 2.4311507937870887e-06, "loss": 0.5744, "mean_token_accuracy": 0.8081125617027283, "num_tokens": 215493504.0, "step": 5789 }, { "epoch": 0.7652656621728786, "grad_norm": 0.7427849207249819, "learning_rate": 2.429632587737476e-06, "loss": 0.5079, "mean_token_accuracy": 0.8307288289070129, "num_tokens": 215559040.0, "step": 5790 }, { "epoch": 0.7653978324081416, "grad_norm": 0.8590977216455752, "learning_rate": 2.4281150353049815e-06, "loss": 0.5236, "mean_token_accuracy": 0.8249297738075256, "num_tokens": 215624576.0, "step": 5791 }, { "epoch": 0.7655300026434047, "grad_norm": 0.827108406885995, "learning_rate": 2.4265981368126606e-06, "loss": 0.5561, "mean_token_accuracy": 0.8151935338973999, "num_tokens": 215690112.0, "step": 5792 }, { "epoch": 0.7656621728786678, "grad_norm": 0.8678743625694412, "learning_rate": 2.425081892583427e-06, "loss": 0.5895, "mean_token_accuracy": 0.8031070828437805, "num_tokens": 215755648.0, "step": 5793 }, { "epoch": 0.7657943431139308, "grad_norm": 0.7555197604433256, "learning_rate": 2.4235663029400587e-06, "loss": 0.5358, "mean_token_accuracy": 0.8209467530250549, "num_tokens": 215821184.0, "step": 5794 }, { "epoch": 0.7659265133491938, "grad_norm": 0.8221241673869543, "learning_rate": 2.422051368205195e-06, "loss": 0.5406, "mean_token_accuracy": 0.8200005888938904, "num_tokens": 215886720.0, "step": 5795 }, { "epoch": 0.7660586835844568, "grad_norm": 0.8071062507096932, "learning_rate": 2.4205370887013287e-06, "loss": 0.577, "mean_token_accuracy": 0.8055182695388794, "num_tokens": 215952256.0, "step": 5796 }, { "epoch": 0.7661908538197199, "grad_norm": 0.7685409065461851, "learning_rate": 2.419023464750825e-06, "loss": 0.5344, "mean_token_accuracy": 0.8229153752326965, "num_tokens": 216017792.0, "step": 5797 }, { "epoch": 0.7663230240549829, "grad_norm": 0.8554921571894372, "learning_rate": 2.4175104966758977e-06, "loss": 0.6084, "mean_token_accuracy": 0.7972469925880432, "num_tokens": 216083328.0, "step": 5798 }, { "epoch": 0.7664551942902459, "grad_norm": 0.7560058547310998, "learning_rate": 2.415998184798628e-06, "loss": 0.5341, "mean_token_accuracy": 0.821541965007782, "num_tokens": 216148864.0, "step": 5799 }, { "epoch": 0.7665873645255089, "grad_norm": 0.7431231292560689, "learning_rate": 2.4144865294409553e-06, "loss": 0.5284, "mean_token_accuracy": 0.82627272605896, "num_tokens": 216214400.0, "step": 5800 }, { "epoch": 0.7667195347607719, "grad_norm": 0.7831031195828498, "learning_rate": 2.412975530924681e-06, "loss": 0.516, "mean_token_accuracy": 0.8289281129837036, "num_tokens": 216279936.0, "step": 5801 }, { "epoch": 0.7668517049960349, "grad_norm": 0.8394704915905234, "learning_rate": 2.4114651895714626e-06, "loss": 0.5922, "mean_token_accuracy": 0.8009400367736816, "num_tokens": 216345472.0, "step": 5802 }, { "epoch": 0.766983875231298, "grad_norm": 0.8174553079605423, "learning_rate": 2.4099555057028214e-06, "loss": 0.5861, "mean_token_accuracy": 0.8040837645530701, "num_tokens": 216411008.0, "step": 5803 }, { "epoch": 0.767116045466561, "grad_norm": 0.8308052807976906, "learning_rate": 2.4084464796401374e-06, "loss": 0.567, "mean_token_accuracy": 0.8103059530258179, "num_tokens": 216473543.0, "step": 5804 }, { "epoch": 0.767248215701824, "grad_norm": 0.777394428041238, "learning_rate": 2.4069381117046523e-06, "loss": 0.5875, "mean_token_accuracy": 0.806403398513794, "num_tokens": 216539079.0, "step": 5805 }, { "epoch": 0.767380385937087, "grad_norm": 0.8457356947904877, "learning_rate": 2.405430402217462e-06, "loss": 0.5987, "mean_token_accuracy": 0.7998260259628296, "num_tokens": 216604615.0, "step": 5806 }, { "epoch": 0.76751255617235, "grad_norm": 0.8276477988981431, "learning_rate": 2.403923351499532e-06, "loss": 0.5261, "mean_token_accuracy": 0.8250366449356079, "num_tokens": 216670151.0, "step": 5807 }, { "epoch": 0.767644726407613, "grad_norm": 0.7940606032424662, "learning_rate": 2.4024169598716772e-06, "loss": 0.5753, "mean_token_accuracy": 0.8088297843933105, "num_tokens": 216735687.0, "step": 5808 }, { "epoch": 0.7677768966428761, "grad_norm": 0.8326846662045059, "learning_rate": 2.4009112276545794e-06, "loss": 0.5265, "mean_token_accuracy": 0.8231748342514038, "num_tokens": 216801223.0, "step": 5809 }, { "epoch": 0.7679090668781391, "grad_norm": 0.8677534954843509, "learning_rate": 2.3994061551687762e-06, "loss": 0.5447, "mean_token_accuracy": 0.8223507404327393, "num_tokens": 216866759.0, "step": 5810 }, { "epoch": 0.7680412371134021, "grad_norm": 0.8659374407364153, "learning_rate": 2.3979017427346686e-06, "loss": 0.5834, "mean_token_accuracy": 0.8073647618293762, "num_tokens": 216932295.0, "step": 5811 }, { "epoch": 0.7681734073486651, "grad_norm": 0.8571308988405062, "learning_rate": 2.3963979906725113e-06, "loss": 0.6236, "mean_token_accuracy": 0.7936607003211975, "num_tokens": 216997831.0, "step": 5812 }, { "epoch": 0.7683055775839281, "grad_norm": 0.774769043721649, "learning_rate": 2.3948948993024233e-06, "loss": 0.5687, "mean_token_accuracy": 0.8113478422164917, "num_tokens": 217063367.0, "step": 5813 }, { "epoch": 0.7684377478191912, "grad_norm": 0.7730269490943967, "learning_rate": 2.393392468944383e-06, "loss": 0.5602, "mean_token_accuracy": 0.8115156888961792, "num_tokens": 217128903.0, "step": 5814 }, { "epoch": 0.7685699180544542, "grad_norm": 0.8175786127265706, "learning_rate": 2.391890699918222e-06, "loss": 0.5806, "mean_token_accuracy": 0.807639479637146, "num_tokens": 217194439.0, "step": 5815 }, { "epoch": 0.7687020882897172, "grad_norm": 0.7876769618155162, "learning_rate": 2.3903895925436405e-06, "loss": 0.5564, "mean_token_accuracy": 0.8158649802207947, "num_tokens": 217259975.0, "step": 5816 }, { "epoch": 0.7688342585249802, "grad_norm": 0.8447872812921196, "learning_rate": 2.38888914714019e-06, "loss": 0.6262, "mean_token_accuracy": 0.7923025488853455, "num_tokens": 217325511.0, "step": 5817 }, { "epoch": 0.7689664287602432, "grad_norm": 0.8411188612513566, "learning_rate": 2.387389364027286e-06, "loss": 0.6087, "mean_token_accuracy": 0.7982694506645203, "num_tokens": 217391047.0, "step": 5818 }, { "epoch": 0.7690985989955063, "grad_norm": 0.8121208572731822, "learning_rate": 2.3858902435241967e-06, "loss": 0.5279, "mean_token_accuracy": 0.8236173987388611, "num_tokens": 217456583.0, "step": 5819 }, { "epoch": 0.7692307692307693, "grad_norm": 0.8521070126431769, "learning_rate": 2.38439178595006e-06, "loss": 0.5569, "mean_token_accuracy": 0.816215991973877, "num_tokens": 217522119.0, "step": 5820 }, { "epoch": 0.7693629394660323, "grad_norm": 0.8044194148432182, "learning_rate": 2.3828939916238605e-06, "loss": 0.5524, "mean_token_accuracy": 0.8169789910316467, "num_tokens": 217587655.0, "step": 5821 }, { "epoch": 0.7694951097012953, "grad_norm": 0.793822080898781, "learning_rate": 2.3813968608644503e-06, "loss": 0.5884, "mean_token_accuracy": 0.8028628826141357, "num_tokens": 217653191.0, "step": 5822 }, { "epoch": 0.7696272799365583, "grad_norm": 0.8150305437902028, "learning_rate": 2.3799003939905368e-06, "loss": 0.5794, "mean_token_accuracy": 0.8073647618293762, "num_tokens": 217718727.0, "step": 5823 }, { "epoch": 0.7697594501718213, "grad_norm": 0.7952111803645945, "learning_rate": 2.378404591320688e-06, "loss": 0.5601, "mean_token_accuracy": 0.8126144409179688, "num_tokens": 217784263.0, "step": 5824 }, { "epoch": 0.7698916204070844, "grad_norm": 0.811230120196056, "learning_rate": 2.376909453173324e-06, "loss": 0.6035, "mean_token_accuracy": 0.797857403755188, "num_tokens": 217849799.0, "step": 5825 }, { "epoch": 0.7700237906423474, "grad_norm": 0.8414103404877296, "learning_rate": 2.375414979866735e-06, "loss": 0.6331, "mean_token_accuracy": 0.7874649167060852, "num_tokens": 217915335.0, "step": 5826 }, { "epoch": 0.7701559608776104, "grad_norm": 0.7993906177729226, "learning_rate": 2.373921171719059e-06, "loss": 0.5501, "mean_token_accuracy": 0.8165669441223145, "num_tokens": 217980871.0, "step": 5827 }, { "epoch": 0.7702881311128734, "grad_norm": 0.8623128308799919, "learning_rate": 2.3724280290482977e-06, "loss": 0.5813, "mean_token_accuracy": 0.8069069981575012, "num_tokens": 218046407.0, "step": 5828 }, { "epoch": 0.7704203013481364, "grad_norm": 0.8561723751940241, "learning_rate": 2.37093555217231e-06, "loss": 0.5952, "mean_token_accuracy": 0.8027560710906982, "num_tokens": 218111943.0, "step": 5829 }, { "epoch": 0.7705524715833995, "grad_norm": 0.7636005616628236, "learning_rate": 2.3694437414088144e-06, "loss": 0.5472, "mean_token_accuracy": 0.8141252398490906, "num_tokens": 218177479.0, "step": 5830 }, { "epoch": 0.7706846418186625, "grad_norm": 0.7768817872252726, "learning_rate": 2.3679525970753833e-06, "loss": 0.5551, "mean_token_accuracy": 0.816338062286377, "num_tokens": 218243015.0, "step": 5831 }, { "epoch": 0.7708168120539255, "grad_norm": 0.8491551369560862, "learning_rate": 2.3664621194894525e-06, "loss": 0.5953, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 218308551.0, "step": 5832 }, { "epoch": 0.7709489822891885, "grad_norm": 0.8895658838955253, "learning_rate": 2.364972308968314e-06, "loss": 0.6194, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 218374087.0, "step": 5833 }, { "epoch": 0.7710811525244515, "grad_norm": 0.8209306231587669, "learning_rate": 2.3634831658291145e-06, "loss": 0.5425, "mean_token_accuracy": 0.8167958855628967, "num_tokens": 218439623.0, "step": 5834 }, { "epoch": 0.7712133227597145, "grad_norm": 0.7875101514269943, "learning_rate": 2.3619946903888635e-06, "loss": 0.5687, "mean_token_accuracy": 0.8105084896087646, "num_tokens": 218505159.0, "step": 5835 }, { "epoch": 0.7713454929949776, "grad_norm": 0.8865341426388655, "learning_rate": 2.3605068829644256e-06, "loss": 0.5539, "mean_token_accuracy": 0.8179709315299988, "num_tokens": 218570695.0, "step": 5836 }, { "epoch": 0.7714776632302406, "grad_norm": 0.7515958520710783, "learning_rate": 2.359019743872527e-06, "loss": 0.5557, "mean_token_accuracy": 0.8149340748786926, "num_tokens": 218636231.0, "step": 5837 }, { "epoch": 0.7716098334655036, "grad_norm": 0.7990375917141043, "learning_rate": 2.357533273429742e-06, "loss": 0.5801, "mean_token_accuracy": 0.8078684210777283, "num_tokens": 218701767.0, "step": 5838 }, { "epoch": 0.7717420037007666, "grad_norm": 0.8558058167121062, "learning_rate": 2.3560474719525164e-06, "loss": 0.5962, "mean_token_accuracy": 0.8041447997093201, "num_tokens": 218767303.0, "step": 5839 }, { "epoch": 0.7718741739360296, "grad_norm": 0.7780559195503981, "learning_rate": 2.3545623397571417e-06, "loss": 0.5353, "mean_token_accuracy": 0.8235411047935486, "num_tokens": 218832839.0, "step": 5840 }, { "epoch": 0.7720063441712927, "grad_norm": 0.8257757834870719, "learning_rate": 2.3530778771597725e-06, "loss": 0.6448, "mean_token_accuracy": 0.7869155406951904, "num_tokens": 218898375.0, "step": 5841 }, { "epoch": 0.7721385144065557, "grad_norm": 0.8415515402512646, "learning_rate": 2.3515940844764202e-06, "loss": 0.6176, "mean_token_accuracy": 0.793218195438385, "num_tokens": 218963911.0, "step": 5842 }, { "epoch": 0.7722706846418187, "grad_norm": 0.8179988955962185, "learning_rate": 2.3501109620229558e-06, "loss": 0.5605, "mean_token_accuracy": 0.8141099810600281, "num_tokens": 219029447.0, "step": 5843 }, { "epoch": 0.7724028548770817, "grad_norm": 0.8672921540336402, "learning_rate": 2.348628510115099e-06, "loss": 0.589, "mean_token_accuracy": 0.807639479637146, "num_tokens": 219094983.0, "step": 5844 }, { "epoch": 0.7725350251123447, "grad_norm": 0.8154370166219741, "learning_rate": 2.347146729068439e-06, "loss": 0.5727, "mean_token_accuracy": 0.8104932308197021, "num_tokens": 219160519.0, "step": 5845 }, { "epoch": 0.7726671953476077, "grad_norm": 0.791884100108931, "learning_rate": 2.3456656191984125e-06, "loss": 0.5352, "mean_token_accuracy": 0.8210841417312622, "num_tokens": 219226055.0, "step": 5846 }, { "epoch": 0.7727993655828708, "grad_norm": 0.7763407387055563, "learning_rate": 2.3441851808203196e-06, "loss": 0.5331, "mean_token_accuracy": 0.8216182589530945, "num_tokens": 219291591.0, "step": 5847 }, { "epoch": 0.7729315358181338, "grad_norm": 0.8382289080362677, "learning_rate": 2.34270541424931e-06, "loss": 0.5545, "mean_token_accuracy": 0.8150561451911926, "num_tokens": 219357127.0, "step": 5848 }, { "epoch": 0.7730637060533968, "grad_norm": 0.813548924911645, "learning_rate": 2.341226319800401e-06, "loss": 0.593, "mean_token_accuracy": 0.8027102947235107, "num_tokens": 219422663.0, "step": 5849 }, { "epoch": 0.7731958762886598, "grad_norm": 0.7885820371337952, "learning_rate": 2.339747897788456e-06, "loss": 0.6027, "mean_token_accuracy": 0.797979474067688, "num_tokens": 219488199.0, "step": 5850 }, { "epoch": 0.7733280465239228, "grad_norm": 0.8041184787877567, "learning_rate": 2.3382701485282032e-06, "loss": 0.5657, "mean_token_accuracy": 0.8119582533836365, "num_tokens": 219553735.0, "step": 5851 }, { "epoch": 0.7734602167591859, "grad_norm": 0.8513398573316402, "learning_rate": 2.336793072334225e-06, "loss": 0.5686, "mean_token_accuracy": 0.811790406703949, "num_tokens": 219619271.0, "step": 5852 }, { "epoch": 0.7735923869944489, "grad_norm": 0.9254943920398188, "learning_rate": 2.3353166695209567e-06, "loss": 0.6237, "mean_token_accuracy": 0.7937675714492798, "num_tokens": 219684807.0, "step": 5853 }, { "epoch": 0.7737245572297119, "grad_norm": 0.8749216027674825, "learning_rate": 2.333840940402696e-06, "loss": 0.5781, "mean_token_accuracy": 0.8075326681137085, "num_tokens": 219750343.0, "step": 5854 }, { "epoch": 0.7738567274649749, "grad_norm": 0.8382863592312486, "learning_rate": 2.3323658852935945e-06, "loss": 0.5795, "mean_token_accuracy": 0.807700514793396, "num_tokens": 219815879.0, "step": 5855 }, { "epoch": 0.7739888977002379, "grad_norm": 0.9195322978757193, "learning_rate": 2.3308915045076617e-06, "loss": 0.6253, "mean_token_accuracy": 0.7981626391410828, "num_tokens": 219881415.0, "step": 5856 }, { "epoch": 0.774121067935501, "grad_norm": 0.7977865524281492, "learning_rate": 2.3294177983587583e-06, "loss": 0.5851, "mean_token_accuracy": 0.8084635734558105, "num_tokens": 219946951.0, "step": 5857 }, { "epoch": 0.774253238170764, "grad_norm": 0.7608985954841836, "learning_rate": 2.327944767160612e-06, "loss": 0.5411, "mean_token_accuracy": 0.8201532363891602, "num_tokens": 220012487.0, "step": 5858 }, { "epoch": 0.774385408406027, "grad_norm": 0.8156565601781869, "learning_rate": 2.3264724112267957e-06, "loss": 0.5874, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 220078023.0, "step": 5859 }, { "epoch": 0.77451757864129, "grad_norm": 0.9215577051605209, "learning_rate": 2.325000730870744e-06, "loss": 0.6048, "mean_token_accuracy": 0.7986204624176025, "num_tokens": 220143559.0, "step": 5860 }, { "epoch": 0.774649748876553, "grad_norm": 0.8683480876453445, "learning_rate": 2.3235297264057477e-06, "loss": 0.5552, "mean_token_accuracy": 0.8176962733268738, "num_tokens": 220209095.0, "step": 5861 }, { "epoch": 0.774781919111816, "grad_norm": 0.8150683307880843, "learning_rate": 2.322059398144954e-06, "loss": 0.5786, "mean_token_accuracy": 0.8085551261901855, "num_tokens": 220274631.0, "step": 5862 }, { "epoch": 0.7749140893470791, "grad_norm": 0.78930660345907, "learning_rate": 2.320589746401363e-06, "loss": 0.594, "mean_token_accuracy": 0.8049994111061096, "num_tokens": 220340167.0, "step": 5863 }, { "epoch": 0.7750462595823421, "grad_norm": 0.7675729524750443, "learning_rate": 2.319120771487833e-06, "loss": 0.5502, "mean_token_accuracy": 0.8168416619300842, "num_tokens": 220405703.0, "step": 5864 }, { "epoch": 0.7751784298176051, "grad_norm": 0.8192508024154799, "learning_rate": 2.317652473717079e-06, "loss": 0.5899, "mean_token_accuracy": 0.8027102947235107, "num_tokens": 220471239.0, "step": 5865 }, { "epoch": 0.7753106000528681, "grad_norm": 0.8875023635339625, "learning_rate": 2.3161848534016723e-06, "loss": 0.5808, "mean_token_accuracy": 0.8080515265464783, "num_tokens": 220536775.0, "step": 5866 }, { "epoch": 0.7754427702881311, "grad_norm": 0.8422474339335818, "learning_rate": 2.3147179108540345e-06, "loss": 0.5819, "mean_token_accuracy": 0.8051977753639221, "num_tokens": 220602311.0, "step": 5867 }, { "epoch": 0.7755749405233942, "grad_norm": 0.9420611126983028, "learning_rate": 2.3132516463864517e-06, "loss": 0.5757, "mean_token_accuracy": 0.8078989386558533, "num_tokens": 220667847.0, "step": 5868 }, { "epoch": 0.7757071107586572, "grad_norm": 0.8185436669721446, "learning_rate": 2.311786060311058e-06, "loss": 0.6182, "mean_token_accuracy": 0.7970027923583984, "num_tokens": 220733383.0, "step": 5869 }, { "epoch": 0.7758392809939202, "grad_norm": 0.8582594131170871, "learning_rate": 2.3103211529398464e-06, "loss": 0.5711, "mean_token_accuracy": 0.8104169368743896, "num_tokens": 220798919.0, "step": 5870 }, { "epoch": 0.7759714512291832, "grad_norm": 0.7767780322978806, "learning_rate": 2.3088569245846675e-06, "loss": 0.5376, "mean_token_accuracy": 0.8221523761749268, "num_tokens": 220864455.0, "step": 5871 }, { "epoch": 0.7761036214644462, "grad_norm": 0.8084470755593932, "learning_rate": 2.307393375557222e-06, "loss": 0.5678, "mean_token_accuracy": 0.8111189007759094, "num_tokens": 220929991.0, "step": 5872 }, { "epoch": 0.7762357916997092, "grad_norm": 0.805668685395082, "learning_rate": 2.305930506169069e-06, "loss": 0.5668, "mean_token_accuracy": 0.812873899936676, "num_tokens": 220995527.0, "step": 5873 }, { "epoch": 0.7763679619349723, "grad_norm": 0.8449666999008634, "learning_rate": 2.3044683167316247e-06, "loss": 0.604, "mean_token_accuracy": 0.8031681180000305, "num_tokens": 221061063.0, "step": 5874 }, { "epoch": 0.7765001321702353, "grad_norm": 0.8218629268963336, "learning_rate": 2.303006807556159e-06, "loss": 0.6096, "mean_token_accuracy": 0.7954156994819641, "num_tokens": 221126599.0, "step": 5875 }, { "epoch": 0.7766323024054983, "grad_norm": 0.7942457203650408, "learning_rate": 2.3015459789537926e-06, "loss": 0.5914, "mean_token_accuracy": 0.8027408123016357, "num_tokens": 221192135.0, "step": 5876 }, { "epoch": 0.7767644726407613, "grad_norm": 0.8476331808834655, "learning_rate": 2.3000858312355118e-06, "loss": 0.5837, "mean_token_accuracy": 0.802466094493866, "num_tokens": 221257671.0, "step": 5877 }, { "epoch": 0.7768966428760243, "grad_norm": 0.8003974505960584, "learning_rate": 2.298626364712146e-06, "loss": 0.5743, "mean_token_accuracy": 0.8078836798667908, "num_tokens": 221323207.0, "step": 5878 }, { "epoch": 0.7770288131112874, "grad_norm": 0.8626517102330488, "learning_rate": 2.2971675796943887e-06, "loss": 0.5961, "mean_token_accuracy": 0.8016268014907837, "num_tokens": 221388743.0, "step": 5879 }, { "epoch": 0.7771609833465504, "grad_norm": 0.8479504673410636, "learning_rate": 2.295709476492781e-06, "loss": 0.5817, "mean_token_accuracy": 0.8071663975715637, "num_tokens": 221454279.0, "step": 5880 }, { "epoch": 0.7772931535818134, "grad_norm": 0.784444814515877, "learning_rate": 2.2942520554177275e-06, "loss": 0.5677, "mean_token_accuracy": 0.8099285960197449, "num_tokens": 221519815.0, "step": 5881 }, { "epoch": 0.7774253238170764, "grad_norm": 0.7986548694366725, "learning_rate": 2.2927953167794794e-06, "loss": 0.5798, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 221585351.0, "step": 5882 }, { "epoch": 0.7775574940523394, "grad_norm": 0.8235565388600175, "learning_rate": 2.291339260888146e-06, "loss": 0.5644, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 221650887.0, "step": 5883 }, { "epoch": 0.7776896642876024, "grad_norm": 0.8655357306415065, "learning_rate": 2.289883888053692e-06, "loss": 0.5972, "mean_token_accuracy": 0.8029850125312805, "num_tokens": 221716423.0, "step": 5884 }, { "epoch": 0.7778218345228655, "grad_norm": 0.7541285353187935, "learning_rate": 2.288429198585938e-06, "loss": 0.5152, "mean_token_accuracy": 0.8268678784370422, "num_tokens": 221781959.0, "step": 5885 }, { "epoch": 0.7779540047581285, "grad_norm": 0.7776742852154876, "learning_rate": 2.2869751927945526e-06, "loss": 0.5254, "mean_token_accuracy": 0.8220303058624268, "num_tokens": 221847495.0, "step": 5886 }, { "epoch": 0.7780861749933915, "grad_norm": 0.7498365108313194, "learning_rate": 2.2855218709890654e-06, "loss": 0.5541, "mean_token_accuracy": 0.816185474395752, "num_tokens": 221913031.0, "step": 5887 }, { "epoch": 0.7782183452286545, "grad_norm": 0.8169913111510191, "learning_rate": 2.284069233478859e-06, "loss": 0.565, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 221978567.0, "step": 5888 }, { "epoch": 0.7783505154639175, "grad_norm": 0.8695811409960383, "learning_rate": 2.282617280573169e-06, "loss": 0.5704, "mean_token_accuracy": 0.811881959438324, "num_tokens": 222044103.0, "step": 5889 }, { "epoch": 0.7784826856991806, "grad_norm": 0.8427308050050442, "learning_rate": 2.281166012581087e-06, "loss": 0.5601, "mean_token_accuracy": 0.8167806267738342, "num_tokens": 222109639.0, "step": 5890 }, { "epoch": 0.7786148559344436, "grad_norm": 0.7621371954458841, "learning_rate": 2.279715429811555e-06, "loss": 0.5302, "mean_token_accuracy": 0.8229764103889465, "num_tokens": 222175175.0, "step": 5891 }, { "epoch": 0.7787470261697066, "grad_norm": 0.8279389506694997, "learning_rate": 2.2782655325733737e-06, "loss": 0.569, "mean_token_accuracy": 0.8110731244087219, "num_tokens": 222240711.0, "step": 5892 }, { "epoch": 0.7788791964049696, "grad_norm": 0.7669122491274085, "learning_rate": 2.276816321175195e-06, "loss": 0.5335, "mean_token_accuracy": 0.8223049640655518, "num_tokens": 222306247.0, "step": 5893 }, { "epoch": 0.7790113666402326, "grad_norm": 0.9077704089088513, "learning_rate": 2.2753677959255284e-06, "loss": 0.6349, "mean_token_accuracy": 0.7866865992546082, "num_tokens": 222371783.0, "step": 5894 }, { "epoch": 0.7791435368754956, "grad_norm": 0.7143891601991226, "learning_rate": 2.2739199571327307e-06, "loss": 0.478, "mean_token_accuracy": 0.8409076929092407, "num_tokens": 222437319.0, "step": 5895 }, { "epoch": 0.7792757071107587, "grad_norm": 0.8981066523274636, "learning_rate": 2.2724728051050183e-06, "loss": 0.6415, "mean_token_accuracy": 0.78789222240448, "num_tokens": 222502855.0, "step": 5896 }, { "epoch": 0.7794078773460217, "grad_norm": 0.795759580921892, "learning_rate": 2.27102634015046e-06, "loss": 0.5518, "mean_token_accuracy": 0.8152393102645874, "num_tokens": 222568391.0, "step": 5897 }, { "epoch": 0.7795400475812847, "grad_norm": 0.812397782694372, "learning_rate": 2.269580562576979e-06, "loss": 0.6081, "mean_token_accuracy": 0.7993529438972473, "num_tokens": 222633927.0, "step": 5898 }, { "epoch": 0.7796722178165477, "grad_norm": 0.7558544211766615, "learning_rate": 2.2681354726923472e-06, "loss": 0.5404, "mean_token_accuracy": 0.8234647512435913, "num_tokens": 222699463.0, "step": 5899 }, { "epoch": 0.7798043880518107, "grad_norm": 0.84637438516823, "learning_rate": 2.2666910708041996e-06, "loss": 0.5779, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 222764999.0, "step": 5900 }, { "epoch": 0.7799365582870738, "grad_norm": 0.8526577306906108, "learning_rate": 2.265247357220015e-06, "loss": 0.6018, "mean_token_accuracy": 0.8005890846252441, "num_tokens": 222830535.0, "step": 5901 }, { "epoch": 0.7800687285223368, "grad_norm": 0.8948371783285379, "learning_rate": 2.2638043322471315e-06, "loss": 0.6456, "mean_token_accuracy": 0.7854047417640686, "num_tokens": 222896071.0, "step": 5902 }, { "epoch": 0.7802008987575998, "grad_norm": 0.7498584174571142, "learning_rate": 2.2623619961927385e-06, "loss": 0.5757, "mean_token_accuracy": 0.8062507510185242, "num_tokens": 222961607.0, "step": 5903 }, { "epoch": 0.7803330689928628, "grad_norm": 0.8364563981256186, "learning_rate": 2.2609203493638814e-06, "loss": 0.6047, "mean_token_accuracy": 0.7969112396240234, "num_tokens": 223027143.0, "step": 5904 }, { "epoch": 0.7804652392281258, "grad_norm": 0.8880124121034758, "learning_rate": 2.2594793920674525e-06, "loss": 0.6184, "mean_token_accuracy": 0.7953546643257141, "num_tokens": 223092679.0, "step": 5905 }, { "epoch": 0.7805974094633888, "grad_norm": 0.7443773424737823, "learning_rate": 2.2580391246102045e-06, "loss": 0.5188, "mean_token_accuracy": 0.8277987837791443, "num_tokens": 223158215.0, "step": 5906 }, { "epoch": 0.7807295796986519, "grad_norm": 0.8849017508578443, "learning_rate": 2.2565995472987395e-06, "loss": 0.5535, "mean_token_accuracy": 0.8142015337944031, "num_tokens": 223223751.0, "step": 5907 }, { "epoch": 0.7808617499339149, "grad_norm": 0.8523319439965902, "learning_rate": 2.255160660439515e-06, "loss": 0.6196, "mean_token_accuracy": 0.7953088879585266, "num_tokens": 223289287.0, "step": 5908 }, { "epoch": 0.7809939201691779, "grad_norm": 0.8677828007680182, "learning_rate": 2.2537224643388368e-06, "loss": 0.5954, "mean_token_accuracy": 0.8032291531562805, "num_tokens": 223354823.0, "step": 5909 }, { "epoch": 0.7811260904044409, "grad_norm": 0.890199223632408, "learning_rate": 2.2522849593028687e-06, "loss": 0.608, "mean_token_accuracy": 0.7974453568458557, "num_tokens": 223420359.0, "step": 5910 }, { "epoch": 0.7812582606397039, "grad_norm": 0.8111375862995615, "learning_rate": 2.250848145637626e-06, "loss": 0.5917, "mean_token_accuracy": 0.802405059337616, "num_tokens": 223485895.0, "step": 5911 }, { "epoch": 0.781390430874967, "grad_norm": 0.7644833077947503, "learning_rate": 2.2494120236489747e-06, "loss": 0.5175, "mean_token_accuracy": 0.8289281129837036, "num_tokens": 223551431.0, "step": 5912 }, { "epoch": 0.78152260111023, "grad_norm": 0.8922897735529349, "learning_rate": 2.247976593642638e-06, "loss": 0.5883, "mean_token_accuracy": 0.8049536347389221, "num_tokens": 223616967.0, "step": 5913 }, { "epoch": 0.781654771345493, "grad_norm": 0.8404271629631631, "learning_rate": 2.2465418559241853e-06, "loss": 0.5855, "mean_token_accuracy": 0.8037937879562378, "num_tokens": 223682503.0, "step": 5914 }, { "epoch": 0.781786941580756, "grad_norm": 0.8040215774584553, "learning_rate": 2.2451078107990448e-06, "loss": 0.5352, "mean_token_accuracy": 0.8208094239234924, "num_tokens": 223748039.0, "step": 5915 }, { "epoch": 0.781919111816019, "grad_norm": 0.8496059014407997, "learning_rate": 2.243674458572494e-06, "loss": 0.6167, "mean_token_accuracy": 0.7965449690818787, "num_tokens": 223813575.0, "step": 5916 }, { "epoch": 0.782051282051282, "grad_norm": 0.8719446090717001, "learning_rate": 2.2422417995496655e-06, "loss": 0.607, "mean_token_accuracy": 0.79891037940979, "num_tokens": 223879111.0, "step": 5917 }, { "epoch": 0.7821834522865451, "grad_norm": 0.7811207673618383, "learning_rate": 2.2408098340355374e-06, "loss": 0.5057, "mean_token_accuracy": 0.832285463809967, "num_tokens": 223944647.0, "step": 5918 }, { "epoch": 0.7823156225218081, "grad_norm": 0.7544888909839704, "learning_rate": 2.2393785623349516e-06, "loss": 0.5453, "mean_token_accuracy": 0.8157734274864197, "num_tokens": 224010183.0, "step": 5919 }, { "epoch": 0.7824477927570711, "grad_norm": 0.7287826547077888, "learning_rate": 2.2379479847525915e-06, "loss": 0.4741, "mean_token_accuracy": 0.8427389860153198, "num_tokens": 224075719.0, "step": 5920 }, { "epoch": 0.7825799629923341, "grad_norm": 0.8969585899492982, "learning_rate": 2.236518101592999e-06, "loss": 0.6015, "mean_token_accuracy": 0.8010163307189941, "num_tokens": 224141255.0, "step": 5921 }, { "epoch": 0.7827121332275971, "grad_norm": 0.8367563575186749, "learning_rate": 2.2350889131605656e-06, "loss": 0.6157, "mean_token_accuracy": 0.7939506769180298, "num_tokens": 224206791.0, "step": 5922 }, { "epoch": 0.7828443034628602, "grad_norm": 0.7643028185447238, "learning_rate": 2.233660419759538e-06, "loss": 0.531, "mean_token_accuracy": 0.8235868811607361, "num_tokens": 224272327.0, "step": 5923 }, { "epoch": 0.7829764736981232, "grad_norm": 0.8262968479228924, "learning_rate": 2.2322326216940093e-06, "loss": 0.5711, "mean_token_accuracy": 0.8092876076698303, "num_tokens": 224337863.0, "step": 5924 }, { "epoch": 0.7831086439333862, "grad_norm": 0.7966824208351965, "learning_rate": 2.230805519267929e-06, "loss": 0.583, "mean_token_accuracy": 0.8062202334403992, "num_tokens": 224403399.0, "step": 5925 }, { "epoch": 0.7832408141686492, "grad_norm": 0.7923747235177361, "learning_rate": 2.2293791127850986e-06, "loss": 0.5526, "mean_token_accuracy": 0.8180014491081238, "num_tokens": 224468935.0, "step": 5926 }, { "epoch": 0.7833729844039122, "grad_norm": 0.8270194546736025, "learning_rate": 2.22795340254917e-06, "loss": 0.5724, "mean_token_accuracy": 0.8096386194229126, "num_tokens": 224534471.0, "step": 5927 }, { "epoch": 0.7835051546391752, "grad_norm": 0.8026345141600045, "learning_rate": 2.2265283888636456e-06, "loss": 0.5905, "mean_token_accuracy": 0.8045110702514648, "num_tokens": 224600007.0, "step": 5928 }, { "epoch": 0.7836373248744383, "grad_norm": 0.7785863383787209, "learning_rate": 2.2251040720318824e-06, "loss": 0.5136, "mean_token_accuracy": 0.8290044069290161, "num_tokens": 224665543.0, "step": 5929 }, { "epoch": 0.7837694951097013, "grad_norm": 0.8086249422404859, "learning_rate": 2.2236804523570864e-06, "loss": 0.5808, "mean_token_accuracy": 0.8051367402076721, "num_tokens": 224731079.0, "step": 5930 }, { "epoch": 0.7839016653449643, "grad_norm": 0.8338368486865412, "learning_rate": 2.2222575301423182e-06, "loss": 0.5744, "mean_token_accuracy": 0.808417797088623, "num_tokens": 224796615.0, "step": 5931 }, { "epoch": 0.7840338355802273, "grad_norm": 0.8047613037012602, "learning_rate": 2.2208353056904886e-06, "loss": 0.5826, "mean_token_accuracy": 0.8086161613464355, "num_tokens": 224862151.0, "step": 5932 }, { "epoch": 0.7841660058154903, "grad_norm": 0.8230806838950271, "learning_rate": 2.2194137793043576e-06, "loss": 0.5901, "mean_token_accuracy": 0.8018251657485962, "num_tokens": 224927687.0, "step": 5933 }, { "epoch": 0.7842981760507534, "grad_norm": 0.7450510099158676, "learning_rate": 2.2179929512865383e-06, "loss": 0.4962, "mean_token_accuracy": 0.8327890634536743, "num_tokens": 224993223.0, "step": 5934 }, { "epoch": 0.7844303462860164, "grad_norm": 0.8460213929656445, "learning_rate": 2.216572821939497e-06, "loss": 0.5943, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 225058759.0, "step": 5935 }, { "epoch": 0.7845625165212794, "grad_norm": 0.7810512014406964, "learning_rate": 2.21515339156555e-06, "loss": 0.5436, "mean_token_accuracy": 0.817513108253479, "num_tokens": 225124295.0, "step": 5936 }, { "epoch": 0.7846946867565424, "grad_norm": 0.8552005543170128, "learning_rate": 2.2137346604668617e-06, "loss": 0.5745, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 225189831.0, "step": 5937 }, { "epoch": 0.7848268569918054, "grad_norm": 0.8025471955830058, "learning_rate": 2.2123166289454524e-06, "loss": 0.543, "mean_token_accuracy": 0.8187797665596008, "num_tokens": 225255367.0, "step": 5938 }, { "epoch": 0.7849590272270685, "grad_norm": 0.8268337305645901, "learning_rate": 2.21089929730319e-06, "loss": 0.5568, "mean_token_accuracy": 0.8154529333114624, "num_tokens": 225320903.0, "step": 5939 }, { "epoch": 0.7850911974623315, "grad_norm": 0.7980034048492956, "learning_rate": 2.2094826658417977e-06, "loss": 0.5798, "mean_token_accuracy": 0.807792067527771, "num_tokens": 225386439.0, "step": 5940 }, { "epoch": 0.7852233676975945, "grad_norm": 0.8315401279203927, "learning_rate": 2.2080667348628427e-06, "loss": 0.5767, "mean_token_accuracy": 0.8091350197792053, "num_tokens": 225451975.0, "step": 5941 }, { "epoch": 0.7853555379328575, "grad_norm": 0.7364519030315604, "learning_rate": 2.206651504667752e-06, "loss": 0.4999, "mean_token_accuracy": 0.832468569278717, "num_tokens": 225517511.0, "step": 5942 }, { "epoch": 0.7854877081681205, "grad_norm": 0.7881012752351284, "learning_rate": 2.2052369755577933e-06, "loss": 0.5588, "mean_token_accuracy": 0.8137131929397583, "num_tokens": 225583047.0, "step": 5943 }, { "epoch": 0.7856198784033835, "grad_norm": 0.728162763799788, "learning_rate": 2.203823147834093e-06, "loss": 0.5294, "mean_token_accuracy": 0.8254486918449402, "num_tokens": 225648583.0, "step": 5944 }, { "epoch": 0.7857520486386466, "grad_norm": 0.7974194689183638, "learning_rate": 2.2024100217976256e-06, "loss": 0.5901, "mean_token_accuracy": 0.8053656220436096, "num_tokens": 225714119.0, "step": 5945 }, { "epoch": 0.7858842188739096, "grad_norm": 0.8396291688953998, "learning_rate": 2.200997597749217e-06, "loss": 0.5917, "mean_token_accuracy": 0.8048162460327148, "num_tokens": 225779655.0, "step": 5946 }, { "epoch": 0.7860163891091726, "grad_norm": 0.7649930765815469, "learning_rate": 2.1995858759895405e-06, "loss": 0.5043, "mean_token_accuracy": 0.831720769405365, "num_tokens": 225845191.0, "step": 5947 }, { "epoch": 0.7861485593444356, "grad_norm": 0.8346491091576103, "learning_rate": 2.1981748568191235e-06, "loss": 0.6264, "mean_token_accuracy": 0.7929587364196777, "num_tokens": 225910727.0, "step": 5948 }, { "epoch": 0.7862807295796986, "grad_norm": 0.7310774040206619, "learning_rate": 2.196764540538342e-06, "loss": 0.5697, "mean_token_accuracy": 0.8097454309463501, "num_tokens": 225976263.0, "step": 5949 }, { "epoch": 0.7864128998149617, "grad_norm": 0.8014606216496604, "learning_rate": 2.1953549274474234e-06, "loss": 0.5547, "mean_token_accuracy": 0.8158192038536072, "num_tokens": 226041799.0, "step": 5950 }, { "epoch": 0.7865450700502247, "grad_norm": 0.7737376000698354, "learning_rate": 2.193946017846446e-06, "loss": 0.5625, "mean_token_accuracy": 0.8124618530273438, "num_tokens": 226107335.0, "step": 5951 }, { "epoch": 0.7866772402854877, "grad_norm": 0.7829456285557276, "learning_rate": 2.1925378120353345e-06, "loss": 0.5477, "mean_token_accuracy": 0.8210994005203247, "num_tokens": 226172871.0, "step": 5952 }, { "epoch": 0.7868094105207507, "grad_norm": 0.8241466668986555, "learning_rate": 2.19113031031387e-06, "loss": 0.5821, "mean_token_accuracy": 0.806617021560669, "num_tokens": 226238407.0, "step": 5953 }, { "epoch": 0.7869415807560137, "grad_norm": 0.7964822298471683, "learning_rate": 2.1897235129816745e-06, "loss": 0.5787, "mean_token_accuracy": 0.8056098222732544, "num_tokens": 226303943.0, "step": 5954 }, { "epoch": 0.7870737509912767, "grad_norm": 0.8332463321247401, "learning_rate": 2.188317420338233e-06, "loss": 0.5903, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 226369479.0, "step": 5955 }, { "epoch": 0.7872059212265398, "grad_norm": 0.8109557367960442, "learning_rate": 2.186912032682868e-06, "loss": 0.5936, "mean_token_accuracy": 0.8017946481704712, "num_tokens": 226435015.0, "step": 5956 }, { "epoch": 0.7873380914618028, "grad_norm": 0.8398182496367605, "learning_rate": 2.1855073503147594e-06, "loss": 0.6166, "mean_token_accuracy": 0.7945763468742371, "num_tokens": 226500551.0, "step": 5957 }, { "epoch": 0.7874702616970658, "grad_norm": 0.8072572591977348, "learning_rate": 2.1841033735329335e-06, "loss": 0.5174, "mean_token_accuracy": 0.8277682662010193, "num_tokens": 226566087.0, "step": 5958 }, { "epoch": 0.7876024319323288, "grad_norm": 0.8378058326521499, "learning_rate": 2.1827001026362704e-06, "loss": 0.5604, "mean_token_accuracy": 0.8149035573005676, "num_tokens": 226631623.0, "step": 5959 }, { "epoch": 0.7877346021675918, "grad_norm": 0.9434762358055526, "learning_rate": 2.1812975379234925e-06, "loss": 0.6244, "mean_token_accuracy": 0.79361492395401, "num_tokens": 226697159.0, "step": 5960 }, { "epoch": 0.7878667724028549, "grad_norm": 0.8457670923910195, "learning_rate": 2.1798956796931815e-06, "loss": 0.626, "mean_token_accuracy": 0.7919667959213257, "num_tokens": 226762695.0, "step": 5961 }, { "epoch": 0.7879989426381179, "grad_norm": 0.8458099103500193, "learning_rate": 2.1784945282437604e-06, "loss": 0.6066, "mean_token_accuracy": 0.7965754866600037, "num_tokens": 226828231.0, "step": 5962 }, { "epoch": 0.7881311128733809, "grad_norm": 0.8318716955619406, "learning_rate": 2.1770940838735062e-06, "loss": 0.6035, "mean_token_accuracy": 0.7985746264457703, "num_tokens": 226893767.0, "step": 5963 }, { "epoch": 0.7882632831086439, "grad_norm": 0.8287459553112765, "learning_rate": 2.1756943468805448e-06, "loss": 0.6274, "mean_token_accuracy": 0.7929892539978027, "num_tokens": 226959303.0, "step": 5964 }, { "epoch": 0.7883954533439069, "grad_norm": 0.7969680668174888, "learning_rate": 2.174295317562851e-06, "loss": 0.5284, "mean_token_accuracy": 0.825998067855835, "num_tokens": 227024839.0, "step": 5965 }, { "epoch": 0.78852762357917, "grad_norm": 0.7794256564385443, "learning_rate": 2.1728969962182475e-06, "loss": 0.5547, "mean_token_accuracy": 0.8140794634819031, "num_tokens": 227090375.0, "step": 5966 }, { "epoch": 0.788659793814433, "grad_norm": 0.8554798797045143, "learning_rate": 2.171499383144408e-06, "loss": 0.648, "mean_token_accuracy": 0.7864729762077332, "num_tokens": 227155911.0, "step": 5967 }, { "epoch": 0.788791964049696, "grad_norm": 0.8171976730527969, "learning_rate": 2.170102478638856e-06, "loss": 0.5701, "mean_token_accuracy": 0.812034547328949, "num_tokens": 227221447.0, "step": 5968 }, { "epoch": 0.788924134284959, "grad_norm": 0.8291151609190912, "learning_rate": 2.1687062829989646e-06, "loss": 0.6134, "mean_token_accuracy": 0.7952631115913391, "num_tokens": 227286983.0, "step": 5969 }, { "epoch": 0.789056304520222, "grad_norm": 0.7644842848896504, "learning_rate": 2.167310796521952e-06, "loss": 0.5279, "mean_token_accuracy": 0.8240447044372559, "num_tokens": 227352519.0, "step": 5970 }, { "epoch": 0.789188474755485, "grad_norm": 0.9261617619090293, "learning_rate": 2.165916019504889e-06, "loss": 0.6154, "mean_token_accuracy": 0.7987425327301025, "num_tokens": 227418055.0, "step": 5971 }, { "epoch": 0.789320644990748, "grad_norm": 0.8064999372010735, "learning_rate": 2.1645219522446966e-06, "loss": 0.5766, "mean_token_accuracy": 0.8093487024307251, "num_tokens": 227483591.0, "step": 5972 }, { "epoch": 0.7894528152260111, "grad_norm": 0.7559254815312413, "learning_rate": 2.1631285950381374e-06, "loss": 0.5185, "mean_token_accuracy": 0.8276461958885193, "num_tokens": 227549127.0, "step": 5973 }, { "epoch": 0.7895849854612741, "grad_norm": 0.8104738891138218, "learning_rate": 2.1617359481818338e-06, "loss": 0.5648, "mean_token_accuracy": 0.808784008026123, "num_tokens": 227614663.0, "step": 5974 }, { "epoch": 0.7897171556965371, "grad_norm": 0.7955439347602683, "learning_rate": 2.160344011972249e-06, "loss": 0.5944, "mean_token_accuracy": 0.802191436290741, "num_tokens": 227680199.0, "step": 5975 }, { "epoch": 0.7898493259318001, "grad_norm": 0.852803188736244, "learning_rate": 2.158952786705696e-06, "loss": 0.5472, "mean_token_accuracy": 0.8168569207191467, "num_tokens": 227745735.0, "step": 5976 }, { "epoch": 0.7899814961670631, "grad_norm": 0.7751692130540527, "learning_rate": 2.157562272678339e-06, "loss": 0.5636, "mean_token_accuracy": 0.812187135219574, "num_tokens": 227811271.0, "step": 5977 }, { "epoch": 0.7901136664023262, "grad_norm": 0.8093378936119868, "learning_rate": 2.15617247018619e-06, "loss": 0.5992, "mean_token_accuracy": 0.8001770377159119, "num_tokens": 227876807.0, "step": 5978 }, { "epoch": 0.7902458366375892, "grad_norm": 0.7829275120283045, "learning_rate": 2.154783379525106e-06, "loss": 0.5214, "mean_token_accuracy": 0.8300726413726807, "num_tokens": 227942343.0, "step": 5979 }, { "epoch": 0.7903780068728522, "grad_norm": 0.8070753732584285, "learning_rate": 2.1533950009907995e-06, "loss": 0.5741, "mean_token_accuracy": 0.8083414435386658, "num_tokens": 228007879.0, "step": 5980 }, { "epoch": 0.7905101771081152, "grad_norm": 0.7963602050500188, "learning_rate": 2.152007334878824e-06, "loss": 0.5452, "mean_token_accuracy": 0.8188408017158508, "num_tokens": 228073415.0, "step": 5981 }, { "epoch": 0.7906423473433782, "grad_norm": 0.7718455222566228, "learning_rate": 2.150620381484586e-06, "loss": 0.5704, "mean_token_accuracy": 0.8071206212043762, "num_tokens": 228138951.0, "step": 5982 }, { "epoch": 0.7907745175786413, "grad_norm": 0.8674087932076558, "learning_rate": 2.149234141103339e-06, "loss": 0.6078, "mean_token_accuracy": 0.7982847094535828, "num_tokens": 228204487.0, "step": 5983 }, { "epoch": 0.7909066878139043, "grad_norm": 0.8261101746947673, "learning_rate": 2.1478486140301852e-06, "loss": 0.5938, "mean_token_accuracy": 0.8031986355781555, "num_tokens": 228270023.0, "step": 5984 }, { "epoch": 0.7910388580491673, "grad_norm": 0.8412501682067658, "learning_rate": 2.1464638005600723e-06, "loss": 0.5865, "mean_token_accuracy": 0.8038700819015503, "num_tokens": 228335559.0, "step": 5985 }, { "epoch": 0.7911710282844303, "grad_norm": 0.8091645246399419, "learning_rate": 2.1450797009878e-06, "loss": 0.5682, "mean_token_accuracy": 0.8106915950775146, "num_tokens": 228401095.0, "step": 5986 }, { "epoch": 0.7913031985196933, "grad_norm": 0.799582554052052, "learning_rate": 2.143696315608013e-06, "loss": 0.5279, "mean_token_accuracy": 0.8237547278404236, "num_tokens": 228466631.0, "step": 5987 }, { "epoch": 0.7914353687549563, "grad_norm": 0.8258224615719042, "learning_rate": 2.142313644715208e-06, "loss": 0.5402, "mean_token_accuracy": 0.8223659992218018, "num_tokens": 228532167.0, "step": 5988 }, { "epoch": 0.7915675389902194, "grad_norm": 0.8395139055519877, "learning_rate": 2.140931688603722e-06, "loss": 0.5662, "mean_token_accuracy": 0.8117293119430542, "num_tokens": 228597703.0, "step": 5989 }, { "epoch": 0.7916997092254824, "grad_norm": 0.7539171375461484, "learning_rate": 2.1395504475677486e-06, "loss": 0.5618, "mean_token_accuracy": 0.8126907348632812, "num_tokens": 228663239.0, "step": 5990 }, { "epoch": 0.7918318794607454, "grad_norm": 0.7768070089335825, "learning_rate": 2.1381699219013245e-06, "loss": 0.5711, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 228728775.0, "step": 5991 }, { "epoch": 0.7919640496960084, "grad_norm": 0.8685202269265823, "learning_rate": 2.1367901118983316e-06, "loss": 0.5754, "mean_token_accuracy": 0.8093944787979126, "num_tokens": 228794311.0, "step": 5992 }, { "epoch": 0.7920962199312714, "grad_norm": 0.8576051701343727, "learning_rate": 2.1354110178525073e-06, "loss": 0.6022, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 228859847.0, "step": 5993 }, { "epoch": 0.7922283901665345, "grad_norm": 0.8077455011685708, "learning_rate": 2.1340326400574285e-06, "loss": 0.5698, "mean_token_accuracy": 0.8100048899650574, "num_tokens": 228925383.0, "step": 5994 }, { "epoch": 0.7923605604017975, "grad_norm": 0.8301899000712544, "learning_rate": 2.1326549788065247e-06, "loss": 0.6033, "mean_token_accuracy": 0.7976285219192505, "num_tokens": 228990919.0, "step": 5995 }, { "epoch": 0.7924927306370605, "grad_norm": 0.8689045097687187, "learning_rate": 2.1312780343930707e-06, "loss": 0.5754, "mean_token_accuracy": 0.807761549949646, "num_tokens": 229056455.0, "step": 5996 }, { "epoch": 0.7926249008723235, "grad_norm": 0.8055178730583333, "learning_rate": 2.1299018071101905e-06, "loss": 0.5722, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 229121991.0, "step": 5997 }, { "epoch": 0.7927570711075865, "grad_norm": 0.8058891088622205, "learning_rate": 2.128526297250852e-06, "loss": 0.58, "mean_token_accuracy": 0.8056098222732544, "num_tokens": 229187527.0, "step": 5998 }, { "epoch": 0.7928892413428495, "grad_norm": 0.796561227710383, "learning_rate": 2.1271515051078733e-06, "loss": 0.5354, "mean_token_accuracy": 0.8227627873420715, "num_tokens": 229253063.0, "step": 5999 }, { "epoch": 0.7930214115781126, "grad_norm": 0.7325497977631084, "learning_rate": 2.1257774309739186e-06, "loss": 0.5095, "mean_token_accuracy": 0.8318734169006348, "num_tokens": 229318599.0, "step": 6000 }, { "epoch": 0.7931535818133756, "grad_norm": 0.8002163786555261, "learning_rate": 2.124404075141502e-06, "loss": 0.5913, "mean_token_accuracy": 0.8034886121749878, "num_tokens": 229384135.0, "step": 6001 }, { "epoch": 0.7932857520486386, "grad_norm": 0.8206502498414319, "learning_rate": 2.123031437902978e-06, "loss": 0.6181, "mean_token_accuracy": 0.7950646877288818, "num_tokens": 229449671.0, "step": 6002 }, { "epoch": 0.7934179222839016, "grad_norm": 0.8067082303774457, "learning_rate": 2.121659519550557e-06, "loss": 0.6172, "mean_token_accuracy": 0.7938743829727173, "num_tokens": 229515207.0, "step": 6003 }, { "epoch": 0.7935500925191646, "grad_norm": 0.8767625785102785, "learning_rate": 2.1202883203762887e-06, "loss": 0.6347, "mean_token_accuracy": 0.7888994216918945, "num_tokens": 229580743.0, "step": 6004 }, { "epoch": 0.7936822627544277, "grad_norm": 0.8269314538753886, "learning_rate": 2.1189178406720733e-06, "loss": 0.5862, "mean_token_accuracy": 0.8046941757202148, "num_tokens": 229646279.0, "step": 6005 }, { "epoch": 0.7938144329896907, "grad_norm": 0.9072312292927354, "learning_rate": 2.1175480807296583e-06, "loss": 0.6421, "mean_token_accuracy": 0.7838481068611145, "num_tokens": 229711815.0, "step": 6006 }, { "epoch": 0.7939466032249537, "grad_norm": 0.8212565873302002, "learning_rate": 2.116179040840637e-06, "loss": 0.5845, "mean_token_accuracy": 0.8052435517311096, "num_tokens": 229777351.0, "step": 6007 }, { "epoch": 0.7940787734602167, "grad_norm": 0.7527103611779946, "learning_rate": 2.114810721296448e-06, "loss": 0.5058, "mean_token_accuracy": 0.8306677937507629, "num_tokens": 229842887.0, "step": 6008 }, { "epoch": 0.7942109436954797, "grad_norm": 0.8544358381653269, "learning_rate": 2.1134431223883785e-06, "loss": 0.5781, "mean_token_accuracy": 0.806464433670044, "num_tokens": 229908423.0, "step": 6009 }, { "epoch": 0.7943431139307428, "grad_norm": 0.8009510453511501, "learning_rate": 2.1120762444075633e-06, "loss": 0.5437, "mean_token_accuracy": 0.817604660987854, "num_tokens": 229973959.0, "step": 6010 }, { "epoch": 0.7944752841660058, "grad_norm": 0.8769150868892328, "learning_rate": 2.1107100876449786e-06, "loss": 0.6072, "mean_token_accuracy": 0.7984067797660828, "num_tokens": 230039495.0, "step": 6011 }, { "epoch": 0.7946074544012688, "grad_norm": 0.7989295186123124, "learning_rate": 2.1093446523914555e-06, "loss": 0.5888, "mean_token_accuracy": 0.8033817410469055, "num_tokens": 230105031.0, "step": 6012 }, { "epoch": 0.7947396246365318, "grad_norm": 0.7522885912051707, "learning_rate": 2.1079799389376628e-06, "loss": 0.5449, "mean_token_accuracy": 0.8213740587234497, "num_tokens": 230170567.0, "step": 6013 }, { "epoch": 0.7948717948717948, "grad_norm": 0.8340904340975517, "learning_rate": 2.1066159475741217e-06, "loss": 0.5493, "mean_token_accuracy": 0.8175283670425415, "num_tokens": 230236103.0, "step": 6014 }, { "epoch": 0.7950039651070578, "grad_norm": 0.8972319039048783, "learning_rate": 2.1052526785911948e-06, "loss": 0.6313, "mean_token_accuracy": 0.7916463017463684, "num_tokens": 230301639.0, "step": 6015 }, { "epoch": 0.7951361353423209, "grad_norm": 0.8471388725877392, "learning_rate": 2.1038901322790974e-06, "loss": 0.6067, "mean_token_accuracy": 0.7990782856941223, "num_tokens": 230367175.0, "step": 6016 }, { "epoch": 0.7952683055775839, "grad_norm": 0.7340417589569026, "learning_rate": 2.1025283089278843e-06, "loss": 0.5499, "mean_token_accuracy": 0.817330002784729, "num_tokens": 230432711.0, "step": 6017 }, { "epoch": 0.7954004758128469, "grad_norm": 0.7882062667930787, "learning_rate": 2.1011672088274592e-06, "loss": 0.5679, "mean_token_accuracy": 0.8111494183540344, "num_tokens": 230498247.0, "step": 6018 }, { "epoch": 0.7955326460481099, "grad_norm": 0.8213407290359599, "learning_rate": 2.0998068322675735e-06, "loss": 0.6052, "mean_token_accuracy": 0.7982389330863953, "num_tokens": 230563783.0, "step": 6019 }, { "epoch": 0.7956648162833729, "grad_norm": 0.7730617144515429, "learning_rate": 2.098447179537824e-06, "loss": 0.558, "mean_token_accuracy": 0.8152393102645874, "num_tokens": 230629319.0, "step": 6020 }, { "epoch": 0.795796986518636, "grad_norm": 0.7976765401540886, "learning_rate": 2.097088250927648e-06, "loss": 0.5947, "mean_token_accuracy": 0.8004364371299744, "num_tokens": 230694855.0, "step": 6021 }, { "epoch": 0.795929156753899, "grad_norm": 0.895615307224629, "learning_rate": 2.0957300467263376e-06, "loss": 0.5933, "mean_token_accuracy": 0.8035954236984253, "num_tokens": 230760391.0, "step": 6022 }, { "epoch": 0.796061326989162, "grad_norm": 0.8090110307629413, "learning_rate": 2.0943725672230233e-06, "loss": 0.5753, "mean_token_accuracy": 0.8082041144371033, "num_tokens": 230825927.0, "step": 6023 }, { "epoch": 0.796193497224425, "grad_norm": 0.8491061487043511, "learning_rate": 2.0930158127066854e-06, "loss": 0.6269, "mean_token_accuracy": 0.7928366661071777, "num_tokens": 230891463.0, "step": 6024 }, { "epoch": 0.796325667459688, "grad_norm": 0.848024716944384, "learning_rate": 2.091659783466149e-06, "loss": 0.6101, "mean_token_accuracy": 0.79891037940979, "num_tokens": 230956999.0, "step": 6025 }, { "epoch": 0.796457837694951, "grad_norm": 0.825446764791611, "learning_rate": 2.0903044797900822e-06, "loss": 0.5557, "mean_token_accuracy": 0.812965452671051, "num_tokens": 231022535.0, "step": 6026 }, { "epoch": 0.7965900079302141, "grad_norm": 0.8009603755853063, "learning_rate": 2.088949901967003e-06, "loss": 0.5681, "mean_token_accuracy": 0.8149951100349426, "num_tokens": 231088071.0, "step": 6027 }, { "epoch": 0.7967221781654771, "grad_norm": 0.7968235342131914, "learning_rate": 2.0875960502852714e-06, "loss": 0.5891, "mean_token_accuracy": 0.8032901883125305, "num_tokens": 231153607.0, "step": 6028 }, { "epoch": 0.7968543484007401, "grad_norm": 0.8223766789759752, "learning_rate": 2.0862429250330967e-06, "loss": 0.5816, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 231219143.0, "step": 6029 }, { "epoch": 0.7969865186360032, "grad_norm": 0.8240913131710594, "learning_rate": 2.0848905264985275e-06, "loss": 0.5966, "mean_token_accuracy": 0.8012300133705139, "num_tokens": 231284679.0, "step": 6030 }, { "epoch": 0.7971186888712662, "grad_norm": 0.7813001969954291, "learning_rate": 2.083538854969463e-06, "loss": 0.57, "mean_token_accuracy": 0.8095775842666626, "num_tokens": 231350215.0, "step": 6031 }, { "epoch": 0.7972508591065293, "grad_norm": 0.7585228820093167, "learning_rate": 2.082187910733646e-06, "loss": 0.5554, "mean_token_accuracy": 0.816429615020752, "num_tokens": 231415751.0, "step": 6032 }, { "epoch": 0.7973830293417923, "grad_norm": 0.8266935752975656, "learning_rate": 2.080837694078665e-06, "loss": 0.5711, "mean_token_accuracy": 0.807635486125946, "num_tokens": 231479357.0, "step": 6033 }, { "epoch": 0.7975151995770553, "grad_norm": 0.8604290137365169, "learning_rate": 2.079488205291949e-06, "loss": 0.6018, "mean_token_accuracy": 0.8034427762031555, "num_tokens": 231544893.0, "step": 6034 }, { "epoch": 0.7976473698123183, "grad_norm": 0.8120143128404913, "learning_rate": 2.0781394446607823e-06, "loss": 0.5721, "mean_token_accuracy": 0.8095623254776001, "num_tokens": 231610429.0, "step": 6035 }, { "epoch": 0.7977795400475813, "grad_norm": 0.7670689671367211, "learning_rate": 2.0767914124722826e-06, "loss": 0.5665, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 231675965.0, "step": 6036 }, { "epoch": 0.7979117102828444, "grad_norm": 0.7910658128585314, "learning_rate": 2.0754441090134203e-06, "loss": 0.5259, "mean_token_accuracy": 0.8266542553901672, "num_tokens": 231741501.0, "step": 6037 }, { "epoch": 0.7980438805181074, "grad_norm": 0.8332595927985675, "learning_rate": 2.0740975345710073e-06, "loss": 0.5602, "mean_token_accuracy": 0.8151629567146301, "num_tokens": 231807037.0, "step": 6038 }, { "epoch": 0.7981760507533704, "grad_norm": 0.8514422166586996, "learning_rate": 2.072751689431703e-06, "loss": 0.5874, "mean_token_accuracy": 0.8047704696655273, "num_tokens": 231872573.0, "step": 6039 }, { "epoch": 0.7983082209886334, "grad_norm": 0.7686036278196747, "learning_rate": 2.071406573882005e-06, "loss": 0.5525, "mean_token_accuracy": 0.8154529333114624, "num_tokens": 231938109.0, "step": 6040 }, { "epoch": 0.7984403912238964, "grad_norm": 0.8656403479061581, "learning_rate": 2.0700621882082667e-06, "loss": 0.5565, "mean_token_accuracy": 0.8159870505332947, "num_tokens": 232003645.0, "step": 6041 }, { "epoch": 0.7985725614591594, "grad_norm": 0.8866739332631685, "learning_rate": 2.0687185326966754e-06, "loss": 0.6385, "mean_token_accuracy": 0.7869918346405029, "num_tokens": 232069181.0, "step": 6042 }, { "epoch": 0.7987047316944225, "grad_norm": 0.820155471995291, "learning_rate": 2.06737560763327e-06, "loss": 0.6041, "mean_token_accuracy": 0.7995818853378296, "num_tokens": 232134717.0, "step": 6043 }, { "epoch": 0.7988369019296855, "grad_norm": 0.8195360033217516, "learning_rate": 2.066033413303928e-06, "loss": 0.5941, "mean_token_accuracy": 0.802160918712616, "num_tokens": 232200253.0, "step": 6044 }, { "epoch": 0.7989690721649485, "grad_norm": 0.875534767444935, "learning_rate": 2.064691949994376e-06, "loss": 0.6512, "mean_token_accuracy": 0.7823678255081177, "num_tokens": 232265789.0, "step": 6045 }, { "epoch": 0.7991012424002115, "grad_norm": 0.8020241389088473, "learning_rate": 2.0633512179901836e-06, "loss": 0.569, "mean_token_accuracy": 0.8120498061180115, "num_tokens": 232331325.0, "step": 6046 }, { "epoch": 0.7992334126354745, "grad_norm": 0.8166631630034681, "learning_rate": 2.0620112175767653e-06, "loss": 0.6072, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 232396861.0, "step": 6047 }, { "epoch": 0.7993655828707376, "grad_norm": 0.804960780513725, "learning_rate": 2.06067194903938e-06, "loss": 0.5696, "mean_token_accuracy": 0.8152545690536499, "num_tokens": 232462397.0, "step": 6048 }, { "epoch": 0.7994977531060006, "grad_norm": 0.7056184155482459, "learning_rate": 2.0593334126631273e-06, "loss": 0.4912, "mean_token_accuracy": 0.8352001905441284, "num_tokens": 232527933.0, "step": 6049 }, { "epoch": 0.7996299233412636, "grad_norm": 0.8201185978177062, "learning_rate": 2.0579956087329556e-06, "loss": 0.5113, "mean_token_accuracy": 0.8297979235649109, "num_tokens": 232593469.0, "step": 6050 }, { "epoch": 0.7997620935765266, "grad_norm": 0.8761650910384156, "learning_rate": 2.056658537533654e-06, "loss": 0.5838, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 232659005.0, "step": 6051 }, { "epoch": 0.7998942638117896, "grad_norm": 0.8530651819943749, "learning_rate": 2.05532219934986e-06, "loss": 0.5908, "mean_token_accuracy": 0.8058081865310669, "num_tokens": 232724541.0, "step": 6052 }, { "epoch": 0.8000264340470526, "grad_norm": 0.8107113925971614, "learning_rate": 2.0539865944660477e-06, "loss": 0.5816, "mean_token_accuracy": 0.8055487871170044, "num_tokens": 232790077.0, "step": 6053 }, { "epoch": 0.8001586042823157, "grad_norm": 0.7837405559498009, "learning_rate": 2.0526517231665438e-06, "loss": 0.5497, "mean_token_accuracy": 0.8145983219146729, "num_tokens": 232855613.0, "step": 6054 }, { "epoch": 0.8002907745175787, "grad_norm": 0.853820119507357, "learning_rate": 2.0513175857355112e-06, "loss": 0.5866, "mean_token_accuracy": 0.8046178817749023, "num_tokens": 232921149.0, "step": 6055 }, { "epoch": 0.8004229447528417, "grad_norm": 0.8798495029397945, "learning_rate": 2.0499841824569614e-06, "loss": 0.5837, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 232986685.0, "step": 6056 }, { "epoch": 0.8005551149881047, "grad_norm": 0.804352843604308, "learning_rate": 2.0486515136147487e-06, "loss": 0.5441, "mean_token_accuracy": 0.8194817304611206, "num_tokens": 233052221.0, "step": 6057 }, { "epoch": 0.8006872852233677, "grad_norm": 0.8719479345834782, "learning_rate": 2.0473195794925703e-06, "loss": 0.5738, "mean_token_accuracy": 0.8080515265464783, "num_tokens": 233117757.0, "step": 6058 }, { "epoch": 0.8008194554586308, "grad_norm": 0.8039199916414874, "learning_rate": 2.045988380373966e-06, "loss": 0.5666, "mean_token_accuracy": 0.8117751479148865, "num_tokens": 233183293.0, "step": 6059 }, { "epoch": 0.8009516256938938, "grad_norm": 0.752382072030485, "learning_rate": 2.044657916542321e-06, "loss": 0.5222, "mean_token_accuracy": 0.8250671625137329, "num_tokens": 233248829.0, "step": 6060 }, { "epoch": 0.8010837959291568, "grad_norm": 0.8210640937822408, "learning_rate": 2.0433281882808627e-06, "loss": 0.5713, "mean_token_accuracy": 0.8092723488807678, "num_tokens": 233314365.0, "step": 6061 }, { "epoch": 0.8012159661644198, "grad_norm": 0.8289777577676233, "learning_rate": 2.0419991958726644e-06, "loss": 0.6211, "mean_token_accuracy": 0.7930808067321777, "num_tokens": 233379901.0, "step": 6062 }, { "epoch": 0.8013481363996828, "grad_norm": 0.7553435549960986, "learning_rate": 2.040670939600638e-06, "loss": 0.495, "mean_token_accuracy": 0.8333994746208191, "num_tokens": 233445437.0, "step": 6063 }, { "epoch": 0.8014803066349458, "grad_norm": 0.795769427965809, "learning_rate": 2.039343419747543e-06, "loss": 0.592, "mean_token_accuracy": 0.8038090467453003, "num_tokens": 233510973.0, "step": 6064 }, { "epoch": 0.8016124768702089, "grad_norm": 0.8093275822157887, "learning_rate": 2.038016636595981e-06, "loss": 0.5409, "mean_token_accuracy": 0.8188102841377258, "num_tokens": 233576509.0, "step": 6065 }, { "epoch": 0.8017446471054719, "grad_norm": 0.8213170003949909, "learning_rate": 2.036690590428396e-06, "loss": 0.6118, "mean_token_accuracy": 0.7953699231147766, "num_tokens": 233642045.0, "step": 6066 }, { "epoch": 0.8018768173407349, "grad_norm": 0.7721124735044131, "learning_rate": 2.035365281527076e-06, "loss": 0.5953, "mean_token_accuracy": 0.7980252504348755, "num_tokens": 233707581.0, "step": 6067 }, { "epoch": 0.8020089875759979, "grad_norm": 0.8361486747997623, "learning_rate": 2.0340407101741512e-06, "loss": 0.6099, "mean_token_accuracy": 0.7969264984130859, "num_tokens": 233773117.0, "step": 6068 }, { "epoch": 0.8021411578112609, "grad_norm": 0.7899410628940979, "learning_rate": 2.0327168766515943e-06, "loss": 0.561, "mean_token_accuracy": 0.8107373714447021, "num_tokens": 233838653.0, "step": 6069 }, { "epoch": 0.802273328046524, "grad_norm": 0.8629869336693599, "learning_rate": 2.031393781241223e-06, "loss": 0.5717, "mean_token_accuracy": 0.8105390071868896, "num_tokens": 233904189.0, "step": 6070 }, { "epoch": 0.802405498281787, "grad_norm": 0.8785500613286866, "learning_rate": 2.030071424224698e-06, "loss": 0.614, "mean_token_accuracy": 0.7962397933006287, "num_tokens": 233969725.0, "step": 6071 }, { "epoch": 0.80253766851705, "grad_norm": 0.8495445791267864, "learning_rate": 2.028749805883517e-06, "loss": 0.4934, "mean_token_accuracy": 0.8371230363845825, "num_tokens": 234035261.0, "step": 6072 }, { "epoch": 0.802669838752313, "grad_norm": 0.8072168463738048, "learning_rate": 2.0274289264990303e-06, "loss": 0.6037, "mean_token_accuracy": 0.7999023199081421, "num_tokens": 234100797.0, "step": 6073 }, { "epoch": 0.802802008987576, "grad_norm": 0.8404418183369067, "learning_rate": 2.026108786352422e-06, "loss": 0.5738, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 234166333.0, "step": 6074 }, { "epoch": 0.802934179222839, "grad_norm": 0.7988574608215577, "learning_rate": 2.0247893857247248e-06, "loss": 0.5523, "mean_token_accuracy": 0.8177267909049988, "num_tokens": 234231869.0, "step": 6075 }, { "epoch": 0.8030663494581021, "grad_norm": 0.7813233136225597, "learning_rate": 2.023470724896808e-06, "loss": 0.5944, "mean_token_accuracy": 0.8017336130142212, "num_tokens": 234297405.0, "step": 6076 }, { "epoch": 0.8031985196933651, "grad_norm": 0.8377928175670695, "learning_rate": 2.0221528041493906e-06, "loss": 0.5757, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 234362941.0, "step": 6077 }, { "epoch": 0.8033306899286281, "grad_norm": 0.8280943614438758, "learning_rate": 2.0208356237630284e-06, "loss": 0.5947, "mean_token_accuracy": 0.8017030954360962, "num_tokens": 234428477.0, "step": 6078 }, { "epoch": 0.8034628601638911, "grad_norm": 0.8801412862205351, "learning_rate": 2.0195191840181223e-06, "loss": 0.6267, "mean_token_accuracy": 0.7897234559059143, "num_tokens": 234494013.0, "step": 6079 }, { "epoch": 0.8035950303991541, "grad_norm": 0.7979933378986707, "learning_rate": 2.0182034851949144e-06, "loss": 0.5951, "mean_token_accuracy": 0.802374541759491, "num_tokens": 234559549.0, "step": 6080 }, { "epoch": 0.8037272006344172, "grad_norm": 0.7711034307334363, "learning_rate": 2.0168885275734914e-06, "loss": 0.5597, "mean_token_accuracy": 0.8131943345069885, "num_tokens": 234625085.0, "step": 6081 }, { "epoch": 0.8038593708696802, "grad_norm": 0.8204510873943885, "learning_rate": 2.0155743114337773e-06, "loss": 0.5458, "mean_token_accuracy": 0.8164448738098145, "num_tokens": 234690621.0, "step": 6082 }, { "epoch": 0.8039915411049432, "grad_norm": 0.7978975478617341, "learning_rate": 2.014260837055543e-06, "loss": 0.5635, "mean_token_accuracy": 0.8133012056350708, "num_tokens": 234756157.0, "step": 6083 }, { "epoch": 0.8041237113402062, "grad_norm": 0.827656299841209, "learning_rate": 2.012948104718399e-06, "loss": 0.5544, "mean_token_accuracy": 0.8150714039802551, "num_tokens": 234821693.0, "step": 6084 }, { "epoch": 0.8042558815754692, "grad_norm": 0.8286655148019978, "learning_rate": 2.0116361147017993e-06, "loss": 0.5524, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 234887229.0, "step": 6085 }, { "epoch": 0.8043880518107323, "grad_norm": 0.878173918687324, "learning_rate": 2.0103248672850403e-06, "loss": 0.5884, "mean_token_accuracy": 0.8046941757202148, "num_tokens": 234952765.0, "step": 6086 }, { "epoch": 0.8045202220459953, "grad_norm": 0.8373931943104039, "learning_rate": 2.009014362747257e-06, "loss": 0.5887, "mean_token_accuracy": 0.8041447997093201, "num_tokens": 235018301.0, "step": 6087 }, { "epoch": 0.8046523922812583, "grad_norm": 0.8207741253729809, "learning_rate": 2.007704601367429e-06, "loss": 0.6207, "mean_token_accuracy": 0.7955835461616516, "num_tokens": 235083837.0, "step": 6088 }, { "epoch": 0.8047845625165213, "grad_norm": 0.7615665824723127, "learning_rate": 2.006395583424377e-06, "loss": 0.5188, "mean_token_accuracy": 0.8251129388809204, "num_tokens": 235149373.0, "step": 6089 }, { "epoch": 0.8049167327517843, "grad_norm": 0.8792802722837425, "learning_rate": 2.005087309196766e-06, "loss": 0.596, "mean_token_accuracy": 0.8010469079017639, "num_tokens": 235214909.0, "step": 6090 }, { "epoch": 0.8050489029870473, "grad_norm": 0.7524368285129784, "learning_rate": 2.0037797789630967e-06, "loss": 0.5329, "mean_token_accuracy": 0.8213130235671997, "num_tokens": 235280445.0, "step": 6091 }, { "epoch": 0.8051810732223104, "grad_norm": 0.8711641061944789, "learning_rate": 2.0024729930017166e-06, "loss": 0.5964, "mean_token_accuracy": 0.803000271320343, "num_tokens": 235345981.0, "step": 6092 }, { "epoch": 0.8053132434575734, "grad_norm": 0.7719418148629199, "learning_rate": 2.001166951590814e-06, "loss": 0.569, "mean_token_accuracy": 0.8084330558776855, "num_tokens": 235411517.0, "step": 6093 }, { "epoch": 0.8054454136928364, "grad_norm": 0.7356778724184616, "learning_rate": 1.999861655008418e-06, "loss": 0.5247, "mean_token_accuracy": 0.82639479637146, "num_tokens": 235477053.0, "step": 6094 }, { "epoch": 0.8055775839280994, "grad_norm": 0.7599540259805989, "learning_rate": 1.9985571035323947e-06, "loss": 0.5496, "mean_token_accuracy": 0.8167653679847717, "num_tokens": 235542589.0, "step": 6095 }, { "epoch": 0.8057097541633624, "grad_norm": 0.870766936832467, "learning_rate": 1.9972532974404624e-06, "loss": 0.5849, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 235608125.0, "step": 6096 }, { "epoch": 0.8058419243986255, "grad_norm": 0.8504305991857573, "learning_rate": 1.9959502370101697e-06, "loss": 0.5934, "mean_token_accuracy": 0.79878830909729, "num_tokens": 235673661.0, "step": 6097 }, { "epoch": 0.8059740946338885, "grad_norm": 0.7905962602908564, "learning_rate": 1.9946479225189118e-06, "loss": 0.5535, "mean_token_accuracy": 0.8158192038536072, "num_tokens": 235739197.0, "step": 6098 }, { "epoch": 0.8061062648691515, "grad_norm": 0.8347859213583894, "learning_rate": 1.993346354243925e-06, "loss": 0.5742, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 235804733.0, "step": 6099 }, { "epoch": 0.8062384351044145, "grad_norm": 0.7752897241292848, "learning_rate": 1.9920455324622866e-06, "loss": 0.575, "mean_token_accuracy": 0.8094860315322876, "num_tokens": 235870269.0, "step": 6100 }, { "epoch": 0.8063706053396775, "grad_norm": 0.8951984033115293, "learning_rate": 1.9907454574509132e-06, "loss": 0.5609, "mean_token_accuracy": 0.8159260153770447, "num_tokens": 235935805.0, "step": 6101 }, { "epoch": 0.8065027755749405, "grad_norm": 0.8889209405438123, "learning_rate": 1.989446129486563e-06, "loss": 0.5554, "mean_token_accuracy": 0.8176352381706238, "num_tokens": 236001341.0, "step": 6102 }, { "epoch": 0.8066349458102036, "grad_norm": 0.8018679236096763, "learning_rate": 1.9881475488458375e-06, "loss": 0.582, "mean_token_accuracy": 0.8062965273857117, "num_tokens": 236066877.0, "step": 6103 }, { "epoch": 0.8067671160454666, "grad_norm": 0.8241159517927287, "learning_rate": 1.986849715805178e-06, "loss": 0.5986, "mean_token_accuracy": 0.8005738258361816, "num_tokens": 236132413.0, "step": 6104 }, { "epoch": 0.8068992862807296, "grad_norm": 0.7992892707786651, "learning_rate": 1.9855526306408626e-06, "loss": 0.556, "mean_token_accuracy": 0.816185474395752, "num_tokens": 236197949.0, "step": 6105 }, { "epoch": 0.8070314565159926, "grad_norm": 0.8381635410437727, "learning_rate": 1.9842562936290164e-06, "loss": 0.5913, "mean_token_accuracy": 0.8047857284545898, "num_tokens": 236263485.0, "step": 6106 }, { "epoch": 0.8071636267512556, "grad_norm": 0.8002912239843291, "learning_rate": 1.9829607050456013e-06, "loss": 0.5938, "mean_token_accuracy": 0.7999938726425171, "num_tokens": 236329021.0, "step": 6107 }, { "epoch": 0.8072957969865187, "grad_norm": 0.8779718320012182, "learning_rate": 1.981665865166422e-06, "loss": 0.6157, "mean_token_accuracy": 0.7954767346382141, "num_tokens": 236394557.0, "step": 6108 }, { "epoch": 0.8074279672217817, "grad_norm": 0.8149825062173731, "learning_rate": 1.9803717742671237e-06, "loss": 0.5561, "mean_token_accuracy": 0.8149340748786926, "num_tokens": 236460093.0, "step": 6109 }, { "epoch": 0.8075601374570447, "grad_norm": 0.8163774538941633, "learning_rate": 1.9790784326231888e-06, "loss": 0.5769, "mean_token_accuracy": 0.8062354922294617, "num_tokens": 236525629.0, "step": 6110 }, { "epoch": 0.8076923076923077, "grad_norm": 0.785676301584355, "learning_rate": 1.977785840509944e-06, "loss": 0.5733, "mean_token_accuracy": 0.8085551261901855, "num_tokens": 236591165.0, "step": 6111 }, { "epoch": 0.8078244779275707, "grad_norm": 0.8381568074296788, "learning_rate": 1.9764939982025553e-06, "loss": 0.5996, "mean_token_accuracy": 0.7990782856941223, "num_tokens": 236656701.0, "step": 6112 }, { "epoch": 0.8079566481628337, "grad_norm": 0.8541027808400465, "learning_rate": 1.97520290597603e-06, "loss": 0.5834, "mean_token_accuracy": 0.8055945634841919, "num_tokens": 236722237.0, "step": 6113 }, { "epoch": 0.8080888183980968, "grad_norm": 0.7941095082457298, "learning_rate": 1.9739125641052113e-06, "loss": 0.5619, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 236787773.0, "step": 6114 }, { "epoch": 0.8082209886333598, "grad_norm": 0.8029022485769095, "learning_rate": 1.9726229728647904e-06, "loss": 0.5912, "mean_token_accuracy": 0.803274929523468, "num_tokens": 236853309.0, "step": 6115 }, { "epoch": 0.8083531588686228, "grad_norm": 0.7895562604124795, "learning_rate": 1.9713341325292916e-06, "loss": 0.5352, "mean_token_accuracy": 0.8210994005203247, "num_tokens": 236918845.0, "step": 6116 }, { "epoch": 0.8084853291038858, "grad_norm": 0.8003822438916585, "learning_rate": 1.970046043373082e-06, "loss": 0.6155, "mean_token_accuracy": 0.7920125722885132, "num_tokens": 236984381.0, "step": 6117 }, { "epoch": 0.8086174993391488, "grad_norm": 0.7877219463653282, "learning_rate": 1.9687587056703704e-06, "loss": 0.5608, "mean_token_accuracy": 0.8130417466163635, "num_tokens": 237049917.0, "step": 6118 }, { "epoch": 0.8087496695744119, "grad_norm": 0.7912939836057863, "learning_rate": 1.9674721196952044e-06, "loss": 0.5778, "mean_token_accuracy": 0.808509349822998, "num_tokens": 237115453.0, "step": 6119 }, { "epoch": 0.8088818398096749, "grad_norm": 0.8477503193271869, "learning_rate": 1.966186285721469e-06, "loss": 0.569, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 237180989.0, "step": 6120 }, { "epoch": 0.8090140100449379, "grad_norm": 0.8204313658306193, "learning_rate": 1.9649012040228925e-06, "loss": 0.5464, "mean_token_accuracy": 0.8173757791519165, "num_tokens": 237246525.0, "step": 6121 }, { "epoch": 0.8091461802802009, "grad_norm": 0.8134223079534258, "learning_rate": 1.963616874873042e-06, "loss": 0.5735, "mean_token_accuracy": 0.807761549949646, "num_tokens": 237312061.0, "step": 6122 }, { "epoch": 0.8092783505154639, "grad_norm": 0.8208210236320045, "learning_rate": 1.9623332985453254e-06, "loss": 0.5819, "mean_token_accuracy": 0.8081125617027283, "num_tokens": 237377597.0, "step": 6123 }, { "epoch": 0.809410520750727, "grad_norm": 0.7557178169964012, "learning_rate": 1.9610504753129878e-06, "loss": 0.5364, "mean_token_accuracy": 0.8205958008766174, "num_tokens": 237443133.0, "step": 6124 }, { "epoch": 0.80954269098599, "grad_norm": 0.790236157604522, "learning_rate": 1.9597684054491163e-06, "loss": 0.5579, "mean_token_accuracy": 0.8149035573005676, "num_tokens": 237508669.0, "step": 6125 }, { "epoch": 0.809674861221253, "grad_norm": 0.8017022869040584, "learning_rate": 1.9584870892266356e-06, "loss": 0.6137, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 237574205.0, "step": 6126 }, { "epoch": 0.809807031456516, "grad_norm": 0.7433403365425724, "learning_rate": 1.9572065269183127e-06, "loss": 0.5386, "mean_token_accuracy": 0.8204126358032227, "num_tokens": 237639741.0, "step": 6127 }, { "epoch": 0.809939201691779, "grad_norm": 0.795336040174671, "learning_rate": 1.955926718796753e-06, "loss": 0.5886, "mean_token_accuracy": 0.8057013750076294, "num_tokens": 237705277.0, "step": 6128 }, { "epoch": 0.810071371927042, "grad_norm": 0.7749244634116828, "learning_rate": 1.954647665134399e-06, "loss": 0.5605, "mean_token_accuracy": 0.8142015337944031, "num_tokens": 237770813.0, "step": 6129 }, { "epoch": 0.8102035421623051, "grad_norm": 0.7752726068322672, "learning_rate": 1.953369366203536e-06, "loss": 0.5549, "mean_token_accuracy": 0.811790406703949, "num_tokens": 237836349.0, "step": 6130 }, { "epoch": 0.8103357123975681, "grad_norm": 0.8954593027678532, "learning_rate": 1.9520918222762853e-06, "loss": 0.6612, "mean_token_accuracy": 0.779132604598999, "num_tokens": 237901885.0, "step": 6131 }, { "epoch": 0.8104678826328311, "grad_norm": 0.8165550367336516, "learning_rate": 1.950815033624613e-06, "loss": 0.6034, "mean_token_accuracy": 0.79860520362854, "num_tokens": 237967421.0, "step": 6132 }, { "epoch": 0.8106000528680941, "grad_norm": 0.8080000092173464, "learning_rate": 1.949539000520315e-06, "loss": 0.5942, "mean_token_accuracy": 0.7999175786972046, "num_tokens": 238032957.0, "step": 6133 }, { "epoch": 0.8107322231033571, "grad_norm": 0.7880580963885903, "learning_rate": 1.9482637232350388e-06, "loss": 0.5619, "mean_token_accuracy": 0.8133927583694458, "num_tokens": 238098493.0, "step": 6134 }, { "epoch": 0.8108643933386201, "grad_norm": 0.8002537716493604, "learning_rate": 1.9469892020402592e-06, "loss": 0.5455, "mean_token_accuracy": 0.8202600479125977, "num_tokens": 238164029.0, "step": 6135 }, { "epoch": 0.8109965635738832, "grad_norm": 0.81049992904919, "learning_rate": 1.945715437207299e-06, "loss": 0.5625, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 238229565.0, "step": 6136 }, { "epoch": 0.8111287338091462, "grad_norm": 0.8285060478547461, "learning_rate": 1.9444424290073104e-06, "loss": 0.6126, "mean_token_accuracy": 0.7978726625442505, "num_tokens": 238295101.0, "step": 6137 }, { "epoch": 0.8112609040444092, "grad_norm": 0.8578788953977341, "learning_rate": 1.943170177711297e-06, "loss": 0.643, "mean_token_accuracy": 0.7873581051826477, "num_tokens": 238360637.0, "step": 6138 }, { "epoch": 0.8113930742796722, "grad_norm": 0.8353497253220507, "learning_rate": 1.9418986835900906e-06, "loss": 0.6282, "mean_token_accuracy": 0.7913716435432434, "num_tokens": 238426173.0, "step": 6139 }, { "epoch": 0.8115252445149352, "grad_norm": 0.8406162950063081, "learning_rate": 1.940627946914367e-06, "loss": 0.592, "mean_token_accuracy": 0.8044805526733398, "num_tokens": 238491709.0, "step": 6140 }, { "epoch": 0.8116574147501983, "grad_norm": 0.8383963046469235, "learning_rate": 1.939357967954638e-06, "loss": 0.5819, "mean_token_accuracy": 0.8076547384262085, "num_tokens": 238557245.0, "step": 6141 }, { "epoch": 0.8117895849854613, "grad_norm": 0.8180861331563528, "learning_rate": 1.938088746981259e-06, "loss": 0.517, "mean_token_accuracy": 0.8259217143058777, "num_tokens": 238622781.0, "step": 6142 }, { "epoch": 0.8119217552207243, "grad_norm": 0.8241915596132532, "learning_rate": 1.9368202842644167e-06, "loss": 0.5974, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 238688317.0, "step": 6143 }, { "epoch": 0.8120539254559873, "grad_norm": 0.7867833499679432, "learning_rate": 1.935552580074142e-06, "loss": 0.5426, "mean_token_accuracy": 0.8190849423408508, "num_tokens": 238753853.0, "step": 6144 }, { "epoch": 0.8121860956912503, "grad_norm": 0.789038998664016, "learning_rate": 1.934285634680304e-06, "loss": 0.5831, "mean_token_accuracy": 0.8091197609901428, "num_tokens": 238819389.0, "step": 6145 }, { "epoch": 0.8123182659265133, "grad_norm": 0.8720653972581653, "learning_rate": 1.9330194483526056e-06, "loss": 0.6052, "mean_token_accuracy": 0.7950952053070068, "num_tokens": 238884925.0, "step": 6146 }, { "epoch": 0.8124504361617764, "grad_norm": 0.8117381668439618, "learning_rate": 1.931754021360595e-06, "loss": 0.5468, "mean_token_accuracy": 0.8191918134689331, "num_tokens": 238950461.0, "step": 6147 }, { "epoch": 0.8125826063970394, "grad_norm": 0.7900167955376648, "learning_rate": 1.9304893539736536e-06, "loss": 0.5635, "mean_token_accuracy": 0.8118056654930115, "num_tokens": 239015997.0, "step": 6148 }, { "epoch": 0.8127147766323024, "grad_norm": 0.8209818725347569, "learning_rate": 1.9292254464610018e-06, "loss": 0.6005, "mean_token_accuracy": 0.8004669547080994, "num_tokens": 239081533.0, "step": 6149 }, { "epoch": 0.8128469468675654, "grad_norm": 0.7498498583823145, "learning_rate": 1.9279622990917004e-06, "loss": 0.5117, "mean_token_accuracy": 0.8279056549072266, "num_tokens": 239147069.0, "step": 6150 }, { "epoch": 0.8129791171028284, "grad_norm": 0.8127468797565872, "learning_rate": 1.926699912134648e-06, "loss": 0.556, "mean_token_accuracy": 0.8147509694099426, "num_tokens": 239212605.0, "step": 6151 }, { "epoch": 0.8131112873380915, "grad_norm": 0.8379793310506517, "learning_rate": 1.9254382858585773e-06, "loss": 0.6471, "mean_token_accuracy": 0.7839854955673218, "num_tokens": 239278141.0, "step": 6152 }, { "epoch": 0.8132434575733545, "grad_norm": 0.8105178242060411, "learning_rate": 1.924177420532065e-06, "loss": 0.6158, "mean_token_accuracy": 0.7943322062492371, "num_tokens": 239343677.0, "step": 6153 }, { "epoch": 0.8133756278086175, "grad_norm": 0.8322668119172965, "learning_rate": 1.922917316423522e-06, "loss": 0.5721, "mean_token_accuracy": 0.8083109259605408, "num_tokens": 239409213.0, "step": 6154 }, { "epoch": 0.8135077980438805, "grad_norm": 0.8175585558857564, "learning_rate": 1.921657973801199e-06, "loss": 0.5615, "mean_token_accuracy": 0.8152087926864624, "num_tokens": 239474749.0, "step": 6155 }, { "epoch": 0.8136399682791435, "grad_norm": 0.7574521190841212, "learning_rate": 1.9203993929331806e-06, "loss": 0.5535, "mean_token_accuracy": 0.8153766393661499, "num_tokens": 239540285.0, "step": 6156 }, { "epoch": 0.8137721385144066, "grad_norm": 0.778143433307058, "learning_rate": 1.919141574087397e-06, "loss": 0.5805, "mean_token_accuracy": 0.8090892434120178, "num_tokens": 239605821.0, "step": 6157 }, { "epoch": 0.8139043087496696, "grad_norm": 0.811179233635122, "learning_rate": 1.9178845175316084e-06, "loss": 0.5779, "mean_token_accuracy": 0.8071663975715637, "num_tokens": 239671357.0, "step": 6158 }, { "epoch": 0.8140364789849326, "grad_norm": 0.7607855355769771, "learning_rate": 1.9166282235334173e-06, "loss": 0.5138, "mean_token_accuracy": 0.8308204412460327, "num_tokens": 239736893.0, "step": 6159 }, { "epoch": 0.8141686492201956, "grad_norm": 0.7512037172509917, "learning_rate": 1.9153726923602613e-06, "loss": 0.5333, "mean_token_accuracy": 0.8236936926841736, "num_tokens": 239802429.0, "step": 6160 }, { "epoch": 0.8143008194554586, "grad_norm": 0.7734047559738425, "learning_rate": 1.914117924279419e-06, "loss": 0.5815, "mean_token_accuracy": 0.8062660098075867, "num_tokens": 239867965.0, "step": 6161 }, { "epoch": 0.8144329896907216, "grad_norm": 0.7679592156535701, "learning_rate": 1.9128639195580016e-06, "loss": 0.5663, "mean_token_accuracy": 0.8142626285552979, "num_tokens": 239933501.0, "step": 6162 }, { "epoch": 0.8145651599259847, "grad_norm": 0.8308063172830374, "learning_rate": 1.9116106784629616e-06, "loss": 0.603, "mean_token_accuracy": 0.7960108518600464, "num_tokens": 239999037.0, "step": 6163 }, { "epoch": 0.8146973301612477, "grad_norm": 0.7807698877768672, "learning_rate": 1.910358201261089e-06, "loss": 0.5427, "mean_token_accuracy": 0.8183677196502686, "num_tokens": 240064573.0, "step": 6164 }, { "epoch": 0.8148295003965107, "grad_norm": 0.8259775623478736, "learning_rate": 1.909106488219008e-06, "loss": 0.5566, "mean_token_accuracy": 0.8161396384239197, "num_tokens": 240130109.0, "step": 6165 }, { "epoch": 0.8149616706317737, "grad_norm": 0.757264268508731, "learning_rate": 1.9078555396031833e-06, "loss": 0.5001, "mean_token_accuracy": 0.832590639591217, "num_tokens": 240195645.0, "step": 6166 }, { "epoch": 0.8150938408670367, "grad_norm": 0.8904233721883763, "learning_rate": 1.9066053556799152e-06, "loss": 0.6119, "mean_token_accuracy": 0.8002228140830994, "num_tokens": 240261181.0, "step": 6167 }, { "epoch": 0.8152260111022998, "grad_norm": 0.729249088249973, "learning_rate": 1.905355936715344e-06, "loss": 0.5026, "mean_token_accuracy": 0.8328043222427368, "num_tokens": 240326717.0, "step": 6168 }, { "epoch": 0.8153581813375628, "grad_norm": 0.806761568791428, "learning_rate": 1.9041072829754402e-06, "loss": 0.6013, "mean_token_accuracy": 0.7998870611190796, "num_tokens": 240392253.0, "step": 6169 }, { "epoch": 0.8154903515728258, "grad_norm": 0.7746523729725217, "learning_rate": 1.902859394726022e-06, "loss": 0.5287, "mean_token_accuracy": 0.8241362571716309, "num_tokens": 240457789.0, "step": 6170 }, { "epoch": 0.8156225218080888, "grad_norm": 0.7400084441472237, "learning_rate": 1.9016122722327341e-06, "loss": 0.4973, "mean_token_accuracy": 0.83692467212677, "num_tokens": 240523325.0, "step": 6171 }, { "epoch": 0.8157546920433518, "grad_norm": 0.8495440628083303, "learning_rate": 1.9003659157610648e-06, "loss": 0.5916, "mean_token_accuracy": 0.8019624948501587, "num_tokens": 240588861.0, "step": 6172 }, { "epoch": 0.8158868622786148, "grad_norm": 0.8644948175601406, "learning_rate": 1.8991203255763368e-06, "loss": 0.6113, "mean_token_accuracy": 0.7965297102928162, "num_tokens": 240654397.0, "step": 6173 }, { "epoch": 0.8160190325138779, "grad_norm": 0.7907001009715654, "learning_rate": 1.8978755019437116e-06, "loss": 0.5684, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 240719933.0, "step": 6174 }, { "epoch": 0.8161512027491409, "grad_norm": 0.8023137039810632, "learning_rate": 1.896631445128183e-06, "loss": 0.5652, "mean_token_accuracy": 0.8126144409179688, "num_tokens": 240785469.0, "step": 6175 }, { "epoch": 0.8162833729844039, "grad_norm": 0.7834649496865499, "learning_rate": 1.8953881553945895e-06, "loss": 0.5576, "mean_token_accuracy": 0.8134843111038208, "num_tokens": 240851005.0, "step": 6176 }, { "epoch": 0.8164155432196669, "grad_norm": 0.8302382162320898, "learning_rate": 1.8941456330075975e-06, "loss": 0.5903, "mean_token_accuracy": 0.8016420602798462, "num_tokens": 240916541.0, "step": 6177 }, { "epoch": 0.8165477134549299, "grad_norm": 0.7988202275495875, "learning_rate": 1.8929038782317152e-06, "loss": 0.6032, "mean_token_accuracy": 0.7990782856941223, "num_tokens": 240982077.0, "step": 6178 }, { "epoch": 0.816679883690193, "grad_norm": 0.7929969781369116, "learning_rate": 1.8916628913312862e-06, "loss": 0.5479, "mean_token_accuracy": 0.8191612958908081, "num_tokens": 241047613.0, "step": 6179 }, { "epoch": 0.816812053925456, "grad_norm": 0.7828332783329977, "learning_rate": 1.8904226725704928e-06, "loss": 0.5615, "mean_token_accuracy": 0.8127670884132385, "num_tokens": 241113149.0, "step": 6180 }, { "epoch": 0.816944224160719, "grad_norm": 0.8123919966742273, "learning_rate": 1.8891832222133483e-06, "loss": 0.5219, "mean_token_accuracy": 0.82602858543396, "num_tokens": 241178685.0, "step": 6181 }, { "epoch": 0.817076394395982, "grad_norm": 0.8088320207301137, "learning_rate": 1.8879445405237073e-06, "loss": 0.5686, "mean_token_accuracy": 0.8100964426994324, "num_tokens": 241244221.0, "step": 6182 }, { "epoch": 0.817208564631245, "grad_norm": 0.8454504647080631, "learning_rate": 1.886706627765261e-06, "loss": 0.6229, "mean_token_accuracy": 0.7949426174163818, "num_tokens": 241309757.0, "step": 6183 }, { "epoch": 0.817340734866508, "grad_norm": 0.7718808981516263, "learning_rate": 1.885469484201532e-06, "loss": 0.5575, "mean_token_accuracy": 0.8142321109771729, "num_tokens": 241375293.0, "step": 6184 }, { "epoch": 0.8174729051017711, "grad_norm": 0.7489233176024225, "learning_rate": 1.8842331100958832e-06, "loss": 0.5415, "mean_token_accuracy": 0.8194664716720581, "num_tokens": 241440829.0, "step": 6185 }, { "epoch": 0.8176050753370341, "grad_norm": 0.8074226991133295, "learning_rate": 1.8829975057115146e-06, "loss": 0.5657, "mean_token_accuracy": 0.8106763362884521, "num_tokens": 241506365.0, "step": 6186 }, { "epoch": 0.8177372455722971, "grad_norm": 0.7837966773764128, "learning_rate": 1.88176267131146e-06, "loss": 0.536, "mean_token_accuracy": 0.8230680227279663, "num_tokens": 241571901.0, "step": 6187 }, { "epoch": 0.8178694158075601, "grad_norm": 0.8304195931462667, "learning_rate": 1.8805286071585872e-06, "loss": 0.5857, "mean_token_accuracy": 0.8037174940109253, "num_tokens": 241637437.0, "step": 6188 }, { "epoch": 0.8180015860428231, "grad_norm": 0.7987285978271877, "learning_rate": 1.8792953135156072e-06, "loss": 0.5708, "mean_token_accuracy": 0.8081583380699158, "num_tokens": 241702973.0, "step": 6189 }, { "epoch": 0.8181337562780862, "grad_norm": 0.798643706642303, "learning_rate": 1.8780627906450588e-06, "loss": 0.5871, "mean_token_accuracy": 0.806800127029419, "num_tokens": 241768509.0, "step": 6190 }, { "epoch": 0.8182659265133492, "grad_norm": 0.8384545559492207, "learning_rate": 1.8768310388093217e-06, "loss": 0.5841, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 241834045.0, "step": 6191 }, { "epoch": 0.8183980967486122, "grad_norm": 0.7561514171031172, "learning_rate": 1.87560005827061e-06, "loss": 0.5523, "mean_token_accuracy": 0.8172231912612915, "num_tokens": 241899581.0, "step": 6192 }, { "epoch": 0.8185302669838752, "grad_norm": 0.7796466802405627, "learning_rate": 1.874369849290975e-06, "loss": 0.5291, "mean_token_accuracy": 0.8219844698905945, "num_tokens": 241965117.0, "step": 6193 }, { "epoch": 0.8186624372191382, "grad_norm": 0.800548904085319, "learning_rate": 1.8731404121322993e-06, "loss": 0.5385, "mean_token_accuracy": 0.817604660987854, "num_tokens": 242030653.0, "step": 6194 }, { "epoch": 0.8187946074544012, "grad_norm": 0.8523173416047355, "learning_rate": 1.8719117470563092e-06, "loss": 0.5738, "mean_token_accuracy": 0.8083261847496033, "num_tokens": 242096189.0, "step": 6195 }, { "epoch": 0.8189267776896643, "grad_norm": 0.8023326594091836, "learning_rate": 1.8706838543245572e-06, "loss": 0.5844, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 242161725.0, "step": 6196 }, { "epoch": 0.8190589479249273, "grad_norm": 0.8273587926214198, "learning_rate": 1.8694567341984393e-06, "loss": 0.6332, "mean_token_accuracy": 0.7893877625465393, "num_tokens": 242227261.0, "step": 6197 }, { "epoch": 0.8191911181601903, "grad_norm": 0.8572986864658347, "learning_rate": 1.8682303869391798e-06, "loss": 0.5911, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 242292797.0, "step": 6198 }, { "epoch": 0.8193232883954533, "grad_norm": 0.8264420572061952, "learning_rate": 1.8670048128078477e-06, "loss": 0.5856, "mean_token_accuracy": 0.8043584227561951, "num_tokens": 242358333.0, "step": 6199 }, { "epoch": 0.8194554586307163, "grad_norm": 0.8421403400361683, "learning_rate": 1.8657800120653392e-06, "loss": 0.6361, "mean_token_accuracy": 0.7874343991279602, "num_tokens": 242423869.0, "step": 6200 }, { "epoch": 0.8195876288659794, "grad_norm": 0.8486335529471711, "learning_rate": 1.8645559849723876e-06, "loss": 0.5833, "mean_token_accuracy": 0.8056555986404419, "num_tokens": 242489405.0, "step": 6201 }, { "epoch": 0.8197197991012424, "grad_norm": 0.8247567096112312, "learning_rate": 1.863332731789566e-06, "loss": 0.608, "mean_token_accuracy": 0.7993529438972473, "num_tokens": 242554941.0, "step": 6202 }, { "epoch": 0.8198519693365054, "grad_norm": 0.8329503173007758, "learning_rate": 1.8621102527772767e-06, "loss": 0.5602, "mean_token_accuracy": 0.8132248520851135, "num_tokens": 242620477.0, "step": 6203 }, { "epoch": 0.8199841395717684, "grad_norm": 0.8806742412565841, "learning_rate": 1.8608885481957603e-06, "loss": 0.6699, "mean_token_accuracy": 0.7763093709945679, "num_tokens": 242686013.0, "step": 6204 }, { "epoch": 0.8201163098070314, "grad_norm": 0.8402069577912357, "learning_rate": 1.8596676183050932e-06, "loss": 0.5589, "mean_token_accuracy": 0.8142168521881104, "num_tokens": 242751549.0, "step": 6205 }, { "epoch": 0.8202484800422944, "grad_norm": 0.7611531930926242, "learning_rate": 1.858447463365187e-06, "loss": 0.5262, "mean_token_accuracy": 0.8261964321136475, "num_tokens": 242817085.0, "step": 6206 }, { "epoch": 0.8203806502775575, "grad_norm": 0.8081335334820581, "learning_rate": 1.8572280836357828e-06, "loss": 0.577, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 242882621.0, "step": 6207 }, { "epoch": 0.8205128205128205, "grad_norm": 0.8093906662246889, "learning_rate": 1.8560094793764668e-06, "loss": 0.5699, "mean_token_accuracy": 0.8094707727432251, "num_tokens": 242948157.0, "step": 6208 }, { "epoch": 0.8206449907480835, "grad_norm": 0.8853689302129129, "learning_rate": 1.85479165084665e-06, "loss": 0.5343, "mean_token_accuracy": 0.8249908685684204, "num_tokens": 243013693.0, "step": 6209 }, { "epoch": 0.8207771609833465, "grad_norm": 0.8093448447972841, "learning_rate": 1.8535745983055852e-06, "loss": 0.551, "mean_token_accuracy": 0.8192223310470581, "num_tokens": 243079229.0, "step": 6210 }, { "epoch": 0.8209093312186095, "grad_norm": 0.8133221278763408, "learning_rate": 1.8523583220123549e-06, "loss": 0.5666, "mean_token_accuracy": 0.813148558139801, "num_tokens": 243144765.0, "step": 6211 }, { "epoch": 0.8210415014538726, "grad_norm": 0.7488063931126551, "learning_rate": 1.8511428222258826e-06, "loss": 0.5139, "mean_token_accuracy": 0.8286839127540588, "num_tokens": 243210301.0, "step": 6212 }, { "epoch": 0.8211736716891356, "grad_norm": 0.8471402530599131, "learning_rate": 1.8499280992049202e-06, "loss": 0.5842, "mean_token_accuracy": 0.8072579503059387, "num_tokens": 243275837.0, "step": 6213 }, { "epoch": 0.8213058419243986, "grad_norm": 0.8374337123434832, "learning_rate": 1.8487141532080567e-06, "loss": 0.5527, "mean_token_accuracy": 0.8179098963737488, "num_tokens": 243341373.0, "step": 6214 }, { "epoch": 0.8214380121596616, "grad_norm": 0.7926686473094078, "learning_rate": 1.847500984493717e-06, "loss": 0.578, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 243406909.0, "step": 6215 }, { "epoch": 0.8215701823949246, "grad_norm": 0.9103354179309905, "learning_rate": 1.8462885933201596e-06, "loss": 0.6442, "mean_token_accuracy": 0.7832224369049072, "num_tokens": 243472445.0, "step": 6216 }, { "epoch": 0.8217023526301876, "grad_norm": 0.8018604053994626, "learning_rate": 1.8450769799454749e-06, "loss": 0.5641, "mean_token_accuracy": 0.8134385347366333, "num_tokens": 243537981.0, "step": 6217 }, { "epoch": 0.8218345228654507, "grad_norm": 0.8722711862466161, "learning_rate": 1.843866144627594e-06, "loss": 0.6506, "mean_token_accuracy": 0.7829630374908447, "num_tokens": 243603517.0, "step": 6218 }, { "epoch": 0.8219666931007137, "grad_norm": 0.7671078537204737, "learning_rate": 1.8426560876242751e-06, "loss": 0.5566, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 243669053.0, "step": 6219 }, { "epoch": 0.8220988633359767, "grad_norm": 0.7855132463958779, "learning_rate": 1.8414468091931154e-06, "loss": 0.5521, "mean_token_accuracy": 0.8176657557487488, "num_tokens": 243734589.0, "step": 6220 }, { "epoch": 0.8222310335712397, "grad_norm": 0.8411408886707885, "learning_rate": 1.8402383095915466e-06, "loss": 0.5802, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 243800125.0, "step": 6221 }, { "epoch": 0.8223632038065027, "grad_norm": 0.8008213718540634, "learning_rate": 1.8390305890768295e-06, "loss": 0.5664, "mean_token_accuracy": 0.8122023940086365, "num_tokens": 243865661.0, "step": 6222 }, { "epoch": 0.8224953740417658, "grad_norm": 0.8254777570316891, "learning_rate": 1.8378236479060645e-06, "loss": 0.5873, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 243931197.0, "step": 6223 }, { "epoch": 0.8226275442770288, "grad_norm": 0.7682297930049226, "learning_rate": 1.8366174863361844e-06, "loss": 0.5779, "mean_token_accuracy": 0.807669997215271, "num_tokens": 243996733.0, "step": 6224 }, { "epoch": 0.8227597145122918, "grad_norm": 0.7842445392698334, "learning_rate": 1.8354121046239566e-06, "loss": 0.5954, "mean_token_accuracy": 0.8034427762031555, "num_tokens": 244062269.0, "step": 6225 }, { "epoch": 0.8228918847475548, "grad_norm": 0.747298681171584, "learning_rate": 1.8342075030259797e-06, "loss": 0.5277, "mean_token_accuracy": 0.8230985403060913, "num_tokens": 244127805.0, "step": 6226 }, { "epoch": 0.8230240549828178, "grad_norm": 0.8524282838048105, "learning_rate": 1.8330036817986885e-06, "loss": 0.6039, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 244193341.0, "step": 6227 }, { "epoch": 0.8231562252180809, "grad_norm": 0.8326497059201796, "learning_rate": 1.831800641198353e-06, "loss": 0.6127, "mean_token_accuracy": 0.7972164750099182, "num_tokens": 244258877.0, "step": 6228 }, { "epoch": 0.8232883954533439, "grad_norm": 0.7406200018011989, "learning_rate": 1.8305983814810752e-06, "loss": 0.5557, "mean_token_accuracy": 0.8135453462600708, "num_tokens": 244324413.0, "step": 6229 }, { "epoch": 0.8234205656886069, "grad_norm": 0.7506797414835306, "learning_rate": 1.829396902902788e-06, "loss": 0.5048, "mean_token_accuracy": 0.8330942392349243, "num_tokens": 244389949.0, "step": 6230 }, { "epoch": 0.8235527359238699, "grad_norm": 0.8786726763573189, "learning_rate": 1.8281962057192657e-06, "loss": 0.6127, "mean_token_accuracy": 0.7962550520896912, "num_tokens": 244455485.0, "step": 6231 }, { "epoch": 0.8236849061591329, "grad_norm": 0.7960459428703934, "learning_rate": 1.8269962901861087e-06, "loss": 0.5483, "mean_token_accuracy": 0.818260908126831, "num_tokens": 244521021.0, "step": 6232 }, { "epoch": 0.8238170763943959, "grad_norm": 0.8529264284595908, "learning_rate": 1.8257971565587546e-06, "loss": 0.6242, "mean_token_accuracy": 0.7940422296524048, "num_tokens": 244586557.0, "step": 6233 }, { "epoch": 0.823949246629659, "grad_norm": 0.8272755896728482, "learning_rate": 1.8245988050924736e-06, "loss": 0.6093, "mean_token_accuracy": 0.797948956489563, "num_tokens": 244652093.0, "step": 6234 }, { "epoch": 0.824081416864922, "grad_norm": 0.7707003409219777, "learning_rate": 1.823401236042372e-06, "loss": 0.5437, "mean_token_accuracy": 0.8186271786689758, "num_tokens": 244717629.0, "step": 6235 }, { "epoch": 0.824213587100185, "grad_norm": 0.7026427607662196, "learning_rate": 1.822204449663383e-06, "loss": 0.5075, "mean_token_accuracy": 0.8310340642929077, "num_tokens": 244783165.0, "step": 6236 }, { "epoch": 0.824345757335448, "grad_norm": 0.8268828230577532, "learning_rate": 1.8210084462102812e-06, "loss": 0.5551, "mean_token_accuracy": 0.8148882985115051, "num_tokens": 244848701.0, "step": 6237 }, { "epoch": 0.824477927570711, "grad_norm": 0.7791742446471815, "learning_rate": 1.8198132259376694e-06, "loss": 0.5166, "mean_token_accuracy": 0.8264864087104797, "num_tokens": 244914237.0, "step": 6238 }, { "epoch": 0.824610097805974, "grad_norm": 0.7977667968161318, "learning_rate": 1.818618789099984e-06, "loss": 0.5903, "mean_token_accuracy": 0.8015199899673462, "num_tokens": 244979773.0, "step": 6239 }, { "epoch": 0.8247422680412371, "grad_norm": 0.8896327325005841, "learning_rate": 1.8174251359514988e-06, "loss": 0.6081, "mean_token_accuracy": 0.7966976165771484, "num_tokens": 245045309.0, "step": 6240 }, { "epoch": 0.8248744382765001, "grad_norm": 0.8330568103394576, "learning_rate": 1.816232266746314e-06, "loss": 0.6031, "mean_token_accuracy": 0.8013063073158264, "num_tokens": 245110845.0, "step": 6241 }, { "epoch": 0.8250066085117631, "grad_norm": 0.7833366036704643, "learning_rate": 1.8150401817383679e-06, "loss": 0.5532, "mean_token_accuracy": 0.8151324391365051, "num_tokens": 245176381.0, "step": 6242 }, { "epoch": 0.8251387787470261, "grad_norm": 0.8084083842148668, "learning_rate": 1.813848881181431e-06, "loss": 0.5328, "mean_token_accuracy": 0.8215877413749695, "num_tokens": 245241917.0, "step": 6243 }, { "epoch": 0.8252709489822891, "grad_norm": 0.776321509852152, "learning_rate": 1.812658365329107e-06, "loss": 0.5764, "mean_token_accuracy": 0.8078684210777283, "num_tokens": 245307453.0, "step": 6244 }, { "epoch": 0.8254031192175522, "grad_norm": 0.83065366169423, "learning_rate": 1.8114686344348297e-06, "loss": 0.5413, "mean_token_accuracy": 0.8213130235671997, "num_tokens": 245372989.0, "step": 6245 }, { "epoch": 0.8255352894528152, "grad_norm": 0.8726203196977618, "learning_rate": 1.8102796887518693e-06, "loss": 0.6172, "mean_token_accuracy": 0.7936301827430725, "num_tokens": 245438525.0, "step": 6246 }, { "epoch": 0.8256674596880782, "grad_norm": 0.763287714354175, "learning_rate": 1.8090915285333268e-06, "loss": 0.5313, "mean_token_accuracy": 0.8235411047935486, "num_tokens": 245504061.0, "step": 6247 }, { "epoch": 0.8257996299233412, "grad_norm": 0.7920364349636289, "learning_rate": 1.807904154032139e-06, "loss": 0.5924, "mean_token_accuracy": 0.802282989025116, "num_tokens": 245569597.0, "step": 6248 }, { "epoch": 0.8259318001586042, "grad_norm": 0.7907229145324567, "learning_rate": 1.8067175655010688e-06, "loss": 0.5703, "mean_token_accuracy": 0.8112409710884094, "num_tokens": 245635133.0, "step": 6249 }, { "epoch": 0.8260639703938673, "grad_norm": 0.8377364191089725, "learning_rate": 1.805531763192721e-06, "loss": 0.5997, "mean_token_accuracy": 0.7987425327301025, "num_tokens": 245700669.0, "step": 6250 }, { "epoch": 0.8261961406291303, "grad_norm": 0.888832935220518, "learning_rate": 1.804346747359525e-06, "loss": 0.6005, "mean_token_accuracy": 0.8010621666908264, "num_tokens": 245766205.0, "step": 6251 }, { "epoch": 0.8263283108643933, "grad_norm": 0.8380159006943808, "learning_rate": 1.8031625182537466e-06, "loss": 0.6036, "mean_token_accuracy": 0.802191436290741, "num_tokens": 245831741.0, "step": 6252 }, { "epoch": 0.8264604810996563, "grad_norm": 0.8171001143991298, "learning_rate": 1.801979076127484e-06, "loss": 0.5805, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 245897277.0, "step": 6253 }, { "epoch": 0.8265926513349193, "grad_norm": 0.8047217791694421, "learning_rate": 1.800796421232668e-06, "loss": 0.5481, "mean_token_accuracy": 0.8170858025550842, "num_tokens": 245962813.0, "step": 6254 }, { "epoch": 0.8267248215701823, "grad_norm": 0.8626769344220454, "learning_rate": 1.7996145538210568e-06, "loss": 0.6245, "mean_token_accuracy": 0.7920736074447632, "num_tokens": 246028349.0, "step": 6255 }, { "epoch": 0.8268569918054454, "grad_norm": 0.8829517882134118, "learning_rate": 1.7984334741442521e-06, "loss": 0.5871, "mean_token_accuracy": 0.8009247779846191, "num_tokens": 246093885.0, "step": 6256 }, { "epoch": 0.8269891620407084, "grad_norm": 0.781183986526424, "learning_rate": 1.797253182453676e-06, "loss": 0.5958, "mean_token_accuracy": 0.8007111549377441, "num_tokens": 246159421.0, "step": 6257 }, { "epoch": 0.8271213322759714, "grad_norm": 0.8372070452289567, "learning_rate": 1.7960736790005906e-06, "loss": 0.6351, "mean_token_accuracy": 0.7900286912918091, "num_tokens": 246224957.0, "step": 6258 }, { "epoch": 0.8272535025112344, "grad_norm": 0.8252613079963614, "learning_rate": 1.794894964036086e-06, "loss": 0.6001, "mean_token_accuracy": 0.8012757897377014, "num_tokens": 246290493.0, "step": 6259 }, { "epoch": 0.8273856727464974, "grad_norm": 0.7968032128695964, "learning_rate": 1.793717037811086e-06, "loss": 0.579, "mean_token_accuracy": 0.8061897158622742, "num_tokens": 246356029.0, "step": 6260 }, { "epoch": 0.8275178429817605, "grad_norm": 0.7808313006909106, "learning_rate": 1.7925399005763471e-06, "loss": 0.5474, "mean_token_accuracy": 0.8181235790252686, "num_tokens": 246421565.0, "step": 6261 }, { "epoch": 0.8276500132170235, "grad_norm": 0.7815907191288494, "learning_rate": 1.7913635525824574e-06, "loss": 0.5799, "mean_token_accuracy": 0.8052130341529846, "num_tokens": 246487101.0, "step": 6262 }, { "epoch": 0.8277821834522865, "grad_norm": 0.8563367089030218, "learning_rate": 1.7901879940798378e-06, "loss": 0.62, "mean_token_accuracy": 0.7932944893836975, "num_tokens": 246552637.0, "step": 6263 }, { "epoch": 0.8279143536875495, "grad_norm": 0.8424349421575054, "learning_rate": 1.7890132253187376e-06, "loss": 0.5377, "mean_token_accuracy": 0.8202295303344727, "num_tokens": 246618173.0, "step": 6264 }, { "epoch": 0.8280465239228125, "grad_norm": 0.8199766489360545, "learning_rate": 1.7878392465492417e-06, "loss": 0.559, "mean_token_accuracy": 0.8150103688240051, "num_tokens": 246683709.0, "step": 6265 }, { "epoch": 0.8281786941580757, "grad_norm": 0.8203332473344246, "learning_rate": 1.7866660580212661e-06, "loss": 0.5556, "mean_token_accuracy": 0.8127365112304688, "num_tokens": 246749245.0, "step": 6266 }, { "epoch": 0.8283108643933387, "grad_norm": 0.7704943878915421, "learning_rate": 1.785493659984559e-06, "loss": 0.5517, "mean_token_accuracy": 0.8166890740394592, "num_tokens": 246814781.0, "step": 6267 }, { "epoch": 0.8284430346286017, "grad_norm": 0.7847722273534313, "learning_rate": 1.7843220526886962e-06, "loss": 0.5684, "mean_token_accuracy": 0.8119582533836365, "num_tokens": 246880317.0, "step": 6268 }, { "epoch": 0.8285752048638647, "grad_norm": 0.8424520431306315, "learning_rate": 1.7831512363830927e-06, "loss": 0.5323, "mean_token_accuracy": 0.8248534798622131, "num_tokens": 246945853.0, "step": 6269 }, { "epoch": 0.8287073750991277, "grad_norm": 0.7814140768472122, "learning_rate": 1.7819812113169879e-06, "loss": 0.4857, "mean_token_accuracy": 0.8363295197486877, "num_tokens": 247011389.0, "step": 6270 }, { "epoch": 0.8288395453343907, "grad_norm": 0.7775723399481036, "learning_rate": 1.7808119777394577e-06, "loss": 0.5493, "mean_token_accuracy": 0.8142931461334229, "num_tokens": 247076925.0, "step": 6271 }, { "epoch": 0.8289717155696538, "grad_norm": 0.8056334027547175, "learning_rate": 1.779643535899404e-06, "loss": 0.5685, "mean_token_accuracy": 0.8096233606338501, "num_tokens": 247142461.0, "step": 6272 }, { "epoch": 0.8291038858049168, "grad_norm": 0.8215025701143904, "learning_rate": 1.778475886045568e-06, "loss": 0.5828, "mean_token_accuracy": 0.8063728213310242, "num_tokens": 247207997.0, "step": 6273 }, { "epoch": 0.8292360560401798, "grad_norm": 0.8010380379755471, "learning_rate": 1.7773090284265154e-06, "loss": 0.5674, "mean_token_accuracy": 0.8118361830711365, "num_tokens": 247273533.0, "step": 6274 }, { "epoch": 0.8293682262754428, "grad_norm": 0.9381409304603304, "learning_rate": 1.7761429632906464e-06, "loss": 0.6303, "mean_token_accuracy": 0.7891435623168945, "num_tokens": 247339069.0, "step": 6275 }, { "epoch": 0.8295003965107058, "grad_norm": 0.8153013326336948, "learning_rate": 1.7749776908861916e-06, "loss": 0.5642, "mean_token_accuracy": 0.8116530179977417, "num_tokens": 247404605.0, "step": 6276 }, { "epoch": 0.8296325667459689, "grad_norm": 0.7707146469597049, "learning_rate": 1.773813211461215e-06, "loss": 0.5777, "mean_token_accuracy": 0.8084635734558105, "num_tokens": 247470141.0, "step": 6277 }, { "epoch": 0.8297647369812319, "grad_norm": 0.8458985433032392, "learning_rate": 1.7726495252636072e-06, "loss": 0.5791, "mean_token_accuracy": 0.8071816563606262, "num_tokens": 247535677.0, "step": 6278 }, { "epoch": 0.8298969072164949, "grad_norm": 0.8574283009230711, "learning_rate": 1.7714866325410943e-06, "loss": 0.6235, "mean_token_accuracy": 0.7912648320198059, "num_tokens": 247601213.0, "step": 6279 }, { "epoch": 0.8300290774517579, "grad_norm": 0.8184593270640159, "learning_rate": 1.7703245335412311e-06, "loss": 0.5638, "mean_token_accuracy": 0.8132706880569458, "num_tokens": 247666749.0, "step": 6280 }, { "epoch": 0.8301612476870209, "grad_norm": 0.8410737912356592, "learning_rate": 1.7691632285114052e-06, "loss": 0.6336, "mean_token_accuracy": 0.7892351150512695, "num_tokens": 247732285.0, "step": 6281 }, { "epoch": 0.830293417922284, "grad_norm": 0.7973834252131583, "learning_rate": 1.7680027176988346e-06, "loss": 0.5234, "mean_token_accuracy": 0.8235105872154236, "num_tokens": 247797821.0, "step": 6282 }, { "epoch": 0.830425588157547, "grad_norm": 0.8381450218365769, "learning_rate": 1.7668430013505662e-06, "loss": 0.5888, "mean_token_accuracy": 0.8050909638404846, "num_tokens": 247863357.0, "step": 6283 }, { "epoch": 0.83055775839281, "grad_norm": 0.7859648477257121, "learning_rate": 1.7656840797134794e-06, "loss": 0.5474, "mean_token_accuracy": 0.8171621561050415, "num_tokens": 247928893.0, "step": 6284 }, { "epoch": 0.830689928628073, "grad_norm": 0.8451759481211684, "learning_rate": 1.7645259530342852e-06, "loss": 0.5843, "mean_token_accuracy": 0.808661937713623, "num_tokens": 247994429.0, "step": 6285 }, { "epoch": 0.830822098863336, "grad_norm": 0.78345892762568, "learning_rate": 1.7633686215595257e-06, "loss": 0.5268, "mean_token_accuracy": 0.8260438442230225, "num_tokens": 248059965.0, "step": 6286 }, { "epoch": 0.830954269098599, "grad_norm": 0.8459122821544637, "learning_rate": 1.7622120855355706e-06, "loss": 0.6426, "mean_token_accuracy": 0.7885026335716248, "num_tokens": 248125501.0, "step": 6287 }, { "epoch": 0.8310864393338621, "grad_norm": 0.8011127786563288, "learning_rate": 1.7610563452086233e-06, "loss": 0.5803, "mean_token_accuracy": 0.8074105978012085, "num_tokens": 248191037.0, "step": 6288 }, { "epoch": 0.8312186095691251, "grad_norm": 0.7729588188871527, "learning_rate": 1.7599014008247162e-06, "loss": 0.5298, "mean_token_accuracy": 0.8231290578842163, "num_tokens": 248256573.0, "step": 6289 }, { "epoch": 0.8313507798043881, "grad_norm": 0.7852754791924386, "learning_rate": 1.7587472526297142e-06, "loss": 0.5687, "mean_token_accuracy": 0.8070138096809387, "num_tokens": 248322109.0, "step": 6290 }, { "epoch": 0.8314829500396511, "grad_norm": 0.8631486596524248, "learning_rate": 1.7575939008693082e-06, "loss": 0.6256, "mean_token_accuracy": 0.7941032648086548, "num_tokens": 248387645.0, "step": 6291 }, { "epoch": 0.8316151202749141, "grad_norm": 0.79745505214407, "learning_rate": 1.7564413457890275e-06, "loss": 0.5447, "mean_token_accuracy": 0.8210077881813049, "num_tokens": 248453181.0, "step": 6292 }, { "epoch": 0.8317472905101772, "grad_norm": 0.8391890887215997, "learning_rate": 1.7552895876342227e-06, "loss": 0.5699, "mean_token_accuracy": 0.8114699125289917, "num_tokens": 248518717.0, "step": 6293 }, { "epoch": 0.8318794607454402, "grad_norm": 0.7794890203797715, "learning_rate": 1.7541386266500815e-06, "loss": 0.5468, "mean_token_accuracy": 0.818444013595581, "num_tokens": 248584253.0, "step": 6294 }, { "epoch": 0.8320116309807032, "grad_norm": 0.7751266045086964, "learning_rate": 1.752988463081618e-06, "loss": 0.5925, "mean_token_accuracy": 0.8028781414031982, "num_tokens": 248649789.0, "step": 6295 }, { "epoch": 0.8321438012159662, "grad_norm": 0.8144050912660914, "learning_rate": 1.7518390971736805e-06, "loss": 0.512, "mean_token_accuracy": 0.8275851607322693, "num_tokens": 248715325.0, "step": 6296 }, { "epoch": 0.8322759714512292, "grad_norm": 0.7984505670997857, "learning_rate": 1.7506905291709417e-06, "loss": 0.5668, "mean_token_accuracy": 0.8116225004196167, "num_tokens": 248780861.0, "step": 6297 }, { "epoch": 0.8324081416864922, "grad_norm": 0.8965228320009322, "learning_rate": 1.7495427593179092e-06, "loss": 0.6113, "mean_token_accuracy": 0.7954919934272766, "num_tokens": 248846397.0, "step": 6298 }, { "epoch": 0.8325403119217553, "grad_norm": 0.8384357257828954, "learning_rate": 1.7483957878589197e-06, "loss": 0.6281, "mean_token_accuracy": 0.7940422296524048, "num_tokens": 248911933.0, "step": 6299 }, { "epoch": 0.8326724821570183, "grad_norm": 0.871961206869255, "learning_rate": 1.747249615038139e-06, "loss": 0.6452, "mean_token_accuracy": 0.7826425433158875, "num_tokens": 248977469.0, "step": 6300 }, { "epoch": 0.8328046523922813, "grad_norm": 0.7531098860572758, "learning_rate": 1.7461042410995644e-06, "loss": 0.5647, "mean_token_accuracy": 0.8119277358055115, "num_tokens": 249043005.0, "step": 6301 }, { "epoch": 0.8329368226275443, "grad_norm": 0.8507712141125047, "learning_rate": 1.7449596662870203e-06, "loss": 0.6092, "mean_token_accuracy": 0.7999480962753296, "num_tokens": 249108541.0, "step": 6302 }, { "epoch": 0.8330689928628073, "grad_norm": 0.8267474377806044, "learning_rate": 1.7438158908441649e-06, "loss": 0.6194, "mean_token_accuracy": 0.7936912178993225, "num_tokens": 249174077.0, "step": 6303 }, { "epoch": 0.8332011630980704, "grad_norm": 0.7234848335780548, "learning_rate": 1.7426729150144802e-06, "loss": 0.5304, "mean_token_accuracy": 0.8243193626403809, "num_tokens": 249239613.0, "step": 6304 }, { "epoch": 0.8333333333333334, "grad_norm": 0.8457660262760566, "learning_rate": 1.7415307390412871e-06, "loss": 0.5992, "mean_token_accuracy": 0.802466094493866, "num_tokens": 249305149.0, "step": 6305 }, { "epoch": 0.8334655035685964, "grad_norm": 0.7722094997147979, "learning_rate": 1.740389363167727e-06, "loss": 0.5467, "mean_token_accuracy": 0.817268967628479, "num_tokens": 249370685.0, "step": 6306 }, { "epoch": 0.8335976738038594, "grad_norm": 0.7975420229808878, "learning_rate": 1.739248787636777e-06, "loss": 0.6172, "mean_token_accuracy": 0.7961482405662537, "num_tokens": 249436221.0, "step": 6307 }, { "epoch": 0.8337298440391224, "grad_norm": 0.7610037736031919, "learning_rate": 1.738109012691241e-06, "loss": 0.5304, "mean_token_accuracy": 0.8240141868591309, "num_tokens": 249501757.0, "step": 6308 }, { "epoch": 0.8338620142743854, "grad_norm": 0.8015663390766887, "learning_rate": 1.7369700385737543e-06, "loss": 0.6032, "mean_token_accuracy": 0.7985135912895203, "num_tokens": 249567293.0, "step": 6309 }, { "epoch": 0.8339941845096485, "grad_norm": 0.851944918997343, "learning_rate": 1.735831865526778e-06, "loss": 0.5645, "mean_token_accuracy": 0.8128281235694885, "num_tokens": 249632829.0, "step": 6310 }, { "epoch": 0.8341263547449115, "grad_norm": 0.7574611425354296, "learning_rate": 1.7346944937926102e-06, "loss": 0.5225, "mean_token_accuracy": 0.8247619271278381, "num_tokens": 249698365.0, "step": 6311 }, { "epoch": 0.8342585249801745, "grad_norm": 0.7582073310903296, "learning_rate": 1.7335579236133693e-06, "loss": 0.5278, "mean_token_accuracy": 0.8252655267715454, "num_tokens": 249763901.0, "step": 6312 }, { "epoch": 0.8343906952154375, "grad_norm": 0.8075408441829715, "learning_rate": 1.7324221552310083e-06, "loss": 0.53, "mean_token_accuracy": 0.8233426809310913, "num_tokens": 249829437.0, "step": 6313 }, { "epoch": 0.8345228654507005, "grad_norm": 0.741057961739002, "learning_rate": 1.7312871888873097e-06, "loss": 0.5087, "mean_token_accuracy": 0.8305457234382629, "num_tokens": 249894973.0, "step": 6314 }, { "epoch": 0.8346550356859636, "grad_norm": 0.8972969811202861, "learning_rate": 1.7301530248238844e-06, "loss": 0.5819, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 249960509.0, "step": 6315 }, { "epoch": 0.8347872059212266, "grad_norm": 0.8174824346831185, "learning_rate": 1.7290196632821703e-06, "loss": 0.5853, "mean_token_accuracy": 0.8043736815452576, "num_tokens": 250026045.0, "step": 6316 }, { "epoch": 0.8349193761564896, "grad_norm": 0.7752425220560067, "learning_rate": 1.727887104503437e-06, "loss": 0.571, "mean_token_accuracy": 0.8095623254776001, "num_tokens": 250091581.0, "step": 6317 }, { "epoch": 0.8350515463917526, "grad_norm": 0.7466048981433032, "learning_rate": 1.7267553487287841e-06, "loss": 0.4825, "mean_token_accuracy": 0.8400378227233887, "num_tokens": 250157117.0, "step": 6318 }, { "epoch": 0.8351837166270156, "grad_norm": 0.8120133447981427, "learning_rate": 1.7256243961991381e-06, "loss": 0.6104, "mean_token_accuracy": 0.7972012162208557, "num_tokens": 250222653.0, "step": 6319 }, { "epoch": 0.8353158868622786, "grad_norm": 0.7986243473213532, "learning_rate": 1.7244942471552539e-06, "loss": 0.5595, "mean_token_accuracy": 0.8134843111038208, "num_tokens": 250288189.0, "step": 6320 }, { "epoch": 0.8354480570975417, "grad_norm": 0.8105919863671986, "learning_rate": 1.7233649018377174e-06, "loss": 0.5919, "mean_token_accuracy": 0.8058539628982544, "num_tokens": 250353725.0, "step": 6321 }, { "epoch": 0.8355802273328047, "grad_norm": 0.7370437687667217, "learning_rate": 1.7222363604869446e-06, "loss": 0.5138, "mean_token_accuracy": 0.8309577703475952, "num_tokens": 250419261.0, "step": 6322 }, { "epoch": 0.8357123975680677, "grad_norm": 0.7848572616513801, "learning_rate": 1.7211086233431739e-06, "loss": 0.5815, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 250484797.0, "step": 6323 }, { "epoch": 0.8358445678033307, "grad_norm": 0.7611940713934258, "learning_rate": 1.7199816906464826e-06, "loss": 0.5489, "mean_token_accuracy": 0.8167653679847717, "num_tokens": 250550333.0, "step": 6324 }, { "epoch": 0.8359767380385937, "grad_norm": 0.8831179238266678, "learning_rate": 1.7188555626367665e-06, "loss": 0.642, "mean_token_accuracy": 0.7882431745529175, "num_tokens": 250615869.0, "step": 6325 }, { "epoch": 0.8361089082738568, "grad_norm": 0.8244398278536895, "learning_rate": 1.7177302395537565e-06, "loss": 0.6016, "mean_token_accuracy": 0.8008332252502441, "num_tokens": 250681405.0, "step": 6326 }, { "epoch": 0.8362410785091198, "grad_norm": 0.8200231414062327, "learning_rate": 1.7166057216370108e-06, "loss": 0.5762, "mean_token_accuracy": 0.8075326681137085, "num_tokens": 250746941.0, "step": 6327 }, { "epoch": 0.8363732487443828, "grad_norm": 0.797334003756221, "learning_rate": 1.7154820091259159e-06, "loss": 0.5604, "mean_token_accuracy": 0.812065064907074, "num_tokens": 250812477.0, "step": 6328 }, { "epoch": 0.8365054189796458, "grad_norm": 0.9064869006873526, "learning_rate": 1.7143591022596846e-06, "loss": 0.597, "mean_token_accuracy": 0.8018862009048462, "num_tokens": 250878013.0, "step": 6329 }, { "epoch": 0.8366375892149088, "grad_norm": 0.7805582960778721, "learning_rate": 1.713237001277364e-06, "loss": 0.5254, "mean_token_accuracy": 0.8254486918449402, "num_tokens": 250943549.0, "step": 6330 }, { "epoch": 0.8367697594501718, "grad_norm": 0.7822958961949028, "learning_rate": 1.7121157064178224e-06, "loss": 0.568, "mean_token_accuracy": 0.8130417466163635, "num_tokens": 251009085.0, "step": 6331 }, { "epoch": 0.8369019296854349, "grad_norm": 0.7779425885339815, "learning_rate": 1.7109952179197631e-06, "loss": 0.5627, "mean_token_accuracy": 0.8093792200088501, "num_tokens": 251074621.0, "step": 6332 }, { "epoch": 0.8370340999206979, "grad_norm": 0.7253399279024325, "learning_rate": 1.7098755360217106e-06, "loss": 0.5087, "mean_token_accuracy": 0.8317665457725525, "num_tokens": 251140157.0, "step": 6333 }, { "epoch": 0.8371662701559609, "grad_norm": 0.8178126524545192, "learning_rate": 1.7087566609620272e-06, "loss": 0.5492, "mean_token_accuracy": 0.8195427656173706, "num_tokens": 251205693.0, "step": 6334 }, { "epoch": 0.8372984403912239, "grad_norm": 0.7696718105861241, "learning_rate": 1.707638592978893e-06, "loss": 0.5973, "mean_token_accuracy": 0.8001312613487244, "num_tokens": 251271229.0, "step": 6335 }, { "epoch": 0.8374306106264869, "grad_norm": 0.8334687923882328, "learning_rate": 1.7065213323103247e-06, "loss": 0.5616, "mean_token_accuracy": 0.8142168521881104, "num_tokens": 251336765.0, "step": 6336 }, { "epoch": 0.83756278086175, "grad_norm": 0.7970216587152018, "learning_rate": 1.7054048791941624e-06, "loss": 0.5548, "mean_token_accuracy": 0.8160328269004822, "num_tokens": 251402301.0, "step": 6337 }, { "epoch": 0.837694951097013, "grad_norm": 0.7916688432310718, "learning_rate": 1.7042892338680766e-06, "loss": 0.538, "mean_token_accuracy": 0.8210077881813049, "num_tokens": 251467837.0, "step": 6338 }, { "epoch": 0.837827121332276, "grad_norm": 0.8315865011657492, "learning_rate": 1.7031743965695635e-06, "loss": 0.5559, "mean_token_accuracy": 0.8146898746490479, "num_tokens": 251533373.0, "step": 6339 }, { "epoch": 0.837959291567539, "grad_norm": 0.7828317241125906, "learning_rate": 1.7020603675359488e-06, "loss": 0.6003, "mean_token_accuracy": 0.79903244972229, "num_tokens": 251598909.0, "step": 6340 }, { "epoch": 0.838091461802802, "grad_norm": 0.7143403788190262, "learning_rate": 1.7009471470043882e-06, "loss": 0.4928, "mean_token_accuracy": 0.8339335918426514, "num_tokens": 251664445.0, "step": 6341 }, { "epoch": 0.838223632038065, "grad_norm": 0.9079641649346915, "learning_rate": 1.6998347352118594e-06, "loss": 0.6347, "mean_token_accuracy": 0.7884415984153748, "num_tokens": 251729981.0, "step": 6342 }, { "epoch": 0.8383558022733281, "grad_norm": 0.8494559473080302, "learning_rate": 1.6987231323951754e-06, "loss": 0.5846, "mean_token_accuracy": 0.8048925399780273, "num_tokens": 251795517.0, "step": 6343 }, { "epoch": 0.8384879725085911, "grad_norm": 0.7488287976195823, "learning_rate": 1.6976123387909705e-06, "loss": 0.4931, "mean_token_accuracy": 0.8377792835235596, "num_tokens": 251861053.0, "step": 6344 }, { "epoch": 0.8386201427438541, "grad_norm": 0.846596708699531, "learning_rate": 1.6965023546357108e-06, "loss": 0.5682, "mean_token_accuracy": 0.8082956671714783, "num_tokens": 251926589.0, "step": 6345 }, { "epoch": 0.8387523129791171, "grad_norm": 0.7803238702677496, "learning_rate": 1.695393180165688e-06, "loss": 0.5631, "mean_token_accuracy": 0.813057005405426, "num_tokens": 251992125.0, "step": 6346 }, { "epoch": 0.8388844832143801, "grad_norm": 0.8662241208510637, "learning_rate": 1.6942848156170244e-06, "loss": 0.6117, "mean_token_accuracy": 0.7937980890274048, "num_tokens": 252057661.0, "step": 6347 }, { "epoch": 0.8390166534496432, "grad_norm": 0.8205003174584813, "learning_rate": 1.6931772612256648e-06, "loss": 0.5731, "mean_token_accuracy": 0.8071816563606262, "num_tokens": 252123197.0, "step": 6348 }, { "epoch": 0.8391488236849062, "grad_norm": 0.8264275020501992, "learning_rate": 1.6920705172273855e-06, "loss": 0.5946, "mean_token_accuracy": 0.8040990233421326, "num_tokens": 252188733.0, "step": 6349 }, { "epoch": 0.8392809939201692, "grad_norm": 0.8651792470553208, "learning_rate": 1.6909645838577894e-06, "loss": 0.6396, "mean_token_accuracy": 0.7842754125595093, "num_tokens": 252254269.0, "step": 6350 }, { "epoch": 0.8394131641554322, "grad_norm": 0.7696212777598227, "learning_rate": 1.6898594613523068e-06, "loss": 0.5614, "mean_token_accuracy": 0.8128586411476135, "num_tokens": 252319805.0, "step": 6351 }, { "epoch": 0.8395453343906952, "grad_norm": 0.8446352042723128, "learning_rate": 1.6887551499461935e-06, "loss": 0.5951, "mean_token_accuracy": 0.7999633550643921, "num_tokens": 252385341.0, "step": 6352 }, { "epoch": 0.8396775046259582, "grad_norm": 0.8766555462806156, "learning_rate": 1.6876516498745372e-06, "loss": 0.569, "mean_token_accuracy": 0.8090434670448303, "num_tokens": 252450877.0, "step": 6353 }, { "epoch": 0.8398096748612213, "grad_norm": 0.8457940704894088, "learning_rate": 1.6865489613722482e-06, "loss": 0.5641, "mean_token_accuracy": 0.812782347202301, "num_tokens": 252516413.0, "step": 6354 }, { "epoch": 0.8399418450964843, "grad_norm": 0.7527017322182157, "learning_rate": 1.6854470846740655e-06, "loss": 0.5506, "mean_token_accuracy": 0.8155444860458374, "num_tokens": 252581949.0, "step": 6355 }, { "epoch": 0.8400740153317473, "grad_norm": 0.7877442439135356, "learning_rate": 1.6843460200145567e-06, "loss": 0.5543, "mean_token_accuracy": 0.8149035573005676, "num_tokens": 252647485.0, "step": 6356 }, { "epoch": 0.8402061855670103, "grad_norm": 0.7691485018555413, "learning_rate": 1.6832457676281166e-06, "loss": 0.5534, "mean_token_accuracy": 0.8162007331848145, "num_tokens": 252713021.0, "step": 6357 }, { "epoch": 0.8403383558022733, "grad_norm": 0.7803075944507253, "learning_rate": 1.6821463277489624e-06, "loss": 0.5451, "mean_token_accuracy": 0.818291425704956, "num_tokens": 252778557.0, "step": 6358 }, { "epoch": 0.8404705260375364, "grad_norm": 0.8812257148192926, "learning_rate": 1.6810477006111448e-06, "loss": 0.6222, "mean_token_accuracy": 0.7930502891540527, "num_tokens": 252844093.0, "step": 6359 }, { "epoch": 0.8406026962727994, "grad_norm": 0.7445362371616105, "learning_rate": 1.6799498864485386e-06, "loss": 0.5357, "mean_token_accuracy": 0.8212519884109497, "num_tokens": 252909629.0, "step": 6360 }, { "epoch": 0.8407348665080624, "grad_norm": 0.751461480540124, "learning_rate": 1.6788528854948428e-06, "loss": 0.506, "mean_token_accuracy": 0.8328653573989868, "num_tokens": 252975165.0, "step": 6361 }, { "epoch": 0.8408670367433254, "grad_norm": 0.8434146874368318, "learning_rate": 1.6777566979835897e-06, "loss": 0.5622, "mean_token_accuracy": 0.8134232759475708, "num_tokens": 253040701.0, "step": 6362 }, { "epoch": 0.8409992069785884, "grad_norm": 0.8241186030525194, "learning_rate": 1.6766613241481313e-06, "loss": 0.5797, "mean_token_accuracy": 0.8093487024307251, "num_tokens": 253106237.0, "step": 6363 }, { "epoch": 0.8411313772138514, "grad_norm": 0.7979478378111361, "learning_rate": 1.6755667642216533e-06, "loss": 0.537, "mean_token_accuracy": 0.8220149874687195, "num_tokens": 253171773.0, "step": 6364 }, { "epoch": 0.8412635474491145, "grad_norm": 0.8752016770531569, "learning_rate": 1.6744730184371605e-06, "loss": 0.6256, "mean_token_accuracy": 0.7937523126602173, "num_tokens": 253237309.0, "step": 6365 }, { "epoch": 0.8413957176843775, "grad_norm": 0.809992780646537, "learning_rate": 1.673380087027493e-06, "loss": 0.5422, "mean_token_accuracy": 0.8192528486251831, "num_tokens": 253302845.0, "step": 6366 }, { "epoch": 0.8415278879196405, "grad_norm": 0.7575134889111783, "learning_rate": 1.6722879702253098e-06, "loss": 0.5538, "mean_token_accuracy": 0.8165669441223145, "num_tokens": 253368381.0, "step": 6367 }, { "epoch": 0.8416600581549035, "grad_norm": 0.7713355910402547, "learning_rate": 1.6711966682631003e-06, "loss": 0.5573, "mean_token_accuracy": 0.8116225004196167, "num_tokens": 253433917.0, "step": 6368 }, { "epoch": 0.8417922283901665, "grad_norm": 0.8508933078526428, "learning_rate": 1.670106181373181e-06, "loss": 0.6115, "mean_token_accuracy": 0.7955225110054016, "num_tokens": 253499453.0, "step": 6369 }, { "epoch": 0.8419243986254296, "grad_norm": 0.7614197214549886, "learning_rate": 1.6690165097876941e-06, "loss": 0.5337, "mean_token_accuracy": 0.8215572237968445, "num_tokens": 253564989.0, "step": 6370 }, { "epoch": 0.8420565688606926, "grad_norm": 0.8477114296183326, "learning_rate": 1.6679276537386053e-06, "loss": 0.6165, "mean_token_accuracy": 0.7936912178993225, "num_tokens": 253630525.0, "step": 6371 }, { "epoch": 0.8421887390959556, "grad_norm": 0.7666861960777254, "learning_rate": 1.666839613457713e-06, "loss": 0.5616, "mean_token_accuracy": 0.8124313354492188, "num_tokens": 253696061.0, "step": 6372 }, { "epoch": 0.8423209093312186, "grad_norm": 0.8414586021095871, "learning_rate": 1.6657523891766358e-06, "loss": 0.6209, "mean_token_accuracy": 0.7923025488853455, "num_tokens": 253761597.0, "step": 6373 }, { "epoch": 0.8424530795664816, "grad_norm": 0.8050143187674464, "learning_rate": 1.6646659811268206e-06, "loss": 0.5601, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 253827133.0, "step": 6374 }, { "epoch": 0.8425852498017447, "grad_norm": 0.8629006606161239, "learning_rate": 1.663580389539543e-06, "loss": 0.6113, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 253892669.0, "step": 6375 }, { "epoch": 0.8427174200370077, "grad_norm": 0.7827455931532429, "learning_rate": 1.6624956146459026e-06, "loss": 0.5668, "mean_token_accuracy": 0.8112562298774719, "num_tokens": 253958205.0, "step": 6376 }, { "epoch": 0.8428495902722707, "grad_norm": 0.8313254362551526, "learning_rate": 1.661411656676823e-06, "loss": 0.5848, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 254023741.0, "step": 6377 }, { "epoch": 0.8429817605075337, "grad_norm": 0.7859581080313089, "learning_rate": 1.6603285158630578e-06, "loss": 0.5447, "mean_token_accuracy": 0.8186119198799133, "num_tokens": 254089277.0, "step": 6378 }, { "epoch": 0.8431139307427967, "grad_norm": 0.80413597213941, "learning_rate": 1.6592461924351867e-06, "loss": 0.5917, "mean_token_accuracy": 0.8025271892547607, "num_tokens": 254154813.0, "step": 6379 }, { "epoch": 0.8432461009780597, "grad_norm": 0.8365980711775084, "learning_rate": 1.6581646866236107e-06, "loss": 0.5751, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 254220349.0, "step": 6380 }, { "epoch": 0.8433782712133228, "grad_norm": 0.9130554381791749, "learning_rate": 1.6570839986585607e-06, "loss": 0.5797, "mean_token_accuracy": 0.8097301721572876, "num_tokens": 254285885.0, "step": 6381 }, { "epoch": 0.8435104414485858, "grad_norm": 0.7941876269667589, "learning_rate": 1.6560041287700934e-06, "loss": 0.5817, "mean_token_accuracy": 0.8074411153793335, "num_tokens": 254351421.0, "step": 6382 }, { "epoch": 0.8436426116838488, "grad_norm": 0.8329902858108561, "learning_rate": 1.6549250771880915e-06, "loss": 0.5954, "mean_token_accuracy": 0.8039922118186951, "num_tokens": 254416957.0, "step": 6383 }, { "epoch": 0.8437747819191118, "grad_norm": 0.8318124665801473, "learning_rate": 1.653846844142259e-06, "loss": 0.6091, "mean_token_accuracy": 0.7975063920021057, "num_tokens": 254482493.0, "step": 6384 }, { "epoch": 0.8439069521543748, "grad_norm": 0.8030052196743239, "learning_rate": 1.6527694298621339e-06, "loss": 0.532, "mean_token_accuracy": 0.8215572237968445, "num_tokens": 254548029.0, "step": 6385 }, { "epoch": 0.8440391223896379, "grad_norm": 0.8297942280539146, "learning_rate": 1.651692834577071e-06, "loss": 0.6016, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 254613565.0, "step": 6386 }, { "epoch": 0.8441712926249009, "grad_norm": 0.8214903826956622, "learning_rate": 1.6506170585162572e-06, "loss": 0.6051, "mean_token_accuracy": 0.7992919087409973, "num_tokens": 254679101.0, "step": 6387 }, { "epoch": 0.8443034628601639, "grad_norm": 0.7849232896410123, "learning_rate": 1.6495421019087026e-06, "loss": 0.5624, "mean_token_accuracy": 0.8168264031410217, "num_tokens": 254744637.0, "step": 6388 }, { "epoch": 0.8444356330954269, "grad_norm": 0.8833311703247952, "learning_rate": 1.6484679649832433e-06, "loss": 0.5997, "mean_token_accuracy": 0.8009552955627441, "num_tokens": 254810173.0, "step": 6389 }, { "epoch": 0.8445678033306899, "grad_norm": 0.8417334842967468, "learning_rate": 1.6473946479685382e-06, "loss": 0.5943, "mean_token_accuracy": 0.8012452721595764, "num_tokens": 254875709.0, "step": 6390 }, { "epoch": 0.8446999735659529, "grad_norm": 0.7547160042649753, "learning_rate": 1.646322151093078e-06, "loss": 0.5456, "mean_token_accuracy": 0.818474531173706, "num_tokens": 254941245.0, "step": 6391 }, { "epoch": 0.844832143801216, "grad_norm": 0.8188059417651312, "learning_rate": 1.645250474585172e-06, "loss": 0.5836, "mean_token_accuracy": 0.8049536347389221, "num_tokens": 255006781.0, "step": 6392 }, { "epoch": 0.844964314036479, "grad_norm": 0.8824238999784171, "learning_rate": 1.6441796186729592e-06, "loss": 0.6266, "mean_token_accuracy": 0.7912342548370361, "num_tokens": 255072317.0, "step": 6393 }, { "epoch": 0.845096484271742, "grad_norm": 0.8034961450570635, "learning_rate": 1.6431095835844001e-06, "loss": 0.5714, "mean_token_accuracy": 0.8092113137245178, "num_tokens": 255137853.0, "step": 6394 }, { "epoch": 0.845228654507005, "grad_norm": 0.7779037073510824, "learning_rate": 1.6420403695472863e-06, "loss": 0.5644, "mean_token_accuracy": 0.8133622407913208, "num_tokens": 255203389.0, "step": 6395 }, { "epoch": 0.845360824742268, "grad_norm": 0.8379227560619175, "learning_rate": 1.6409719767892291e-06, "loss": 0.6423, "mean_token_accuracy": 0.7860456705093384, "num_tokens": 255268925.0, "step": 6396 }, { "epoch": 0.845492994977531, "grad_norm": 0.8358040711030829, "learning_rate": 1.6399044055376668e-06, "loss": 0.5731, "mean_token_accuracy": 0.8089976906776428, "num_tokens": 255334461.0, "step": 6397 }, { "epoch": 0.8456251652127941, "grad_norm": 0.7959315340631629, "learning_rate": 1.638837656019865e-06, "loss": 0.555, "mean_token_accuracy": 0.8161243796348572, "num_tokens": 255399997.0, "step": 6398 }, { "epoch": 0.8457573354480571, "grad_norm": 0.8001005478474402, "learning_rate": 1.6377717284629092e-06, "loss": 0.5948, "mean_token_accuracy": 0.8023898005485535, "num_tokens": 255465533.0, "step": 6399 }, { "epoch": 0.8458895056833201, "grad_norm": 0.8495115941901491, "learning_rate": 1.636706623093715e-06, "loss": 0.5885, "mean_token_accuracy": 0.8059455752372742, "num_tokens": 255531069.0, "step": 6400 }, { "epoch": 0.8460216759185831, "grad_norm": 0.853985282881911, "learning_rate": 1.635642340139022e-06, "loss": 0.5891, "mean_token_accuracy": 0.8038395643234253, "num_tokens": 255596605.0, "step": 6401 }, { "epoch": 0.8461538461538461, "grad_norm": 0.8370166271312279, "learning_rate": 1.6345788798253924e-06, "loss": 0.5861, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 255662141.0, "step": 6402 }, { "epoch": 0.8462860163891092, "grad_norm": 0.7551560630624123, "learning_rate": 1.6335162423792145e-06, "loss": 0.569, "mean_token_accuracy": 0.8104474544525146, "num_tokens": 255727677.0, "step": 6403 }, { "epoch": 0.8464181866243722, "grad_norm": 0.8198202259600147, "learning_rate": 1.632454428026703e-06, "loss": 0.5917, "mean_token_accuracy": 0.802466094493866, "num_tokens": 255793213.0, "step": 6404 }, { "epoch": 0.8465503568596352, "grad_norm": 0.7773022003174018, "learning_rate": 1.631393436993895e-06, "loss": 0.5651, "mean_token_accuracy": 0.8122023940086365, "num_tokens": 255858749.0, "step": 6405 }, { "epoch": 0.8466825270948982, "grad_norm": 0.8416997474886609, "learning_rate": 1.6303332695066531e-06, "loss": 0.5812, "mean_token_accuracy": 0.8053267002105713, "num_tokens": 255919695.0, "step": 6406 }, { "epoch": 0.8468146973301612, "grad_norm": 0.8097574956172436, "learning_rate": 1.6292739257906654e-06, "loss": 0.5679, "mean_token_accuracy": 0.8091350197792053, "num_tokens": 255985231.0, "step": 6407 }, { "epoch": 0.8469468675654243, "grad_norm": 0.779268369166679, "learning_rate": 1.6282154060714457e-06, "loss": 0.5734, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 256050767.0, "step": 6408 }, { "epoch": 0.8470790378006873, "grad_norm": 0.7861018811120635, "learning_rate": 1.6271577105743268e-06, "loss": 0.5503, "mean_token_accuracy": 0.8132248520851135, "num_tokens": 256116303.0, "step": 6409 }, { "epoch": 0.8472112080359503, "grad_norm": 0.7564631615723539, "learning_rate": 1.6261008395244732e-06, "loss": 0.5011, "mean_token_accuracy": 0.8346660733222961, "num_tokens": 256181839.0, "step": 6410 }, { "epoch": 0.8473433782712133, "grad_norm": 0.8233592729133024, "learning_rate": 1.6250447931468694e-06, "loss": 0.541, "mean_token_accuracy": 0.8222439289093018, "num_tokens": 256247375.0, "step": 6411 }, { "epoch": 0.8474755485064763, "grad_norm": 0.8182503090728271, "learning_rate": 1.6239895716663267e-06, "loss": 0.5643, "mean_token_accuracy": 0.8118361830711365, "num_tokens": 256312911.0, "step": 6412 }, { "epoch": 0.8476077187417393, "grad_norm": 0.8634759663164472, "learning_rate": 1.6229351753074772e-06, "loss": 0.5956, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 256378447.0, "step": 6413 }, { "epoch": 0.8477398889770024, "grad_norm": 0.7769220985069873, "learning_rate": 1.621881604294783e-06, "loss": 0.5547, "mean_token_accuracy": 0.8172231912612915, "num_tokens": 256443983.0, "step": 6414 }, { "epoch": 0.8478720592122654, "grad_norm": 1.028047156748289, "learning_rate": 1.6208288588525256e-06, "loss": 0.5703, "mean_token_accuracy": 0.8077160716056824, "num_tokens": 256502233.0, "step": 6415 }, { "epoch": 0.8480042294475284, "grad_norm": 0.7646500892898503, "learning_rate": 1.6197769392048124e-06, "loss": 0.5172, "mean_token_accuracy": 0.826990008354187, "num_tokens": 256567769.0, "step": 6416 }, { "epoch": 0.8481363996827914, "grad_norm": 0.7991598278196431, "learning_rate": 1.618725845575577e-06, "loss": 0.5833, "mean_token_accuracy": 0.8037174940109253, "num_tokens": 256633305.0, "step": 6417 }, { "epoch": 0.8482685699180544, "grad_norm": 0.8081178407431028, "learning_rate": 1.6176755781885732e-06, "loss": 0.5564, "mean_token_accuracy": 0.8141710162162781, "num_tokens": 256698841.0, "step": 6418 }, { "epoch": 0.8484007401533175, "grad_norm": 0.7816421927358598, "learning_rate": 1.6166261372673807e-06, "loss": 0.5556, "mean_token_accuracy": 0.8134690523147583, "num_tokens": 256764377.0, "step": 6419 }, { "epoch": 0.8485329103885805, "grad_norm": 0.859623624722691, "learning_rate": 1.6155775230354055e-06, "loss": 0.5927, "mean_token_accuracy": 0.8024813532829285, "num_tokens": 256829913.0, "step": 6420 }, { "epoch": 0.8486650806238435, "grad_norm": 0.8263900808945274, "learning_rate": 1.6145297357158751e-06, "loss": 0.567, "mean_token_accuracy": 0.8124008178710938, "num_tokens": 256895449.0, "step": 6421 }, { "epoch": 0.8487972508591065, "grad_norm": 0.8626172183856169, "learning_rate": 1.6134827755318399e-06, "loss": 0.589, "mean_token_accuracy": 0.8047399520874023, "num_tokens": 256960985.0, "step": 6422 }, { "epoch": 0.8489294210943695, "grad_norm": 0.7791078918023644, "learning_rate": 1.6124366427061783e-06, "loss": 0.5442, "mean_token_accuracy": 0.8193444013595581, "num_tokens": 257026521.0, "step": 6423 }, { "epoch": 0.8490615913296325, "grad_norm": 0.7752728645473983, "learning_rate": 1.6113913374615887e-06, "loss": 0.5311, "mean_token_accuracy": 0.8240752220153809, "num_tokens": 257092057.0, "step": 6424 }, { "epoch": 0.8491937615648956, "grad_norm": 0.7690210366666553, "learning_rate": 1.6103468600205954e-06, "loss": 0.5556, "mean_token_accuracy": 0.817513108253479, "num_tokens": 257157593.0, "step": 6425 }, { "epoch": 0.8493259318001586, "grad_norm": 0.8270772752339123, "learning_rate": 1.609303210605544e-06, "loss": 0.556, "mean_token_accuracy": 0.8160328269004822, "num_tokens": 257223129.0, "step": 6426 }, { "epoch": 0.8494581020354216, "grad_norm": 0.7769231360800792, "learning_rate": 1.608260389438609e-06, "loss": 0.5787, "mean_token_accuracy": 0.8090434670448303, "num_tokens": 257288665.0, "step": 6427 }, { "epoch": 0.8495902722706846, "grad_norm": 0.803723700347437, "learning_rate": 1.6072183967417826e-06, "loss": 0.533, "mean_token_accuracy": 0.8212825059890747, "num_tokens": 257354201.0, "step": 6428 }, { "epoch": 0.8497224425059476, "grad_norm": 0.7907201622118613, "learning_rate": 1.6061772327368841e-06, "loss": 0.5593, "mean_token_accuracy": 0.8108900189399719, "num_tokens": 257419737.0, "step": 6429 }, { "epoch": 0.8498546127412107, "grad_norm": 0.7539033189781478, "learning_rate": 1.6051368976455556e-06, "loss": 0.5595, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 257485273.0, "step": 6430 }, { "epoch": 0.8499867829764737, "grad_norm": 0.8358826479611986, "learning_rate": 1.6040973916892629e-06, "loss": 0.5956, "mean_token_accuracy": 0.802221953868866, "num_tokens": 257550809.0, "step": 6431 }, { "epoch": 0.8501189532117367, "grad_norm": 0.8545829057427884, "learning_rate": 1.6030587150892938e-06, "loss": 0.6313, "mean_token_accuracy": 0.7900286912918091, "num_tokens": 257616345.0, "step": 6432 }, { "epoch": 0.8502511234469997, "grad_norm": 0.7988415453386185, "learning_rate": 1.6020208680667637e-06, "loss": 0.5596, "mean_token_accuracy": 0.8123397827148438, "num_tokens": 257681881.0, "step": 6433 }, { "epoch": 0.8503832936822627, "grad_norm": 0.8146678467906948, "learning_rate": 1.6009838508426062e-06, "loss": 0.6013, "mean_token_accuracy": 0.8003754019737244, "num_tokens": 257747417.0, "step": 6434 }, { "epoch": 0.8505154639175257, "grad_norm": 0.8110034329459527, "learning_rate": 1.5999476636375807e-06, "loss": 0.5331, "mean_token_accuracy": 0.8209162354469299, "num_tokens": 257812953.0, "step": 6435 }, { "epoch": 0.8506476341527888, "grad_norm": 0.8111460725032998, "learning_rate": 1.5989123066722716e-06, "loss": 0.5446, "mean_token_accuracy": 0.8174062967300415, "num_tokens": 257878489.0, "step": 6436 }, { "epoch": 0.8507798043880518, "grad_norm": 0.8276283306502427, "learning_rate": 1.5978777801670828e-06, "loss": 0.6187, "mean_token_accuracy": 0.7948358058929443, "num_tokens": 257944025.0, "step": 6437 }, { "epoch": 0.8509119746233148, "grad_norm": 0.8943412839017596, "learning_rate": 1.596844084342244e-06, "loss": 0.6916, "mean_token_accuracy": 0.7694268226623535, "num_tokens": 258009561.0, "step": 6438 }, { "epoch": 0.8510441448585778, "grad_norm": 0.852998247378773, "learning_rate": 1.5958112194178074e-06, "loss": 0.5974, "mean_token_accuracy": 0.8003754019737244, "num_tokens": 258075097.0, "step": 6439 }, { "epoch": 0.8511763150938408, "grad_norm": 0.7918752847080406, "learning_rate": 1.59477918561365e-06, "loss": 0.5956, "mean_token_accuracy": 0.7996429204940796, "num_tokens": 258140633.0, "step": 6440 }, { "epoch": 0.8513084853291039, "grad_norm": 0.7835121667042735, "learning_rate": 1.5937479831494676e-06, "loss": 0.5847, "mean_token_accuracy": 0.8035649061203003, "num_tokens": 258206169.0, "step": 6441 }, { "epoch": 0.8514406555643669, "grad_norm": 0.8783220547207291, "learning_rate": 1.592717612244783e-06, "loss": 0.6436, "mean_token_accuracy": 0.7856336236000061, "num_tokens": 258271705.0, "step": 6442 }, { "epoch": 0.8515728257996299, "grad_norm": 0.745379627780492, "learning_rate": 1.5916880731189406e-06, "loss": 0.5297, "mean_token_accuracy": 0.8236631751060486, "num_tokens": 258337241.0, "step": 6443 }, { "epoch": 0.8517049960348929, "grad_norm": 0.7626342551527977, "learning_rate": 1.5906593659911085e-06, "loss": 0.5414, "mean_token_accuracy": 0.8216792941093445, "num_tokens": 258402777.0, "step": 6444 }, { "epoch": 0.8518371662701559, "grad_norm": 0.8504515334349826, "learning_rate": 1.5896314910802744e-06, "loss": 0.66, "mean_token_accuracy": 0.7825509905815125, "num_tokens": 258468313.0, "step": 6445 }, { "epoch": 0.851969336505419, "grad_norm": 0.8021236886744628, "learning_rate": 1.5886044486052548e-06, "loss": 0.535, "mean_token_accuracy": 0.8200311064720154, "num_tokens": 258533849.0, "step": 6446 }, { "epoch": 0.852101506740682, "grad_norm": 0.777345152746412, "learning_rate": 1.5875782387846842e-06, "loss": 0.5677, "mean_token_accuracy": 0.8107526302337646, "num_tokens": 258599385.0, "step": 6447 }, { "epoch": 0.852233676975945, "grad_norm": 0.7785553371412508, "learning_rate": 1.5865528618370197e-06, "loss": 0.5428, "mean_token_accuracy": 0.8194817304611206, "num_tokens": 258664921.0, "step": 6448 }, { "epoch": 0.852365847211208, "grad_norm": 0.8076174297644941, "learning_rate": 1.585528317980545e-06, "loss": 0.6121, "mean_token_accuracy": 0.7962855696678162, "num_tokens": 258730457.0, "step": 6449 }, { "epoch": 0.852498017446471, "grad_norm": 0.8001210224189984, "learning_rate": 1.5845046074333636e-06, "loss": 0.5963, "mean_token_accuracy": 0.7997802495956421, "num_tokens": 258795993.0, "step": 6450 }, { "epoch": 0.852630187681734, "grad_norm": 0.7865512370254193, "learning_rate": 1.583481730413401e-06, "loss": 0.5494, "mean_token_accuracy": 0.816521167755127, "num_tokens": 258861529.0, "step": 6451 }, { "epoch": 0.8527623579169971, "grad_norm": 0.7751220417668337, "learning_rate": 1.5824596871384067e-06, "loss": 0.5687, "mean_token_accuracy": 0.8098828196525574, "num_tokens": 258927065.0, "step": 6452 }, { "epoch": 0.8528945281522601, "grad_norm": 0.7812786671502128, "learning_rate": 1.5814384778259526e-06, "loss": 0.5511, "mean_token_accuracy": 0.8163838386535645, "num_tokens": 258992601.0, "step": 6453 }, { "epoch": 0.8530266983875231, "grad_norm": 0.8114086725685176, "learning_rate": 1.5804181026934343e-06, "loss": 0.5902, "mean_token_accuracy": 0.8035801649093628, "num_tokens": 259058137.0, "step": 6454 }, { "epoch": 0.8531588686227861, "grad_norm": 0.7817832706910454, "learning_rate": 1.5793985619580655e-06, "loss": 0.5768, "mean_token_accuracy": 0.8057929277420044, "num_tokens": 259123673.0, "step": 6455 }, { "epoch": 0.8532910388580491, "grad_norm": 0.7884670641247282, "learning_rate": 1.5783798558368868e-06, "loss": 0.513, "mean_token_accuracy": 0.8308204412460327, "num_tokens": 259189209.0, "step": 6456 }, { "epoch": 0.8534232090933122, "grad_norm": 0.8153530376771448, "learning_rate": 1.5773619845467596e-06, "loss": 0.547, "mean_token_accuracy": 0.8179861903190613, "num_tokens": 259254745.0, "step": 6457 }, { "epoch": 0.8535553793285752, "grad_norm": 0.8094983566674093, "learning_rate": 1.5763449483043667e-06, "loss": 0.559, "mean_token_accuracy": 0.8131027817726135, "num_tokens": 259320281.0, "step": 6458 }, { "epoch": 0.8536875495638382, "grad_norm": 0.7776318740106727, "learning_rate": 1.5753287473262152e-06, "loss": 0.5265, "mean_token_accuracy": 0.8253875970840454, "num_tokens": 259385817.0, "step": 6459 }, { "epoch": 0.8538197197991012, "grad_norm": 0.7454322710714507, "learning_rate": 1.5743133818286312e-06, "loss": 0.5239, "mean_token_accuracy": 0.8265474438667297, "num_tokens": 259451353.0, "step": 6460 }, { "epoch": 0.8539518900343642, "grad_norm": 0.780939007695054, "learning_rate": 1.5732988520277659e-06, "loss": 0.5586, "mean_token_accuracy": 0.8142626285552979, "num_tokens": 259516889.0, "step": 6461 }, { "epoch": 0.8540840602696272, "grad_norm": 0.8031506269442018, "learning_rate": 1.5722851581395912e-06, "loss": 0.5835, "mean_token_accuracy": 0.8045568466186523, "num_tokens": 259582425.0, "step": 6462 }, { "epoch": 0.8542162305048903, "grad_norm": 0.7737826591166892, "learning_rate": 1.5712723003799027e-06, "loss": 0.5784, "mean_token_accuracy": 0.8079294562339783, "num_tokens": 259647961.0, "step": 6463 }, { "epoch": 0.8543484007401533, "grad_norm": 0.7010808937907067, "learning_rate": 1.5702602789643137e-06, "loss": 0.5442, "mean_token_accuracy": 0.8174062967300415, "num_tokens": 259713497.0, "step": 6464 }, { "epoch": 0.8544805709754163, "grad_norm": 0.7702797767488535, "learning_rate": 1.5692490941082664e-06, "loss": 0.5569, "mean_token_accuracy": 0.8140336871147156, "num_tokens": 259779033.0, "step": 6465 }, { "epoch": 0.8546127412106793, "grad_norm": 0.8342236783143446, "learning_rate": 1.5682387460270174e-06, "loss": 0.529, "mean_token_accuracy": 0.8214656114578247, "num_tokens": 259844569.0, "step": 6466 }, { "epoch": 0.8547449114459423, "grad_norm": 0.8223000187222295, "learning_rate": 1.56722923493565e-06, "loss": 0.5271, "mean_token_accuracy": 0.8235868811607361, "num_tokens": 259910105.0, "step": 6467 }, { "epoch": 0.8548770816812054, "grad_norm": 0.8048209114970898, "learning_rate": 1.5662205610490683e-06, "loss": 0.5809, "mean_token_accuracy": 0.8065407276153564, "num_tokens": 259975641.0, "step": 6468 }, { "epoch": 0.8550092519164684, "grad_norm": 0.8463071724571498, "learning_rate": 1.565212724581999e-06, "loss": 0.5807, "mean_token_accuracy": 0.808570384979248, "num_tokens": 260041177.0, "step": 6469 }, { "epoch": 0.8551414221517314, "grad_norm": 0.806136997716362, "learning_rate": 1.5642057257489867e-06, "loss": 0.5627, "mean_token_accuracy": 0.8133317232131958, "num_tokens": 260106713.0, "step": 6470 }, { "epoch": 0.8552735923869944, "grad_norm": 0.8941130453564531, "learning_rate": 1.5631995647644022e-06, "loss": 0.5951, "mean_token_accuracy": 0.803061306476593, "num_tokens": 260172249.0, "step": 6471 }, { "epoch": 0.8554057626222574, "grad_norm": 0.7762455385366542, "learning_rate": 1.562194241842436e-06, "loss": 0.557, "mean_token_accuracy": 0.8151629567146301, "num_tokens": 260237785.0, "step": 6472 }, { "epoch": 0.8555379328575204, "grad_norm": 0.8331880567492473, "learning_rate": 1.5611897571971014e-06, "loss": 0.6161, "mean_token_accuracy": 0.7921651601791382, "num_tokens": 260303321.0, "step": 6473 }, { "epoch": 0.8556701030927835, "grad_norm": 0.8432385153147218, "learning_rate": 1.5601861110422297e-06, "loss": 0.5681, "mean_token_accuracy": 0.8135758638381958, "num_tokens": 260368857.0, "step": 6474 }, { "epoch": 0.8558022733280465, "grad_norm": 0.8407643529290255, "learning_rate": 1.559183303591477e-06, "loss": 0.5848, "mean_token_accuracy": 0.8061439394950867, "num_tokens": 260434393.0, "step": 6475 }, { "epoch": 0.8559344435633095, "grad_norm": 0.8282439746399861, "learning_rate": 1.558181335058322e-06, "loss": 0.5747, "mean_token_accuracy": 0.8109968304634094, "num_tokens": 260499929.0, "step": 6476 }, { "epoch": 0.8560666137985725, "grad_norm": 0.7771612685812679, "learning_rate": 1.5571802056560607e-06, "loss": 0.5612, "mean_token_accuracy": 0.8138810992240906, "num_tokens": 260565465.0, "step": 6477 }, { "epoch": 0.8561987840338355, "grad_norm": 0.7802309291005022, "learning_rate": 1.5561799155978146e-06, "loss": 0.5452, "mean_token_accuracy": 0.8176199197769165, "num_tokens": 260631001.0, "step": 6478 }, { "epoch": 0.8563309542690986, "grad_norm": 0.8337500529131768, "learning_rate": 1.5551804650965225e-06, "loss": 0.5327, "mean_token_accuracy": 0.8233731985092163, "num_tokens": 260696537.0, "step": 6479 }, { "epoch": 0.8564631245043616, "grad_norm": 0.9096897552266252, "learning_rate": 1.5541818543649472e-06, "loss": 0.5707, "mean_token_accuracy": 0.8103253841400146, "num_tokens": 260762073.0, "step": 6480 }, { "epoch": 0.8565952947396246, "grad_norm": 0.7997646125865785, "learning_rate": 1.5531840836156729e-06, "loss": 0.5552, "mean_token_accuracy": 0.8153003454208374, "num_tokens": 260827609.0, "step": 6481 }, { "epoch": 0.8567274649748876, "grad_norm": 0.7969414617475254, "learning_rate": 1.5521871530611041e-06, "loss": 0.5767, "mean_token_accuracy": 0.8069527745246887, "num_tokens": 260893145.0, "step": 6482 }, { "epoch": 0.8568596352101506, "grad_norm": 0.7809487384488794, "learning_rate": 1.5511910629134647e-06, "loss": 0.5446, "mean_token_accuracy": 0.8203821182250977, "num_tokens": 260958681.0, "step": 6483 }, { "epoch": 0.8569918054454136, "grad_norm": 0.7566809754144629, "learning_rate": 1.5501958133848036e-06, "loss": 0.5724, "mean_token_accuracy": 0.8088451027870178, "num_tokens": 261024217.0, "step": 6484 }, { "epoch": 0.8571239756806767, "grad_norm": 0.7993871866569255, "learning_rate": 1.5492014046869877e-06, "loss": 0.5657, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 261089753.0, "step": 6485 }, { "epoch": 0.8572561459159397, "grad_norm": 0.7789282539370821, "learning_rate": 1.5482078370317066e-06, "loss": 0.544, "mean_token_accuracy": 0.8179709315299988, "num_tokens": 261155289.0, "step": 6486 }, { "epoch": 0.8573883161512027, "grad_norm": 0.8447928488476384, "learning_rate": 1.5472151106304677e-06, "loss": 0.6002, "mean_token_accuracy": 0.7998565435409546, "num_tokens": 261220825.0, "step": 6487 }, { "epoch": 0.8575204863864657, "grad_norm": 0.8725788625729163, "learning_rate": 1.5462232256946058e-06, "loss": 0.6249, "mean_token_accuracy": 0.7938591241836548, "num_tokens": 261286361.0, "step": 6488 }, { "epoch": 0.8576526566217287, "grad_norm": 0.8293721508172746, "learning_rate": 1.545232182435269e-06, "loss": 0.5578, "mean_token_accuracy": 0.8116377592086792, "num_tokens": 261351897.0, "step": 6489 }, { "epoch": 0.8577848268569918, "grad_norm": 0.7969866080990624, "learning_rate": 1.5442419810634312e-06, "loss": 0.5431, "mean_token_accuracy": 0.8196954131126404, "num_tokens": 261417433.0, "step": 6490 }, { "epoch": 0.8579169970922548, "grad_norm": 0.8129950940840838, "learning_rate": 1.5432526217898852e-06, "loss": 0.5929, "mean_token_accuracy": 0.8017183542251587, "num_tokens": 261482969.0, "step": 6491 }, { "epoch": 0.8580491673275178, "grad_norm": 0.7984113499121684, "learning_rate": 1.5422641048252463e-06, "loss": 0.5641, "mean_token_accuracy": 0.8134843111038208, "num_tokens": 261548505.0, "step": 6492 }, { "epoch": 0.8581813375627808, "grad_norm": 0.7828405359302332, "learning_rate": 1.5412764303799466e-06, "loss": 0.5914, "mean_token_accuracy": 0.8015505075454712, "num_tokens": 261614041.0, "step": 6493 }, { "epoch": 0.8583135077980438, "grad_norm": 0.7562457659627169, "learning_rate": 1.5402895986642425e-06, "loss": 0.5767, "mean_token_accuracy": 0.807761549949646, "num_tokens": 261679577.0, "step": 6494 }, { "epoch": 0.8584456780333068, "grad_norm": 0.8166380183671341, "learning_rate": 1.5393036098882105e-06, "loss": 0.5437, "mean_token_accuracy": 0.8198937773704529, "num_tokens": 261745113.0, "step": 6495 }, { "epoch": 0.8585778482685699, "grad_norm": 0.8057139538678697, "learning_rate": 1.5383184642617461e-06, "loss": 0.5545, "mean_token_accuracy": 0.8140336871147156, "num_tokens": 261810649.0, "step": 6496 }, { "epoch": 0.8587100185038329, "grad_norm": 0.7612356193228419, "learning_rate": 1.5373341619945681e-06, "loss": 0.5646, "mean_token_accuracy": 0.8129196763038635, "num_tokens": 261876185.0, "step": 6497 }, { "epoch": 0.8588421887390959, "grad_norm": 0.7763669231542261, "learning_rate": 1.536350703296211e-06, "loss": 0.5296, "mean_token_accuracy": 0.8241209983825684, "num_tokens": 261941721.0, "step": 6498 }, { "epoch": 0.8589743589743589, "grad_norm": 0.7972682787675769, "learning_rate": 1.5353680883760358e-06, "loss": 0.555, "mean_token_accuracy": 0.8156207799911499, "num_tokens": 262007257.0, "step": 6499 }, { "epoch": 0.8591065292096219, "grad_norm": 0.7867039621741176, "learning_rate": 1.5343863174432169e-06, "loss": 0.5703, "mean_token_accuracy": 0.8089519143104553, "num_tokens": 262072793.0, "step": 6500 }, { "epoch": 0.859238699444885, "grad_norm": 0.8213490770295432, "learning_rate": 1.5334053907067563e-06, "loss": 0.5863, "mean_token_accuracy": 0.8027255535125732, "num_tokens": 262138329.0, "step": 6501 }, { "epoch": 0.859370869680148, "grad_norm": 0.8385776969937971, "learning_rate": 1.5324253083754711e-06, "loss": 0.6251, "mean_token_accuracy": 0.7913563847541809, "num_tokens": 262203865.0, "step": 6502 }, { "epoch": 0.8595030399154111, "grad_norm": 0.8298242304393808, "learning_rate": 1.531446070658001e-06, "loss": 0.6024, "mean_token_accuracy": 0.7982084155082703, "num_tokens": 262269401.0, "step": 6503 }, { "epoch": 0.8596352101506741, "grad_norm": 0.7992868579852459, "learning_rate": 1.5304676777628042e-06, "loss": 0.5761, "mean_token_accuracy": 0.8092113137245178, "num_tokens": 262334937.0, "step": 6504 }, { "epoch": 0.8597673803859371, "grad_norm": 0.8024458762228699, "learning_rate": 1.5294901298981624e-06, "loss": 0.5702, "mean_token_accuracy": 0.8099133372306824, "num_tokens": 262400473.0, "step": 6505 }, { "epoch": 0.8598995506212002, "grad_norm": 0.8049053341169194, "learning_rate": 1.528513427272172e-06, "loss": 0.5797, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 262466009.0, "step": 6506 }, { "epoch": 0.8600317208564632, "grad_norm": 0.8058150442710469, "learning_rate": 1.5275375700927555e-06, "loss": 0.5576, "mean_token_accuracy": 0.8142473697662354, "num_tokens": 262531545.0, "step": 6507 }, { "epoch": 0.8601638910917262, "grad_norm": 0.9158599206020639, "learning_rate": 1.5265625585676508e-06, "loss": 0.6151, "mean_token_accuracy": 0.7993224263191223, "num_tokens": 262597081.0, "step": 6508 }, { "epoch": 0.8602960613269892, "grad_norm": 0.8416468924564843, "learning_rate": 1.525588392904418e-06, "loss": 0.5661, "mean_token_accuracy": 0.8125381469726562, "num_tokens": 262662617.0, "step": 6509 }, { "epoch": 0.8604282315622522, "grad_norm": 0.7500969752570453, "learning_rate": 1.5246150733104364e-06, "loss": 0.5178, "mean_token_accuracy": 0.8281192779541016, "num_tokens": 262728153.0, "step": 6510 }, { "epoch": 0.8605604017975153, "grad_norm": 0.7877633471401009, "learning_rate": 1.523642599992907e-06, "loss": 0.5892, "mean_token_accuracy": 0.8008179664611816, "num_tokens": 262793689.0, "step": 6511 }, { "epoch": 0.8606925720327783, "grad_norm": 0.8708914711441628, "learning_rate": 1.5226709731588459e-06, "loss": 0.6349, "mean_token_accuracy": 0.7906848788261414, "num_tokens": 262859225.0, "step": 6512 }, { "epoch": 0.8608247422680413, "grad_norm": 0.7639660364673077, "learning_rate": 1.5217001930150945e-06, "loss": 0.5602, "mean_token_accuracy": 0.8126754760742188, "num_tokens": 262924761.0, "step": 6513 }, { "epoch": 0.8609569125033043, "grad_norm": 0.8253443073225231, "learning_rate": 1.5207302597683113e-06, "loss": 0.5984, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 262990297.0, "step": 6514 }, { "epoch": 0.8610890827385673, "grad_norm": 0.9034506486464982, "learning_rate": 1.5197611736249758e-06, "loss": 0.6713, "mean_token_accuracy": 0.7775607109069824, "num_tokens": 263055833.0, "step": 6515 }, { "epoch": 0.8612212529738303, "grad_norm": 0.7721483882765338, "learning_rate": 1.5187929347913835e-06, "loss": 0.5961, "mean_token_accuracy": 0.8045415878295898, "num_tokens": 263121369.0, "step": 6516 }, { "epoch": 0.8613534232090934, "grad_norm": 0.8223397061106715, "learning_rate": 1.517825543473654e-06, "loss": 0.5769, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 263186905.0, "step": 6517 }, { "epoch": 0.8614855934443564, "grad_norm": 0.7430151769683621, "learning_rate": 1.516858999877725e-06, "loss": 0.5236, "mean_token_accuracy": 0.8245483040809631, "num_tokens": 263252441.0, "step": 6518 }, { "epoch": 0.8616177636796194, "grad_norm": 0.7425397980574664, "learning_rate": 1.5158933042093512e-06, "loss": 0.5224, "mean_token_accuracy": 0.8248077034950256, "num_tokens": 263317977.0, "step": 6519 }, { "epoch": 0.8617499339148824, "grad_norm": 0.7611263443594267, "learning_rate": 1.5149284566741133e-06, "loss": 0.5558, "mean_token_accuracy": 0.8147509694099426, "num_tokens": 263383513.0, "step": 6520 }, { "epoch": 0.8618821041501454, "grad_norm": 0.8422497150982801, "learning_rate": 1.5139644574774032e-06, "loss": 0.5799, "mean_token_accuracy": 0.8083414435386658, "num_tokens": 263449049.0, "step": 6521 }, { "epoch": 0.8620142743854085, "grad_norm": 0.8218721043230718, "learning_rate": 1.513001306824438e-06, "loss": 0.5918, "mean_token_accuracy": 0.8039769530296326, "num_tokens": 263514585.0, "step": 6522 }, { "epoch": 0.8621464446206715, "grad_norm": 0.8810764015105534, "learning_rate": 1.5120390049202512e-06, "loss": 0.6066, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 263580121.0, "step": 6523 }, { "epoch": 0.8622786148559345, "grad_norm": 0.7939679479875823, "learning_rate": 1.511077551969699e-06, "loss": 0.5674, "mean_token_accuracy": 0.8127976059913635, "num_tokens": 263645657.0, "step": 6524 }, { "epoch": 0.8624107850911975, "grad_norm": 0.7737827315951957, "learning_rate": 1.5101169481774522e-06, "loss": 0.5475, "mean_token_accuracy": 0.8148272633552551, "num_tokens": 263711193.0, "step": 6525 }, { "epoch": 0.8625429553264605, "grad_norm": 0.8235624643827733, "learning_rate": 1.5091571937480052e-06, "loss": 0.5988, "mean_token_accuracy": 0.8012147545814514, "num_tokens": 263776729.0, "step": 6526 }, { "epoch": 0.8626751255617235, "grad_norm": 0.8596772965294377, "learning_rate": 1.5081982888856684e-06, "loss": 0.6141, "mean_token_accuracy": 0.7953851819038391, "num_tokens": 263842265.0, "step": 6527 }, { "epoch": 0.8628072957969866, "grad_norm": 0.8347730737770209, "learning_rate": 1.507240233794573e-06, "loss": 0.6181, "mean_token_accuracy": 0.7958735227584839, "num_tokens": 263907801.0, "step": 6528 }, { "epoch": 0.8629394660322496, "grad_norm": 0.7904018562388484, "learning_rate": 1.506283028678669e-06, "loss": 0.5613, "mean_token_accuracy": 0.8140184283256531, "num_tokens": 263973337.0, "step": 6529 }, { "epoch": 0.8630716362675126, "grad_norm": 0.8179462382135037, "learning_rate": 1.5053266737417265e-06, "loss": 0.6078, "mean_token_accuracy": 0.798757791519165, "num_tokens": 264038873.0, "step": 6530 }, { "epoch": 0.8632038065027756, "grad_norm": 0.7960803158384995, "learning_rate": 1.5043711691873312e-06, "loss": 0.55, "mean_token_accuracy": 0.8157886862754822, "num_tokens": 264104409.0, "step": 6531 }, { "epoch": 0.8633359767380386, "grad_norm": 0.813783064612544, "learning_rate": 1.5034165152188913e-06, "loss": 0.532, "mean_token_accuracy": 0.821572482585907, "num_tokens": 264169945.0, "step": 6532 }, { "epoch": 0.8634681469733017, "grad_norm": 0.8771474833012244, "learning_rate": 1.5024627120396323e-06, "loss": 0.5779, "mean_token_accuracy": 0.8092418313026428, "num_tokens": 264235481.0, "step": 6533 }, { "epoch": 0.8636003172085647, "grad_norm": 0.8426803310751914, "learning_rate": 1.5015097598526008e-06, "loss": 0.6247, "mean_token_accuracy": 0.7942864298820496, "num_tokens": 264301017.0, "step": 6534 }, { "epoch": 0.8637324874438277, "grad_norm": 0.8192995631656836, "learning_rate": 1.5005576588606583e-06, "loss": 0.5814, "mean_token_accuracy": 0.8055793046951294, "num_tokens": 264366553.0, "step": 6535 }, { "epoch": 0.8638646576790907, "grad_norm": 0.8023036220481021, "learning_rate": 1.4996064092664874e-06, "loss": 0.572, "mean_token_accuracy": 0.8095470666885376, "num_tokens": 264432089.0, "step": 6536 }, { "epoch": 0.8639968279143537, "grad_norm": 0.7647611449788689, "learning_rate": 1.4986560112725906e-06, "loss": 0.5624, "mean_token_accuracy": 0.8108289837837219, "num_tokens": 264497625.0, "step": 6537 }, { "epoch": 0.8641289981496167, "grad_norm": 0.8250602777913403, "learning_rate": 1.497706465081285e-06, "loss": 0.6027, "mean_token_accuracy": 0.802099883556366, "num_tokens": 264563161.0, "step": 6538 }, { "epoch": 0.8642611683848798, "grad_norm": 0.8177112427262179, "learning_rate": 1.4967577708947134e-06, "loss": 0.5733, "mean_token_accuracy": 0.8069985508918762, "num_tokens": 264628697.0, "step": 6539 }, { "epoch": 0.8643933386201428, "grad_norm": 0.8832913157460722, "learning_rate": 1.4958099289148298e-06, "loss": 0.5923, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 264694233.0, "step": 6540 }, { "epoch": 0.8645255088554058, "grad_norm": 0.7555602361409239, "learning_rate": 1.4948629393434105e-06, "loss": 0.4942, "mean_token_accuracy": 0.8366194367408752, "num_tokens": 264759769.0, "step": 6541 }, { "epoch": 0.8646576790906688, "grad_norm": 0.7223636183320865, "learning_rate": 1.4939168023820506e-06, "loss": 0.5059, "mean_token_accuracy": 0.8296605944633484, "num_tokens": 264825305.0, "step": 6542 }, { "epoch": 0.8647898493259318, "grad_norm": 0.8173851024999326, "learning_rate": 1.4929715182321628e-06, "loss": 0.5473, "mean_token_accuracy": 0.8190086483955383, "num_tokens": 264890841.0, "step": 6543 }, { "epoch": 0.8649220195611949, "grad_norm": 0.7948092778431036, "learning_rate": 1.4920270870949773e-06, "loss": 0.5743, "mean_token_accuracy": 0.808753490447998, "num_tokens": 264956377.0, "step": 6544 }, { "epoch": 0.8650541897964579, "grad_norm": 0.8627545761356333, "learning_rate": 1.4910835091715447e-06, "loss": 0.5918, "mean_token_accuracy": 0.8041142821311951, "num_tokens": 265021913.0, "step": 6545 }, { "epoch": 0.8651863600317209, "grad_norm": 0.8836711213368457, "learning_rate": 1.490140784662733e-06, "loss": 0.6018, "mean_token_accuracy": 0.7991698384284973, "num_tokens": 265087449.0, "step": 6546 }, { "epoch": 0.8653185302669839, "grad_norm": 0.8065548670852827, "learning_rate": 1.4891989137692292e-06, "loss": 0.5779, "mean_token_accuracy": 0.8047246932983398, "num_tokens": 265152985.0, "step": 6547 }, { "epoch": 0.8654507005022469, "grad_norm": 0.7731530436750083, "learning_rate": 1.4882578966915356e-06, "loss": 0.5791, "mean_token_accuracy": 0.8067848682403564, "num_tokens": 265218521.0, "step": 6548 }, { "epoch": 0.86558287073751, "grad_norm": 0.7293036728763306, "learning_rate": 1.4873177336299782e-06, "loss": 0.4901, "mean_token_accuracy": 0.8343914151191711, "num_tokens": 265284057.0, "step": 6549 }, { "epoch": 0.865715040972773, "grad_norm": 0.81721920429129, "learning_rate": 1.4863784247846955e-06, "loss": 0.5748, "mean_token_accuracy": 0.8091807961463928, "num_tokens": 265349593.0, "step": 6550 }, { "epoch": 0.865847211208036, "grad_norm": 0.7627519913029733, "learning_rate": 1.4854399703556475e-06, "loss": 0.5453, "mean_token_accuracy": 0.8187797665596008, "num_tokens": 265415129.0, "step": 6551 }, { "epoch": 0.865979381443299, "grad_norm": 0.8359453034461124, "learning_rate": 1.4845023705426125e-06, "loss": 0.5886, "mean_token_accuracy": 0.8051214814186096, "num_tokens": 265480665.0, "step": 6552 }, { "epoch": 0.866111551678562, "grad_norm": 0.8652784764774334, "learning_rate": 1.4835656255451858e-06, "loss": 0.6042, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 265546201.0, "step": 6553 }, { "epoch": 0.866243721913825, "grad_norm": 0.7618293088907018, "learning_rate": 1.4826297355627795e-06, "loss": 0.5452, "mean_token_accuracy": 0.8183372020721436, "num_tokens": 265611737.0, "step": 6554 }, { "epoch": 0.8663758921490881, "grad_norm": 0.8277837538284488, "learning_rate": 1.481694700794626e-06, "loss": 0.5755, "mean_token_accuracy": 0.808356761932373, "num_tokens": 265677273.0, "step": 6555 }, { "epoch": 0.8665080623843511, "grad_norm": 0.8209859034779428, "learning_rate": 1.4807605214397752e-06, "loss": 0.5755, "mean_token_accuracy": 0.807822585105896, "num_tokens": 265742809.0, "step": 6556 }, { "epoch": 0.8666402326196141, "grad_norm": 0.8511824867010205, "learning_rate": 1.479827197697092e-06, "loss": 0.5977, "mean_token_accuracy": 0.8017183542251587, "num_tokens": 265808345.0, "step": 6557 }, { "epoch": 0.8667724028548771, "grad_norm": 0.8010774311962501, "learning_rate": 1.478894729765265e-06, "loss": 0.5617, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 265873881.0, "step": 6558 }, { "epoch": 0.8669045730901401, "grad_norm": 0.8422257730484674, "learning_rate": 1.4779631178427944e-06, "loss": 0.5887, "mean_token_accuracy": 0.8052282929420471, "num_tokens": 265939417.0, "step": 6559 }, { "epoch": 0.8670367433254031, "grad_norm": 0.8431322926522828, "learning_rate": 1.4770323621280024e-06, "loss": 0.5958, "mean_token_accuracy": 0.8010316491127014, "num_tokens": 266004953.0, "step": 6560 }, { "epoch": 0.8671689135606662, "grad_norm": 0.8021582750987519, "learning_rate": 1.476102462819026e-06, "loss": 0.5547, "mean_token_accuracy": 0.8147051930427551, "num_tokens": 266070489.0, "step": 6561 }, { "epoch": 0.8673010837959292, "grad_norm": 0.7598730961576724, "learning_rate": 1.4751734201138237e-06, "loss": 0.5582, "mean_token_accuracy": 0.8105695247650146, "num_tokens": 266136025.0, "step": 6562 }, { "epoch": 0.8674332540311922, "grad_norm": 0.8346805156784197, "learning_rate": 1.4742452342101673e-06, "loss": 0.5944, "mean_token_accuracy": 0.8012910485267639, "num_tokens": 266201561.0, "step": 6563 }, { "epoch": 0.8675654242664552, "grad_norm": 0.8334554926059993, "learning_rate": 1.473317905305649e-06, "loss": 0.5492, "mean_token_accuracy": 0.8186424374580383, "num_tokens": 266267097.0, "step": 6564 }, { "epoch": 0.8676975945017182, "grad_norm": 0.7785276430435131, "learning_rate": 1.4723914335976769e-06, "loss": 0.5913, "mean_token_accuracy": 0.8053656220436096, "num_tokens": 266332633.0, "step": 6565 }, { "epoch": 0.8678297647369813, "grad_norm": 0.7552597469798549, "learning_rate": 1.47146581928348e-06, "loss": 0.5439, "mean_token_accuracy": 0.8182151317596436, "num_tokens": 266398169.0, "step": 6566 }, { "epoch": 0.8679619349722443, "grad_norm": 0.7643374380481157, "learning_rate": 1.4705410625600978e-06, "loss": 0.5605, "mean_token_accuracy": 0.8151935338973999, "num_tokens": 266463705.0, "step": 6567 }, { "epoch": 0.8680941052075073, "grad_norm": 0.7955104891268239, "learning_rate": 1.4696171636243965e-06, "loss": 0.5418, "mean_token_accuracy": 0.8197259306907654, "num_tokens": 266529241.0, "step": 6568 }, { "epoch": 0.8682262754427703, "grad_norm": 0.8241679454032553, "learning_rate": 1.4686941226730516e-06, "loss": 0.6354, "mean_token_accuracy": 0.7900439500808716, "num_tokens": 266594777.0, "step": 6569 }, { "epoch": 0.8683584456780333, "grad_norm": 0.8758293336309075, "learning_rate": 1.4677719399025607e-06, "loss": 0.5823, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 266660313.0, "step": 6570 }, { "epoch": 0.8684906159132963, "grad_norm": 0.8617464919179812, "learning_rate": 1.4668506155092366e-06, "loss": 0.6116, "mean_token_accuracy": 0.7943474650382996, "num_tokens": 266725849.0, "step": 6571 }, { "epoch": 0.8686227861485594, "grad_norm": 0.7668179506466292, "learning_rate": 1.4659301496892114e-06, "loss": 0.5417, "mean_token_accuracy": 0.8177115321159363, "num_tokens": 266791385.0, "step": 6572 }, { "epoch": 0.8687549563838224, "grad_norm": 0.7732843437172159, "learning_rate": 1.4650105426384317e-06, "loss": 0.5436, "mean_token_accuracy": 0.8161396384239197, "num_tokens": 266856921.0, "step": 6573 }, { "epoch": 0.8688871266190854, "grad_norm": 0.7741240329976211, "learning_rate": 1.4640917945526625e-06, "loss": 0.5514, "mean_token_accuracy": 0.818169355392456, "num_tokens": 266922457.0, "step": 6574 }, { "epoch": 0.8690192968543484, "grad_norm": 0.8065995444737314, "learning_rate": 1.4631739056274868e-06, "loss": 0.5611, "mean_token_accuracy": 0.8108595013618469, "num_tokens": 266987993.0, "step": 6575 }, { "epoch": 0.8691514670896114, "grad_norm": 0.7924616329740825, "learning_rate": 1.4622568760583036e-06, "loss": 0.5291, "mean_token_accuracy": 0.8242583274841309, "num_tokens": 267053529.0, "step": 6576 }, { "epoch": 0.8692836373248745, "grad_norm": 0.8205760705556145, "learning_rate": 1.4613407060403295e-06, "loss": 0.5884, "mean_token_accuracy": 0.8013063073158264, "num_tokens": 267119065.0, "step": 6577 }, { "epoch": 0.8694158075601375, "grad_norm": 0.7689311924912573, "learning_rate": 1.4604253957685978e-06, "loss": 0.5525, "mean_token_accuracy": 0.8178946375846863, "num_tokens": 267184601.0, "step": 6578 }, { "epoch": 0.8695479777954005, "grad_norm": 0.8634079783914436, "learning_rate": 1.45951094543796e-06, "loss": 0.6305, "mean_token_accuracy": 0.7899676561355591, "num_tokens": 267250137.0, "step": 6579 }, { "epoch": 0.8696801480306635, "grad_norm": 0.8690700759198122, "learning_rate": 1.4585973552430804e-06, "loss": 0.5698, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 267315673.0, "step": 6580 }, { "epoch": 0.8698123182659265, "grad_norm": 0.8574616312494692, "learning_rate": 1.4576846253784476e-06, "loss": 0.662, "mean_token_accuracy": 0.7808417677879333, "num_tokens": 267381209.0, "step": 6581 }, { "epoch": 0.8699444885011896, "grad_norm": 0.791033016778154, "learning_rate": 1.4567727560383598e-06, "loss": 0.6062, "mean_token_accuracy": 0.7981473803520203, "num_tokens": 267446745.0, "step": 6582 }, { "epoch": 0.8700766587364526, "grad_norm": 0.861769335514247, "learning_rate": 1.4558617474169356e-06, "loss": 0.6062, "mean_token_accuracy": 0.797887921333313, "num_tokens": 267512281.0, "step": 6583 }, { "epoch": 0.8702088289717156, "grad_norm": 0.8622998001454932, "learning_rate": 1.4549515997081101e-06, "loss": 0.5999, "mean_token_accuracy": 0.8003296256065369, "num_tokens": 267577817.0, "step": 6584 }, { "epoch": 0.8703409992069786, "grad_norm": 0.9603714953809513, "learning_rate": 1.4540423131056352e-06, "loss": 0.602, "mean_token_accuracy": 0.7993987202644348, "num_tokens": 267643353.0, "step": 6585 }, { "epoch": 0.8704731694422416, "grad_norm": 0.7984680097496102, "learning_rate": 1.4531338878030759e-06, "loss": 0.5473, "mean_token_accuracy": 0.816399097442627, "num_tokens": 267708889.0, "step": 6586 }, { "epoch": 0.8706053396775046, "grad_norm": 0.8635397794861196, "learning_rate": 1.4522263239938222e-06, "loss": 0.5955, "mean_token_accuracy": 0.798879861831665, "num_tokens": 267774425.0, "step": 6587 }, { "epoch": 0.8707375099127677, "grad_norm": 0.7630164676144512, "learning_rate": 1.4513196218710712e-06, "loss": 0.5368, "mean_token_accuracy": 0.8192223310470581, "num_tokens": 267839961.0, "step": 6588 }, { "epoch": 0.8708696801480307, "grad_norm": 0.8062074678269013, "learning_rate": 1.4504137816278435e-06, "loss": 0.5935, "mean_token_accuracy": 0.8007721900939941, "num_tokens": 267905497.0, "step": 6589 }, { "epoch": 0.8710018503832937, "grad_norm": 0.7906116171523819, "learning_rate": 1.4495088034569712e-06, "loss": 0.5349, "mean_token_accuracy": 0.8210383057594299, "num_tokens": 267971033.0, "step": 6590 }, { "epoch": 0.8711340206185567, "grad_norm": 0.8676405919401048, "learning_rate": 1.4486046875511067e-06, "loss": 0.5792, "mean_token_accuracy": 0.8103101253509521, "num_tokens": 268036569.0, "step": 6591 }, { "epoch": 0.8712661908538197, "grad_norm": 0.7799444100742132, "learning_rate": 1.4477014341027176e-06, "loss": 0.5308, "mean_token_accuracy": 0.8247619271278381, "num_tokens": 268102105.0, "step": 6592 }, { "epoch": 0.8713983610890828, "grad_norm": 0.8061497297449536, "learning_rate": 1.4467990433040866e-06, "loss": 0.6197, "mean_token_accuracy": 0.7953241467475891, "num_tokens": 268167641.0, "step": 6593 }, { "epoch": 0.8715305313243458, "grad_norm": 0.8100517386298802, "learning_rate": 1.4458975153473157e-06, "loss": 0.5587, "mean_token_accuracy": 0.8151782155036926, "num_tokens": 268233177.0, "step": 6594 }, { "epoch": 0.8716627015596088, "grad_norm": 0.809170501891993, "learning_rate": 1.4449968504243194e-06, "loss": 0.5666, "mean_token_accuracy": 0.8149035573005676, "num_tokens": 268298713.0, "step": 6595 }, { "epoch": 0.8717948717948718, "grad_norm": 0.8143001288777496, "learning_rate": 1.444097048726831e-06, "loss": 0.5671, "mean_token_accuracy": 0.8105237483978271, "num_tokens": 268364249.0, "step": 6596 }, { "epoch": 0.8719270420301348, "grad_norm": 0.8124714854023948, "learning_rate": 1.4431981104463998e-06, "loss": 0.5781, "mean_token_accuracy": 0.8059150576591492, "num_tokens": 268429785.0, "step": 6597 }, { "epoch": 0.8720592122653978, "grad_norm": 0.8721252285369316, "learning_rate": 1.4423000357743926e-06, "loss": 0.6133, "mean_token_accuracy": 0.7964076399803162, "num_tokens": 268495321.0, "step": 6598 }, { "epoch": 0.8721913825006609, "grad_norm": 0.8025793771713866, "learning_rate": 1.4414028249019865e-06, "loss": 0.5748, "mean_token_accuracy": 0.8073037266731262, "num_tokens": 268560857.0, "step": 6599 }, { "epoch": 0.8723235527359239, "grad_norm": 0.7443268694029903, "learning_rate": 1.4405064780201835e-06, "loss": 0.5149, "mean_token_accuracy": 0.8253570795059204, "num_tokens": 268626393.0, "step": 6600 }, { "epoch": 0.8724557229711869, "grad_norm": 0.7906849589560694, "learning_rate": 1.4396109953197945e-06, "loss": 0.5855, "mean_token_accuracy": 0.8045110702514648, "num_tokens": 268691929.0, "step": 6601 }, { "epoch": 0.8725878932064499, "grad_norm": 0.8650824591748039, "learning_rate": 1.43871637699145e-06, "loss": 0.6073, "mean_token_accuracy": 0.7973995804786682, "num_tokens": 268757465.0, "step": 6602 }, { "epoch": 0.8727200634417129, "grad_norm": 0.8216586791858299, "learning_rate": 1.4378226232255948e-06, "loss": 0.6123, "mean_token_accuracy": 0.7960566282272339, "num_tokens": 268823001.0, "step": 6603 }, { "epoch": 0.872852233676976, "grad_norm": 0.8402364026631035, "learning_rate": 1.4369297342124918e-06, "loss": 0.5862, "mean_token_accuracy": 0.806525468826294, "num_tokens": 268888537.0, "step": 6604 }, { "epoch": 0.872984403912239, "grad_norm": 0.8236049200884591, "learning_rate": 1.4360377101422171e-06, "loss": 0.5793, "mean_token_accuracy": 0.8092570900917053, "num_tokens": 268954073.0, "step": 6605 }, { "epoch": 0.873116574147502, "grad_norm": 0.8296515399796852, "learning_rate": 1.4351465512046644e-06, "loss": 0.5916, "mean_token_accuracy": 0.803213894367218, "num_tokens": 269019609.0, "step": 6606 }, { "epoch": 0.873248744382765, "grad_norm": 0.8119084184063292, "learning_rate": 1.4342562575895432e-06, "loss": 0.5761, "mean_token_accuracy": 0.8060676455497742, "num_tokens": 269085145.0, "step": 6607 }, { "epoch": 0.873380914618028, "grad_norm": 0.8118379957770522, "learning_rate": 1.4333668294863784e-06, "loss": 0.5913, "mean_token_accuracy": 0.8012910485267639, "num_tokens": 269150681.0, "step": 6608 }, { "epoch": 0.873513084853291, "grad_norm": 0.7999326069871909, "learning_rate": 1.43247826708451e-06, "loss": 0.5122, "mean_token_accuracy": 0.8279514312744141, "num_tokens": 269216217.0, "step": 6609 }, { "epoch": 0.8736452550885541, "grad_norm": 0.7765060742636605, "learning_rate": 1.4315905705730947e-06, "loss": 0.5365, "mean_token_accuracy": 0.8235411047935486, "num_tokens": 269281753.0, "step": 6610 }, { "epoch": 0.8737774253238171, "grad_norm": 0.7814454949695459, "learning_rate": 1.430703740141105e-06, "loss": 0.5624, "mean_token_accuracy": 0.812095582485199, "num_tokens": 269347289.0, "step": 6611 }, { "epoch": 0.8739095955590801, "grad_norm": 0.8257082171697214, "learning_rate": 1.4298177759773276e-06, "loss": 0.6118, "mean_token_accuracy": 0.7964839339256287, "num_tokens": 269412825.0, "step": 6612 }, { "epoch": 0.8740417657943431, "grad_norm": 0.8083688537056856, "learning_rate": 1.4289326782703676e-06, "loss": 0.5937, "mean_token_accuracy": 0.8017488718032837, "num_tokens": 269478361.0, "step": 6613 }, { "epoch": 0.8741739360296061, "grad_norm": 0.795198667442434, "learning_rate": 1.4280484472086418e-06, "loss": 0.5832, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 269543897.0, "step": 6614 }, { "epoch": 0.8743061062648692, "grad_norm": 0.846208950859376, "learning_rate": 1.4271650829803848e-06, "loss": 0.6523, "mean_token_accuracy": 0.7841075658798218, "num_tokens": 269609433.0, "step": 6615 }, { "epoch": 0.8744382765001322, "grad_norm": 0.7856221634689429, "learning_rate": 1.426282585773648e-06, "loss": 0.537, "mean_token_accuracy": 0.8194206953048706, "num_tokens": 269674969.0, "step": 6616 }, { "epoch": 0.8745704467353952, "grad_norm": 0.8022048149157275, "learning_rate": 1.4254009557762954e-06, "loss": 0.5535, "mean_token_accuracy": 0.8151324391365051, "num_tokens": 269740505.0, "step": 6617 }, { "epoch": 0.8747026169706582, "grad_norm": 0.766889538936138, "learning_rate": 1.4245201931760066e-06, "loss": 0.5543, "mean_token_accuracy": 0.8149798512458801, "num_tokens": 269806041.0, "step": 6618 }, { "epoch": 0.8748347872059212, "grad_norm": 0.8077952242207236, "learning_rate": 1.4236402981602807e-06, "loss": 0.5713, "mean_token_accuracy": 0.8089671730995178, "num_tokens": 269871577.0, "step": 6619 }, { "epoch": 0.8749669574411842, "grad_norm": 0.8548067921208027, "learning_rate": 1.4227612709164262e-06, "loss": 0.6464, "mean_token_accuracy": 0.7842601537704468, "num_tokens": 269937113.0, "step": 6620 }, { "epoch": 0.8750991276764473, "grad_norm": 0.8704265491427039, "learning_rate": 1.4218831116315706e-06, "loss": 0.599, "mean_token_accuracy": 0.7995513081550598, "num_tokens": 270002649.0, "step": 6621 }, { "epoch": 0.8752312979117103, "grad_norm": 0.7512248251182502, "learning_rate": 1.4210058204926547e-06, "loss": 0.5411, "mean_token_accuracy": 0.8199700713157654, "num_tokens": 270068185.0, "step": 6622 }, { "epoch": 0.8753634681469733, "grad_norm": 0.8601485895741318, "learning_rate": 1.4201293976864378e-06, "loss": 0.5748, "mean_token_accuracy": 0.8080210089683533, "num_tokens": 270133721.0, "step": 6623 }, { "epoch": 0.8754956383822363, "grad_norm": 0.8094324055777966, "learning_rate": 1.4192538433994896e-06, "loss": 0.6048, "mean_token_accuracy": 0.8001312613487244, "num_tokens": 270199257.0, "step": 6624 }, { "epoch": 0.8756278086174993, "grad_norm": 0.7604311319144385, "learning_rate": 1.4183791578181986e-06, "loss": 0.5531, "mean_token_accuracy": 0.8159412741661072, "num_tokens": 270264793.0, "step": 6625 }, { "epoch": 0.8757599788527624, "grad_norm": 0.8688435681204334, "learning_rate": 1.4175053411287661e-06, "loss": 0.5603, "mean_token_accuracy": 0.8117445707321167, "num_tokens": 270330329.0, "step": 6626 }, { "epoch": 0.8758921490880254, "grad_norm": 0.8111043144755401, "learning_rate": 1.4166323935172117e-06, "loss": 0.5819, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 270395865.0, "step": 6627 }, { "epoch": 0.8760243193232884, "grad_norm": 0.7866808450683196, "learning_rate": 1.4157603151693653e-06, "loss": 0.5336, "mean_token_accuracy": 0.8243498802185059, "num_tokens": 270461401.0, "step": 6628 }, { "epoch": 0.8761564895585514, "grad_norm": 0.8289103308834377, "learning_rate": 1.4148891062708747e-06, "loss": 0.592, "mean_token_accuracy": 0.8057471513748169, "num_tokens": 270526937.0, "step": 6629 }, { "epoch": 0.8762886597938144, "grad_norm": 0.8795492763389721, "learning_rate": 1.4140187670072022e-06, "loss": 0.6047, "mean_token_accuracy": 0.7999023199081421, "num_tokens": 270592473.0, "step": 6630 }, { "epoch": 0.8764208300290774, "grad_norm": 0.8657272116663467, "learning_rate": 1.4131492975636255e-06, "loss": 0.5779, "mean_token_accuracy": 0.8056555986404419, "num_tokens": 270658009.0, "step": 6631 }, { "epoch": 0.8765530002643405, "grad_norm": 0.8022861935133521, "learning_rate": 1.4122806981252362e-06, "loss": 0.5418, "mean_token_accuracy": 0.818199872970581, "num_tokens": 270723545.0, "step": 6632 }, { "epoch": 0.8766851704996035, "grad_norm": 0.8043017870276415, "learning_rate": 1.411412968876941e-06, "loss": 0.5206, "mean_token_accuracy": 0.8252960443496704, "num_tokens": 270789081.0, "step": 6633 }, { "epoch": 0.8768173407348665, "grad_norm": 0.8129367397269824, "learning_rate": 1.4105461100034606e-06, "loss": 0.5888, "mean_token_accuracy": 0.8044958114624023, "num_tokens": 270854617.0, "step": 6634 }, { "epoch": 0.8769495109701295, "grad_norm": 0.7470400398678397, "learning_rate": 1.409680121689332e-06, "loss": 0.5276, "mean_token_accuracy": 0.8248687386512756, "num_tokens": 270920153.0, "step": 6635 }, { "epoch": 0.8770816812053925, "grad_norm": 0.7583655557824246, "learning_rate": 1.4088150041189065e-06, "loss": 0.5308, "mean_token_accuracy": 0.8247314095497131, "num_tokens": 270985689.0, "step": 6636 }, { "epoch": 0.8772138514406556, "grad_norm": 0.830370791538257, "learning_rate": 1.4079507574763477e-06, "loss": 0.5852, "mean_token_accuracy": 0.8070290684700012, "num_tokens": 271051225.0, "step": 6637 }, { "epoch": 0.8773460216759186, "grad_norm": 0.7931007547511931, "learning_rate": 1.407087381945637e-06, "loss": 0.5789, "mean_token_accuracy": 0.8068153858184814, "num_tokens": 271116761.0, "step": 6638 }, { "epoch": 0.8774781919111816, "grad_norm": 0.8155391067033279, "learning_rate": 1.406224877710569e-06, "loss": 0.5749, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 271182297.0, "step": 6639 }, { "epoch": 0.8776103621464446, "grad_norm": 0.7770413395911244, "learning_rate": 1.4053632449547533e-06, "loss": 0.5377, "mean_token_accuracy": 0.821786105632782, "num_tokens": 271247833.0, "step": 6640 }, { "epoch": 0.8777425323817076, "grad_norm": 0.8241986988632913, "learning_rate": 1.404502483861611e-06, "loss": 0.5781, "mean_token_accuracy": 0.8062812685966492, "num_tokens": 271313369.0, "step": 6641 }, { "epoch": 0.8778747026169706, "grad_norm": 0.7863363418042457, "learning_rate": 1.4036425946143838e-06, "loss": 0.5765, "mean_token_accuracy": 0.8073037266731262, "num_tokens": 271378905.0, "step": 6642 }, { "epoch": 0.8780068728522337, "grad_norm": 0.8898305172378742, "learning_rate": 1.4027835773961217e-06, "loss": 0.6603, "mean_token_accuracy": 0.7797430157661438, "num_tokens": 271444441.0, "step": 6643 }, { "epoch": 0.8781390430874967, "grad_norm": 0.9253960993126582, "learning_rate": 1.401925432389693e-06, "loss": 0.6558, "mean_token_accuracy": 0.7806128859519958, "num_tokens": 271509977.0, "step": 6644 }, { "epoch": 0.8782712133227597, "grad_norm": 0.8338103951590464, "learning_rate": 1.4010681597777772e-06, "loss": 0.6106, "mean_token_accuracy": 0.7988646030426025, "num_tokens": 271575513.0, "step": 6645 }, { "epoch": 0.8784033835580227, "grad_norm": 0.8677291898814928, "learning_rate": 1.4002117597428715e-06, "loss": 0.6279, "mean_token_accuracy": 0.7895861268043518, "num_tokens": 271641049.0, "step": 6646 }, { "epoch": 0.8785355537932857, "grad_norm": 0.821047797139435, "learning_rate": 1.3993562324672843e-06, "loss": 0.5935, "mean_token_accuracy": 0.8035954236984253, "num_tokens": 271706585.0, "step": 6647 }, { "epoch": 0.8786677240285488, "grad_norm": 0.7970723194166766, "learning_rate": 1.3985015781331396e-06, "loss": 0.5705, "mean_token_accuracy": 0.8094097375869751, "num_tokens": 271772121.0, "step": 6648 }, { "epoch": 0.8787998942638118, "grad_norm": 0.780294547025239, "learning_rate": 1.3976477969223762e-06, "loss": 0.5582, "mean_token_accuracy": 0.8138658404350281, "num_tokens": 271837657.0, "step": 6649 }, { "epoch": 0.8789320644990748, "grad_norm": 0.7746377286520203, "learning_rate": 1.3967948890167462e-06, "loss": 0.5597, "mean_token_accuracy": 0.8137285113334656, "num_tokens": 271903193.0, "step": 6650 }, { "epoch": 0.8790642347343378, "grad_norm": 0.8269771452031334, "learning_rate": 1.3959428545978154e-06, "loss": 0.5737, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 271968729.0, "step": 6651 }, { "epoch": 0.8791964049696008, "grad_norm": 0.8523772266635262, "learning_rate": 1.395091693846964e-06, "loss": 0.5738, "mean_token_accuracy": 0.8087992668151855, "num_tokens": 272034265.0, "step": 6652 }, { "epoch": 0.8793285752048638, "grad_norm": 0.816904840926288, "learning_rate": 1.3942414069453864e-06, "loss": 0.5696, "mean_token_accuracy": 0.8104779720306396, "num_tokens": 272099801.0, "step": 6653 }, { "epoch": 0.8794607454401269, "grad_norm": 0.8421263228435854, "learning_rate": 1.3933919940740915e-06, "loss": 0.5964, "mean_token_accuracy": 0.8015810251235962, "num_tokens": 272165337.0, "step": 6654 }, { "epoch": 0.8795929156753899, "grad_norm": 0.7533985020713411, "learning_rate": 1.392543455413902e-06, "loss": 0.5553, "mean_token_accuracy": 0.8148882985115051, "num_tokens": 272230873.0, "step": 6655 }, { "epoch": 0.8797250859106529, "grad_norm": 0.8398985643301846, "learning_rate": 1.3916957911454521e-06, "loss": 0.5732, "mean_token_accuracy": 0.8093028664588928, "num_tokens": 272296409.0, "step": 6656 }, { "epoch": 0.8798572561459159, "grad_norm": 0.7892573058993951, "learning_rate": 1.3908490014491937e-06, "loss": 0.5438, "mean_token_accuracy": 0.818596601486206, "num_tokens": 272361945.0, "step": 6657 }, { "epoch": 0.8799894263811789, "grad_norm": 0.8614358598754167, "learning_rate": 1.3900030865053896e-06, "loss": 0.5879, "mean_token_accuracy": 0.8028934001922607, "num_tokens": 272427481.0, "step": 6658 }, { "epoch": 0.880121596616442, "grad_norm": 0.7889267646608843, "learning_rate": 1.3891580464941188e-06, "loss": 0.5527, "mean_token_accuracy": 0.8180624842643738, "num_tokens": 272493017.0, "step": 6659 }, { "epoch": 0.880253766851705, "grad_norm": 0.7115253604638229, "learning_rate": 1.38831388159527e-06, "loss": 0.5013, "mean_token_accuracy": 0.8329263925552368, "num_tokens": 272558553.0, "step": 6660 }, { "epoch": 0.880385937086968, "grad_norm": 0.8062403646913461, "learning_rate": 1.3874705919885518e-06, "loss": 0.5633, "mean_token_accuracy": 0.8111341595649719, "num_tokens": 272624089.0, "step": 6661 }, { "epoch": 0.880518107322231, "grad_norm": 0.8438828871649597, "learning_rate": 1.38662817785348e-06, "loss": 0.6753, "mean_token_accuracy": 0.7764772176742554, "num_tokens": 272689625.0, "step": 6662 }, { "epoch": 0.880650277557494, "grad_norm": 0.8191957578208234, "learning_rate": 1.3857866393693884e-06, "loss": 0.5989, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 272755161.0, "step": 6663 }, { "epoch": 0.880782447792757, "grad_norm": 0.8121255327012489, "learning_rate": 1.3849459767154226e-06, "loss": 0.5908, "mean_token_accuracy": 0.8039311170578003, "num_tokens": 272820697.0, "step": 6664 }, { "epoch": 0.8809146180280201, "grad_norm": 0.774059211211643, "learning_rate": 1.3841061900705426e-06, "loss": 0.5793, "mean_token_accuracy": 0.8061134219169617, "num_tokens": 272886233.0, "step": 6665 }, { "epoch": 0.8810467882632831, "grad_norm": 0.7687712313409153, "learning_rate": 1.383267279613521e-06, "loss": 0.5425, "mean_token_accuracy": 0.8171926736831665, "num_tokens": 272951769.0, "step": 6666 }, { "epoch": 0.8811789584985461, "grad_norm": 0.8182489682436493, "learning_rate": 1.3824292455229437e-06, "loss": 0.5835, "mean_token_accuracy": 0.8068153858184814, "num_tokens": 273017305.0, "step": 6667 }, { "epoch": 0.8813111287338091, "grad_norm": 0.852666361672423, "learning_rate": 1.3815920879772116e-06, "loss": 0.5934, "mean_token_accuracy": 0.803030788898468, "num_tokens": 273082841.0, "step": 6668 }, { "epoch": 0.8814432989690721, "grad_norm": 0.7605972761212771, "learning_rate": 1.3807558071545387e-06, "loss": 0.5197, "mean_token_accuracy": 0.8272188901901245, "num_tokens": 273148377.0, "step": 6669 }, { "epoch": 0.8815754692043352, "grad_norm": 0.8515598914366664, "learning_rate": 1.3799204032329503e-06, "loss": 0.6176, "mean_token_accuracy": 0.7954919934272766, "num_tokens": 273213913.0, "step": 6670 }, { "epoch": 0.8817076394395982, "grad_norm": 0.8334627686453867, "learning_rate": 1.3790858763902864e-06, "loss": 0.5359, "mean_token_accuracy": 0.8221676349639893, "num_tokens": 273279449.0, "step": 6671 }, { "epoch": 0.8818398096748612, "grad_norm": 0.8193425994073227, "learning_rate": 1.3782522268042018e-06, "loss": 0.5857, "mean_token_accuracy": 0.8027560710906982, "num_tokens": 273344985.0, "step": 6672 }, { "epoch": 0.8819719799101242, "grad_norm": 0.8144460275204833, "learning_rate": 1.3774194546521614e-06, "loss": 0.5502, "mean_token_accuracy": 0.816643238067627, "num_tokens": 273410521.0, "step": 6673 }, { "epoch": 0.8821041501453872, "grad_norm": 0.856466368071581, "learning_rate": 1.3765875601114465e-06, "loss": 0.5911, "mean_token_accuracy": 0.8041600584983826, "num_tokens": 273476057.0, "step": 6674 }, { "epoch": 0.8822363203806503, "grad_norm": 0.8535719191409519, "learning_rate": 1.3757565433591494e-06, "loss": 0.6123, "mean_token_accuracy": 0.7920278310775757, "num_tokens": 273541593.0, "step": 6675 }, { "epoch": 0.8823684906159133, "grad_norm": 0.7887608933014489, "learning_rate": 1.3749264045721754e-06, "loss": 0.5845, "mean_token_accuracy": 0.8063575625419617, "num_tokens": 273607129.0, "step": 6676 }, { "epoch": 0.8825006608511763, "grad_norm": 0.8348188477890944, "learning_rate": 1.3740971439272453e-06, "loss": 0.5972, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 273672665.0, "step": 6677 }, { "epoch": 0.8826328310864393, "grad_norm": 0.7442086756011285, "learning_rate": 1.3732687616008908e-06, "loss": 0.5326, "mean_token_accuracy": 0.8211756944656372, "num_tokens": 273738201.0, "step": 6678 }, { "epoch": 0.8827650013217023, "grad_norm": 0.8236699747076112, "learning_rate": 1.3724412577694558e-06, "loss": 0.5783, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 273803737.0, "step": 6679 }, { "epoch": 0.8828971715569653, "grad_norm": 0.7854539801630493, "learning_rate": 1.371614632609101e-06, "loss": 0.589, "mean_token_accuracy": 0.8009552955627441, "num_tokens": 273869273.0, "step": 6680 }, { "epoch": 0.8830293417922284, "grad_norm": 0.8177064550844579, "learning_rate": 1.3707888862957958e-06, "loss": 0.5851, "mean_token_accuracy": 0.8051214814186096, "num_tokens": 273934809.0, "step": 6681 }, { "epoch": 0.8831615120274914, "grad_norm": 0.7921947213936137, "learning_rate": 1.369964019005326e-06, "loss": 0.5648, "mean_token_accuracy": 0.8107984662055969, "num_tokens": 274000345.0, "step": 6682 }, { "epoch": 0.8832936822627544, "grad_norm": 0.7574843518147719, "learning_rate": 1.3691400309132856e-06, "loss": 0.542, "mean_token_accuracy": 0.8221523761749268, "num_tokens": 274065881.0, "step": 6683 }, { "epoch": 0.8834258524980174, "grad_norm": 0.8121935292206071, "learning_rate": 1.3683169221950887e-06, "loss": 0.5965, "mean_token_accuracy": 0.802374541759491, "num_tokens": 274131417.0, "step": 6684 }, { "epoch": 0.8835580227332804, "grad_norm": 0.8198574094703651, "learning_rate": 1.3674946930259545e-06, "loss": 0.5937, "mean_token_accuracy": 0.8022982478141785, "num_tokens": 274196953.0, "step": 6685 }, { "epoch": 0.8836901929685435, "grad_norm": 0.8491650881308263, "learning_rate": 1.36667334358092e-06, "loss": 0.6213, "mean_token_accuracy": 0.7919210195541382, "num_tokens": 274262489.0, "step": 6686 }, { "epoch": 0.8838223632038065, "grad_norm": 0.7733986656742937, "learning_rate": 1.3658528740348326e-06, "loss": 0.5623, "mean_token_accuracy": 0.8139421343803406, "num_tokens": 274328025.0, "step": 6687 }, { "epoch": 0.8839545334390695, "grad_norm": 0.8152161708766587, "learning_rate": 1.365033284562354e-06, "loss": 0.561, "mean_token_accuracy": 0.8136674165725708, "num_tokens": 274393561.0, "step": 6688 }, { "epoch": 0.8840867036743325, "grad_norm": 0.856932144052673, "learning_rate": 1.364214575337957e-06, "loss": 0.612, "mean_token_accuracy": 0.7977505922317505, "num_tokens": 274459097.0, "step": 6689 }, { "epoch": 0.8842188739095955, "grad_norm": 0.7813415872171877, "learning_rate": 1.3633967465359283e-06, "loss": 0.5343, "mean_token_accuracy": 0.822686493396759, "num_tokens": 274524633.0, "step": 6690 }, { "epoch": 0.8843510441448585, "grad_norm": 0.8433572206530611, "learning_rate": 1.3625797983303654e-06, "loss": 0.5692, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 274590169.0, "step": 6691 }, { "epoch": 0.8844832143801216, "grad_norm": 0.8307412063259975, "learning_rate": 1.3617637308951808e-06, "loss": 0.5859, "mean_token_accuracy": 0.8038700819015503, "num_tokens": 274655705.0, "step": 6692 }, { "epoch": 0.8846153846153846, "grad_norm": 0.7535018974644692, "learning_rate": 1.3609485444040985e-06, "loss": 0.5194, "mean_token_accuracy": 0.8255402445793152, "num_tokens": 274721241.0, "step": 6693 }, { "epoch": 0.8847475548506476, "grad_norm": 0.7989401683477798, "learning_rate": 1.360134239030653e-06, "loss": 0.5774, "mean_token_accuracy": 0.8089061379432678, "num_tokens": 274786777.0, "step": 6694 }, { "epoch": 0.8848797250859106, "grad_norm": 0.8149654238976991, "learning_rate": 1.359320814948194e-06, "loss": 0.5394, "mean_token_accuracy": 0.8196648955345154, "num_tokens": 274852313.0, "step": 6695 }, { "epoch": 0.8850118953211736, "grad_norm": 0.7802748951967379, "learning_rate": 1.3585082723298818e-06, "loss": 0.5275, "mean_token_accuracy": 0.8267458081245422, "num_tokens": 274917849.0, "step": 6696 }, { "epoch": 0.8851440655564367, "grad_norm": 0.8392312963922476, "learning_rate": 1.357696611348692e-06, "loss": 0.5559, "mean_token_accuracy": 0.813118040561676, "num_tokens": 274983385.0, "step": 6697 }, { "epoch": 0.8852762357916997, "grad_norm": 0.8041592154785919, "learning_rate": 1.356885832177407e-06, "loss": 0.5893, "mean_token_accuracy": 0.8031070828437805, "num_tokens": 275048921.0, "step": 6698 }, { "epoch": 0.8854084060269627, "grad_norm": 0.7849080485680126, "learning_rate": 1.3560759349886268e-06, "loss": 0.5797, "mean_token_accuracy": 0.8062660098075867, "num_tokens": 275114457.0, "step": 6699 }, { "epoch": 0.8855405762622257, "grad_norm": 0.7386265374463364, "learning_rate": 1.3552669199547609e-06, "loss": 0.5341, "mean_token_accuracy": 0.8213587999343872, "num_tokens": 275179993.0, "step": 6700 }, { "epoch": 0.8856727464974887, "grad_norm": 0.7871786006359635, "learning_rate": 1.3544587872480329e-06, "loss": 0.5388, "mean_token_accuracy": 0.8230832815170288, "num_tokens": 275245529.0, "step": 6701 }, { "epoch": 0.8858049167327517, "grad_norm": 0.8465647035539471, "learning_rate": 1.3536515370404743e-06, "loss": 0.5954, "mean_token_accuracy": 0.8019014596939087, "num_tokens": 275311065.0, "step": 6702 }, { "epoch": 0.8859370869680148, "grad_norm": 0.7288240306801457, "learning_rate": 1.352845169503936e-06, "loss": 0.5188, "mean_token_accuracy": 0.8273409605026245, "num_tokens": 275376601.0, "step": 6703 }, { "epoch": 0.8860692572032778, "grad_norm": 0.8411456308438475, "learning_rate": 1.3520396848100742e-06, "loss": 0.6055, "mean_token_accuracy": 0.7962397933006287, "num_tokens": 275442137.0, "step": 6704 }, { "epoch": 0.8862014274385408, "grad_norm": 0.7901617203675823, "learning_rate": 1.35123508313036e-06, "loss": 0.562, "mean_token_accuracy": 0.8140947222709656, "num_tokens": 275507673.0, "step": 6705 }, { "epoch": 0.8863335976738038, "grad_norm": 0.7742865028661917, "learning_rate": 1.3504313646360765e-06, "loss": 0.5259, "mean_token_accuracy": 0.8250824213027954, "num_tokens": 275573209.0, "step": 6706 }, { "epoch": 0.8864657679090668, "grad_norm": 0.7972701900563657, "learning_rate": 1.3496285294983197e-06, "loss": 0.5424, "mean_token_accuracy": 0.818321943283081, "num_tokens": 275638745.0, "step": 6707 }, { "epoch": 0.8865979381443299, "grad_norm": 0.8365696731846739, "learning_rate": 1.3488265778879949e-06, "loss": 0.6019, "mean_token_accuracy": 0.798009991645813, "num_tokens": 275704281.0, "step": 6708 }, { "epoch": 0.8867301083795929, "grad_norm": 0.778946371026722, "learning_rate": 1.348025509975821e-06, "loss": 0.5269, "mean_token_accuracy": 0.8239530920982361, "num_tokens": 275769817.0, "step": 6709 }, { "epoch": 0.8868622786148559, "grad_norm": 0.7662323471926378, "learning_rate": 1.3472253259323299e-06, "loss": 0.5605, "mean_token_accuracy": 0.8145220279693604, "num_tokens": 275835353.0, "step": 6710 }, { "epoch": 0.8869944488501189, "grad_norm": 0.7928860353366541, "learning_rate": 1.3464260259278628e-06, "loss": 0.5876, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 275900889.0, "step": 6711 }, { "epoch": 0.8871266190853819, "grad_norm": 0.7643545459986039, "learning_rate": 1.3456276101325739e-06, "loss": 0.5336, "mean_token_accuracy": 0.8214808702468872, "num_tokens": 275966425.0, "step": 6712 }, { "epoch": 0.887258789320645, "grad_norm": 0.8321894290748038, "learning_rate": 1.3448300787164299e-06, "loss": 0.6036, "mean_token_accuracy": 0.7959040403366089, "num_tokens": 276031961.0, "step": 6713 }, { "epoch": 0.887390959555908, "grad_norm": 0.8782099687456618, "learning_rate": 1.3440334318492096e-06, "loss": 0.6183, "mean_token_accuracy": 0.797979474067688, "num_tokens": 276097497.0, "step": 6714 }, { "epoch": 0.887523129791171, "grad_norm": 0.8240277051297532, "learning_rate": 1.3432376697004993e-06, "loss": 0.5948, "mean_token_accuracy": 0.8004669547080994, "num_tokens": 276163033.0, "step": 6715 }, { "epoch": 0.887655300026434, "grad_norm": 0.7797553108541305, "learning_rate": 1.3424427924397038e-06, "loss": 0.5377, "mean_token_accuracy": 0.8201684951782227, "num_tokens": 276228569.0, "step": 6716 }, { "epoch": 0.887787470261697, "grad_norm": 0.8011716518925597, "learning_rate": 1.3416488002360333e-06, "loss": 0.566, "mean_token_accuracy": 0.8100506663322449, "num_tokens": 276294105.0, "step": 6717 }, { "epoch": 0.88791964049696, "grad_norm": 0.872688478728439, "learning_rate": 1.3408556932585137e-06, "loss": 0.6196, "mean_token_accuracy": 0.7947137355804443, "num_tokens": 276359641.0, "step": 6718 }, { "epoch": 0.8880518107322231, "grad_norm": 0.8154449593120722, "learning_rate": 1.34006347167598e-06, "loss": 0.6101, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 276425177.0, "step": 6719 }, { "epoch": 0.8881839809674861, "grad_norm": 0.7862169535554304, "learning_rate": 1.339272135657081e-06, "loss": 0.537, "mean_token_accuracy": 0.8207483887672424, "num_tokens": 276490713.0, "step": 6720 }, { "epoch": 0.8883161512027491, "grad_norm": 0.7950908992852512, "learning_rate": 1.3384816853702728e-06, "loss": 0.5988, "mean_token_accuracy": 0.8019167184829712, "num_tokens": 276556249.0, "step": 6721 }, { "epoch": 0.8884483214380121, "grad_norm": 0.8355471927383276, "learning_rate": 1.337692120983829e-06, "loss": 0.6079, "mean_token_accuracy": 0.7968196868896484, "num_tokens": 276621785.0, "step": 6722 }, { "epoch": 0.8885804916732751, "grad_norm": 0.8680941179970556, "learning_rate": 1.336903442665829e-06, "loss": 0.6025, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 276687321.0, "step": 6723 }, { "epoch": 0.8887126619085381, "grad_norm": 0.8592270410547577, "learning_rate": 1.3361156505841669e-06, "loss": 0.5892, "mean_token_accuracy": 0.8044805526733398, "num_tokens": 276752857.0, "step": 6724 }, { "epoch": 0.8888448321438012, "grad_norm": 0.8426415948104806, "learning_rate": 1.3353287449065478e-06, "loss": 0.6012, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 276818393.0, "step": 6725 }, { "epoch": 0.8889770023790642, "grad_norm": 0.7846265512702306, "learning_rate": 1.334542725800487e-06, "loss": 0.5478, "mean_token_accuracy": 0.8180624842643738, "num_tokens": 276883929.0, "step": 6726 }, { "epoch": 0.8891091726143272, "grad_norm": 0.7928220689130178, "learning_rate": 1.3337575934333106e-06, "loss": 0.5859, "mean_token_accuracy": 0.802252471446991, "num_tokens": 276949465.0, "step": 6727 }, { "epoch": 0.8892413428495902, "grad_norm": 0.8282714562305735, "learning_rate": 1.3329733479721579e-06, "loss": 0.5984, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 277015001.0, "step": 6728 }, { "epoch": 0.8893735130848532, "grad_norm": 0.8158036141424535, "learning_rate": 1.3321899895839785e-06, "loss": 0.6178, "mean_token_accuracy": 0.7956293225288391, "num_tokens": 277080537.0, "step": 6729 }, { "epoch": 0.8895056833201163, "grad_norm": 0.8352651073801521, "learning_rate": 1.3314075184355327e-06, "loss": 0.6237, "mean_token_accuracy": 0.7923635840415955, "num_tokens": 277146073.0, "step": 6730 }, { "epoch": 0.8896378535553793, "grad_norm": 0.7783643556499371, "learning_rate": 1.3306259346933916e-06, "loss": 0.5276, "mean_token_accuracy": 0.8237699866294861, "num_tokens": 277211609.0, "step": 6731 }, { "epoch": 0.8897700237906423, "grad_norm": 0.8236029191102032, "learning_rate": 1.3298452385239387e-06, "loss": 0.6022, "mean_token_accuracy": 0.7991088032722473, "num_tokens": 277277145.0, "step": 6732 }, { "epoch": 0.8899021940259053, "grad_norm": 0.7790385585353721, "learning_rate": 1.3290654300933692e-06, "loss": 0.5704, "mean_token_accuracy": 0.8133927583694458, "num_tokens": 277342681.0, "step": 6733 }, { "epoch": 0.8900343642611683, "grad_norm": 0.7500414184297312, "learning_rate": 1.3282865095676856e-06, "loss": 0.5405, "mean_token_accuracy": 0.8206415772438049, "num_tokens": 277408217.0, "step": 6734 }, { "epoch": 0.8901665344964313, "grad_norm": 0.7965188449941685, "learning_rate": 1.3275084771127055e-06, "loss": 0.5973, "mean_token_accuracy": 0.8037480115890503, "num_tokens": 277473753.0, "step": 6735 }, { "epoch": 0.8902987047316944, "grad_norm": 0.7886316725631252, "learning_rate": 1.326731332894055e-06, "loss": 0.5353, "mean_token_accuracy": 0.8212672472000122, "num_tokens": 277539289.0, "step": 6736 }, { "epoch": 0.8904308749669574, "grad_norm": 0.8047331705590502, "learning_rate": 1.3259550770771722e-06, "loss": 0.56, "mean_token_accuracy": 0.8122787475585938, "num_tokens": 277604825.0, "step": 6737 }, { "epoch": 0.8905630452022204, "grad_norm": 0.8443217327243896, "learning_rate": 1.3251797098273058e-06, "loss": 0.5535, "mean_token_accuracy": 0.8150561451911926, "num_tokens": 277670361.0, "step": 6738 }, { "epoch": 0.8906952154374835, "grad_norm": 0.7728930361957397, "learning_rate": 1.3244052313095158e-06, "loss": 0.5657, "mean_token_accuracy": 0.8107984662055969, "num_tokens": 277735897.0, "step": 6739 }, { "epoch": 0.8908273856727466, "grad_norm": 0.8021845670885924, "learning_rate": 1.3236316416886704e-06, "loss": 0.5394, "mean_token_accuracy": 0.8190696835517883, "num_tokens": 277801433.0, "step": 6740 }, { "epoch": 0.8909595559080096, "grad_norm": 0.8003555686543574, "learning_rate": 1.3228589411294536e-06, "loss": 0.5595, "mean_token_accuracy": 0.8140794634819031, "num_tokens": 277866969.0, "step": 6741 }, { "epoch": 0.8910917261432726, "grad_norm": 0.7718074485893742, "learning_rate": 1.3220871297963556e-06, "loss": 0.5291, "mean_token_accuracy": 0.8233274221420288, "num_tokens": 277932505.0, "step": 6742 }, { "epoch": 0.8912238963785356, "grad_norm": 0.7730050590196396, "learning_rate": 1.3213162078536796e-06, "loss": 0.5416, "mean_token_accuracy": 0.8197411894798279, "num_tokens": 277998041.0, "step": 6743 }, { "epoch": 0.8913560666137986, "grad_norm": 0.8209179518909262, "learning_rate": 1.320546175465537e-06, "loss": 0.566, "mean_token_accuracy": 0.8121718764305115, "num_tokens": 278063577.0, "step": 6744 }, { "epoch": 0.8914882368490616, "grad_norm": 0.8140879577926332, "learning_rate": 1.3197770327958535e-06, "loss": 0.5559, "mean_token_accuracy": 0.8164143562316895, "num_tokens": 278129113.0, "step": 6745 }, { "epoch": 0.8916204070843247, "grad_norm": 0.7628693369547634, "learning_rate": 1.319008780008363e-06, "loss": 0.5652, "mean_token_accuracy": 0.8113173246383667, "num_tokens": 278194649.0, "step": 6746 }, { "epoch": 0.8917525773195877, "grad_norm": 0.7759052842683635, "learning_rate": 1.3182414172666103e-06, "loss": 0.5154, "mean_token_accuracy": 0.8277987837791443, "num_tokens": 278260185.0, "step": 6747 }, { "epoch": 0.8918847475548507, "grad_norm": 0.8316373585788102, "learning_rate": 1.3174749447339513e-06, "loss": 0.5972, "mean_token_accuracy": 0.8017183542251587, "num_tokens": 278325721.0, "step": 6748 }, { "epoch": 0.8920169177901137, "grad_norm": 0.7643772326803076, "learning_rate": 1.3167093625735522e-06, "loss": 0.5487, "mean_token_accuracy": 0.8161702156066895, "num_tokens": 278391257.0, "step": 6749 }, { "epoch": 0.8921490880253767, "grad_norm": 0.7573674130883938, "learning_rate": 1.3159446709483879e-06, "loss": 0.517, "mean_token_accuracy": 0.8252807855606079, "num_tokens": 278456793.0, "step": 6750 }, { "epoch": 0.8922812582606398, "grad_norm": 0.8155278868748413, "learning_rate": 1.3151808700212467e-06, "loss": 0.5526, "mean_token_accuracy": 0.8144609928131104, "num_tokens": 278522329.0, "step": 6751 }, { "epoch": 0.8924134284959028, "grad_norm": 0.7734767107084408, "learning_rate": 1.314417959954726e-06, "loss": 0.5728, "mean_token_accuracy": 0.8102642893791199, "num_tokens": 278587865.0, "step": 6752 }, { "epoch": 0.8925455987311658, "grad_norm": 0.8280724280465835, "learning_rate": 1.313655940911232e-06, "loss": 0.6187, "mean_token_accuracy": 0.7939049005508423, "num_tokens": 278653401.0, "step": 6753 }, { "epoch": 0.8926777689664288, "grad_norm": 0.7535102289389232, "learning_rate": 1.3128948130529848e-06, "loss": 0.5819, "mean_token_accuracy": 0.8083720207214355, "num_tokens": 278718937.0, "step": 6754 }, { "epoch": 0.8928099392016918, "grad_norm": 0.7880719413327635, "learning_rate": 1.3121345765420104e-06, "loss": 0.5888, "mean_token_accuracy": 0.8060523867607117, "num_tokens": 278784473.0, "step": 6755 }, { "epoch": 0.8929421094369548, "grad_norm": 0.7836352078871264, "learning_rate": 1.3113752315401498e-06, "loss": 0.581, "mean_token_accuracy": 0.8109815716743469, "num_tokens": 278850009.0, "step": 6756 }, { "epoch": 0.8930742796722179, "grad_norm": 0.7567699026474133, "learning_rate": 1.3106167782090482e-06, "loss": 0.5269, "mean_token_accuracy": 0.8246551156044006, "num_tokens": 278915545.0, "step": 6757 }, { "epoch": 0.8932064499074809, "grad_norm": 0.7881416191116927, "learning_rate": 1.3098592167101681e-06, "loss": 0.5786, "mean_token_accuracy": 0.8053808808326721, "num_tokens": 278981081.0, "step": 6758 }, { "epoch": 0.8933386201427439, "grad_norm": 0.8007051761578967, "learning_rate": 1.3091025472047763e-06, "loss": 0.5804, "mean_token_accuracy": 0.8081278204917908, "num_tokens": 279046617.0, "step": 6759 }, { "epoch": 0.8934707903780069, "grad_norm": 0.8300383007257632, "learning_rate": 1.3083467698539523e-06, "loss": 0.6255, "mean_token_accuracy": 0.7921346426010132, "num_tokens": 279112153.0, "step": 6760 }, { "epoch": 0.8936029606132699, "grad_norm": 0.8185456750075275, "learning_rate": 1.3075918848185862e-06, "loss": 0.5589, "mean_token_accuracy": 0.8147204518318176, "num_tokens": 279177689.0, "step": 6761 }, { "epoch": 0.893735130848533, "grad_norm": 0.7702737211657521, "learning_rate": 1.3068378922593763e-06, "loss": 0.5838, "mean_token_accuracy": 0.8034886121749878, "num_tokens": 279243225.0, "step": 6762 }, { "epoch": 0.893867301083796, "grad_norm": 0.834690452855459, "learning_rate": 1.306084792336832e-06, "loss": 0.5843, "mean_token_accuracy": 0.8044652938842773, "num_tokens": 279308761.0, "step": 6763 }, { "epoch": 0.893999471319059, "grad_norm": 0.7604038915756673, "learning_rate": 1.305332585211273e-06, "loss": 0.5586, "mean_token_accuracy": 0.8132706880569458, "num_tokens": 279374297.0, "step": 6764 }, { "epoch": 0.894131641554322, "grad_norm": 0.7533035584146864, "learning_rate": 1.3045812710428284e-06, "loss": 0.5575, "mean_token_accuracy": 0.8145830631256104, "num_tokens": 279439833.0, "step": 6765 }, { "epoch": 0.894263811789585, "grad_norm": 0.7994833197054205, "learning_rate": 1.3038308499914371e-06, "loss": 0.5524, "mean_token_accuracy": 0.8169942498207092, "num_tokens": 279505369.0, "step": 6766 }, { "epoch": 0.894395982024848, "grad_norm": 0.7574140078610762, "learning_rate": 1.3030813222168483e-06, "loss": 0.5389, "mean_token_accuracy": 0.8186576962471008, "num_tokens": 279570905.0, "step": 6767 }, { "epoch": 0.8945281522601111, "grad_norm": 0.7974619127293463, "learning_rate": 1.3023326878786204e-06, "loss": 0.5652, "mean_token_accuracy": 0.812751829624176, "num_tokens": 279636441.0, "step": 6768 }, { "epoch": 0.8946603224953741, "grad_norm": 0.8163658242434639, "learning_rate": 1.3015849471361223e-06, "loss": 0.5675, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 279701977.0, "step": 6769 }, { "epoch": 0.8947924927306371, "grad_norm": 0.7825376438499694, "learning_rate": 1.3008381001485321e-06, "loss": 0.5473, "mean_token_accuracy": 0.817146897315979, "num_tokens": 279767513.0, "step": 6770 }, { "epoch": 0.8949246629659001, "grad_norm": 0.8691466990234269, "learning_rate": 1.300092147074839e-06, "loss": 0.6137, "mean_token_accuracy": 0.7954462170600891, "num_tokens": 279833049.0, "step": 6771 }, { "epoch": 0.8950568332011631, "grad_norm": 0.8365461763622939, "learning_rate": 1.299347088073839e-06, "loss": 0.5929, "mean_token_accuracy": 0.8046483993530273, "num_tokens": 279898585.0, "step": 6772 }, { "epoch": 0.8951890034364262, "grad_norm": 0.7845362802341194, "learning_rate": 1.2986029233041412e-06, "loss": 0.5274, "mean_token_accuracy": 0.8277530074119568, "num_tokens": 279964121.0, "step": 6773 }, { "epoch": 0.8953211736716892, "grad_norm": 0.795118433033234, "learning_rate": 1.297859652924162e-06, "loss": 0.6061, "mean_token_accuracy": 0.7973232865333557, "num_tokens": 280029657.0, "step": 6774 }, { "epoch": 0.8954533439069522, "grad_norm": 0.74210877574433, "learning_rate": 1.297117277092129e-06, "loss": 0.5135, "mean_token_accuracy": 0.8288975954055786, "num_tokens": 280095193.0, "step": 6775 }, { "epoch": 0.8955855141422152, "grad_norm": 0.8341388688842262, "learning_rate": 1.2963757959660764e-06, "loss": 0.6171, "mean_token_accuracy": 0.7966823577880859, "num_tokens": 280160729.0, "step": 6776 }, { "epoch": 0.8957176843774782, "grad_norm": 0.7186267914265283, "learning_rate": 1.2956352097038527e-06, "loss": 0.4985, "mean_token_accuracy": 0.8337046504020691, "num_tokens": 280226265.0, "step": 6777 }, { "epoch": 0.8958498546127412, "grad_norm": 0.810975733796253, "learning_rate": 1.2948955184631116e-06, "loss": 0.5917, "mean_token_accuracy": 0.8059303164482117, "num_tokens": 280291801.0, "step": 6778 }, { "epoch": 0.8959820248480043, "grad_norm": 0.8008199005682216, "learning_rate": 1.2941567224013182e-06, "loss": 0.5751, "mean_token_accuracy": 0.808631420135498, "num_tokens": 280357337.0, "step": 6779 }, { "epoch": 0.8961141950832673, "grad_norm": 0.7244545236807256, "learning_rate": 1.2934188216757474e-06, "loss": 0.5033, "mean_token_accuracy": 0.8317665457725525, "num_tokens": 280422873.0, "step": 6780 }, { "epoch": 0.8962463653185303, "grad_norm": 0.8723628574091785, "learning_rate": 1.2926818164434823e-06, "loss": 0.6906, "mean_token_accuracy": 0.7740965485572815, "num_tokens": 280488409.0, "step": 6781 }, { "epoch": 0.8963785355537933, "grad_norm": 1.1650327319970584, "learning_rate": 1.2919457068614152e-06, "loss": 0.5583, "mean_token_accuracy": 0.8147662281990051, "num_tokens": 280553945.0, "step": 6782 }, { "epoch": 0.8965107057890563, "grad_norm": 0.8240585909880606, "learning_rate": 1.2912104930862505e-06, "loss": 0.559, "mean_token_accuracy": 0.8141099810600281, "num_tokens": 280619481.0, "step": 6783 }, { "epoch": 0.8966428760243194, "grad_norm": 0.7768288436979256, "learning_rate": 1.290476175274498e-06, "loss": 0.5365, "mean_token_accuracy": 0.8217403292655945, "num_tokens": 280685017.0, "step": 6784 }, { "epoch": 0.8967750462595824, "grad_norm": 0.7727366491815673, "learning_rate": 1.2897427535824788e-06, "loss": 0.5601, "mean_token_accuracy": 0.8165059089660645, "num_tokens": 280750553.0, "step": 6785 }, { "epoch": 0.8969072164948454, "grad_norm": 0.802528464660676, "learning_rate": 1.2890102281663243e-06, "loss": 0.548, "mean_token_accuracy": 0.8150408864021301, "num_tokens": 280816089.0, "step": 6786 }, { "epoch": 0.8970393867301084, "grad_norm": 0.7713981564961938, "learning_rate": 1.2882785991819727e-06, "loss": 0.5716, "mean_token_accuracy": 0.8110731244087219, "num_tokens": 280881625.0, "step": 6787 }, { "epoch": 0.8971715569653714, "grad_norm": 0.7364649973477226, "learning_rate": 1.2875478667851732e-06, "loss": 0.5262, "mean_token_accuracy": 0.8261353969573975, "num_tokens": 280947161.0, "step": 6788 }, { "epoch": 0.8973037272006344, "grad_norm": 0.7651025890420232, "learning_rate": 1.2868180311314824e-06, "loss": 0.5712, "mean_token_accuracy": 0.8090282082557678, "num_tokens": 281012697.0, "step": 6789 }, { "epoch": 0.8974358974358975, "grad_norm": 0.7756527419141382, "learning_rate": 1.2860890923762693e-06, "loss": 0.5332, "mean_token_accuracy": 0.8241515159606934, "num_tokens": 281078233.0, "step": 6790 }, { "epoch": 0.8975680676711605, "grad_norm": 0.8244098167649152, "learning_rate": 1.2853610506747074e-06, "loss": 0.6099, "mean_token_accuracy": 0.8006806373596191, "num_tokens": 281143769.0, "step": 6791 }, { "epoch": 0.8977002379064235, "grad_norm": 0.8184031011931584, "learning_rate": 1.2846339061817823e-06, "loss": 0.594, "mean_token_accuracy": 0.8019624948501587, "num_tokens": 281209305.0, "step": 6792 }, { "epoch": 0.8978324081416865, "grad_norm": 0.7934238416801649, "learning_rate": 1.2839076590522878e-06, "loss": 0.5571, "mean_token_accuracy": 0.8147814869880676, "num_tokens": 281274841.0, "step": 6793 }, { "epoch": 0.8979645783769495, "grad_norm": 0.7789987054095131, "learning_rate": 1.2831823094408283e-06, "loss": 0.5674, "mean_token_accuracy": 0.8110578656196594, "num_tokens": 281340377.0, "step": 6794 }, { "epoch": 0.8980967486122126, "grad_norm": 0.7590749402488082, "learning_rate": 1.2824578575018127e-06, "loss": 0.555, "mean_token_accuracy": 0.8125839233398438, "num_tokens": 281405913.0, "step": 6795 }, { "epoch": 0.8982289188474756, "grad_norm": 0.8324046233863449, "learning_rate": 1.2817343033894647e-06, "loss": 0.5997, "mean_token_accuracy": 0.798040509223938, "num_tokens": 281471449.0, "step": 6796 }, { "epoch": 0.8983610890827386, "grad_norm": 0.8031972266902873, "learning_rate": 1.281011647257812e-06, "loss": 0.5843, "mean_token_accuracy": 0.8044958114624023, "num_tokens": 281536985.0, "step": 6797 }, { "epoch": 0.8984932593180016, "grad_norm": 0.8258151017824366, "learning_rate": 1.2802898892606933e-06, "loss": 0.5575, "mean_token_accuracy": 0.8152545690536499, "num_tokens": 281602521.0, "step": 6798 }, { "epoch": 0.8986254295532646, "grad_norm": 0.7553948651091037, "learning_rate": 1.279569029551756e-06, "loss": 0.5345, "mean_token_accuracy": 0.8209467530250549, "num_tokens": 281668057.0, "step": 6799 }, { "epoch": 0.8987575997885277, "grad_norm": 0.7354855045983566, "learning_rate": 1.2788490682844562e-06, "loss": 0.5005, "mean_token_accuracy": 0.8325753808021545, "num_tokens": 281733593.0, "step": 6800 }, { "epoch": 0.8988897700237907, "grad_norm": 0.7949837655043474, "learning_rate": 1.2781300056120573e-06, "loss": 0.5703, "mean_token_accuracy": 0.8093181252479553, "num_tokens": 281799129.0, "step": 6801 }, { "epoch": 0.8990219402590537, "grad_norm": 0.7728113626740293, "learning_rate": 1.2774118416876359e-06, "loss": 0.5292, "mean_token_accuracy": 0.8242430686950684, "num_tokens": 281864665.0, "step": 6802 }, { "epoch": 0.8991541104943167, "grad_norm": 0.8549477579725306, "learning_rate": 1.2766945766640709e-06, "loss": 0.6408, "mean_token_accuracy": 0.7888689041137695, "num_tokens": 281930201.0, "step": 6803 }, { "epoch": 0.8992862807295797, "grad_norm": 0.7889458919728869, "learning_rate": 1.2759782106940553e-06, "loss": 0.5601, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 281995737.0, "step": 6804 }, { "epoch": 0.8994184509648427, "grad_norm": 0.8112526714361794, "learning_rate": 1.2752627439300866e-06, "loss": 0.5783, "mean_token_accuracy": 0.8051825165748596, "num_tokens": 282061273.0, "step": 6805 }, { "epoch": 0.8995506212001058, "grad_norm": 0.8050234790238661, "learning_rate": 1.2745481765244734e-06, "loss": 0.5484, "mean_token_accuracy": 0.8184897899627686, "num_tokens": 282126809.0, "step": 6806 }, { "epoch": 0.8996827914353688, "grad_norm": 0.8191589363438689, "learning_rate": 1.2738345086293324e-06, "loss": 0.565, "mean_token_accuracy": 0.8138353228569031, "num_tokens": 282192345.0, "step": 6807 }, { "epoch": 0.8998149616706318, "grad_norm": 0.8109415491169552, "learning_rate": 1.2731217403965887e-06, "loss": 0.5488, "mean_token_accuracy": 0.8156971335411072, "num_tokens": 282257881.0, "step": 6808 }, { "epoch": 0.8999471319058948, "grad_norm": 0.7916046251290391, "learning_rate": 1.2724098719779762e-06, "loss": 0.6167, "mean_token_accuracy": 0.7955988049507141, "num_tokens": 282323417.0, "step": 6809 }, { "epoch": 0.9000793021411578, "grad_norm": 0.8081840506347974, "learning_rate": 1.2716989035250353e-06, "loss": 0.6021, "mean_token_accuracy": 0.8007416725158691, "num_tokens": 282388953.0, "step": 6810 }, { "epoch": 0.9002114723764209, "grad_norm": 0.808903284243127, "learning_rate": 1.2709888351891175e-06, "loss": 0.5746, "mean_token_accuracy": 0.8112104535102844, "num_tokens": 282454489.0, "step": 6811 }, { "epoch": 0.9003436426116839, "grad_norm": 0.7605412434976915, "learning_rate": 1.270279667121381e-06, "loss": 0.5528, "mean_token_accuracy": 0.8154376745223999, "num_tokens": 282520025.0, "step": 6812 }, { "epoch": 0.9004758128469469, "grad_norm": 0.7915810558610363, "learning_rate": 1.2695713994727943e-06, "loss": 0.5218, "mean_token_accuracy": 0.8248382210731506, "num_tokens": 282585561.0, "step": 6813 }, { "epoch": 0.9006079830822099, "grad_norm": 0.8553774299404842, "learning_rate": 1.2688640323941302e-06, "loss": 0.582, "mean_token_accuracy": 0.8050146698951721, "num_tokens": 282651097.0, "step": 6814 }, { "epoch": 0.9007401533174729, "grad_norm": 0.7963916424478007, "learning_rate": 1.268157566035975e-06, "loss": 0.5852, "mean_token_accuracy": 0.8023592829704285, "num_tokens": 282716633.0, "step": 6815 }, { "epoch": 0.9008723235527359, "grad_norm": 0.813262483824224, "learning_rate": 1.267452000548719e-06, "loss": 0.5706, "mean_token_accuracy": 0.8099285960197449, "num_tokens": 282782169.0, "step": 6816 }, { "epoch": 0.901004493787999, "grad_norm": 0.8205920690459463, "learning_rate": 1.2667473360825635e-06, "loss": 0.6168, "mean_token_accuracy": 0.7952936291694641, "num_tokens": 282847705.0, "step": 6817 }, { "epoch": 0.901136664023262, "grad_norm": 0.8095564263043036, "learning_rate": 1.2660435727875148e-06, "loss": 0.5822, "mean_token_accuracy": 0.807547926902771, "num_tokens": 282913241.0, "step": 6818 }, { "epoch": 0.901268834258525, "grad_norm": 0.7907839535129795, "learning_rate": 1.265340710813392e-06, "loss": 0.574, "mean_token_accuracy": 0.8080362677574158, "num_tokens": 282978777.0, "step": 6819 }, { "epoch": 0.901401004493788, "grad_norm": 0.8387125828460538, "learning_rate": 1.2646387503098179e-06, "loss": 0.582, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 283044313.0, "step": 6820 }, { "epoch": 0.901533174729051, "grad_norm": 0.8178268110151647, "learning_rate": 1.263937691426226e-06, "loss": 0.5897, "mean_token_accuracy": 0.8035343885421753, "num_tokens": 283109849.0, "step": 6821 }, { "epoch": 0.901665344964314, "grad_norm": 0.8160139515781192, "learning_rate": 1.263237534311857e-06, "loss": 0.5805, "mean_token_accuracy": 0.8056708574295044, "num_tokens": 283175385.0, "step": 6822 }, { "epoch": 0.9017975151995771, "grad_norm": 0.804767050915747, "learning_rate": 1.2625382791157604e-06, "loss": 0.5902, "mean_token_accuracy": 0.8049688935279846, "num_tokens": 283240921.0, "step": 6823 }, { "epoch": 0.9019296854348401, "grad_norm": 0.7740885779539927, "learning_rate": 1.261839925986792e-06, "loss": 0.5569, "mean_token_accuracy": 0.8153613805770874, "num_tokens": 283306457.0, "step": 6824 }, { "epoch": 0.9020618556701031, "grad_norm": 0.8250288103026598, "learning_rate": 1.2611424750736164e-06, "loss": 0.5847, "mean_token_accuracy": 0.8055793046951294, "num_tokens": 283371993.0, "step": 6825 }, { "epoch": 0.9021940259053661, "grad_norm": 0.8054973461508083, "learning_rate": 1.260445926524707e-06, "loss": 0.5772, "mean_token_accuracy": 0.8071969151496887, "num_tokens": 283437529.0, "step": 6826 }, { "epoch": 0.9023261961406291, "grad_norm": 0.7801092864901829, "learning_rate": 1.2597502804883444e-06, "loss": 0.5784, "mean_token_accuracy": 0.808570384979248, "num_tokens": 283503065.0, "step": 6827 }, { "epoch": 0.9024583663758922, "grad_norm": 0.7891968191439803, "learning_rate": 1.2590555371126176e-06, "loss": 0.5523, "mean_token_accuracy": 0.8154681921005249, "num_tokens": 283568601.0, "step": 6828 }, { "epoch": 0.9025905366111552, "grad_norm": 0.822172135103826, "learning_rate": 1.2583616965454224e-06, "loss": 0.5665, "mean_token_accuracy": 0.8092418313026428, "num_tokens": 283634137.0, "step": 6829 }, { "epoch": 0.9027227068464182, "grad_norm": 0.7532899569114393, "learning_rate": 1.2576687589344622e-06, "loss": 0.5746, "mean_token_accuracy": 0.807517409324646, "num_tokens": 283699673.0, "step": 6830 }, { "epoch": 0.9028548770816812, "grad_norm": 0.8504760401013369, "learning_rate": 1.2569767244272498e-06, "loss": 0.5799, "mean_token_accuracy": 0.8065407276153564, "num_tokens": 283765209.0, "step": 6831 }, { "epoch": 0.9029870473169442, "grad_norm": 0.7679734173960521, "learning_rate": 1.2562855931711056e-06, "loss": 0.5348, "mean_token_accuracy": 0.8221828937530518, "num_tokens": 283830745.0, "step": 6832 }, { "epoch": 0.9031192175522073, "grad_norm": 0.8600442573836178, "learning_rate": 1.2555953653131556e-06, "loss": 0.634, "mean_token_accuracy": 0.7894793152809143, "num_tokens": 283896281.0, "step": 6833 }, { "epoch": 0.9032513877874703, "grad_norm": 0.8126090613958706, "learning_rate": 1.2549060410003359e-06, "loss": 0.5756, "mean_token_accuracy": 0.8092418313026428, "num_tokens": 283961817.0, "step": 6834 }, { "epoch": 0.9033835580227333, "grad_norm": 0.7810890125978344, "learning_rate": 1.2542176203793886e-06, "loss": 0.5824, "mean_token_accuracy": 0.8042668700218201, "num_tokens": 284027353.0, "step": 6835 }, { "epoch": 0.9035157282579963, "grad_norm": 0.7936101218006693, "learning_rate": 1.2535301035968645e-06, "loss": 0.5908, "mean_token_accuracy": 0.8024203181266785, "num_tokens": 284092889.0, "step": 6836 }, { "epoch": 0.9036478984932593, "grad_norm": 0.8026353267005358, "learning_rate": 1.2528434907991208e-06, "loss": 0.4934, "mean_token_accuracy": 0.832651674747467, "num_tokens": 284158425.0, "step": 6837 }, { "epoch": 0.9037800687285223, "grad_norm": 0.8220150126466291, "learning_rate": 1.252157782132325e-06, "loss": 0.5836, "mean_token_accuracy": 0.8044500350952148, "num_tokens": 284223961.0, "step": 6838 }, { "epoch": 0.9039122389637854, "grad_norm": 0.8341540467279804, "learning_rate": 1.2514729777424476e-06, "loss": 0.5883, "mean_token_accuracy": 0.8055335283279419, "num_tokens": 284289497.0, "step": 6839 }, { "epoch": 0.9040444091990484, "grad_norm": 0.7998110127116546, "learning_rate": 1.250789077775271e-06, "loss": 0.5925, "mean_token_accuracy": 0.8004822134971619, "num_tokens": 284355033.0, "step": 6840 }, { "epoch": 0.9041765794343114, "grad_norm": 0.781358280356795, "learning_rate": 1.250106082376382e-06, "loss": 0.537, "mean_token_accuracy": 0.8196954131126404, "num_tokens": 284420569.0, "step": 6841 }, { "epoch": 0.9043087496695744, "grad_norm": 0.8025977585394751, "learning_rate": 1.2494239916911782e-06, "loss": 0.5555, "mean_token_accuracy": 0.8159260153770447, "num_tokens": 284486105.0, "step": 6842 }, { "epoch": 0.9044409199048374, "grad_norm": 0.7619348889127989, "learning_rate": 1.2487428058648593e-06, "loss": 0.5134, "mean_token_accuracy": 0.8290349245071411, "num_tokens": 284551641.0, "step": 6843 }, { "epoch": 0.9045730901401005, "grad_norm": 0.8283204725987695, "learning_rate": 1.2480625250424377e-06, "loss": 0.562, "mean_token_accuracy": 0.8158649802207947, "num_tokens": 284617177.0, "step": 6844 }, { "epoch": 0.9047052603753635, "grad_norm": 0.7577779753999134, "learning_rate": 1.2473831493687305e-06, "loss": 0.5548, "mean_token_accuracy": 0.8151629567146301, "num_tokens": 284682713.0, "step": 6845 }, { "epoch": 0.9048374306106265, "grad_norm": 0.8356003648535887, "learning_rate": 1.2467046789883624e-06, "loss": 0.5719, "mean_token_accuracy": 0.8106915950775146, "num_tokens": 284748249.0, "step": 6846 }, { "epoch": 0.9049696008458895, "grad_norm": 0.7879883427403409, "learning_rate": 1.2460271140457662e-06, "loss": 0.5374, "mean_token_accuracy": 0.8205347061157227, "num_tokens": 284813785.0, "step": 6847 }, { "epoch": 0.9051017710811525, "grad_norm": 0.9179955349681811, "learning_rate": 1.2453504546851802e-06, "loss": 0.6273, "mean_token_accuracy": 0.792500913143158, "num_tokens": 284879321.0, "step": 6848 }, { "epoch": 0.9052339413164155, "grad_norm": 0.8072172357672476, "learning_rate": 1.2446747010506531e-06, "loss": 0.6107, "mean_token_accuracy": 0.798940896987915, "num_tokens": 284944857.0, "step": 6849 }, { "epoch": 0.9053661115516786, "grad_norm": 0.823188232851923, "learning_rate": 1.2439998532860355e-06, "loss": 0.5571, "mean_token_accuracy": 0.8136063814163208, "num_tokens": 285010393.0, "step": 6850 }, { "epoch": 0.9054982817869416, "grad_norm": 0.7809851125571159, "learning_rate": 1.2433259115349914e-06, "loss": 0.5697, "mean_token_accuracy": 0.8103253841400146, "num_tokens": 285075929.0, "step": 6851 }, { "epoch": 0.9056304520222046, "grad_norm": 0.8727771072416646, "learning_rate": 1.2426528759409874e-06, "loss": 0.6031, "mean_token_accuracy": 0.7993376851081848, "num_tokens": 285141465.0, "step": 6852 }, { "epoch": 0.9057626222574676, "grad_norm": 0.8138311203378642, "learning_rate": 1.2419807466472987e-06, "loss": 0.6166, "mean_token_accuracy": 0.7940880060195923, "num_tokens": 285207001.0, "step": 6853 }, { "epoch": 0.9058947924927306, "grad_norm": 0.8087730966628268, "learning_rate": 1.2413095237970089e-06, "loss": 0.5776, "mean_token_accuracy": 0.8088603615760803, "num_tokens": 285272537.0, "step": 6854 }, { "epoch": 0.9060269627279937, "grad_norm": 0.7995258990877939, "learning_rate": 1.2406392075330063e-06, "loss": 0.5553, "mean_token_accuracy": 0.812934935092926, "num_tokens": 285338073.0, "step": 6855 }, { "epoch": 0.9061591329632567, "grad_norm": 0.8296251429812587, "learning_rate": 1.2399697979979859e-06, "loss": 0.5902, "mean_token_accuracy": 0.8044652938842773, "num_tokens": 285403609.0, "step": 6856 }, { "epoch": 0.9062913031985197, "grad_norm": 0.826860678490439, "learning_rate": 1.2393012953344544e-06, "loss": 0.5626, "mean_token_accuracy": 0.8131332993507385, "num_tokens": 285469145.0, "step": 6857 }, { "epoch": 0.9064234734337827, "grad_norm": 0.7499961963190721, "learning_rate": 1.2386336996847192e-06, "loss": 0.5572, "mean_token_accuracy": 0.8143846988677979, "num_tokens": 285534681.0, "step": 6858 }, { "epoch": 0.9065556436690457, "grad_norm": 0.7843757733293983, "learning_rate": 1.237967011190898e-06, "loss": 0.5734, "mean_token_accuracy": 0.8107068538665771, "num_tokens": 285600217.0, "step": 6859 }, { "epoch": 0.9066878139043087, "grad_norm": 0.7121056239369065, "learning_rate": 1.237301229994916e-06, "loss": 0.4805, "mean_token_accuracy": 0.84246426820755, "num_tokens": 285665753.0, "step": 6860 }, { "epoch": 0.9068199841395718, "grad_norm": 0.8014503594497385, "learning_rate": 1.2366363562385043e-06, "loss": 0.5866, "mean_token_accuracy": 0.8043126463890076, "num_tokens": 285731289.0, "step": 6861 }, { "epoch": 0.9069521543748348, "grad_norm": 0.8229142901583276, "learning_rate": 1.235972390063198e-06, "loss": 0.5933, "mean_token_accuracy": 0.802252471446991, "num_tokens": 285796825.0, "step": 6862 }, { "epoch": 0.9070843246100978, "grad_norm": 0.8303840501767213, "learning_rate": 1.235309331610344e-06, "loss": 0.5672, "mean_token_accuracy": 0.8139726519584656, "num_tokens": 285862361.0, "step": 6863 }, { "epoch": 0.9072164948453608, "grad_norm": 0.8670589886678565, "learning_rate": 1.2346471810210928e-06, "loss": 0.6029, "mean_token_accuracy": 0.7997192144393921, "num_tokens": 285927897.0, "step": 6864 }, { "epoch": 0.9073486650806238, "grad_norm": 0.8177035674266558, "learning_rate": 1.2339859384364034e-06, "loss": 0.6259, "mean_token_accuracy": 0.7905933260917664, "num_tokens": 285993433.0, "step": 6865 }, { "epoch": 0.9074808353158869, "grad_norm": 0.8960091796279571, "learning_rate": 1.2333256039970392e-06, "loss": 0.5902, "mean_token_accuracy": 0.8048620223999023, "num_tokens": 286058969.0, "step": 6866 }, { "epoch": 0.9076130055511499, "grad_norm": 0.7482109642614382, "learning_rate": 1.2326661778435719e-06, "loss": 0.542, "mean_token_accuracy": 0.8195275068283081, "num_tokens": 286124505.0, "step": 6867 }, { "epoch": 0.9077451757864129, "grad_norm": 0.867468525107464, "learning_rate": 1.2320076601163805e-06, "loss": 0.6342, "mean_token_accuracy": 0.7886247038841248, "num_tokens": 286190041.0, "step": 6868 }, { "epoch": 0.9078773460216759, "grad_norm": 0.7981895962502873, "learning_rate": 1.2313500509556476e-06, "loss": 0.5662, "mean_token_accuracy": 0.8089061379432678, "num_tokens": 286255577.0, "step": 6869 }, { "epoch": 0.9080095162569389, "grad_norm": 0.8173694720405723, "learning_rate": 1.2306933505013676e-06, "loss": 0.6257, "mean_token_accuracy": 0.7920888662338257, "num_tokens": 286321113.0, "step": 6870 }, { "epoch": 0.908141686492202, "grad_norm": 0.8246822571554009, "learning_rate": 1.2300375588933355e-06, "loss": 0.5716, "mean_token_accuracy": 0.8081888556480408, "num_tokens": 286386649.0, "step": 6871 }, { "epoch": 0.908273856727465, "grad_norm": 0.7574289968392967, "learning_rate": 1.2293826762711571e-06, "loss": 0.5624, "mean_token_accuracy": 0.8171010613441467, "num_tokens": 286452185.0, "step": 6872 }, { "epoch": 0.908406026962728, "grad_norm": 0.8439915653019349, "learning_rate": 1.2287287027742425e-06, "loss": 0.6045, "mean_token_accuracy": 0.7990477085113525, "num_tokens": 286517721.0, "step": 6873 }, { "epoch": 0.908538197197991, "grad_norm": 0.7898704716352896, "learning_rate": 1.2280756385418104e-06, "loss": 0.5763, "mean_token_accuracy": 0.8085551261901855, "num_tokens": 286583257.0, "step": 6874 }, { "epoch": 0.908670367433254, "grad_norm": 0.7555693366805176, "learning_rate": 1.2274234837128818e-06, "loss": 0.5688, "mean_token_accuracy": 0.8124160766601562, "num_tokens": 286648793.0, "step": 6875 }, { "epoch": 0.908802537668517, "grad_norm": 0.8524820307864304, "learning_rate": 1.2267722384262898e-06, "loss": 0.5868, "mean_token_accuracy": 0.8036412000656128, "num_tokens": 286714329.0, "step": 6876 }, { "epoch": 0.9089347079037801, "grad_norm": 0.8047861115441365, "learning_rate": 1.226121902820669e-06, "loss": 0.5791, "mean_token_accuracy": 0.8054572343826294, "num_tokens": 286779865.0, "step": 6877 }, { "epoch": 0.9090668781390431, "grad_norm": 0.7870391384062053, "learning_rate": 1.2254724770344638e-06, "loss": 0.5983, "mean_token_accuracy": 0.8002685904502869, "num_tokens": 286845401.0, "step": 6878 }, { "epoch": 0.9091990483743061, "grad_norm": 0.7966689695905197, "learning_rate": 1.2248239612059214e-06, "loss": 0.5458, "mean_token_accuracy": 0.8176199197769165, "num_tokens": 286910937.0, "step": 6879 }, { "epoch": 0.9093312186095691, "grad_norm": 0.8041607602689901, "learning_rate": 1.2241763554730996e-06, "loss": 0.5706, "mean_token_accuracy": 0.807608962059021, "num_tokens": 286976473.0, "step": 6880 }, { "epoch": 0.9094633888448321, "grad_norm": 0.746089789739891, "learning_rate": 1.2235296599738585e-06, "loss": 0.5153, "mean_token_accuracy": 0.8276919722557068, "num_tokens": 287042009.0, "step": 6881 }, { "epoch": 0.9095955590800952, "grad_norm": 0.8043671283600947, "learning_rate": 1.222883874845866e-06, "loss": 0.5714, "mean_token_accuracy": 0.8117445707321167, "num_tokens": 287107545.0, "step": 6882 }, { "epoch": 0.9097277293153582, "grad_norm": 0.8099156509100972, "learning_rate": 1.2222390002265971e-06, "loss": 0.576, "mean_token_accuracy": 0.8081735968589783, "num_tokens": 287173081.0, "step": 6883 }, { "epoch": 0.9098598995506212, "grad_norm": 0.7772649315852086, "learning_rate": 1.2215950362533324e-06, "loss": 0.5037, "mean_token_accuracy": 0.8310493230819702, "num_tokens": 287238617.0, "step": 6884 }, { "epoch": 0.9099920697858842, "grad_norm": 0.7910012633808419, "learning_rate": 1.2209519830631576e-06, "loss": 0.5576, "mean_token_accuracy": 0.8147967457771301, "num_tokens": 287304153.0, "step": 6885 }, { "epoch": 0.9101242400211472, "grad_norm": 0.8354243220717306, "learning_rate": 1.2203098407929644e-06, "loss": 0.5525, "mean_token_accuracy": 0.8152545690536499, "num_tokens": 287369689.0, "step": 6886 }, { "epoch": 0.9102564102564102, "grad_norm": 0.8461750718188311, "learning_rate": 1.2196686095794543e-06, "loss": 0.5724, "mean_token_accuracy": 0.8066627979278564, "num_tokens": 287435225.0, "step": 6887 }, { "epoch": 0.9103885804916733, "grad_norm": 0.738751819296413, "learning_rate": 1.2190282895591287e-06, "loss": 0.5697, "mean_token_accuracy": 0.8122634887695312, "num_tokens": 287500761.0, "step": 6888 }, { "epoch": 0.9105207507269363, "grad_norm": 0.7701208257767163, "learning_rate": 1.218388880868301e-06, "loss": 0.5268, "mean_token_accuracy": 0.8235868811607361, "num_tokens": 287566297.0, "step": 6889 }, { "epoch": 0.9106529209621993, "grad_norm": 0.8371793304414238, "learning_rate": 1.217750383643087e-06, "loss": 0.6255, "mean_token_accuracy": 0.7937980890274048, "num_tokens": 287631833.0, "step": 6890 }, { "epoch": 0.9107850911974623, "grad_norm": 0.8454838564269279, "learning_rate": 1.2171127980194087e-06, "loss": 0.6185, "mean_token_accuracy": 0.7929282188415527, "num_tokens": 287697369.0, "step": 6891 }, { "epoch": 0.9109172614327253, "grad_norm": 0.8946273617282117, "learning_rate": 1.2164761241329962e-06, "loss": 0.5914, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 287762905.0, "step": 6892 }, { "epoch": 0.9110494316679884, "grad_norm": 0.8177766842157445, "learning_rate": 1.2158403621193833e-06, "loss": 0.6117, "mean_token_accuracy": 0.7962092757225037, "num_tokens": 287828441.0, "step": 6893 }, { "epoch": 0.9111816019032514, "grad_norm": 0.7652724404828075, "learning_rate": 1.21520551211391e-06, "loss": 0.5089, "mean_token_accuracy": 0.8319802284240723, "num_tokens": 287893977.0, "step": 6894 }, { "epoch": 0.9113137721385144, "grad_norm": 0.7705577421371915, "learning_rate": 1.2145715742517236e-06, "loss": 0.5337, "mean_token_accuracy": 0.8216182589530945, "num_tokens": 287959513.0, "step": 6895 }, { "epoch": 0.9114459423737774, "grad_norm": 0.883129253856394, "learning_rate": 1.213938548667775e-06, "loss": 0.6069, "mean_token_accuracy": 0.7969112396240234, "num_tokens": 288025049.0, "step": 6896 }, { "epoch": 0.9115781126090404, "grad_norm": 0.8092287564898217, "learning_rate": 1.213306435496824e-06, "loss": 0.5789, "mean_token_accuracy": 0.8074105978012085, "num_tokens": 288090585.0, "step": 6897 }, { "epoch": 0.9117102828443034, "grad_norm": 0.7842003707378011, "learning_rate": 1.2126752348734318e-06, "loss": 0.5665, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 288156121.0, "step": 6898 }, { "epoch": 0.9118424530795665, "grad_norm": 0.7711198283643762, "learning_rate": 1.21204494693197e-06, "loss": 0.547, "mean_token_accuracy": 0.8191460371017456, "num_tokens": 288221657.0, "step": 6899 }, { "epoch": 0.9119746233148295, "grad_norm": 0.8640483466864699, "learning_rate": 1.2114155718066128e-06, "loss": 0.6002, "mean_token_accuracy": 0.8008179664611816, "num_tokens": 288287193.0, "step": 6900 }, { "epoch": 0.9121067935500925, "grad_norm": 0.7916245159918175, "learning_rate": 1.2107871096313414e-06, "loss": 0.6011, "mean_token_accuracy": 0.7979642152786255, "num_tokens": 288352729.0, "step": 6901 }, { "epoch": 0.9122389637853555, "grad_norm": 0.8477646676588952, "learning_rate": 1.2101595605399417e-06, "loss": 0.6507, "mean_token_accuracy": 0.7863814234733582, "num_tokens": 288418265.0, "step": 6902 }, { "epoch": 0.9123711340206185, "grad_norm": 0.8810065430558699, "learning_rate": 1.2095329246660068e-06, "loss": 0.6649, "mean_token_accuracy": 0.7797277569770813, "num_tokens": 288483801.0, "step": 6903 }, { "epoch": 0.9125033042558816, "grad_norm": 0.794368219592678, "learning_rate": 1.2089072021429328e-06, "loss": 0.5337, "mean_token_accuracy": 0.8231443166732788, "num_tokens": 288549337.0, "step": 6904 }, { "epoch": 0.9126354744911446, "grad_norm": 0.8002434995308597, "learning_rate": 1.2082823931039243e-06, "loss": 0.5624, "mean_token_accuracy": 0.8132554292678833, "num_tokens": 288614873.0, "step": 6905 }, { "epoch": 0.9127676447264076, "grad_norm": 0.7766543633345133, "learning_rate": 1.2076584976819903e-06, "loss": 0.5474, "mean_token_accuracy": 0.8179709315299988, "num_tokens": 288680409.0, "step": 6906 }, { "epoch": 0.9128998149616706, "grad_norm": 0.7650654374179398, "learning_rate": 1.2070355160099435e-06, "loss": 0.5048, "mean_token_accuracy": 0.8328653573989868, "num_tokens": 288745945.0, "step": 6907 }, { "epoch": 0.9130319851969336, "grad_norm": 0.7979302859203596, "learning_rate": 1.206413448220405e-06, "loss": 0.5752, "mean_token_accuracy": 0.8098675608634949, "num_tokens": 288811481.0, "step": 6908 }, { "epoch": 0.9131641554321966, "grad_norm": 0.7701457537982913, "learning_rate": 1.2057922944457997e-06, "loss": 0.5616, "mean_token_accuracy": 0.8126602172851562, "num_tokens": 288877017.0, "step": 6909 }, { "epoch": 0.9132963256674597, "grad_norm": 0.8048838804535126, "learning_rate": 1.2051720548183594e-06, "loss": 0.5576, "mean_token_accuracy": 0.8133469820022583, "num_tokens": 288942553.0, "step": 6910 }, { "epoch": 0.9134284959027227, "grad_norm": 0.8355381346733409, "learning_rate": 1.204552729470117e-06, "loss": 0.5365, "mean_token_accuracy": 0.8201074600219727, "num_tokens": 289008089.0, "step": 6911 }, { "epoch": 0.9135606661379857, "grad_norm": 0.8609504276576709, "learning_rate": 1.203934318532918e-06, "loss": 0.6079, "mean_token_accuracy": 0.7950494289398193, "num_tokens": 289073625.0, "step": 6912 }, { "epoch": 0.9136928363732487, "grad_norm": 0.8522214414578345, "learning_rate": 1.2033168221384063e-06, "loss": 0.5926, "mean_token_accuracy": 0.8048162460327148, "num_tokens": 289139161.0, "step": 6913 }, { "epoch": 0.9138250066085117, "grad_norm": 0.8071768121787605, "learning_rate": 1.2027002404180346e-06, "loss": 0.5108, "mean_token_accuracy": 0.8314308524131775, "num_tokens": 289204697.0, "step": 6914 }, { "epoch": 0.9139571768437748, "grad_norm": 0.7979800271589292, "learning_rate": 1.2020845735030609e-06, "loss": 0.562, "mean_token_accuracy": 0.8094707727432251, "num_tokens": 289270233.0, "step": 6915 }, { "epoch": 0.9140893470790378, "grad_norm": 0.8115840332251755, "learning_rate": 1.2014698215245482e-06, "loss": 0.5593, "mean_token_accuracy": 0.8150714039802551, "num_tokens": 289335769.0, "step": 6916 }, { "epoch": 0.9142215173143008, "grad_norm": 0.7523465618351749, "learning_rate": 1.200855984613362e-06, "loss": 0.5462, "mean_token_accuracy": 0.816643238067627, "num_tokens": 289401305.0, "step": 6917 }, { "epoch": 0.9143536875495638, "grad_norm": 0.8578546874697824, "learning_rate": 1.2002430629001784e-06, "loss": 0.5734, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 289466841.0, "step": 6918 }, { "epoch": 0.9144858577848268, "grad_norm": 0.7503591618673211, "learning_rate": 1.1996310565154737e-06, "loss": 0.5136, "mean_token_accuracy": 0.8310493230819702, "num_tokens": 289532377.0, "step": 6919 }, { "epoch": 0.9146180280200898, "grad_norm": 0.8743296505473898, "learning_rate": 1.1990199655895315e-06, "loss": 0.5595, "mean_token_accuracy": 0.8111705780029297, "num_tokens": 289593080.0, "step": 6920 }, { "epoch": 0.9147501982553529, "grad_norm": 0.8527721913331153, "learning_rate": 1.1984097902524412e-06, "loss": 0.6072, "mean_token_accuracy": 0.7969417572021484, "num_tokens": 289658616.0, "step": 6921 }, { "epoch": 0.9148823684906159, "grad_norm": 0.825317103164756, "learning_rate": 1.1978005306340963e-06, "loss": 0.6134, "mean_token_accuracy": 0.7959192991256714, "num_tokens": 289724152.0, "step": 6922 }, { "epoch": 0.9150145387258789, "grad_norm": 0.8376227032050024, "learning_rate": 1.1971921868641942e-06, "loss": 0.5863, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 289789688.0, "step": 6923 }, { "epoch": 0.9151467089611419, "grad_norm": 0.7259992268855976, "learning_rate": 1.1965847590722394e-06, "loss": 0.4908, "mean_token_accuracy": 0.8344066739082336, "num_tokens": 289855224.0, "step": 6924 }, { "epoch": 0.9152788791964049, "grad_norm": 0.766154668263769, "learning_rate": 1.195978247387541e-06, "loss": 0.5931, "mean_token_accuracy": 0.8026187419891357, "num_tokens": 289920760.0, "step": 6925 }, { "epoch": 0.915411049431668, "grad_norm": 0.7672394496327167, "learning_rate": 1.1953726519392118e-06, "loss": 0.5282, "mean_token_accuracy": 0.8223354816436768, "num_tokens": 289986296.0, "step": 6926 }, { "epoch": 0.915543219666931, "grad_norm": 0.8228213559137952, "learning_rate": 1.194767972856171e-06, "loss": 0.5987, "mean_token_accuracy": 0.8039311170578003, "num_tokens": 290051832.0, "step": 6927 }, { "epoch": 0.915675389902194, "grad_norm": 0.87683610440394, "learning_rate": 1.1941642102671416e-06, "loss": 0.5503, "mean_token_accuracy": 0.8177115321159363, "num_tokens": 290117368.0, "step": 6928 }, { "epoch": 0.915807560137457, "grad_norm": 0.8493863723900991, "learning_rate": 1.1935613643006532e-06, "loss": 0.6017, "mean_token_accuracy": 0.8014131188392639, "num_tokens": 290182904.0, "step": 6929 }, { "epoch": 0.91593973037272, "grad_norm": 0.7865805473847921, "learning_rate": 1.1929594350850374e-06, "loss": 0.53, "mean_token_accuracy": 0.8250824213027954, "num_tokens": 290248440.0, "step": 6930 }, { "epoch": 0.916071900607983, "grad_norm": 0.7953805375408237, "learning_rate": 1.1923584227484345e-06, "loss": 0.5778, "mean_token_accuracy": 0.8055487871170044, "num_tokens": 290313976.0, "step": 6931 }, { "epoch": 0.9162040708432461, "grad_norm": 0.7710442143838908, "learning_rate": 1.1917583274187853e-06, "loss": 0.5833, "mean_token_accuracy": 0.8064796924591064, "num_tokens": 290379512.0, "step": 6932 }, { "epoch": 0.9163362410785091, "grad_norm": 0.8655524588797561, "learning_rate": 1.1911591492238395e-06, "loss": 0.613, "mean_token_accuracy": 0.7968807220458984, "num_tokens": 290445048.0, "step": 6933 }, { "epoch": 0.9164684113137721, "grad_norm": 0.7947244043760002, "learning_rate": 1.1905608882911475e-06, "loss": 0.537, "mean_token_accuracy": 0.8205958008766174, "num_tokens": 290510584.0, "step": 6934 }, { "epoch": 0.9166005815490351, "grad_norm": 0.8219060488106654, "learning_rate": 1.1899635447480692e-06, "loss": 0.5793, "mean_token_accuracy": 0.8080667853355408, "num_tokens": 290576120.0, "step": 6935 }, { "epoch": 0.9167327517842981, "grad_norm": 0.804228960457871, "learning_rate": 1.1893671187217637e-06, "loss": 0.5714, "mean_token_accuracy": 0.8105542659759521, "num_tokens": 290641656.0, "step": 6936 }, { "epoch": 0.9168649220195612, "grad_norm": 0.821568736572767, "learning_rate": 1.1887716103392002e-06, "loss": 0.5974, "mean_token_accuracy": 0.8004974722862244, "num_tokens": 290707192.0, "step": 6937 }, { "epoch": 0.9169970922548242, "grad_norm": 0.8339118279373631, "learning_rate": 1.1881770197271489e-06, "loss": 0.5452, "mean_token_accuracy": 0.8194206953048706, "num_tokens": 290772728.0, "step": 6938 }, { "epoch": 0.9171292624900872, "grad_norm": 0.8725265637415364, "learning_rate": 1.1875833470121855e-06, "loss": 0.5995, "mean_token_accuracy": 0.8011689782142639, "num_tokens": 290838264.0, "step": 6939 }, { "epoch": 0.9172614327253502, "grad_norm": 0.7746412637698751, "learning_rate": 1.18699059232069e-06, "loss": 0.5497, "mean_token_accuracy": 0.8192528486251831, "num_tokens": 290903800.0, "step": 6940 }, { "epoch": 0.9173936029606132, "grad_norm": 0.8889086876453152, "learning_rate": 1.1863987557788491e-06, "loss": 0.6059, "mean_token_accuracy": 0.7991088032722473, "num_tokens": 290969336.0, "step": 6941 }, { "epoch": 0.9175257731958762, "grad_norm": 0.831051074689173, "learning_rate": 1.1858078375126518e-06, "loss": 0.5387, "mean_token_accuracy": 0.8209620118141174, "num_tokens": 291034872.0, "step": 6942 }, { "epoch": 0.9176579434311393, "grad_norm": 0.8125498303830377, "learning_rate": 1.1852178376478912e-06, "loss": 0.5191, "mean_token_accuracy": 0.8267000317573547, "num_tokens": 291100408.0, "step": 6943 }, { "epoch": 0.9177901136664023, "grad_norm": 0.7721709799425079, "learning_rate": 1.1846287563101676e-06, "loss": 0.5158, "mean_token_accuracy": 0.8288212418556213, "num_tokens": 291165944.0, "step": 6944 }, { "epoch": 0.9179222839016653, "grad_norm": 0.7802657641670389, "learning_rate": 1.1840405936248825e-06, "loss": 0.5429, "mean_token_accuracy": 0.8197869658470154, "num_tokens": 291231480.0, "step": 6945 }, { "epoch": 0.9180544541369283, "grad_norm": 0.8071589170578632, "learning_rate": 1.1834533497172449e-06, "loss": 0.5433, "mean_token_accuracy": 0.8187492489814758, "num_tokens": 291297016.0, "step": 6946 }, { "epoch": 0.9181866243721913, "grad_norm": 0.7652753400913389, "learning_rate": 1.182867024712266e-06, "loss": 0.5335, "mean_token_accuracy": 0.8218623995780945, "num_tokens": 291362552.0, "step": 6947 }, { "epoch": 0.9183187946074544, "grad_norm": 0.8423156077838234, "learning_rate": 1.1822816187347625e-06, "loss": 0.5919, "mean_token_accuracy": 0.8016268014907837, "num_tokens": 291428088.0, "step": 6948 }, { "epoch": 0.9184509648427174, "grad_norm": 0.7687441086101725, "learning_rate": 1.181697131909354e-06, "loss": 0.5373, "mean_token_accuracy": 0.8198632597923279, "num_tokens": 291493624.0, "step": 6949 }, { "epoch": 0.9185831350779804, "grad_norm": 0.8087586661446029, "learning_rate": 1.1811135643604674e-06, "loss": 0.5546, "mean_token_accuracy": 0.8146898746490479, "num_tokens": 291559160.0, "step": 6950 }, { "epoch": 0.9187153053132434, "grad_norm": 0.9376746290409456, "learning_rate": 1.1805309162123307e-06, "loss": 0.6848, "mean_token_accuracy": 0.7720363736152649, "num_tokens": 291624696.0, "step": 6951 }, { "epoch": 0.9188474755485064, "grad_norm": 0.7926765135659646, "learning_rate": 1.1799491875889775e-06, "loss": 0.5781, "mean_token_accuracy": 0.8066322803497314, "num_tokens": 291690232.0, "step": 6952 }, { "epoch": 0.9189796457837694, "grad_norm": 0.7822002050904273, "learning_rate": 1.1793683786142464e-06, "loss": 0.532, "mean_token_accuracy": 0.8201074600219727, "num_tokens": 291755768.0, "step": 6953 }, { "epoch": 0.9191118160190325, "grad_norm": 0.8312604707820546, "learning_rate": 1.1787884894117799e-06, "loss": 0.5561, "mean_token_accuracy": 0.8165974617004395, "num_tokens": 291821304.0, "step": 6954 }, { "epoch": 0.9192439862542955, "grad_norm": 0.7957631096678106, "learning_rate": 1.1782095201050234e-06, "loss": 0.5524, "mean_token_accuracy": 0.8179556727409363, "num_tokens": 291886840.0, "step": 6955 }, { "epoch": 0.9193761564895585, "grad_norm": 0.7957688725579634, "learning_rate": 1.1776314708172267e-06, "loss": 0.5537, "mean_token_accuracy": 0.8131638169288635, "num_tokens": 291952376.0, "step": 6956 }, { "epoch": 0.9195083267248215, "grad_norm": 0.8196299325102354, "learning_rate": 1.1770543416714458e-06, "loss": 0.5793, "mean_token_accuracy": 0.806678056716919, "num_tokens": 292017912.0, "step": 6957 }, { "epoch": 0.9196404969600845, "grad_norm": 0.8008095644756126, "learning_rate": 1.1764781327905398e-06, "loss": 0.5469, "mean_token_accuracy": 0.8187339901924133, "num_tokens": 292083448.0, "step": 6958 }, { "epoch": 0.9197726671953476, "grad_norm": 0.84646051352696, "learning_rate": 1.1759028442971692e-06, "loss": 0.5898, "mean_token_accuracy": 0.8047857284545898, "num_tokens": 292148984.0, "step": 6959 }, { "epoch": 0.9199048374306106, "grad_norm": 0.8280520954151015, "learning_rate": 1.1753284763138034e-06, "loss": 0.5738, "mean_token_accuracy": 0.812187135219574, "num_tokens": 292214520.0, "step": 6960 }, { "epoch": 0.9200370076658736, "grad_norm": 0.796303204540926, "learning_rate": 1.1747550289627121e-06, "loss": 0.563, "mean_token_accuracy": 0.8119277358055115, "num_tokens": 292280056.0, "step": 6961 }, { "epoch": 0.9201691779011366, "grad_norm": 0.8278292029998086, "learning_rate": 1.17418250236597e-06, "loss": 0.6096, "mean_token_accuracy": 0.7970943450927734, "num_tokens": 292345592.0, "step": 6962 }, { "epoch": 0.9203013481363996, "grad_norm": 0.7212981400528174, "learning_rate": 1.1736108966454577e-06, "loss": 0.5156, "mean_token_accuracy": 0.8286228775978088, "num_tokens": 292411128.0, "step": 6963 }, { "epoch": 0.9204335183716627, "grad_norm": 0.8193593350942555, "learning_rate": 1.1730402119228565e-06, "loss": 0.5395, "mean_token_accuracy": 0.818260908126831, "num_tokens": 292476664.0, "step": 6964 }, { "epoch": 0.9205656886069257, "grad_norm": 0.8140971294286744, "learning_rate": 1.1724704483196532e-06, "loss": 0.5943, "mean_token_accuracy": 0.8009705543518066, "num_tokens": 292542200.0, "step": 6965 }, { "epoch": 0.9206978588421887, "grad_norm": 0.8200899041468394, "learning_rate": 1.1719016059571396e-06, "loss": 0.5611, "mean_token_accuracy": 0.8137590289115906, "num_tokens": 292607736.0, "step": 6966 }, { "epoch": 0.9208300290774517, "grad_norm": 0.7723174714135206, "learning_rate": 1.1713336849564104e-06, "loss": 0.5438, "mean_token_accuracy": 0.818291425704956, "num_tokens": 292673272.0, "step": 6967 }, { "epoch": 0.9209621993127147, "grad_norm": 0.7502595633267325, "learning_rate": 1.170766685438363e-06, "loss": 0.5008, "mean_token_accuracy": 0.8337046504020691, "num_tokens": 292738808.0, "step": 6968 }, { "epoch": 0.9210943695479777, "grad_norm": 0.7638546649540577, "learning_rate": 1.1702006075237004e-06, "loss": 0.5337, "mean_token_accuracy": 0.822747528553009, "num_tokens": 292804344.0, "step": 6969 }, { "epoch": 0.9212265397832408, "grad_norm": 0.7264783182474376, "learning_rate": 1.1696354513329284e-06, "loss": 0.5057, "mean_token_accuracy": 0.8309425115585327, "num_tokens": 292869880.0, "step": 6970 }, { "epoch": 0.9213587100185038, "grad_norm": 0.7808488316562704, "learning_rate": 1.1690712169863584e-06, "loss": 0.5716, "mean_token_accuracy": 0.8090739846229553, "num_tokens": 292935416.0, "step": 6971 }, { "epoch": 0.9214908802537668, "grad_norm": 0.8102452646813958, "learning_rate": 1.1685079046041018e-06, "loss": 0.579, "mean_token_accuracy": 0.8060066103935242, "num_tokens": 293000952.0, "step": 6972 }, { "epoch": 0.9216230504890298, "grad_norm": 0.7834230916421561, "learning_rate": 1.1679455143060785e-06, "loss": 0.5762, "mean_token_accuracy": 0.8092876076698303, "num_tokens": 293066488.0, "step": 6973 }, { "epoch": 0.9217552207242928, "grad_norm": 0.8175125482842281, "learning_rate": 1.1673840462120077e-06, "loss": 0.6056, "mean_token_accuracy": 0.7979947328567505, "num_tokens": 293132024.0, "step": 6974 }, { "epoch": 0.921887390959556, "grad_norm": 0.7506375822884619, "learning_rate": 1.1668235004414146e-06, "loss": 0.5774, "mean_token_accuracy": 0.8077768087387085, "num_tokens": 293197560.0, "step": 6975 }, { "epoch": 0.922019561194819, "grad_norm": 0.8363614534323859, "learning_rate": 1.1662638771136283e-06, "loss": 0.6342, "mean_token_accuracy": 0.7867323756217957, "num_tokens": 293263096.0, "step": 6976 }, { "epoch": 0.922151731430082, "grad_norm": 0.7889835747186271, "learning_rate": 1.1657051763477814e-06, "loss": 0.5934, "mean_token_accuracy": 0.8042363524436951, "num_tokens": 293328632.0, "step": 6977 }, { "epoch": 0.922283901665345, "grad_norm": 0.8105708663672525, "learning_rate": 1.165147398262808e-06, "loss": 0.5874, "mean_token_accuracy": 0.8064186573028564, "num_tokens": 293394168.0, "step": 6978 }, { "epoch": 0.922416071900608, "grad_norm": 0.7512472224953757, "learning_rate": 1.1645905429774487e-06, "loss": 0.5389, "mean_token_accuracy": 0.8223659992218018, "num_tokens": 293459704.0, "step": 6979 }, { "epoch": 0.922548242135871, "grad_norm": 0.7901294032993794, "learning_rate": 1.1640346106102457e-06, "loss": 0.5233, "mean_token_accuracy": 0.8218929171562195, "num_tokens": 293525240.0, "step": 6980 }, { "epoch": 0.9226804123711341, "grad_norm": 0.7543773126646033, "learning_rate": 1.1634796012795458e-06, "loss": 0.5552, "mean_token_accuracy": 0.817360520362854, "num_tokens": 293590776.0, "step": 6981 }, { "epoch": 0.9228125826063971, "grad_norm": 0.768921982462381, "learning_rate": 1.1629255151034992e-06, "loss": 0.5389, "mean_token_accuracy": 0.8189323544502258, "num_tokens": 293656312.0, "step": 6982 }, { "epoch": 0.9229447528416601, "grad_norm": 0.8025292444632043, "learning_rate": 1.1623723522000585e-06, "loss": 0.6098, "mean_token_accuracy": 0.7999480962753296, "num_tokens": 293721848.0, "step": 6983 }, { "epoch": 0.9230769230769231, "grad_norm": 0.7605384342804723, "learning_rate": 1.161820112686981e-06, "loss": 0.5852, "mean_token_accuracy": 0.8037022352218628, "num_tokens": 293787384.0, "step": 6984 }, { "epoch": 0.9232090933121861, "grad_norm": 0.8556047848195346, "learning_rate": 1.1612687966818265e-06, "loss": 0.5884, "mean_token_accuracy": 0.8040074706077576, "num_tokens": 293852920.0, "step": 6985 }, { "epoch": 0.9233412635474492, "grad_norm": 0.8060580553565084, "learning_rate": 1.1607184043019596e-06, "loss": 0.5847, "mean_token_accuracy": 0.8067543506622314, "num_tokens": 293918456.0, "step": 6986 }, { "epoch": 0.9234734337827122, "grad_norm": 0.7894311869037173, "learning_rate": 1.160168935664546e-06, "loss": 0.5798, "mean_token_accuracy": 0.8088297843933105, "num_tokens": 293983992.0, "step": 6987 }, { "epoch": 0.9236056040179752, "grad_norm": 0.7384493888875767, "learning_rate": 1.1596203908865573e-06, "loss": 0.5635, "mean_token_accuracy": 0.8122634887695312, "num_tokens": 294049528.0, "step": 6988 }, { "epoch": 0.9237377742532382, "grad_norm": 0.7414965738724107, "learning_rate": 1.1590727700847666e-06, "loss": 0.5227, "mean_token_accuracy": 0.8277377486228943, "num_tokens": 294115064.0, "step": 6989 }, { "epoch": 0.9238699444885012, "grad_norm": 0.7968397079495579, "learning_rate": 1.1585260733757513e-06, "loss": 0.6133, "mean_token_accuracy": 0.7956140637397766, "num_tokens": 294180600.0, "step": 6990 }, { "epoch": 0.9240021147237643, "grad_norm": 0.8899675086022268, "learning_rate": 1.157980300875891e-06, "loss": 0.6375, "mean_token_accuracy": 0.7900286912918091, "num_tokens": 294246136.0, "step": 6991 }, { "epoch": 0.9241342849590273, "grad_norm": 0.8437726064821701, "learning_rate": 1.1574354527013708e-06, "loss": 0.6255, "mean_token_accuracy": 0.7909748554229736, "num_tokens": 294311672.0, "step": 6992 }, { "epoch": 0.9242664551942903, "grad_norm": 0.7813236513563739, "learning_rate": 1.1568915289681755e-06, "loss": 0.538, "mean_token_accuracy": 0.8198937773704529, "num_tokens": 294377208.0, "step": 6993 }, { "epoch": 0.9243986254295533, "grad_norm": 0.8077817817144756, "learning_rate": 1.1563485297920963e-06, "loss": 0.573, "mean_token_accuracy": 0.807761549949646, "num_tokens": 294442744.0, "step": 6994 }, { "epoch": 0.9245307956648163, "grad_norm": 0.8866107535093434, "learning_rate": 1.1558064552887267e-06, "loss": 0.627, "mean_token_accuracy": 0.792256772518158, "num_tokens": 294508280.0, "step": 6995 }, { "epoch": 0.9246629659000793, "grad_norm": 0.7710076451194415, "learning_rate": 1.1552653055734627e-06, "loss": 0.5858, "mean_token_accuracy": 0.8055182695388794, "num_tokens": 294573816.0, "step": 6996 }, { "epoch": 0.9247951361353424, "grad_norm": 0.7915117252777665, "learning_rate": 1.1547250807615022e-06, "loss": 0.5655, "mean_token_accuracy": 0.8102337718009949, "num_tokens": 294639352.0, "step": 6997 }, { "epoch": 0.9249273063706054, "grad_norm": 0.7628539146382644, "learning_rate": 1.154185780967851e-06, "loss": 0.5315, "mean_token_accuracy": 0.8206415772438049, "num_tokens": 294704888.0, "step": 6998 }, { "epoch": 0.9250594766058684, "grad_norm": 0.764609411306798, "learning_rate": 1.1536474063073119e-06, "loss": 0.5682, "mean_token_accuracy": 0.8104779720306396, "num_tokens": 294770424.0, "step": 6999 }, { "epoch": 0.9251916468411314, "grad_norm": 0.8029767417339665, "learning_rate": 1.1531099568944953e-06, "loss": 0.5604, "mean_token_accuracy": 0.8137131929397583, "num_tokens": 294835960.0, "step": 7000 }, { "epoch": 0.9253238170763944, "grad_norm": 0.7582984528715253, "learning_rate": 1.1525734328438116e-06, "loss": 0.5368, "mean_token_accuracy": 0.8209467530250549, "num_tokens": 294901496.0, "step": 7001 }, { "epoch": 0.9254559873116575, "grad_norm": 0.7762084413832052, "learning_rate": 1.1520378342694765e-06, "loss": 0.5677, "mean_token_accuracy": 0.8101269602775574, "num_tokens": 294967032.0, "step": 7002 }, { "epoch": 0.9255881575469205, "grad_norm": 0.8905163697390528, "learning_rate": 1.1515031612855077e-06, "loss": 0.6355, "mean_token_accuracy": 0.7871291637420654, "num_tokens": 295032568.0, "step": 7003 }, { "epoch": 0.9257203277821835, "grad_norm": 0.8193926219611489, "learning_rate": 1.150969414005726e-06, "loss": 0.559, "mean_token_accuracy": 0.8114241361618042, "num_tokens": 295098104.0, "step": 7004 }, { "epoch": 0.9258524980174465, "grad_norm": 0.8078650963907992, "learning_rate": 1.150436592543755e-06, "loss": 0.6183, "mean_token_accuracy": 0.7959345579147339, "num_tokens": 295163640.0, "step": 7005 }, { "epoch": 0.9259846682527095, "grad_norm": 0.7646414742459172, "learning_rate": 1.1499046970130206e-06, "loss": 0.5727, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 295229176.0, "step": 7006 }, { "epoch": 0.9261168384879725, "grad_norm": 0.8128199303902931, "learning_rate": 1.1493737275267527e-06, "loss": 0.5865, "mean_token_accuracy": 0.806494951248169, "num_tokens": 295294712.0, "step": 7007 }, { "epoch": 0.9262490087232356, "grad_norm": 0.803744181752357, "learning_rate": 1.1488436841979842e-06, "loss": 0.5771, "mean_token_accuracy": 0.8078378438949585, "num_tokens": 295360248.0, "step": 7008 }, { "epoch": 0.9263811789584986, "grad_norm": 0.8439487649020755, "learning_rate": 1.1483145671395496e-06, "loss": 0.5986, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 295425784.0, "step": 7009 }, { "epoch": 0.9265133491937616, "grad_norm": 0.7453307751484396, "learning_rate": 1.1477863764640864e-06, "loss": 0.5543, "mean_token_accuracy": 0.8135911226272583, "num_tokens": 295491320.0, "step": 7010 }, { "epoch": 0.9266455194290246, "grad_norm": 0.8757362865894038, "learning_rate": 1.147259112284037e-06, "loss": 0.6158, "mean_token_accuracy": 0.7921499013900757, "num_tokens": 295556856.0, "step": 7011 }, { "epoch": 0.9267776896642876, "grad_norm": 0.7714776813453439, "learning_rate": 1.1467327747116433e-06, "loss": 0.5982, "mean_token_accuracy": 0.8013826012611389, "num_tokens": 295622392.0, "step": 7012 }, { "epoch": 0.9269098598995507, "grad_norm": 0.8266061166109235, "learning_rate": 1.1462073638589518e-06, "loss": 0.5644, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 295687928.0, "step": 7013 }, { "epoch": 0.9270420301348137, "grad_norm": 0.8329052526292713, "learning_rate": 1.1456828798378125e-06, "loss": 0.5915, "mean_token_accuracy": 0.8025577068328857, "num_tokens": 295753464.0, "step": 7014 }, { "epoch": 0.9271742003700767, "grad_norm": 0.8262741959406091, "learning_rate": 1.1451593227598765e-06, "loss": 0.5865, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 295819000.0, "step": 7015 }, { "epoch": 0.9273063706053397, "grad_norm": 0.7707132388109045, "learning_rate": 1.144636692736598e-06, "loss": 0.5615, "mean_token_accuracy": 0.812187135219574, "num_tokens": 295884536.0, "step": 7016 }, { "epoch": 0.9274385408406027, "grad_norm": 0.8296594237791242, "learning_rate": 1.1441149898792338e-06, "loss": 0.6107, "mean_token_accuracy": 0.7969417572021484, "num_tokens": 295950072.0, "step": 7017 }, { "epoch": 0.9275707110758658, "grad_norm": 0.9453535314627001, "learning_rate": 1.1435942142988444e-06, "loss": 0.6581, "mean_token_accuracy": 0.7813911437988281, "num_tokens": 296015608.0, "step": 7018 }, { "epoch": 0.9277028813111288, "grad_norm": 0.7633387254859867, "learning_rate": 1.1430743661062917e-06, "loss": 0.5232, "mean_token_accuracy": 0.8261353969573975, "num_tokens": 296081144.0, "step": 7019 }, { "epoch": 0.9278350515463918, "grad_norm": 0.7754974656330337, "learning_rate": 1.1425554454122398e-06, "loss": 0.5366, "mean_token_accuracy": 0.8210383057594299, "num_tokens": 296146680.0, "step": 7020 }, { "epoch": 0.9279672217816548, "grad_norm": 0.7748317619683958, "learning_rate": 1.1420374523271573e-06, "loss": 0.5815, "mean_token_accuracy": 0.8071816563606262, "num_tokens": 296212216.0, "step": 7021 }, { "epoch": 0.9280993920169178, "grad_norm": 0.7922880247304876, "learning_rate": 1.141520386961313e-06, "loss": 0.5813, "mean_token_accuracy": 0.8094707727432251, "num_tokens": 296277752.0, "step": 7022 }, { "epoch": 0.9282315622521808, "grad_norm": 0.7940873565073242, "learning_rate": 1.14100424942478e-06, "loss": 0.5509, "mean_token_accuracy": 0.8138810992240906, "num_tokens": 296343288.0, "step": 7023 }, { "epoch": 0.9283637324874439, "grad_norm": 0.7980071130505089, "learning_rate": 1.1404890398274333e-06, "loss": 0.5758, "mean_token_accuracy": 0.8062965273857117, "num_tokens": 296408824.0, "step": 7024 }, { "epoch": 0.9284959027227069, "grad_norm": 0.8246426954323888, "learning_rate": 1.1399747582789498e-06, "loss": 0.6284, "mean_token_accuracy": 0.7916920781135559, "num_tokens": 296474360.0, "step": 7025 }, { "epoch": 0.9286280729579699, "grad_norm": 0.745032062273785, "learning_rate": 1.1394614048888093e-06, "loss": 0.5539, "mean_token_accuracy": 0.8161243796348572, "num_tokens": 296539896.0, "step": 7026 }, { "epoch": 0.9287602431932329, "grad_norm": 0.8090056574600498, "learning_rate": 1.138948979766294e-06, "loss": 0.5403, "mean_token_accuracy": 0.818566083908081, "num_tokens": 296605432.0, "step": 7027 }, { "epoch": 0.9288924134284959, "grad_norm": 0.830200834681011, "learning_rate": 1.1384374830204894e-06, "loss": 0.608, "mean_token_accuracy": 0.7962703108787537, "num_tokens": 296670968.0, "step": 7028 }, { "epoch": 0.929024583663759, "grad_norm": 0.8377684856161204, "learning_rate": 1.1379269147602817e-06, "loss": 0.602, "mean_token_accuracy": 0.8000854849815369, "num_tokens": 296736504.0, "step": 7029 }, { "epoch": 0.929156753899022, "grad_norm": 0.7594657764320475, "learning_rate": 1.13741727509436e-06, "loss": 0.5428, "mean_token_accuracy": 0.8233579397201538, "num_tokens": 296802040.0, "step": 7030 }, { "epoch": 0.929288924134285, "grad_norm": 0.7853058607526489, "learning_rate": 1.1369085641312164e-06, "loss": 0.5585, "mean_token_accuracy": 0.8150408864021301, "num_tokens": 296867576.0, "step": 7031 }, { "epoch": 0.929421094369548, "grad_norm": 0.781091149072293, "learning_rate": 1.136400781979145e-06, "loss": 0.5566, "mean_token_accuracy": 0.816185474395752, "num_tokens": 296933112.0, "step": 7032 }, { "epoch": 0.929553264604811, "grad_norm": 0.7426876293889596, "learning_rate": 1.1358939287462405e-06, "loss": 0.538, "mean_token_accuracy": 0.8223202228546143, "num_tokens": 296998648.0, "step": 7033 }, { "epoch": 0.929685434840074, "grad_norm": 0.8624122511214783, "learning_rate": 1.1353880045404037e-06, "loss": 0.5615, "mean_token_accuracy": 0.8126754760742188, "num_tokens": 297064184.0, "step": 7034 }, { "epoch": 0.9298176050753371, "grad_norm": 0.7744898023702068, "learning_rate": 1.1348830094693342e-06, "loss": 0.5562, "mean_token_accuracy": 0.8129196763038635, "num_tokens": 297129720.0, "step": 7035 }, { "epoch": 0.9299497753106001, "grad_norm": 0.8798013386651967, "learning_rate": 1.1343789436405344e-06, "loss": 0.5939, "mean_token_accuracy": 0.8013978600502014, "num_tokens": 297195256.0, "step": 7036 }, { "epoch": 0.9300819455458631, "grad_norm": 0.8461807394963561, "learning_rate": 1.13387580716131e-06, "loss": 0.6268, "mean_token_accuracy": 0.7929129600524902, "num_tokens": 297260792.0, "step": 7037 }, { "epoch": 0.9302141157811261, "grad_norm": 0.7436138959042432, "learning_rate": 1.1333736001387687e-06, "loss": 0.5804, "mean_token_accuracy": 0.8065712451934814, "num_tokens": 297326328.0, "step": 7038 }, { "epoch": 0.9303462860163891, "grad_norm": 0.8180035937041076, "learning_rate": 1.1328723226798193e-06, "loss": 0.554, "mean_token_accuracy": 0.8148120045661926, "num_tokens": 297391864.0, "step": 7039 }, { "epoch": 0.9304784562516522, "grad_norm": 0.7994966736392031, "learning_rate": 1.1323719748911726e-06, "loss": 0.5919, "mean_token_accuracy": 0.8029544353485107, "num_tokens": 297457400.0, "step": 7040 }, { "epoch": 0.9306106264869152, "grad_norm": 0.7581603977710148, "learning_rate": 1.1318725568793435e-06, "loss": 0.5199, "mean_token_accuracy": 0.8273104429244995, "num_tokens": 297522936.0, "step": 7041 }, { "epoch": 0.9307427967221782, "grad_norm": 0.804282857736541, "learning_rate": 1.131374068750647e-06, "loss": 0.5968, "mean_token_accuracy": 0.8029696941375732, "num_tokens": 297588472.0, "step": 7042 }, { "epoch": 0.9308749669574412, "grad_norm": 0.7966061741435123, "learning_rate": 1.130876510611201e-06, "loss": 0.5447, "mean_token_accuracy": 0.8175588846206665, "num_tokens": 297654008.0, "step": 7043 }, { "epoch": 0.9310071371927042, "grad_norm": 0.8407344914680729, "learning_rate": 1.1303798825669247e-06, "loss": 0.5781, "mean_token_accuracy": 0.8073342442512512, "num_tokens": 297719544.0, "step": 7044 }, { "epoch": 0.9311393074279672, "grad_norm": 0.8641916516199738, "learning_rate": 1.1298841847235408e-06, "loss": 0.5512, "mean_token_accuracy": 0.8123855590820312, "num_tokens": 297785080.0, "step": 7045 }, { "epoch": 0.9312714776632303, "grad_norm": 0.7974646370649043, "learning_rate": 1.129389417186572e-06, "loss": 0.5552, "mean_token_accuracy": 0.817268967628479, "num_tokens": 297850616.0, "step": 7046 }, { "epoch": 0.9314036478984933, "grad_norm": 0.7949917249611836, "learning_rate": 1.1288955800613453e-06, "loss": 0.5987, "mean_token_accuracy": 0.7995513081550598, "num_tokens": 297916152.0, "step": 7047 }, { "epoch": 0.9315358181337563, "grad_norm": 0.8800818166989992, "learning_rate": 1.128402673452987e-06, "loss": 0.5974, "mean_token_accuracy": 0.8009247779846191, "num_tokens": 297981688.0, "step": 7048 }, { "epoch": 0.9316679883690193, "grad_norm": 0.8241692849178966, "learning_rate": 1.127910697466427e-06, "loss": 0.5809, "mean_token_accuracy": 0.8046636581420898, "num_tokens": 298047224.0, "step": 7049 }, { "epoch": 0.9318001586042823, "grad_norm": 0.877062822987858, "learning_rate": 1.127419652206397e-06, "loss": 0.6404, "mean_token_accuracy": 0.7878311276435852, "num_tokens": 298112760.0, "step": 7050 }, { "epoch": 0.9319323288395454, "grad_norm": 0.8310496396578412, "learning_rate": 1.1269295377774306e-06, "loss": 0.5992, "mean_token_accuracy": 0.7992613911628723, "num_tokens": 298178296.0, "step": 7051 }, { "epoch": 0.9320644990748084, "grad_norm": 0.7896354637578447, "learning_rate": 1.1264403542838615e-06, "loss": 0.5511, "mean_token_accuracy": 0.8154529333114624, "num_tokens": 298243832.0, "step": 7052 }, { "epoch": 0.9321966693100714, "grad_norm": 0.7633179601165756, "learning_rate": 1.1259521018298285e-06, "loss": 0.5534, "mean_token_accuracy": 0.8153613805770874, "num_tokens": 298309368.0, "step": 7053 }, { "epoch": 0.9323288395453344, "grad_norm": 0.8034528854706986, "learning_rate": 1.1254647805192693e-06, "loss": 0.6191, "mean_token_accuracy": 0.7944695353507996, "num_tokens": 298374904.0, "step": 7054 }, { "epoch": 0.9324610097805974, "grad_norm": 0.7940765827762244, "learning_rate": 1.1249783904559246e-06, "loss": 0.5573, "mean_token_accuracy": 0.812812864780426, "num_tokens": 298440440.0, "step": 7055 }, { "epoch": 0.9325931800158604, "grad_norm": 0.7966104754617602, "learning_rate": 1.1244929317433365e-06, "loss": 0.5761, "mean_token_accuracy": 0.8078684210777283, "num_tokens": 298505976.0, "step": 7056 }, { "epoch": 0.9327253502511235, "grad_norm": 0.7771638840211892, "learning_rate": 1.1240084044848497e-06, "loss": 0.5426, "mean_token_accuracy": 0.8202600479125977, "num_tokens": 298571512.0, "step": 7057 }, { "epoch": 0.9328575204863865, "grad_norm": 0.8542259946077844, "learning_rate": 1.123524808783609e-06, "loss": 0.6431, "mean_token_accuracy": 0.787983775138855, "num_tokens": 298637048.0, "step": 7058 }, { "epoch": 0.9329896907216495, "grad_norm": 0.7344732049080661, "learning_rate": 1.1230421447425619e-06, "loss": 0.4947, "mean_token_accuracy": 0.8331857919692993, "num_tokens": 298702584.0, "step": 7059 }, { "epoch": 0.9331218609569125, "grad_norm": 0.8051670550552028, "learning_rate": 1.1225604124644583e-06, "loss": 0.6044, "mean_token_accuracy": 0.7996429204940796, "num_tokens": 298768120.0, "step": 7060 }, { "epoch": 0.9332540311921755, "grad_norm": 0.7912919464993898, "learning_rate": 1.122079612051849e-06, "loss": 0.5492, "mean_token_accuracy": 0.8178336024284363, "num_tokens": 298833656.0, "step": 7061 }, { "epoch": 0.9333862014274386, "grad_norm": 0.825484847103988, "learning_rate": 1.1215997436070855e-06, "loss": 0.6584, "mean_token_accuracy": 0.7797277569770813, "num_tokens": 298899192.0, "step": 7062 }, { "epoch": 0.9335183716627016, "grad_norm": 0.7998550261545294, "learning_rate": 1.1211208072323218e-06, "loss": 0.5783, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 298964728.0, "step": 7063 }, { "epoch": 0.9336505418979646, "grad_norm": 0.7896927702310729, "learning_rate": 1.1206428030295147e-06, "loss": 0.5565, "mean_token_accuracy": 0.816277027130127, "num_tokens": 299030264.0, "step": 7064 }, { "epoch": 0.9337827121332276, "grad_norm": 0.8360950946356027, "learning_rate": 1.1201657311004193e-06, "loss": 0.5435, "mean_token_accuracy": 0.8206415772438049, "num_tokens": 299095800.0, "step": 7065 }, { "epoch": 0.9339148823684906, "grad_norm": 0.7934188569123068, "learning_rate": 1.1196895915465964e-06, "loss": 0.5634, "mean_token_accuracy": 0.8127976059913635, "num_tokens": 299161336.0, "step": 7066 }, { "epoch": 0.9340470526037536, "grad_norm": 0.7532025769129772, "learning_rate": 1.1192143844694048e-06, "loss": 0.505, "mean_token_accuracy": 0.8316749930381775, "num_tokens": 299226872.0, "step": 7067 }, { "epoch": 0.9341792228390167, "grad_norm": 0.7947979849664392, "learning_rate": 1.1187401099700064e-06, "loss": 0.564, "mean_token_accuracy": 0.8082956671714783, "num_tokens": 299292408.0, "step": 7068 }, { "epoch": 0.9343113930742797, "grad_norm": 0.7710539424739746, "learning_rate": 1.1182667681493644e-06, "loss": 0.5165, "mean_token_accuracy": 0.8280582427978516, "num_tokens": 299357944.0, "step": 7069 }, { "epoch": 0.9344435633095427, "grad_norm": 0.7263514521458249, "learning_rate": 1.1177943591082437e-06, "loss": 0.5164, "mean_token_accuracy": 0.8274020552635193, "num_tokens": 299423480.0, "step": 7070 }, { "epoch": 0.9345757335448057, "grad_norm": 0.7953608526365825, "learning_rate": 1.1173228829472097e-06, "loss": 0.5337, "mean_token_accuracy": 0.8220913410186768, "num_tokens": 299489016.0, "step": 7071 }, { "epoch": 0.9347079037800687, "grad_norm": 0.7993581420108193, "learning_rate": 1.1168523397666312e-06, "loss": 0.5422, "mean_token_accuracy": 0.8202447891235352, "num_tokens": 299554552.0, "step": 7072 }, { "epoch": 0.9348400740153318, "grad_norm": 0.829206823870767, "learning_rate": 1.1163827296666748e-06, "loss": 0.576, "mean_token_accuracy": 0.8082956671714783, "num_tokens": 299620088.0, "step": 7073 }, { "epoch": 0.9349722442505948, "grad_norm": 0.8357506260178074, "learning_rate": 1.1159140527473127e-06, "loss": 0.59, "mean_token_accuracy": 0.8009705543518066, "num_tokens": 299685624.0, "step": 7074 }, { "epoch": 0.9351044144858578, "grad_norm": 0.8520978602622795, "learning_rate": 1.1154463091083153e-06, "loss": 0.5955, "mean_token_accuracy": 0.8014283776283264, "num_tokens": 299751160.0, "step": 7075 }, { "epoch": 0.9352365847211208, "grad_norm": 0.8103885221088931, "learning_rate": 1.114979498849256e-06, "loss": 0.6128, "mean_token_accuracy": 0.7968044281005859, "num_tokens": 299816696.0, "step": 7076 }, { "epoch": 0.9353687549563838, "grad_norm": 0.8071414272632844, "learning_rate": 1.1145136220695087e-06, "loss": 0.5787, "mean_token_accuracy": 0.8069985508918762, "num_tokens": 299882232.0, "step": 7077 }, { "epoch": 0.9355009251916468, "grad_norm": 0.7885537718355052, "learning_rate": 1.1140486788682488e-06, "loss": 0.5624, "mean_token_accuracy": 0.8143236637115479, "num_tokens": 299947768.0, "step": 7078 }, { "epoch": 0.9356330954269099, "grad_norm": 0.8548867109140216, "learning_rate": 1.1135846693444534e-06, "loss": 0.6383, "mean_token_accuracy": 0.7863508462905884, "num_tokens": 300013304.0, "step": 7079 }, { "epoch": 0.9357652656621729, "grad_norm": 0.8011636845720017, "learning_rate": 1.1131215935969004e-06, "loss": 0.6039, "mean_token_accuracy": 0.7999328374862671, "num_tokens": 300078840.0, "step": 7080 }, { "epoch": 0.9358974358974359, "grad_norm": 0.8388255261138939, "learning_rate": 1.1126594517241676e-06, "loss": 0.6006, "mean_token_accuracy": 0.8002075552940369, "num_tokens": 300144376.0, "step": 7081 }, { "epoch": 0.9360296061326989, "grad_norm": 0.819128397176786, "learning_rate": 1.1121982438246372e-06, "loss": 0.5522, "mean_token_accuracy": 0.8156818747520447, "num_tokens": 300209912.0, "step": 7082 }, { "epoch": 0.9361617763679619, "grad_norm": 0.8457042871176963, "learning_rate": 1.11173796999649e-06, "loss": 0.5925, "mean_token_accuracy": 0.8020846247673035, "num_tokens": 300275448.0, "step": 7083 }, { "epoch": 0.936293946603225, "grad_norm": 0.8153951990875739, "learning_rate": 1.1112786303377083e-06, "loss": 0.6155, "mean_token_accuracy": 0.7943627238273621, "num_tokens": 300340984.0, "step": 7084 }, { "epoch": 0.936426116838488, "grad_norm": 0.8583687262255704, "learning_rate": 1.110820224946077e-06, "loss": 0.5652, "mean_token_accuracy": 0.8112562298774719, "num_tokens": 300406520.0, "step": 7085 }, { "epoch": 0.936558287073751, "grad_norm": 0.772278803873643, "learning_rate": 1.1103627539191795e-06, "loss": 0.5582, "mean_token_accuracy": 0.812751829624176, "num_tokens": 300472056.0, "step": 7086 }, { "epoch": 0.936690457309014, "grad_norm": 0.7817638540018631, "learning_rate": 1.109906217354403e-06, "loss": 0.5376, "mean_token_accuracy": 0.8189476132392883, "num_tokens": 300537592.0, "step": 7087 }, { "epoch": 0.936822627544277, "grad_norm": 0.7897924262419317, "learning_rate": 1.109450615348934e-06, "loss": 0.5492, "mean_token_accuracy": 0.818321943283081, "num_tokens": 300603128.0, "step": 7088 }, { "epoch": 0.93695479777954, "grad_norm": 0.7918749518344356, "learning_rate": 1.108995947999762e-06, "loss": 0.5371, "mean_token_accuracy": 0.8207789063453674, "num_tokens": 300668664.0, "step": 7089 }, { "epoch": 0.9370869680148031, "grad_norm": 0.7534087473232283, "learning_rate": 1.1085422154036737e-06, "loss": 0.5586, "mean_token_accuracy": 0.8139421343803406, "num_tokens": 300734200.0, "step": 7090 }, { "epoch": 0.9372191382500661, "grad_norm": 0.753767598268375, "learning_rate": 1.1080894176572612e-06, "loss": 0.5566, "mean_token_accuracy": 0.8155444860458374, "num_tokens": 300799736.0, "step": 7091 }, { "epoch": 0.9373513084853291, "grad_norm": 0.8142514413263708, "learning_rate": 1.1076375548569153e-06, "loss": 0.5778, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 300865272.0, "step": 7092 }, { "epoch": 0.9374834787205921, "grad_norm": 0.7622084249460754, "learning_rate": 1.1071866270988285e-06, "loss": 0.5107, "mean_token_accuracy": 0.8279056549072266, "num_tokens": 300930808.0, "step": 7093 }, { "epoch": 0.9376156489558551, "grad_norm": 0.8380691356160967, "learning_rate": 1.1067366344789925e-06, "loss": 0.5866, "mean_token_accuracy": 0.8037632703781128, "num_tokens": 300996344.0, "step": 7094 }, { "epoch": 0.9377478191911182, "grad_norm": 0.8191300393743052, "learning_rate": 1.1062875770932033e-06, "loss": 0.5674, "mean_token_accuracy": 0.8109052777290344, "num_tokens": 301061880.0, "step": 7095 }, { "epoch": 0.9378799894263812, "grad_norm": 0.8504970202734703, "learning_rate": 1.1058394550370545e-06, "loss": 0.5895, "mean_token_accuracy": 0.8037327527999878, "num_tokens": 301127416.0, "step": 7096 }, { "epoch": 0.9380121596616442, "grad_norm": 0.8419780213539783, "learning_rate": 1.1053922684059418e-06, "loss": 0.5759, "mean_token_accuracy": 0.808570384979248, "num_tokens": 301192952.0, "step": 7097 }, { "epoch": 0.9381443298969072, "grad_norm": 0.769096227317464, "learning_rate": 1.1049460172950624e-06, "loss": 0.5543, "mean_token_accuracy": 0.8148730397224426, "num_tokens": 301258488.0, "step": 7098 }, { "epoch": 0.9382765001321702, "grad_norm": 0.8021933450102215, "learning_rate": 1.1045007017994145e-06, "loss": 0.5984, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 301324024.0, "step": 7099 }, { "epoch": 0.9384086703674333, "grad_norm": 0.8454553386624349, "learning_rate": 1.1040563220137953e-06, "loss": 0.6271, "mean_token_accuracy": 0.7909290790557861, "num_tokens": 301389560.0, "step": 7100 }, { "epoch": 0.9385408406026963, "grad_norm": 0.851574884402662, "learning_rate": 1.1036128780328046e-06, "loss": 0.6535, "mean_token_accuracy": 0.7815895676612854, "num_tokens": 301455096.0, "step": 7101 }, { "epoch": 0.9386730108379593, "grad_norm": 0.8829561870239582, "learning_rate": 1.103170369950842e-06, "loss": 0.6347, "mean_token_accuracy": 0.7885636687278748, "num_tokens": 301520632.0, "step": 7102 }, { "epoch": 0.9388051810732223, "grad_norm": 0.761979239288848, "learning_rate": 1.1027287978621078e-06, "loss": 0.5154, "mean_token_accuracy": 0.8290654420852661, "num_tokens": 301586168.0, "step": 7103 }, { "epoch": 0.9389373513084853, "grad_norm": 0.8367780393112092, "learning_rate": 1.1022881618606055e-06, "loss": 0.5921, "mean_token_accuracy": 0.8037785291671753, "num_tokens": 301651704.0, "step": 7104 }, { "epoch": 0.9390695215437483, "grad_norm": 0.8360654874421868, "learning_rate": 1.1018484620401354e-06, "loss": 0.5637, "mean_token_accuracy": 0.8115309476852417, "num_tokens": 301717240.0, "step": 7105 }, { "epoch": 0.9392016917790114, "grad_norm": 0.9231456527386163, "learning_rate": 1.101409698494301e-06, "loss": 0.6624, "mean_token_accuracy": 0.7809790968894958, "num_tokens": 301782776.0, "step": 7106 }, { "epoch": 0.9393338620142744, "grad_norm": 0.7669935302261862, "learning_rate": 1.100971871316505e-06, "loss": 0.5536, "mean_token_accuracy": 0.8142778873443604, "num_tokens": 301848312.0, "step": 7107 }, { "epoch": 0.9394660322495374, "grad_norm": 0.8319773177501354, "learning_rate": 1.1005349805999538e-06, "loss": 0.595, "mean_token_accuracy": 0.8004211783409119, "num_tokens": 301913848.0, "step": 7108 }, { "epoch": 0.9395982024848004, "grad_norm": 0.8284599123039199, "learning_rate": 1.1000990264376502e-06, "loss": 0.6184, "mean_token_accuracy": 0.7937675714492798, "num_tokens": 301979384.0, "step": 7109 }, { "epoch": 0.9397303727200634, "grad_norm": 0.8033292638021344, "learning_rate": 1.0996640089224006e-06, "loss": 0.5574, "mean_token_accuracy": 0.8130722641944885, "num_tokens": 302044920.0, "step": 7110 }, { "epoch": 0.9398625429553265, "grad_norm": 0.8938505945821996, "learning_rate": 1.0992299281468109e-06, "loss": 0.6618, "mean_token_accuracy": 0.7808265089988708, "num_tokens": 302110456.0, "step": 7111 }, { "epoch": 0.9399947131905895, "grad_norm": 0.7901394136226219, "learning_rate": 1.098796784203289e-06, "loss": 0.5933, "mean_token_accuracy": 0.7993529438972473, "num_tokens": 302175992.0, "step": 7112 }, { "epoch": 0.9401268834258525, "grad_norm": 0.8535152579836474, "learning_rate": 1.0983645771840398e-06, "loss": 0.5457, "mean_token_accuracy": 0.8171926736831665, "num_tokens": 302241528.0, "step": 7113 }, { "epoch": 0.9402590536611155, "grad_norm": 0.7996501088695888, "learning_rate": 1.0979333071810737e-06, "loss": 0.5422, "mean_token_accuracy": 0.8187950253486633, "num_tokens": 302307064.0, "step": 7114 }, { "epoch": 0.9403912238963785, "grad_norm": 0.810121208389307, "learning_rate": 1.0975029742861977e-06, "loss": 0.5727, "mean_token_accuracy": 0.8092265725135803, "num_tokens": 302372600.0, "step": 7115 }, { "epoch": 0.9405233941316415, "grad_norm": 0.7919660603545958, "learning_rate": 1.0970735785910212e-06, "loss": 0.5952, "mean_token_accuracy": 0.8021151423454285, "num_tokens": 302438136.0, "step": 7116 }, { "epoch": 0.9406555643669046, "grad_norm": 0.8364862979370783, "learning_rate": 1.0966451201869527e-06, "loss": 0.564, "mean_token_accuracy": 0.812034547328949, "num_tokens": 302503672.0, "step": 7117 }, { "epoch": 0.9407877346021676, "grad_norm": 0.8092534546681746, "learning_rate": 1.0962175991652036e-06, "loss": 0.592, "mean_token_accuracy": 0.8042973875999451, "num_tokens": 302569208.0, "step": 7118 }, { "epoch": 0.9409199048374306, "grad_norm": 0.7360023208960117, "learning_rate": 1.0957910156167824e-06, "loss": 0.5225, "mean_token_accuracy": 0.8239378333091736, "num_tokens": 302634744.0, "step": 7119 }, { "epoch": 0.9410520750726936, "grad_norm": 0.8023453232815442, "learning_rate": 1.095365369632501e-06, "loss": 0.6061, "mean_token_accuracy": 0.8010621666908264, "num_tokens": 302700280.0, "step": 7120 }, { "epoch": 0.9411842453079566, "grad_norm": 0.8194290976444573, "learning_rate": 1.0949406613029704e-06, "loss": 0.5615, "mean_token_accuracy": 0.8092418313026428, "num_tokens": 302765816.0, "step": 7121 }, { "epoch": 0.9413164155432197, "grad_norm": 0.8148063018656114, "learning_rate": 1.094516890718602e-06, "loss": 0.5991, "mean_token_accuracy": 0.8026950359344482, "num_tokens": 302831352.0, "step": 7122 }, { "epoch": 0.9414485857784827, "grad_norm": 0.8561281370224005, "learning_rate": 1.0940940579696075e-06, "loss": 0.6684, "mean_token_accuracy": 0.7789189219474792, "num_tokens": 302896888.0, "step": 7123 }, { "epoch": 0.9415807560137457, "grad_norm": 0.8030498688471288, "learning_rate": 1.0936721631459992e-06, "loss": 0.5959, "mean_token_accuracy": 0.8011994957923889, "num_tokens": 302962424.0, "step": 7124 }, { "epoch": 0.9417129262490087, "grad_norm": 0.8076323614461741, "learning_rate": 1.0932512063375903e-06, "loss": 0.5662, "mean_token_accuracy": 0.812965452671051, "num_tokens": 303027960.0, "step": 7125 }, { "epoch": 0.9418450964842717, "grad_norm": 0.8441109249019983, "learning_rate": 1.092831187633993e-06, "loss": 0.5946, "mean_token_accuracy": 0.8016878366470337, "num_tokens": 303093496.0, "step": 7126 }, { "epoch": 0.9419772667195347, "grad_norm": 0.81265822305372, "learning_rate": 1.0924121071246207e-06, "loss": 0.5806, "mean_token_accuracy": 0.8061744570732117, "num_tokens": 303159032.0, "step": 7127 }, { "epoch": 0.9421094369547978, "grad_norm": 0.781784224280926, "learning_rate": 1.0919939648986871e-06, "loss": 0.551, "mean_token_accuracy": 0.8189170956611633, "num_tokens": 303224568.0, "step": 7128 }, { "epoch": 0.9422416071900608, "grad_norm": 0.8108835052887886, "learning_rate": 1.091576761045206e-06, "loss": 0.5742, "mean_token_accuracy": 0.8081735968589783, "num_tokens": 303290104.0, "step": 7129 }, { "epoch": 0.9423737774253238, "grad_norm": 0.7837521668762236, "learning_rate": 1.091160495652991e-06, "loss": 0.5393, "mean_token_accuracy": 0.8193291425704956, "num_tokens": 303355640.0, "step": 7130 }, { "epoch": 0.9425059476605868, "grad_norm": 0.8178757517086058, "learning_rate": 1.090745168810657e-06, "loss": 0.5935, "mean_token_accuracy": 0.8016115427017212, "num_tokens": 303421176.0, "step": 7131 }, { "epoch": 0.9426381178958498, "grad_norm": 0.7545972255896549, "learning_rate": 1.0903307806066173e-06, "loss": 0.5502, "mean_token_accuracy": 0.8167348504066467, "num_tokens": 303486712.0, "step": 7132 }, { "epoch": 0.9427702881311129, "grad_norm": 0.8138037701968068, "learning_rate": 1.089917331129088e-06, "loss": 0.614, "mean_token_accuracy": 0.7918142080307007, "num_tokens": 303552248.0, "step": 7133 }, { "epoch": 0.9429024583663759, "grad_norm": 0.7675067896104419, "learning_rate": 1.0895048204660822e-06, "loss": 0.5359, "mean_token_accuracy": 0.8259064555168152, "num_tokens": 303617784.0, "step": 7134 }, { "epoch": 0.9430346286016389, "grad_norm": 0.7592528652977282, "learning_rate": 1.089093248705416e-06, "loss": 0.5242, "mean_token_accuracy": 0.8236631751060486, "num_tokens": 303683320.0, "step": 7135 }, { "epoch": 0.9431667988369019, "grad_norm": 0.7534002798145674, "learning_rate": 1.0886826159347034e-06, "loss": 0.5295, "mean_token_accuracy": 0.8232511281967163, "num_tokens": 303748856.0, "step": 7136 }, { "epoch": 0.9432989690721649, "grad_norm": 0.8377567085288697, "learning_rate": 1.0882729222413612e-06, "loss": 0.621, "mean_token_accuracy": 0.7946526408195496, "num_tokens": 303814392.0, "step": 7137 }, { "epoch": 0.943431139307428, "grad_norm": 0.7979552650314581, "learning_rate": 1.0878641677126032e-06, "loss": 0.5679, "mean_token_accuracy": 0.8091350197792053, "num_tokens": 303879928.0, "step": 7138 }, { "epoch": 0.943563309542691, "grad_norm": 0.8287308367827896, "learning_rate": 1.0874563524354446e-06, "loss": 0.6115, "mean_token_accuracy": 0.7974453568458557, "num_tokens": 303945464.0, "step": 7139 }, { "epoch": 0.943695479777954, "grad_norm": 0.7938982157941094, "learning_rate": 1.087049476496702e-06, "loss": 0.5865, "mean_token_accuracy": 0.806586503982544, "num_tokens": 304011000.0, "step": 7140 }, { "epoch": 0.943827650013217, "grad_norm": 0.8116935835542834, "learning_rate": 1.08664353998299e-06, "loss": 0.5749, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 304076536.0, "step": 7141 }, { "epoch": 0.94395982024848, "grad_norm": 0.7852552911734881, "learning_rate": 1.0862385429807239e-06, "loss": 0.6353, "mean_token_accuracy": 0.788945198059082, "num_tokens": 304142072.0, "step": 7142 }, { "epoch": 0.944091990483743, "grad_norm": 0.749069093131756, "learning_rate": 1.0858344855761193e-06, "loss": 0.5666, "mean_token_accuracy": 0.8124160766601562, "num_tokens": 304207608.0, "step": 7143 }, { "epoch": 0.9442241607190061, "grad_norm": 0.8321497219777867, "learning_rate": 1.0854313678551918e-06, "loss": 0.5745, "mean_token_accuracy": 0.808387279510498, "num_tokens": 304273144.0, "step": 7144 }, { "epoch": 0.9443563309542691, "grad_norm": 0.7831480468687252, "learning_rate": 1.085029189903756e-06, "loss": 0.531, "mean_token_accuracy": 0.8212672472000122, "num_tokens": 304338680.0, "step": 7145 }, { "epoch": 0.9444885011895321, "grad_norm": 0.8466265201269115, "learning_rate": 1.0846279518074285e-06, "loss": 0.5697, "mean_token_accuracy": 0.8095470666885376, "num_tokens": 304404216.0, "step": 7146 }, { "epoch": 0.9446206714247951, "grad_norm": 0.8569629972263518, "learning_rate": 1.0842276536516233e-06, "loss": 0.5886, "mean_token_accuracy": 0.8032901883125305, "num_tokens": 304469752.0, "step": 7147 }, { "epoch": 0.9447528416600581, "grad_norm": 0.7832543301764251, "learning_rate": 1.083828295521556e-06, "loss": 0.5855, "mean_token_accuracy": 0.8038090467453003, "num_tokens": 304535288.0, "step": 7148 }, { "epoch": 0.9448850118953211, "grad_norm": 0.9154501095518622, "learning_rate": 1.0834298775022411e-06, "loss": 0.5563, "mean_token_accuracy": 0.8154376745223999, "num_tokens": 304600824.0, "step": 7149 }, { "epoch": 0.9450171821305842, "grad_norm": 0.8797325351632891, "learning_rate": 1.0830323996784945e-06, "loss": 0.6217, "mean_token_accuracy": 0.7958124876022339, "num_tokens": 304666360.0, "step": 7150 }, { "epoch": 0.9451493523658472, "grad_norm": 0.7898808361606791, "learning_rate": 1.0826358621349303e-06, "loss": 0.5292, "mean_token_accuracy": 0.8239683508872986, "num_tokens": 304731896.0, "step": 7151 }, { "epoch": 0.9452815226011102, "grad_norm": 0.7414641385462925, "learning_rate": 1.082240264955963e-06, "loss": 0.539, "mean_token_accuracy": 0.8197411894798279, "num_tokens": 304797432.0, "step": 7152 }, { "epoch": 0.9454136928363732, "grad_norm": 0.8098585496711689, "learning_rate": 1.081845608225807e-06, "loss": 0.5757, "mean_token_accuracy": 0.8073495030403137, "num_tokens": 304862968.0, "step": 7153 }, { "epoch": 0.9455458630716362, "grad_norm": 0.7879695617756665, "learning_rate": 1.0814518920284772e-06, "loss": 0.5729, "mean_token_accuracy": 0.807700514793396, "num_tokens": 304928504.0, "step": 7154 }, { "epoch": 0.9456780333068993, "grad_norm": 0.7767946470035727, "learning_rate": 1.0810591164477863e-06, "loss": 0.5715, "mean_token_accuracy": 0.8110426068305969, "num_tokens": 304994040.0, "step": 7155 }, { "epoch": 0.9458102035421623, "grad_norm": 0.8055602700006794, "learning_rate": 1.0806672815673485e-06, "loss": 0.5682, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 305059576.0, "step": 7156 }, { "epoch": 0.9459423737774253, "grad_norm": 0.8045974253281368, "learning_rate": 1.080276387470578e-06, "loss": 0.58, "mean_token_accuracy": 0.8069222569465637, "num_tokens": 305125112.0, "step": 7157 }, { "epoch": 0.9460745440126883, "grad_norm": 0.7842571098971914, "learning_rate": 1.0798864342406871e-06, "loss": 0.562, "mean_token_accuracy": 0.8107526302337646, "num_tokens": 305190648.0, "step": 7158 }, { "epoch": 0.9462067142479513, "grad_norm": 0.7847979877843767, "learning_rate": 1.0794974219606894e-06, "loss": 0.5688, "mean_token_accuracy": 0.8099896311759949, "num_tokens": 305256184.0, "step": 7159 }, { "epoch": 0.9463388844832143, "grad_norm": 0.82139270552041, "learning_rate": 1.0791093507133968e-06, "loss": 0.5953, "mean_token_accuracy": 0.8009552955627441, "num_tokens": 305321720.0, "step": 7160 }, { "epoch": 0.9464710547184774, "grad_norm": 0.8717469134943988, "learning_rate": 1.0787222205814215e-06, "loss": 0.648, "mean_token_accuracy": 0.7842906713485718, "num_tokens": 305387256.0, "step": 7161 }, { "epoch": 0.9466032249537404, "grad_norm": 0.7840829110718639, "learning_rate": 1.0783360316471764e-06, "loss": 0.5977, "mean_token_accuracy": 0.8013520836830139, "num_tokens": 305452792.0, "step": 7162 }, { "epoch": 0.9467353951890034, "grad_norm": 0.8067790442706806, "learning_rate": 1.0779507839928723e-06, "loss": 0.5633, "mean_token_accuracy": 0.8144915103912354, "num_tokens": 305518328.0, "step": 7163 }, { "epoch": 0.9468675654242664, "grad_norm": 0.8053798756482892, "learning_rate": 1.07756647770052e-06, "loss": 0.5754, "mean_token_accuracy": 0.807761549949646, "num_tokens": 305583864.0, "step": 7164 }, { "epoch": 0.9469997356595294, "grad_norm": 0.8587375109847146, "learning_rate": 1.077183112851932e-06, "loss": 0.5458, "mean_token_accuracy": 0.816582202911377, "num_tokens": 305649400.0, "step": 7165 }, { "epoch": 0.9471319058947925, "grad_norm": 0.8135741713428573, "learning_rate": 1.0768006895287168e-06, "loss": 0.5838, "mean_token_accuracy": 0.806678056716919, "num_tokens": 305714936.0, "step": 7166 }, { "epoch": 0.9472640761300555, "grad_norm": 0.7541986263475444, "learning_rate": 1.0764192078122852e-06, "loss": 0.5142, "mean_token_accuracy": 0.8279666900634766, "num_tokens": 305780472.0, "step": 7167 }, { "epoch": 0.9473962463653185, "grad_norm": 0.8014140661885683, "learning_rate": 1.0760386677838463e-06, "loss": 0.5723, "mean_token_accuracy": 0.8062812685966492, "num_tokens": 305846008.0, "step": 7168 }, { "epoch": 0.9475284166005815, "grad_norm": 0.771821323841364, "learning_rate": 1.0756590695244095e-06, "loss": 0.5367, "mean_token_accuracy": 0.8209009766578674, "num_tokens": 305911544.0, "step": 7169 }, { "epoch": 0.9476605868358445, "grad_norm": 0.8017050124106969, "learning_rate": 1.0752804131147837e-06, "loss": 0.5458, "mean_token_accuracy": 0.8176962733268738, "num_tokens": 305977080.0, "step": 7170 }, { "epoch": 0.9477927570711076, "grad_norm": 0.7820269694103591, "learning_rate": 1.0749026986355757e-06, "loss": 0.5453, "mean_token_accuracy": 0.8161091208457947, "num_tokens": 306042616.0, "step": 7171 }, { "epoch": 0.9479249273063706, "grad_norm": 0.7735766260623581, "learning_rate": 1.0745259261671942e-06, "loss": 0.5277, "mean_token_accuracy": 0.822686493396759, "num_tokens": 306108152.0, "step": 7172 }, { "epoch": 0.9480570975416336, "grad_norm": 0.856686284952824, "learning_rate": 1.0741500957898459e-06, "loss": 0.6103, "mean_token_accuracy": 0.7965297102928162, "num_tokens": 306173688.0, "step": 7173 }, { "epoch": 0.9481892677768966, "grad_norm": 0.8568524163703741, "learning_rate": 1.0737752075835366e-06, "loss": 0.6164, "mean_token_accuracy": 0.7938591241836548, "num_tokens": 306239224.0, "step": 7174 }, { "epoch": 0.9483214380121596, "grad_norm": 0.7750554383206639, "learning_rate": 1.0734012616280728e-06, "loss": 0.5538, "mean_token_accuracy": 0.8119887709617615, "num_tokens": 306304760.0, "step": 7175 }, { "epoch": 0.9484536082474226, "grad_norm": 0.8253173439809265, "learning_rate": 1.0730282580030596e-06, "loss": 0.6016, "mean_token_accuracy": 0.7996429204940796, "num_tokens": 306370296.0, "step": 7176 }, { "epoch": 0.9485857784826857, "grad_norm": 0.8489449661059439, "learning_rate": 1.0726561967879023e-06, "loss": 0.5743, "mean_token_accuracy": 0.8073495030403137, "num_tokens": 306435832.0, "step": 7177 }, { "epoch": 0.9487179487179487, "grad_norm": 0.8188558399299892, "learning_rate": 1.0722850780618046e-06, "loss": 0.5216, "mean_token_accuracy": 0.8234037160873413, "num_tokens": 306501368.0, "step": 7178 }, { "epoch": 0.9488501189532117, "grad_norm": 0.8297176132411352, "learning_rate": 1.0719149019037691e-06, "loss": 0.575, "mean_token_accuracy": 0.8080515265464783, "num_tokens": 306566904.0, "step": 7179 }, { "epoch": 0.9489822891884747, "grad_norm": 0.8952007704086856, "learning_rate": 1.0715456683925994e-06, "loss": 0.6059, "mean_token_accuracy": 0.7985593676567078, "num_tokens": 306632440.0, "step": 7180 }, { "epoch": 0.9491144594237377, "grad_norm": 0.7997820835952847, "learning_rate": 1.0711773776068972e-06, "loss": 0.5724, "mean_token_accuracy": 0.8051977753639221, "num_tokens": 306697976.0, "step": 7181 }, { "epoch": 0.9492466296590008, "grad_norm": 0.7213568897369379, "learning_rate": 1.0708100296250648e-06, "loss": 0.5143, "mean_token_accuracy": 0.8291875123977661, "num_tokens": 306763512.0, "step": 7182 }, { "epoch": 0.9493787998942638, "grad_norm": 0.8350259957031132, "learning_rate": 1.0704436245253019e-06, "loss": 0.5459, "mean_token_accuracy": 0.8174368143081665, "num_tokens": 306829048.0, "step": 7183 }, { "epoch": 0.9495109701295268, "grad_norm": 0.8217924381875641, "learning_rate": 1.0700781623856085e-06, "loss": 0.5797, "mean_token_accuracy": 0.802374541759491, "num_tokens": 306894584.0, "step": 7184 }, { "epoch": 0.9496431403647898, "grad_norm": 0.8075948769881708, "learning_rate": 1.0697136432837844e-06, "loss": 0.5546, "mean_token_accuracy": 0.8159107565879822, "num_tokens": 306960120.0, "step": 7185 }, { "epoch": 0.9497753106000528, "grad_norm": 0.7986111394097148, "learning_rate": 1.0693500672974285e-06, "loss": 0.5409, "mean_token_accuracy": 0.8203210830688477, "num_tokens": 307025656.0, "step": 7186 }, { "epoch": 0.9499074808353158, "grad_norm": 0.8489627294990983, "learning_rate": 1.068987434503937e-06, "loss": 0.5599, "mean_token_accuracy": 0.812995970249176, "num_tokens": 307091192.0, "step": 7187 }, { "epoch": 0.9500396510705789, "grad_norm": 0.8555752984972091, "learning_rate": 1.0686257449805084e-06, "loss": 0.6214, "mean_token_accuracy": 0.7982236742973328, "num_tokens": 307156728.0, "step": 7188 }, { "epoch": 0.9501718213058419, "grad_norm": 0.823499193222039, "learning_rate": 1.0682649988041382e-06, "loss": 0.5659, "mean_token_accuracy": 0.8102337718009949, "num_tokens": 307222264.0, "step": 7189 }, { "epoch": 0.9503039915411049, "grad_norm": 0.8503105195090451, "learning_rate": 1.0679051960516218e-06, "loss": 0.5639, "mean_token_accuracy": 0.812934935092926, "num_tokens": 307287800.0, "step": 7190 }, { "epoch": 0.9504361617763679, "grad_norm": 0.8582292954520863, "learning_rate": 1.0675463367995535e-06, "loss": 0.6036, "mean_token_accuracy": 0.7985746264457703, "num_tokens": 307353336.0, "step": 7191 }, { "epoch": 0.9505683320116309, "grad_norm": 0.8531947379878484, "learning_rate": 1.0671884211243274e-06, "loss": 0.6611, "mean_token_accuracy": 0.7800635099411011, "num_tokens": 307418872.0, "step": 7192 }, { "epoch": 0.950700502246894, "grad_norm": 0.7854665877896, "learning_rate": 1.0668314491021362e-06, "loss": 0.5441, "mean_token_accuracy": 0.8170705437660217, "num_tokens": 307484408.0, "step": 7193 }, { "epoch": 0.950832672482157, "grad_norm": 0.7893082515558175, "learning_rate": 1.0664754208089712e-06, "loss": 0.5554, "mean_token_accuracy": 0.8143999576568604, "num_tokens": 307549944.0, "step": 7194 }, { "epoch": 0.95096484271742, "grad_norm": 0.8623311866720893, "learning_rate": 1.0661203363206235e-06, "loss": 0.6105, "mean_token_accuracy": 0.7986204624176025, "num_tokens": 307615480.0, "step": 7195 }, { "epoch": 0.951097012952683, "grad_norm": 0.7805446134812281, "learning_rate": 1.0657661957126841e-06, "loss": 0.5222, "mean_token_accuracy": 0.826059103012085, "num_tokens": 307681016.0, "step": 7196 }, { "epoch": 0.951229183187946, "grad_norm": 0.815340546796893, "learning_rate": 1.0654129990605413e-06, "loss": 0.6071, "mean_token_accuracy": 0.7983457446098328, "num_tokens": 307746552.0, "step": 7197 }, { "epoch": 0.951361353423209, "grad_norm": 0.7567909552726998, "learning_rate": 1.0650607464393834e-06, "loss": 0.5386, "mean_token_accuracy": 0.8187950253486633, "num_tokens": 307812088.0, "step": 7198 }, { "epoch": 0.9514935236584721, "grad_norm": 0.8204586378700031, "learning_rate": 1.0647094379241974e-06, "loss": 0.6002, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 307877624.0, "step": 7199 }, { "epoch": 0.9516256938937351, "grad_norm": 0.827344273472354, "learning_rate": 1.0643590735897706e-06, "loss": 0.5694, "mean_token_accuracy": 0.8084330558776855, "num_tokens": 307943160.0, "step": 7200 }, { "epoch": 0.9517578641289981, "grad_norm": 0.7728024039367389, "learning_rate": 1.064009653510687e-06, "loss": 0.5488, "mean_token_accuracy": 0.8179251551628113, "num_tokens": 308008696.0, "step": 7201 }, { "epoch": 0.9518900343642611, "grad_norm": 0.7991182106175605, "learning_rate": 1.0636611777613314e-06, "loss": 0.5448, "mean_token_accuracy": 0.8207789063453674, "num_tokens": 308074232.0, "step": 7202 }, { "epoch": 0.9520222045995241, "grad_norm": 0.7949338229873392, "learning_rate": 1.0633136464158868e-06, "loss": 0.5618, "mean_token_accuracy": 0.811942994594574, "num_tokens": 308139768.0, "step": 7203 }, { "epoch": 0.9521543748347872, "grad_norm": 0.8252516071840663, "learning_rate": 1.0629670595483364e-06, "loss": 0.5739, "mean_token_accuracy": 0.8107678890228271, "num_tokens": 308205304.0, "step": 7204 }, { "epoch": 0.9522865450700502, "grad_norm": 0.8167441499834638, "learning_rate": 1.0626214172324596e-06, "loss": 0.5712, "mean_token_accuracy": 0.8134690523147583, "num_tokens": 308270840.0, "step": 7205 }, { "epoch": 0.9524187153053132, "grad_norm": 0.8542372383787848, "learning_rate": 1.0622767195418374e-06, "loss": 0.5596, "mean_token_accuracy": 0.812843382358551, "num_tokens": 308336376.0, "step": 7206 }, { "epoch": 0.9525508855405762, "grad_norm": 0.7898387301637548, "learning_rate": 1.061932966549849e-06, "loss": 0.5609, "mean_token_accuracy": 0.8134080171585083, "num_tokens": 308401912.0, "step": 7207 }, { "epoch": 0.9526830557758392, "grad_norm": 0.8010057301886838, "learning_rate": 1.0615901583296714e-06, "loss": 0.574, "mean_token_accuracy": 0.8111036419868469, "num_tokens": 308467448.0, "step": 7208 }, { "epoch": 0.9528152260111022, "grad_norm": 0.7723462113885547, "learning_rate": 1.061248294954282e-06, "loss": 0.5511, "mean_token_accuracy": 0.8165974617004395, "num_tokens": 308532984.0, "step": 7209 }, { "epoch": 0.9529473962463653, "grad_norm": 0.7785081323195818, "learning_rate": 1.0609073764964561e-06, "loss": 0.5458, "mean_token_accuracy": 0.8187950253486633, "num_tokens": 308598520.0, "step": 7210 }, { "epoch": 0.9530795664816283, "grad_norm": 0.7774771035417666, "learning_rate": 1.0605674030287683e-06, "loss": 0.5252, "mean_token_accuracy": 0.8249603509902954, "num_tokens": 308664056.0, "step": 7211 }, { "epoch": 0.9532117367168914, "grad_norm": 0.8507484750659461, "learning_rate": 1.0602283746235916e-06, "loss": 0.5556, "mean_token_accuracy": 0.8153308629989624, "num_tokens": 308729592.0, "step": 7212 }, { "epoch": 0.9533439069521544, "grad_norm": 0.8615236140996848, "learning_rate": 1.059890291353098e-06, "loss": 0.5662, "mean_token_accuracy": 0.8106153011322021, "num_tokens": 308795128.0, "step": 7213 }, { "epoch": 0.9534760771874174, "grad_norm": 0.8021506552111549, "learning_rate": 1.0595531532892588e-06, "loss": 0.589, "mean_token_accuracy": 0.803305447101593, "num_tokens": 308860664.0, "step": 7214 }, { "epoch": 0.9536082474226805, "grad_norm": 0.8013858726568993, "learning_rate": 1.0592169605038436e-06, "loss": 0.5724, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 308926200.0, "step": 7215 }, { "epoch": 0.9537404176579435, "grad_norm": 0.8158899826526281, "learning_rate": 1.0588817130684203e-06, "loss": 0.5756, "mean_token_accuracy": 0.8087687492370605, "num_tokens": 308991736.0, "step": 7216 }, { "epoch": 0.9538725878932065, "grad_norm": 0.7898380961852774, "learning_rate": 1.058547411054356e-06, "loss": 0.5529, "mean_token_accuracy": 0.816338062286377, "num_tokens": 309057272.0, "step": 7217 }, { "epoch": 0.9540047581284695, "grad_norm": 0.8495439289094308, "learning_rate": 1.0582140545328179e-06, "loss": 0.6057, "mean_token_accuracy": 0.8012452721595764, "num_tokens": 309122808.0, "step": 7218 }, { "epoch": 0.9541369283637325, "grad_norm": 0.8004933646952486, "learning_rate": 1.057881643574769e-06, "loss": 0.5375, "mean_token_accuracy": 0.8221981525421143, "num_tokens": 309188344.0, "step": 7219 }, { "epoch": 0.9542690985989956, "grad_norm": 0.8036742129518907, "learning_rate": 1.057550178250974e-06, "loss": 0.5875, "mean_token_accuracy": 0.8052130341529846, "num_tokens": 309253880.0, "step": 7220 }, { "epoch": 0.9544012688342586, "grad_norm": 0.7770973716074849, "learning_rate": 1.0572196586319943e-06, "loss": 0.6017, "mean_token_accuracy": 0.7997344732284546, "num_tokens": 309319416.0, "step": 7221 }, { "epoch": 0.9545334390695216, "grad_norm": 0.740353128461022, "learning_rate": 1.0568900847881907e-06, "loss": 0.5313, "mean_token_accuracy": 0.8214961290359497, "num_tokens": 309384952.0, "step": 7222 }, { "epoch": 0.9546656093047846, "grad_norm": 0.8294490914885101, "learning_rate": 1.0565614567897222e-06, "loss": 0.5542, "mean_token_accuracy": 0.8201227188110352, "num_tokens": 309450488.0, "step": 7223 }, { "epoch": 0.9547977795400476, "grad_norm": 0.8262367461815735, "learning_rate": 1.0562337747065477e-06, "loss": 0.5933, "mean_token_accuracy": 0.8006653785705566, "num_tokens": 309516024.0, "step": 7224 }, { "epoch": 0.9549299497753106, "grad_norm": 0.8228425626423123, "learning_rate": 1.0559070386084234e-06, "loss": 0.6022, "mean_token_accuracy": 0.7982847094535828, "num_tokens": 309581560.0, "step": 7225 }, { "epoch": 0.9550621200105737, "grad_norm": 0.8070767438945846, "learning_rate": 1.0555812485649045e-06, "loss": 0.6071, "mean_token_accuracy": 0.7955682873725891, "num_tokens": 309647096.0, "step": 7226 }, { "epoch": 0.9551942902458367, "grad_norm": 0.7817015530573015, "learning_rate": 1.0552564046453446e-06, "loss": 0.5816, "mean_token_accuracy": 0.8043431639671326, "num_tokens": 309712632.0, "step": 7227 }, { "epoch": 0.9553264604810997, "grad_norm": 0.8427064435560495, "learning_rate": 1.0549325069188974e-06, "loss": 0.5727, "mean_token_accuracy": 0.8069069981575012, "num_tokens": 309778168.0, "step": 7228 }, { "epoch": 0.9554586307163627, "grad_norm": 0.7929390427023101, "learning_rate": 1.0546095554545122e-06, "loss": 0.59, "mean_token_accuracy": 0.8030460476875305, "num_tokens": 309843704.0, "step": 7229 }, { "epoch": 0.9555908009516257, "grad_norm": 0.8475297622619041, "learning_rate": 1.0542875503209405e-06, "loss": 0.5998, "mean_token_accuracy": 0.7997955083847046, "num_tokens": 309909240.0, "step": 7230 }, { "epoch": 0.9557229711868888, "grad_norm": 0.8253509591194838, "learning_rate": 1.0539664915867286e-06, "loss": 0.6312, "mean_token_accuracy": 0.7884110808372498, "num_tokens": 309974776.0, "step": 7231 }, { "epoch": 0.9558551414221518, "grad_norm": 0.7486737175319219, "learning_rate": 1.053646379320225e-06, "loss": 0.5189, "mean_token_accuracy": 0.8257386088371277, "num_tokens": 310040312.0, "step": 7232 }, { "epoch": 0.9559873116574148, "grad_norm": 0.8030336249701185, "learning_rate": 1.0533272135895733e-06, "loss": 0.5466, "mean_token_accuracy": 0.8169942498207092, "num_tokens": 310105848.0, "step": 7233 }, { "epoch": 0.9561194818926778, "grad_norm": 0.823796429107081, "learning_rate": 1.0530089944627188e-06, "loss": 0.5254, "mean_token_accuracy": 0.8228543400764465, "num_tokens": 310171384.0, "step": 7234 }, { "epoch": 0.9562516521279408, "grad_norm": 0.7883382279945708, "learning_rate": 1.0526917220074024e-06, "loss": 0.5883, "mean_token_accuracy": 0.8028934001922607, "num_tokens": 310236920.0, "step": 7235 }, { "epoch": 0.9563838223632039, "grad_norm": 0.7944764255967303, "learning_rate": 1.0523753962911654e-06, "loss": 0.5801, "mean_token_accuracy": 0.8067238330841064, "num_tokens": 310302456.0, "step": 7236 }, { "epoch": 0.9565159925984669, "grad_norm": 0.850749476080457, "learning_rate": 1.052060017381347e-06, "loss": 0.6359, "mean_token_accuracy": 0.7881516218185425, "num_tokens": 310367992.0, "step": 7237 }, { "epoch": 0.9566481628337299, "grad_norm": 0.8717633766065295, "learning_rate": 1.0517455853450839e-06, "loss": 0.5957, "mean_token_accuracy": 0.8036106824874878, "num_tokens": 310433528.0, "step": 7238 }, { "epoch": 0.9567803330689929, "grad_norm": 0.8071020264974159, "learning_rate": 1.0514321002493133e-06, "loss": 0.5677, "mean_token_accuracy": 0.8101574778556824, "num_tokens": 310499064.0, "step": 7239 }, { "epoch": 0.9569125033042559, "grad_norm": 0.7946313037481042, "learning_rate": 1.0511195621607694e-06, "loss": 0.6124, "mean_token_accuracy": 0.7950799465179443, "num_tokens": 310564600.0, "step": 7240 }, { "epoch": 0.9570446735395189, "grad_norm": 0.7991192982098748, "learning_rate": 1.0508079711459846e-06, "loss": 0.5747, "mean_token_accuracy": 0.8102185130119324, "num_tokens": 310630136.0, "step": 7241 }, { "epoch": 0.957176843774782, "grad_norm": 0.8513967918341756, "learning_rate": 1.05049732727129e-06, "loss": 0.5497, "mean_token_accuracy": 0.8156360387802124, "num_tokens": 310695672.0, "step": 7242 }, { "epoch": 0.957309014010045, "grad_norm": 0.7603140288032364, "learning_rate": 1.0501876306028157e-06, "loss": 0.5447, "mean_token_accuracy": 0.817207932472229, "num_tokens": 310761208.0, "step": 7243 }, { "epoch": 0.957441184245308, "grad_norm": 0.804090700982706, "learning_rate": 1.04987888120649e-06, "loss": 0.5645, "mean_token_accuracy": 0.8096843957901001, "num_tokens": 310826744.0, "step": 7244 }, { "epoch": 0.957573354480571, "grad_norm": 0.8187610881972137, "learning_rate": 1.0495710791480375e-06, "loss": 0.6228, "mean_token_accuracy": 0.7927908897399902, "num_tokens": 310892280.0, "step": 7245 }, { "epoch": 0.957705524715834, "grad_norm": 0.7901382998777142, "learning_rate": 1.0492642244929847e-06, "loss": 0.5572, "mean_token_accuracy": 0.8133927583694458, "num_tokens": 310957816.0, "step": 7246 }, { "epoch": 0.957837694951097, "grad_norm": 0.7984845040223862, "learning_rate": 1.048958317306654e-06, "loss": 0.5904, "mean_token_accuracy": 0.8008027076721191, "num_tokens": 311023352.0, "step": 7247 }, { "epoch": 0.9579698651863601, "grad_norm": 0.8070240433352851, "learning_rate": 1.0486533576541658e-06, "loss": 0.564, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 311088888.0, "step": 7248 }, { "epoch": 0.9581020354216231, "grad_norm": 0.8537878728822852, "learning_rate": 1.048349345600441e-06, "loss": 0.5512, "mean_token_accuracy": 0.8153461217880249, "num_tokens": 311154424.0, "step": 7249 }, { "epoch": 0.9582342056568861, "grad_norm": 0.7476355742946929, "learning_rate": 1.0480462812101965e-06, "loss": 0.5394, "mean_token_accuracy": 0.8190391659736633, "num_tokens": 311219960.0, "step": 7250 }, { "epoch": 0.9583663758921491, "grad_norm": 0.8334297938887655, "learning_rate": 1.0477441645479486e-06, "loss": 0.5499, "mean_token_accuracy": 0.8163533210754395, "num_tokens": 311285496.0, "step": 7251 }, { "epoch": 0.9584985461274121, "grad_norm": 0.7408316495340896, "learning_rate": 1.0474429956780119e-06, "loss": 0.523, "mean_token_accuracy": 0.8276614546775818, "num_tokens": 311351032.0, "step": 7252 }, { "epoch": 0.9586307163626752, "grad_norm": 0.8306851357637589, "learning_rate": 1.0471427746644985e-06, "loss": 0.5915, "mean_token_accuracy": 0.8042516112327576, "num_tokens": 311416568.0, "step": 7253 }, { "epoch": 0.9587628865979382, "grad_norm": 0.7470877445303716, "learning_rate": 1.046843501571319e-06, "loss": 0.5377, "mean_token_accuracy": 0.8220760822296143, "num_tokens": 311482104.0, "step": 7254 }, { "epoch": 0.9588950568332012, "grad_norm": 0.8257671666221463, "learning_rate": 1.0465451764621838e-06, "loss": 0.6068, "mean_token_accuracy": 0.7959345579147339, "num_tokens": 311547640.0, "step": 7255 }, { "epoch": 0.9590272270684642, "grad_norm": 0.7482372439814358, "learning_rate": 1.0462477994005982e-06, "loss": 0.5133, "mean_token_accuracy": 0.8284397721290588, "num_tokens": 311613176.0, "step": 7256 }, { "epoch": 0.9591593973037272, "grad_norm": 0.8049043961681738, "learning_rate": 1.045951370449869e-06, "loss": 0.5857, "mean_token_accuracy": 0.8028476238250732, "num_tokens": 311678712.0, "step": 7257 }, { "epoch": 0.9592915675389903, "grad_norm": 0.8366703543784668, "learning_rate": 1.0456558896730991e-06, "loss": 0.6136, "mean_token_accuracy": 0.7964381575584412, "num_tokens": 311744248.0, "step": 7258 }, { "epoch": 0.9594237377742533, "grad_norm": 0.8425670117643449, "learning_rate": 1.0453613571331899e-06, "loss": 0.5881, "mean_token_accuracy": 0.8019167184829712, "num_tokens": 311809784.0, "step": 7259 }, { "epoch": 0.9595559080095163, "grad_norm": 0.7376509871850112, "learning_rate": 1.045067772892842e-06, "loss": 0.5107, "mean_token_accuracy": 0.8279972076416016, "num_tokens": 311875320.0, "step": 7260 }, { "epoch": 0.9596880782447793, "grad_norm": 0.779807209972782, "learning_rate": 1.044775137014553e-06, "loss": 0.5185, "mean_token_accuracy": 0.8263490200042725, "num_tokens": 311940856.0, "step": 7261 }, { "epoch": 0.9598202484800423, "grad_norm": 0.7500870742547895, "learning_rate": 1.0444834495606184e-06, "loss": 0.5517, "mean_token_accuracy": 0.813179075717926, "num_tokens": 312006392.0, "step": 7262 }, { "epoch": 0.9599524187153053, "grad_norm": 0.7969012994428025, "learning_rate": 1.0441927105931333e-06, "loss": 0.5621, "mean_token_accuracy": 0.8121413588523865, "num_tokens": 312071928.0, "step": 7263 }, { "epoch": 0.9600845889505684, "grad_norm": 0.7753274973589548, "learning_rate": 1.0439029201739895e-06, "loss": 0.5601, "mean_token_accuracy": 0.8116072416305542, "num_tokens": 312137464.0, "step": 7264 }, { "epoch": 0.9602167591858314, "grad_norm": 0.8241796490424338, "learning_rate": 1.0436140783648773e-06, "loss": 0.6305, "mean_token_accuracy": 0.7885026335716248, "num_tokens": 312203000.0, "step": 7265 }, { "epoch": 0.9603489294210944, "grad_norm": 0.8223800992685033, "learning_rate": 1.0433261852272855e-06, "loss": 0.5663, "mean_token_accuracy": 0.8104321956634521, "num_tokens": 312268536.0, "step": 7266 }, { "epoch": 0.9604810996563574, "grad_norm": 0.747760873415913, "learning_rate": 1.0430392408225e-06, "loss": 0.5816, "mean_token_accuracy": 0.8043126463890076, "num_tokens": 312334072.0, "step": 7267 }, { "epoch": 0.9606132698916204, "grad_norm": 0.7784215411719527, "learning_rate": 1.0427532452116055e-06, "loss": 0.5854, "mean_token_accuracy": 0.8036564588546753, "num_tokens": 312399608.0, "step": 7268 }, { "epoch": 0.9607454401268835, "grad_norm": 0.820641037805614, "learning_rate": 1.0424681984554846e-06, "loss": 0.6104, "mean_token_accuracy": 0.798071026802063, "num_tokens": 312465144.0, "step": 7269 }, { "epoch": 0.9608776103621465, "grad_norm": 0.7980896653426995, "learning_rate": 1.0421841006148175e-06, "loss": 0.5388, "mean_token_accuracy": 0.8190239071846008, "num_tokens": 312530680.0, "step": 7270 }, { "epoch": 0.9610097805974095, "grad_norm": 0.7423003716703193, "learning_rate": 1.0419009517500831e-06, "loss": 0.5061, "mean_token_accuracy": 0.8293706774711609, "num_tokens": 312596216.0, "step": 7271 }, { "epoch": 0.9611419508326725, "grad_norm": 0.8539793025605933, "learning_rate": 1.0416187519215575e-06, "loss": 0.6382, "mean_token_accuracy": 0.7856793999671936, "num_tokens": 312661752.0, "step": 7272 }, { "epoch": 0.9612741210679355, "grad_norm": 0.8254485333002308, "learning_rate": 1.0413375011893153e-06, "loss": 0.604, "mean_token_accuracy": 0.7961634993553162, "num_tokens": 312727288.0, "step": 7273 }, { "epoch": 0.9614062913031985, "grad_norm": 0.7655892536670124, "learning_rate": 1.041057199613229e-06, "loss": 0.5799, "mean_token_accuracy": 0.8092876076698303, "num_tokens": 312792824.0, "step": 7274 }, { "epoch": 0.9615384615384616, "grad_norm": 0.8213730625093363, "learning_rate": 1.040777847252969e-06, "loss": 0.6131, "mean_token_accuracy": 0.7937675714492798, "num_tokens": 312858360.0, "step": 7275 }, { "epoch": 0.9616706317737246, "grad_norm": 0.7276296200573094, "learning_rate": 1.0404994441680034e-06, "loss": 0.5341, "mean_token_accuracy": 0.8214503526687622, "num_tokens": 312923896.0, "step": 7276 }, { "epoch": 0.9618028020089876, "grad_norm": 0.7804917930058415, "learning_rate": 1.040221990417598e-06, "loss": 0.5569, "mean_token_accuracy": 0.8125686645507812, "num_tokens": 312989432.0, "step": 7277 }, { "epoch": 0.9619349722442506, "grad_norm": 0.8375515809889811, "learning_rate": 1.0399454860608176e-06, "loss": 0.5778, "mean_token_accuracy": 0.806525468826294, "num_tokens": 313054968.0, "step": 7278 }, { "epoch": 0.9620671424795136, "grad_norm": 0.8087282869830145, "learning_rate": 1.0396699311565248e-06, "loss": 0.5838, "mean_token_accuracy": 0.8073647618293762, "num_tokens": 313120504.0, "step": 7279 }, { "epoch": 0.9621993127147767, "grad_norm": 0.8397849108793652, "learning_rate": 1.0393953257633777e-06, "loss": 0.5807, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 313186040.0, "step": 7280 }, { "epoch": 0.9623314829500397, "grad_norm": 0.8289139608769361, "learning_rate": 1.0391216699398356e-06, "loss": 0.5568, "mean_token_accuracy": 0.8141099810600281, "num_tokens": 313251576.0, "step": 7281 }, { "epoch": 0.9624636531853027, "grad_norm": 0.7817303956701656, "learning_rate": 1.0388489637441531e-06, "loss": 0.5792, "mean_token_accuracy": 0.8071358799934387, "num_tokens": 313317112.0, "step": 7282 }, { "epoch": 0.9625958234205657, "grad_norm": 0.7832094649685192, "learning_rate": 1.0385772072343843e-06, "loss": 0.5761, "mean_token_accuracy": 0.8072274327278137, "num_tokens": 313382648.0, "step": 7283 }, { "epoch": 0.9627279936558287, "grad_norm": 0.7951426832993749, "learning_rate": 1.03830640046838e-06, "loss": 0.5495, "mean_token_accuracy": 0.8170247673988342, "num_tokens": 313448184.0, "step": 7284 }, { "epoch": 0.9628601638910917, "grad_norm": 0.7374738161271408, "learning_rate": 1.0380365435037904e-06, "loss": 0.5256, "mean_token_accuracy": 0.8253265619277954, "num_tokens": 313513720.0, "step": 7285 }, { "epoch": 0.9629923341263548, "grad_norm": 0.8001264783607893, "learning_rate": 1.0377676363980605e-06, "loss": 0.5652, "mean_token_accuracy": 0.8109510540962219, "num_tokens": 313579256.0, "step": 7286 }, { "epoch": 0.9631245043616178, "grad_norm": 0.7685833388566157, "learning_rate": 1.0374996792084366e-06, "loss": 0.5353, "mean_token_accuracy": 0.8195275068283081, "num_tokens": 313644792.0, "step": 7287 }, { "epoch": 0.9632566745968808, "grad_norm": 0.7731128043269057, "learning_rate": 1.037232671991961e-06, "loss": 0.4989, "mean_token_accuracy": 0.8337810039520264, "num_tokens": 313710328.0, "step": 7288 }, { "epoch": 0.9633888448321438, "grad_norm": 0.7912676912499821, "learning_rate": 1.0369666148054733e-06, "loss": 0.5346, "mean_token_accuracy": 0.8217708468437195, "num_tokens": 313775864.0, "step": 7289 }, { "epoch": 0.9635210150674068, "grad_norm": 0.7928314770685378, "learning_rate": 1.0367015077056121e-06, "loss": 0.5791, "mean_token_accuracy": 0.8060218691825867, "num_tokens": 313841400.0, "step": 7290 }, { "epoch": 0.9636531853026699, "grad_norm": 0.8533107779313345, "learning_rate": 1.0364373507488126e-06, "loss": 0.6402, "mean_token_accuracy": 0.7861677408218384, "num_tokens": 313906936.0, "step": 7291 }, { "epoch": 0.9637853555379329, "grad_norm": 0.7614310326197542, "learning_rate": 1.036174143991309e-06, "loss": 0.5222, "mean_token_accuracy": 0.8243498802185059, "num_tokens": 313972472.0, "step": 7292 }, { "epoch": 0.9639175257731959, "grad_norm": 0.7673228081809855, "learning_rate": 1.035911887489132e-06, "loss": 0.5838, "mean_token_accuracy": 0.8052893280982971, "num_tokens": 314038008.0, "step": 7293 }, { "epoch": 0.9640496960084589, "grad_norm": 0.8348429114066127, "learning_rate": 1.035650581298111e-06, "loss": 0.6064, "mean_token_accuracy": 0.7968196868896484, "num_tokens": 314103544.0, "step": 7294 }, { "epoch": 0.9641818662437219, "grad_norm": 0.8466657333756358, "learning_rate": 1.0353902254738722e-06, "loss": 0.6013, "mean_token_accuracy": 0.8007569313049316, "num_tokens": 314169080.0, "step": 7295 }, { "epoch": 0.964314036478985, "grad_norm": 0.7673890055567572, "learning_rate": 1.0351308200718399e-06, "loss": 0.5766, "mean_token_accuracy": 0.8060829043388367, "num_tokens": 314234616.0, "step": 7296 }, { "epoch": 0.964446206714248, "grad_norm": 0.781414312480242, "learning_rate": 1.0348723651472362e-06, "loss": 0.5876, "mean_token_accuracy": 0.803030788898468, "num_tokens": 314300152.0, "step": 7297 }, { "epoch": 0.964578376949511, "grad_norm": 0.7594253841381003, "learning_rate": 1.034614860755081e-06, "loss": 0.5891, "mean_token_accuracy": 0.8015046715736389, "num_tokens": 314365688.0, "step": 7298 }, { "epoch": 0.964710547184774, "grad_norm": 0.7551021320895671, "learning_rate": 1.0343583069501911e-06, "loss": 0.5829, "mean_token_accuracy": 0.8051672577857971, "num_tokens": 314431224.0, "step": 7299 }, { "epoch": 0.964842717420037, "grad_norm": 0.7778992191898761, "learning_rate": 1.0341027037871824e-06, "loss": 0.5466, "mean_token_accuracy": 0.8157276511192322, "num_tokens": 314496760.0, "step": 7300 }, { "epoch": 0.9649748876553, "grad_norm": 0.7936692594332005, "learning_rate": 1.0338480513204663e-06, "loss": 0.5591, "mean_token_accuracy": 0.8163838386535645, "num_tokens": 314562296.0, "step": 7301 }, { "epoch": 0.9651070578905631, "grad_norm": 0.8081706472706276, "learning_rate": 1.0335943496042535e-06, "loss": 0.564, "mean_token_accuracy": 0.8131332993507385, "num_tokens": 314627832.0, "step": 7302 }, { "epoch": 0.9652392281258261, "grad_norm": 0.8178841820795072, "learning_rate": 1.0333415986925522e-06, "loss": 0.589, "mean_token_accuracy": 0.8043889403343201, "num_tokens": 314693368.0, "step": 7303 }, { "epoch": 0.9653713983610891, "grad_norm": 0.8435152802724627, "learning_rate": 1.033089798639167e-06, "loss": 0.6261, "mean_token_accuracy": 0.7934165596961975, "num_tokens": 314758904.0, "step": 7304 }, { "epoch": 0.9655035685963521, "grad_norm": 0.7836206160305582, "learning_rate": 1.0328389494977018e-06, "loss": 0.5657, "mean_token_accuracy": 0.8120803236961365, "num_tokens": 314824440.0, "step": 7305 }, { "epoch": 0.9656357388316151, "grad_norm": 0.8022703704615924, "learning_rate": 1.0325890513215566e-06, "loss": 0.5847, "mean_token_accuracy": 0.8043279051780701, "num_tokens": 314889976.0, "step": 7306 }, { "epoch": 0.9657679090668781, "grad_norm": 0.8330367145249932, "learning_rate": 1.032340104163929e-06, "loss": 0.5637, "mean_token_accuracy": 0.8115462064743042, "num_tokens": 314955512.0, "step": 7307 }, { "epoch": 0.9659000793021412, "grad_norm": 0.8353252550706239, "learning_rate": 1.0320921080778159e-06, "loss": 0.5916, "mean_token_accuracy": 0.7996734380722046, "num_tokens": 315021048.0, "step": 7308 }, { "epoch": 0.9660322495374042, "grad_norm": 0.8477379286468503, "learning_rate": 1.031845063116009e-06, "loss": 0.5889, "mean_token_accuracy": 0.8046178817749023, "num_tokens": 315086584.0, "step": 7309 }, { "epoch": 0.9661644197726672, "grad_norm": 0.8235536500617505, "learning_rate": 1.0315989693311005e-06, "loss": 0.5653, "mean_token_accuracy": 0.808722972869873, "num_tokens": 315152120.0, "step": 7310 }, { "epoch": 0.9662965900079302, "grad_norm": 0.8095511197395496, "learning_rate": 1.0313538267754777e-06, "loss": 0.5906, "mean_token_accuracy": 0.802466094493866, "num_tokens": 315217656.0, "step": 7311 }, { "epoch": 0.9664287602431932, "grad_norm": 0.8011988699320693, "learning_rate": 1.0311096355013265e-06, "loss": 0.6012, "mean_token_accuracy": 0.8000702261924744, "num_tokens": 315283192.0, "step": 7312 }, { "epoch": 0.9665609304784563, "grad_norm": 0.8492845544903234, "learning_rate": 1.03086639556063e-06, "loss": 0.5736, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 315348728.0, "step": 7313 }, { "epoch": 0.9666931007137193, "grad_norm": 0.7719885341064912, "learning_rate": 1.0306241070051695e-06, "loss": 0.5389, "mean_token_accuracy": 0.8205194473266602, "num_tokens": 315414264.0, "step": 7314 }, { "epoch": 0.9668252709489823, "grad_norm": 0.7788439523300992, "learning_rate": 1.0303827698865224e-06, "loss": 0.5691, "mean_token_accuracy": 0.8122329115867615, "num_tokens": 315479800.0, "step": 7315 }, { "epoch": 0.9669574411842453, "grad_norm": 0.7986947282939442, "learning_rate": 1.0301423842560647e-06, "loss": 0.5565, "mean_token_accuracy": 0.8133622407913208, "num_tokens": 315545336.0, "step": 7316 }, { "epoch": 0.9670896114195083, "grad_norm": 0.7733149293446702, "learning_rate": 1.02990295016497e-06, "loss": 0.5408, "mean_token_accuracy": 0.8196801543235779, "num_tokens": 315610872.0, "step": 7317 }, { "epoch": 0.9672217816547714, "grad_norm": 0.7583436578471339, "learning_rate": 1.0296644676642081e-06, "loss": 0.5328, "mean_token_accuracy": 0.8212214708328247, "num_tokens": 315676408.0, "step": 7318 }, { "epoch": 0.9673539518900344, "grad_norm": 0.8362141231386014, "learning_rate": 1.0294269368045468e-06, "loss": 0.5323, "mean_token_accuracy": 0.8248839974403381, "num_tokens": 315741944.0, "step": 7319 }, { "epoch": 0.9674861221252974, "grad_norm": 0.8299385397789529, "learning_rate": 1.0291903576365523e-06, "loss": 0.5959, "mean_token_accuracy": 0.8014131188392639, "num_tokens": 315807480.0, "step": 7320 }, { "epoch": 0.9676182923605604, "grad_norm": 0.8294035199038629, "learning_rate": 1.0289547302105867e-06, "loss": 0.5293, "mean_token_accuracy": 0.8262879848480225, "num_tokens": 315873016.0, "step": 7321 }, { "epoch": 0.9677504625958234, "grad_norm": 0.8243431462607049, "learning_rate": 1.02872005457681e-06, "loss": 0.5865, "mean_token_accuracy": 0.8008790016174316, "num_tokens": 315938552.0, "step": 7322 }, { "epoch": 0.9678826328310864, "grad_norm": 0.8077269438914599, "learning_rate": 1.0284863307851804e-06, "loss": 0.5413, "mean_token_accuracy": 0.8216487765312195, "num_tokens": 316004088.0, "step": 7323 }, { "epoch": 0.9680148030663495, "grad_norm": 0.8069664527654927, "learning_rate": 1.0282535588854527e-06, "loss": 0.5722, "mean_token_accuracy": 0.8106000423431396, "num_tokens": 316069624.0, "step": 7324 }, { "epoch": 0.9681469733016125, "grad_norm": 0.7956800962283077, "learning_rate": 1.0280217389271786e-06, "loss": 0.5682, "mean_token_accuracy": 0.8096081018447876, "num_tokens": 316135160.0, "step": 7325 }, { "epoch": 0.9682791435368755, "grad_norm": 0.8401924321036602, "learning_rate": 1.027790870959708e-06, "loss": 0.6391, "mean_token_accuracy": 0.7868544459342957, "num_tokens": 316200696.0, "step": 7326 }, { "epoch": 0.9684113137721385, "grad_norm": 0.7933033576624303, "learning_rate": 1.0275609550321881e-06, "loss": 0.5648, "mean_token_accuracy": 0.812126100063324, "num_tokens": 316266232.0, "step": 7327 }, { "epoch": 0.9685434840074015, "grad_norm": 0.80995750672725, "learning_rate": 1.0273319911935633e-06, "loss": 0.5888, "mean_token_accuracy": 0.8050299286842346, "num_tokens": 316331768.0, "step": 7328 }, { "epoch": 0.9686756542426646, "grad_norm": 0.8085455250062562, "learning_rate": 1.0271039794925749e-06, "loss": 0.5807, "mean_token_accuracy": 0.8048009872436523, "num_tokens": 316397304.0, "step": 7329 }, { "epoch": 0.9688078244779276, "grad_norm": 0.8460096098036942, "learning_rate": 1.0268769199777616e-06, "loss": 0.6211, "mean_token_accuracy": 0.7901507616043091, "num_tokens": 316462840.0, "step": 7330 }, { "epoch": 0.9689399947131906, "grad_norm": 0.7608254460511407, "learning_rate": 1.0266508126974602e-06, "loss": 0.517, "mean_token_accuracy": 0.8296148180961609, "num_tokens": 316528376.0, "step": 7331 }, { "epoch": 0.9690721649484536, "grad_norm": 0.7870395619055452, "learning_rate": 1.026425657699804e-06, "loss": 0.5769, "mean_token_accuracy": 0.8081125617027283, "num_tokens": 316593912.0, "step": 7332 }, { "epoch": 0.9692043351837166, "grad_norm": 0.803106007956432, "learning_rate": 1.0262014550327237e-06, "loss": 0.5974, "mean_token_accuracy": 0.8004211783409119, "num_tokens": 316659448.0, "step": 7333 }, { "epoch": 0.9693365054189796, "grad_norm": 0.8424998736119736, "learning_rate": 1.0259782047439477e-06, "loss": 0.6201, "mean_token_accuracy": 0.792348325252533, "num_tokens": 316724984.0, "step": 7334 }, { "epoch": 0.9694686756542427, "grad_norm": 0.7672872251833971, "learning_rate": 1.0257559068810008e-06, "loss": 0.5104, "mean_token_accuracy": 0.8311713933944702, "num_tokens": 316790520.0, "step": 7335 }, { "epoch": 0.9696008458895057, "grad_norm": 0.7869679177180339, "learning_rate": 1.0255345614912063e-06, "loss": 0.5451, "mean_token_accuracy": 0.8192833662033081, "num_tokens": 316856056.0, "step": 7336 }, { "epoch": 0.9697330161247687, "grad_norm": 0.7973451004427002, "learning_rate": 1.025314168621683e-06, "loss": 0.5666, "mean_token_accuracy": 0.8107678890228271, "num_tokens": 316921592.0, "step": 7337 }, { "epoch": 0.9698651863600317, "grad_norm": 0.7749296182624715, "learning_rate": 1.0250947283193482e-06, "loss": 0.5495, "mean_token_accuracy": 0.8143694400787354, "num_tokens": 316987128.0, "step": 7338 }, { "epoch": 0.9699973565952947, "grad_norm": 0.7222153928400741, "learning_rate": 1.0248762406309173e-06, "loss": 0.5131, "mean_token_accuracy": 0.8279361724853516, "num_tokens": 317052664.0, "step": 7339 }, { "epoch": 0.9701295268305578, "grad_norm": 0.7804601503556108, "learning_rate": 1.0246587056029004e-06, "loss": 0.5897, "mean_token_accuracy": 0.8033512234687805, "num_tokens": 317118200.0, "step": 7340 }, { "epoch": 0.9702616970658208, "grad_norm": 0.8250729982702499, "learning_rate": 1.0244421232816064e-06, "loss": 0.5596, "mean_token_accuracy": 0.8113631010055542, "num_tokens": 317183736.0, "step": 7341 }, { "epoch": 0.9703938673010838, "grad_norm": 0.855187356993962, "learning_rate": 1.0242264937131422e-06, "loss": 0.6418, "mean_token_accuracy": 0.7862898111343384, "num_tokens": 317249272.0, "step": 7342 }, { "epoch": 0.9705260375363468, "grad_norm": 0.7696385370612443, "learning_rate": 1.0240118169434094e-06, "loss": 0.5289, "mean_token_accuracy": 0.822655975818634, "num_tokens": 317314808.0, "step": 7343 }, { "epoch": 0.9706582077716098, "grad_norm": 0.7254475066785017, "learning_rate": 1.0237980930181094e-06, "loss": 0.5112, "mean_token_accuracy": 0.8312324285507202, "num_tokens": 317380344.0, "step": 7344 }, { "epoch": 0.9707903780068728, "grad_norm": 0.8489833859445624, "learning_rate": 1.0235853219827383e-06, "loss": 0.5942, "mean_token_accuracy": 0.8011079430580139, "num_tokens": 317445880.0, "step": 7345 }, { "epoch": 0.9709225482421359, "grad_norm": 0.808831306631279, "learning_rate": 1.023373503882592e-06, "loss": 0.5947, "mean_token_accuracy": 0.8001922965049744, "num_tokens": 317511416.0, "step": 7346 }, { "epoch": 0.9710547184773989, "grad_norm": 0.7704432846165499, "learning_rate": 1.0231626387627614e-06, "loss": 0.5322, "mean_token_accuracy": 0.8201990127563477, "num_tokens": 317576952.0, "step": 7347 }, { "epoch": 0.9711868887126619, "grad_norm": 0.796761412340538, "learning_rate": 1.0229527266681354e-06, "loss": 0.5972, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 317642488.0, "step": 7348 }, { "epoch": 0.9713190589479249, "grad_norm": 0.7801145886228, "learning_rate": 1.0227437676433995e-06, "loss": 0.5757, "mean_token_accuracy": 0.807700514793396, "num_tokens": 317708024.0, "step": 7349 }, { "epoch": 0.9714512291831879, "grad_norm": 0.8340106657855945, "learning_rate": 1.0225357617330375e-06, "loss": 0.5666, "mean_token_accuracy": 0.8115919828414917, "num_tokens": 317773560.0, "step": 7350 }, { "epoch": 0.971583399418451, "grad_norm": 0.8290762992178669, "learning_rate": 1.022328708981329e-06, "loss": 0.5635, "mean_token_accuracy": 0.8136979341506958, "num_tokens": 317839096.0, "step": 7351 }, { "epoch": 0.971715569653714, "grad_norm": 0.7826023530482715, "learning_rate": 1.0221226094323513e-06, "loss": 0.5528, "mean_token_accuracy": 0.816460132598877, "num_tokens": 317904632.0, "step": 7352 }, { "epoch": 0.971847739888977, "grad_norm": 0.7666135976697571, "learning_rate": 1.0219174631299783e-06, "loss": 0.5566, "mean_token_accuracy": 0.8157123923301697, "num_tokens": 317970168.0, "step": 7353 }, { "epoch": 0.97197991012424, "grad_norm": 0.8377836721626891, "learning_rate": 1.0217132701178826e-06, "loss": 0.5583, "mean_token_accuracy": 0.8173147439956665, "num_tokens": 318035704.0, "step": 7354 }, { "epoch": 0.972112080359503, "grad_norm": 0.8129502511754871, "learning_rate": 1.021510030439531e-06, "loss": 0.5739, "mean_token_accuracy": 0.8092265725135803, "num_tokens": 318101240.0, "step": 7355 }, { "epoch": 0.972244250594766, "grad_norm": 0.8098482990765985, "learning_rate": 1.02130774413819e-06, "loss": 0.5698, "mean_token_accuracy": 0.8114393949508667, "num_tokens": 318166776.0, "step": 7356 }, { "epoch": 0.9723764208300291, "grad_norm": 0.8172591439672289, "learning_rate": 1.0211064112569224e-06, "loss": 0.598, "mean_token_accuracy": 0.8014131188392639, "num_tokens": 318232312.0, "step": 7357 }, { "epoch": 0.9725085910652921, "grad_norm": 0.7782819959369548, "learning_rate": 1.0209060318385868e-06, "loss": 0.5658, "mean_token_accuracy": 0.8095470666885376, "num_tokens": 318297848.0, "step": 7358 }, { "epoch": 0.9726407613005551, "grad_norm": 0.7760579311379129, "learning_rate": 1.0207066059258402e-06, "loss": 0.5579, "mean_token_accuracy": 0.8142931461334229, "num_tokens": 318363384.0, "step": 7359 }, { "epoch": 0.9727729315358181, "grad_norm": 0.8028873248811177, "learning_rate": 1.020508133561136e-06, "loss": 0.5552, "mean_token_accuracy": 0.8176962733268738, "num_tokens": 318428920.0, "step": 7360 }, { "epoch": 0.9729051017710811, "grad_norm": 0.7906788419905566, "learning_rate": 1.0203106147867263e-06, "loss": 0.5827, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 318494456.0, "step": 7361 }, { "epoch": 0.9730372720063442, "grad_norm": 0.8456912505342385, "learning_rate": 1.0201140496446565e-06, "loss": 0.5902, "mean_token_accuracy": 0.8048467636108398, "num_tokens": 318559992.0, "step": 7362 }, { "epoch": 0.9731694422416072, "grad_norm": 0.7794951977715902, "learning_rate": 1.019918438176773e-06, "loss": 0.5509, "mean_token_accuracy": 0.8155292272567749, "num_tokens": 318625528.0, "step": 7363 }, { "epoch": 0.9733016124768702, "grad_norm": 0.7699977112468575, "learning_rate": 1.0197237804247157e-06, "loss": 0.5643, "mean_token_accuracy": 0.8157581686973572, "num_tokens": 318691064.0, "step": 7364 }, { "epoch": 0.9734337827121332, "grad_norm": 0.8127092253009432, "learning_rate": 1.0195300764299245e-06, "loss": 0.6094, "mean_token_accuracy": 0.798879861831665, "num_tokens": 318756600.0, "step": 7365 }, { "epoch": 0.9735659529473962, "grad_norm": 0.7883202841956097, "learning_rate": 1.019337326233635e-06, "loss": 0.5934, "mean_token_accuracy": 0.8024966716766357, "num_tokens": 318822136.0, "step": 7366 }, { "epoch": 0.9736981231826592, "grad_norm": 0.8327221085233262, "learning_rate": 1.019145529876879e-06, "loss": 0.5768, "mean_token_accuracy": 0.8049536347389221, "num_tokens": 318887672.0, "step": 7367 }, { "epoch": 0.9738302934179223, "grad_norm": 0.8219134239176741, "learning_rate": 1.0189546874004856e-06, "loss": 0.5793, "mean_token_accuracy": 0.8091655373573303, "num_tokens": 318953208.0, "step": 7368 }, { "epoch": 0.9739624636531853, "grad_norm": 0.7567670633084064, "learning_rate": 1.0187647988450825e-06, "loss": 0.529, "mean_token_accuracy": 0.8245940804481506, "num_tokens": 319018744.0, "step": 7369 }, { "epoch": 0.9740946338884483, "grad_norm": 0.8245800106765855, "learning_rate": 1.018575864251092e-06, "loss": 0.6046, "mean_token_accuracy": 0.7972164750099182, "num_tokens": 319084280.0, "step": 7370 }, { "epoch": 0.9742268041237113, "grad_norm": 0.7701820122842344, "learning_rate": 1.0183878836587348e-06, "loss": 0.583, "mean_token_accuracy": 0.8062965273857117, "num_tokens": 319149816.0, "step": 7371 }, { "epoch": 0.9743589743589743, "grad_norm": 0.7715042346779899, "learning_rate": 1.0182008571080275e-06, "loss": 0.5281, "mean_token_accuracy": 0.8226712346076965, "num_tokens": 319215352.0, "step": 7372 }, { "epoch": 0.9744911445942374, "grad_norm": 0.7818943525624414, "learning_rate": 1.018014784638785e-06, "loss": 0.552, "mean_token_accuracy": 0.8148882985115051, "num_tokens": 319280888.0, "step": 7373 }, { "epoch": 0.9746233148295004, "grad_norm": 0.7734808095134393, "learning_rate": 1.0178296662906175e-06, "loss": 0.5657, "mean_token_accuracy": 0.8104474544525146, "num_tokens": 319346424.0, "step": 7374 }, { "epoch": 0.9747554850647634, "grad_norm": 0.8544050510162629, "learning_rate": 1.0176455021029326e-06, "loss": 0.5783, "mean_token_accuracy": 0.8050451874732971, "num_tokens": 319411960.0, "step": 7375 }, { "epoch": 0.9748876553000264, "grad_norm": 0.7419821335490772, "learning_rate": 1.017462292114936e-06, "loss": 0.5744, "mean_token_accuracy": 0.8102185130119324, "num_tokens": 319477496.0, "step": 7376 }, { "epoch": 0.9750198255352894, "grad_norm": 0.8158197872960884, "learning_rate": 1.0172800363656285e-06, "loss": 0.6008, "mean_token_accuracy": 0.802344024181366, "num_tokens": 319543032.0, "step": 7377 }, { "epoch": 0.9751519957705524, "grad_norm": 0.8258909280540617, "learning_rate": 1.017098734893809e-06, "loss": 0.5894, "mean_token_accuracy": 0.8051062226295471, "num_tokens": 319608568.0, "step": 7378 }, { "epoch": 0.9752841660058155, "grad_norm": 0.787527491541834, "learning_rate": 1.0169183877380726e-06, "loss": 0.6085, "mean_token_accuracy": 0.79884934425354, "num_tokens": 319674104.0, "step": 7379 }, { "epoch": 0.9754163362410785, "grad_norm": 0.8259116170664905, "learning_rate": 1.0167389949368112e-06, "loss": 0.5815, "mean_token_accuracy": 0.8072426915168762, "num_tokens": 319739640.0, "step": 7380 }, { "epoch": 0.9755485064763415, "grad_norm": 0.7502259584018148, "learning_rate": 1.0165605565282142e-06, "loss": 0.5515, "mean_token_accuracy": 0.8166890740394592, "num_tokens": 319805176.0, "step": 7381 }, { "epoch": 0.9756806767116045, "grad_norm": 0.7286758137695245, "learning_rate": 1.0163830725502673e-06, "loss": 0.5245, "mean_token_accuracy": 0.825937032699585, "num_tokens": 319870712.0, "step": 7382 }, { "epoch": 0.9758128469468675, "grad_norm": 0.8050310480319659, "learning_rate": 1.0162065430407531e-06, "loss": 0.5966, "mean_token_accuracy": 0.8005433082580566, "num_tokens": 319936248.0, "step": 7383 }, { "epoch": 0.9759450171821306, "grad_norm": 0.8442330291785238, "learning_rate": 1.0160309680372512e-06, "loss": 0.6283, "mean_token_accuracy": 0.7913563847541809, "num_tokens": 320001784.0, "step": 7384 }, { "epoch": 0.9760771874173936, "grad_norm": 0.7617173352087798, "learning_rate": 1.015856347577137e-06, "loss": 0.5687, "mean_token_accuracy": 0.8085246086120605, "num_tokens": 320067320.0, "step": 7385 }, { "epoch": 0.9762093576526566, "grad_norm": 0.729986310182902, "learning_rate": 1.015682681697585e-06, "loss": 0.542, "mean_token_accuracy": 0.8198174834251404, "num_tokens": 320132856.0, "step": 7386 }, { "epoch": 0.9763415278879196, "grad_norm": 0.7865458121821051, "learning_rate": 1.0155099704355634e-06, "loss": 0.5679, "mean_token_accuracy": 0.8111646771430969, "num_tokens": 320198392.0, "step": 7387 }, { "epoch": 0.9764736981231826, "grad_norm": 0.7915585518983124, "learning_rate": 1.0153382138278405e-06, "loss": 0.602, "mean_token_accuracy": 0.7989561557769775, "num_tokens": 320263928.0, "step": 7388 }, { "epoch": 0.9766058683584457, "grad_norm": 0.7715755154442595, "learning_rate": 1.0151674119109785e-06, "loss": 0.5702, "mean_token_accuracy": 0.8105084896087646, "num_tokens": 320329464.0, "step": 7389 }, { "epoch": 0.9767380385937087, "grad_norm": 0.8343267526001402, "learning_rate": 1.0149975647213382e-06, "loss": 0.5902, "mean_token_accuracy": 0.8066017627716064, "num_tokens": 320395000.0, "step": 7390 }, { "epoch": 0.9768702088289717, "grad_norm": 0.7770120220442425, "learning_rate": 1.0148286722950758e-06, "loss": 0.5312, "mean_token_accuracy": 0.8223507404327393, "num_tokens": 320460536.0, "step": 7391 }, { "epoch": 0.9770023790642347, "grad_norm": 0.8517157770628799, "learning_rate": 1.0146607346681456e-06, "loss": 0.6408, "mean_token_accuracy": 0.7893572449684143, "num_tokens": 320526072.0, "step": 7392 }, { "epoch": 0.9771345492994977, "grad_norm": 0.8152362029731574, "learning_rate": 1.0144937518762982e-06, "loss": 0.5763, "mean_token_accuracy": 0.8054267168045044, "num_tokens": 320591608.0, "step": 7393 }, { "epoch": 0.9772667195347607, "grad_norm": 0.7657899392296796, "learning_rate": 1.01432772395508e-06, "loss": 0.5508, "mean_token_accuracy": 0.8160633444786072, "num_tokens": 320657144.0, "step": 7394 }, { "epoch": 0.9773988897700238, "grad_norm": 0.813177524135791, "learning_rate": 1.0141626509398356e-06, "loss": 0.6021, "mean_token_accuracy": 0.8000702261924744, "num_tokens": 320722680.0, "step": 7395 }, { "epoch": 0.9775310600052868, "grad_norm": 0.7944066630464662, "learning_rate": 1.0139985328657049e-06, "loss": 0.5479, "mean_token_accuracy": 0.8162922859191895, "num_tokens": 320788216.0, "step": 7396 }, { "epoch": 0.9776632302405498, "grad_norm": 0.7631640910718817, "learning_rate": 1.013835369767626e-06, "loss": 0.5527, "mean_token_accuracy": 0.8184287548065186, "num_tokens": 320853752.0, "step": 7397 }, { "epoch": 0.9777954004758128, "grad_norm": 0.8339421454883487, "learning_rate": 1.0136731616803323e-06, "loss": 0.5662, "mean_token_accuracy": 0.8098523020744324, "num_tokens": 320919288.0, "step": 7398 }, { "epoch": 0.9779275707110758, "grad_norm": 0.7750759279081983, "learning_rate": 1.0135119086383546e-06, "loss": 0.5511, "mean_token_accuracy": 0.8168721795082092, "num_tokens": 320984824.0, "step": 7399 }, { "epoch": 0.9780597409463389, "grad_norm": 0.86551878612326, "learning_rate": 1.0133516106760207e-06, "loss": 0.5977, "mean_token_accuracy": 0.8006196022033691, "num_tokens": 321050360.0, "step": 7400 }, { "epoch": 0.9781919111816019, "grad_norm": 0.7956942962384976, "learning_rate": 1.0131922678274545e-06, "loss": 0.5407, "mean_token_accuracy": 0.8188713192939758, "num_tokens": 321115896.0, "step": 7401 }, { "epoch": 0.9783240814168649, "grad_norm": 0.7663889657207524, "learning_rate": 1.0130338801265761e-06, "loss": 0.5218, "mean_token_accuracy": 0.8257843852043152, "num_tokens": 321181432.0, "step": 7402 }, { "epoch": 0.9784562516521279, "grad_norm": 0.8334986416056667, "learning_rate": 1.0128764476071037e-06, "loss": 0.547, "mean_token_accuracy": 0.8144152164459229, "num_tokens": 321246968.0, "step": 7403 }, { "epoch": 0.9785884218873909, "grad_norm": 0.7738271479789393, "learning_rate": 1.0127199703025516e-06, "loss": 0.5762, "mean_token_accuracy": 0.8093639612197876, "num_tokens": 321312504.0, "step": 7404 }, { "epoch": 0.9787205921226539, "grad_norm": 0.7820298279167553, "learning_rate": 1.01256444824623e-06, "loss": 0.5346, "mean_token_accuracy": 0.822900116443634, "num_tokens": 321378040.0, "step": 7405 }, { "epoch": 0.978852762357917, "grad_norm": 0.8696120785136425, "learning_rate": 1.0124098814712461e-06, "loss": 0.5809, "mean_token_accuracy": 0.8082193732261658, "num_tokens": 321443576.0, "step": 7406 }, { "epoch": 0.97898493259318, "grad_norm": 0.7251603849956483, "learning_rate": 1.0122562700105047e-06, "loss": 0.5267, "mean_token_accuracy": 0.8251739740371704, "num_tokens": 321509112.0, "step": 7407 }, { "epoch": 0.979117102828443, "grad_norm": 0.7738915037927889, "learning_rate": 1.0121036138967058e-06, "loss": 0.5255, "mean_token_accuracy": 0.8254181146621704, "num_tokens": 321574648.0, "step": 7408 }, { "epoch": 0.979249273063706, "grad_norm": 0.863155709039469, "learning_rate": 1.0119519131623476e-06, "loss": 0.5823, "mean_token_accuracy": 0.8062660098075867, "num_tokens": 321640184.0, "step": 7409 }, { "epoch": 0.979381443298969, "grad_norm": 0.8238574147757799, "learning_rate": 1.0118011678397225e-06, "loss": 0.5714, "mean_token_accuracy": 0.8084025382995605, "num_tokens": 321705720.0, "step": 7410 }, { "epoch": 0.979513613534232, "grad_norm": 0.7926110422702639, "learning_rate": 1.0116513779609224e-06, "loss": 0.583, "mean_token_accuracy": 0.8064491748809814, "num_tokens": 321771256.0, "step": 7411 }, { "epoch": 0.9796457837694951, "grad_norm": 0.8032864593445544, "learning_rate": 1.0115025435578339e-06, "loss": 0.5951, "mean_token_accuracy": 0.8010163307189941, "num_tokens": 321836792.0, "step": 7412 }, { "epoch": 0.9797779540047581, "grad_norm": 0.824422112418059, "learning_rate": 1.0113546646621412e-06, "loss": 0.5982, "mean_token_accuracy": 0.8001160025596619, "num_tokens": 321902328.0, "step": 7413 }, { "epoch": 0.9799101242400211, "grad_norm": 0.8527217586050228, "learning_rate": 1.0112077413053238e-06, "loss": 0.6312, "mean_token_accuracy": 0.7893419861793518, "num_tokens": 321967864.0, "step": 7414 }, { "epoch": 0.9800422944752841, "grad_norm": 0.8427600257007329, "learning_rate": 1.0110617735186585e-06, "loss": 0.6245, "mean_token_accuracy": 0.7901049852371216, "num_tokens": 322033400.0, "step": 7415 }, { "epoch": 0.9801744647105471, "grad_norm": 0.7987774468600825, "learning_rate": 1.0109167613332205e-06, "loss": 0.5794, "mean_token_accuracy": 0.8072274327278137, "num_tokens": 322098936.0, "step": 7416 }, { "epoch": 0.9803066349458102, "grad_norm": 0.8210354063490841, "learning_rate": 1.0107727047798779e-06, "loss": 0.5638, "mean_token_accuracy": 0.8106610774993896, "num_tokens": 322164472.0, "step": 7417 }, { "epoch": 0.9804388051810732, "grad_norm": 0.7968173411111125, "learning_rate": 1.010629603889298e-06, "loss": 0.5435, "mean_token_accuracy": 0.8205499649047852, "num_tokens": 322230008.0, "step": 7418 }, { "epoch": 0.9805709754163362, "grad_norm": 0.8007878738556418, "learning_rate": 1.010487458691945e-06, "loss": 0.5881, "mean_token_accuracy": 0.8029239177703857, "num_tokens": 322295544.0, "step": 7419 }, { "epoch": 0.9807031456515992, "grad_norm": 0.876989077370773, "learning_rate": 1.010346269218077e-06, "loss": 0.658, "mean_token_accuracy": 0.7822457551956177, "num_tokens": 322361080.0, "step": 7420 }, { "epoch": 0.9808353158868622, "grad_norm": 0.7249785800416072, "learning_rate": 1.0102060354977511e-06, "loss": 0.5056, "mean_token_accuracy": 0.8298437595367432, "num_tokens": 322426616.0, "step": 7421 }, { "epoch": 0.9809674861221253, "grad_norm": 0.7914995736381182, "learning_rate": 1.0100667575608206e-06, "loss": 0.523, "mean_token_accuracy": 0.826059103012085, "num_tokens": 322492152.0, "step": 7422 }, { "epoch": 0.9810996563573883, "grad_norm": 0.8393996274733284, "learning_rate": 1.009928435436934e-06, "loss": 0.6032, "mean_token_accuracy": 0.7977811098098755, "num_tokens": 322557688.0, "step": 7423 }, { "epoch": 0.9812318265926513, "grad_norm": 0.776018028691492, "learning_rate": 1.0097910691555379e-06, "loss": 0.5473, "mean_token_accuracy": 0.8200311064720154, "num_tokens": 322623224.0, "step": 7424 }, { "epoch": 0.9813639968279143, "grad_norm": 0.7974245902595373, "learning_rate": 1.009654658745874e-06, "loss": 0.5396, "mean_token_accuracy": 0.8190544247627258, "num_tokens": 322688760.0, "step": 7425 }, { "epoch": 0.9814961670631773, "grad_norm": 0.8163953166220732, "learning_rate": 1.0095192042369816e-06, "loss": 0.5703, "mean_token_accuracy": 0.8106915950775146, "num_tokens": 322754296.0, "step": 7426 }, { "epoch": 0.9816283372984403, "grad_norm": 0.8031543265189095, "learning_rate": 1.009384705657696e-06, "loss": 0.5649, "mean_token_accuracy": 0.8125534057617188, "num_tokens": 322819832.0, "step": 7427 }, { "epoch": 0.9817605075337034, "grad_norm": 0.7811103533548641, "learning_rate": 1.0092511630366496e-06, "loss": 0.5776, "mean_token_accuracy": 0.8074716329574585, "num_tokens": 322885368.0, "step": 7428 }, { "epoch": 0.9818926777689664, "grad_norm": 0.8027004021430237, "learning_rate": 1.0091185764022702e-06, "loss": 0.569, "mean_token_accuracy": 0.8125991821289062, "num_tokens": 322950904.0, "step": 7429 }, { "epoch": 0.9820248480042294, "grad_norm": 0.8272770064440735, "learning_rate": 1.0089869457827836e-06, "loss": 0.5963, "mean_token_accuracy": 0.8033207058906555, "num_tokens": 323016440.0, "step": 7430 }, { "epoch": 0.9821570182394924, "grad_norm": 0.8122939632446181, "learning_rate": 1.00885627120621e-06, "loss": 0.5553, "mean_token_accuracy": 0.8163789510726929, "num_tokens": 323081849.0, "step": 7431 }, { "epoch": 0.9822891884747554, "grad_norm": 0.8060170361474318, "learning_rate": 1.008726552700368e-06, "loss": 0.6094, "mean_token_accuracy": 0.7972012162208557, "num_tokens": 323147385.0, "step": 7432 }, { "epoch": 0.9824213587100185, "grad_norm": 0.836040987426292, "learning_rate": 1.0085977902928722e-06, "loss": 0.59, "mean_token_accuracy": 0.8025577068328857, "num_tokens": 323212921.0, "step": 7433 }, { "epoch": 0.9825535289452815, "grad_norm": 0.7422593914351392, "learning_rate": 1.008469984011133e-06, "loss": 0.5137, "mean_token_accuracy": 0.8286991715431213, "num_tokens": 323278457.0, "step": 7434 }, { "epoch": 0.9826856991805445, "grad_norm": 0.8149226790999526, "learning_rate": 1.0083431338823578e-06, "loss": 0.5491, "mean_token_accuracy": 0.8170705437660217, "num_tokens": 323343993.0, "step": 7435 }, { "epoch": 0.9828178694158075, "grad_norm": 0.7796869311100182, "learning_rate": 1.0082172399335507e-06, "loss": 0.5717, "mean_token_accuracy": 0.8063117861747742, "num_tokens": 323409529.0, "step": 7436 }, { "epoch": 0.9829500396510705, "grad_norm": 0.7846886732767729, "learning_rate": 1.0080923021915111e-06, "loss": 0.562, "mean_token_accuracy": 0.8106915950775146, "num_tokens": 323475065.0, "step": 7437 }, { "epoch": 0.9830822098863335, "grad_norm": 0.8917250002609042, "learning_rate": 1.007968320682836e-06, "loss": 0.6009, "mean_token_accuracy": 0.8030460476875305, "num_tokens": 323540601.0, "step": 7438 }, { "epoch": 0.9832143801215966, "grad_norm": 0.796409015124315, "learning_rate": 1.0078452954339189e-06, "loss": 0.5896, "mean_token_accuracy": 0.8038853406906128, "num_tokens": 323606137.0, "step": 7439 }, { "epoch": 0.9833465503568596, "grad_norm": 0.7929798861699933, "learning_rate": 1.0077232264709491e-06, "loss": 0.5847, "mean_token_accuracy": 0.8038395643234253, "num_tokens": 323671673.0, "step": 7440 }, { "epoch": 0.9834787205921226, "grad_norm": 0.7737050137629106, "learning_rate": 1.007602113819912e-06, "loss": 0.5596, "mean_token_accuracy": 0.816551685333252, "num_tokens": 323737209.0, "step": 7441 }, { "epoch": 0.9836108908273856, "grad_norm": 0.8411581281398778, "learning_rate": 1.0074819575065905e-06, "loss": 0.5922, "mean_token_accuracy": 0.8008487224578857, "num_tokens": 323799898.0, "step": 7442 }, { "epoch": 0.9837430610626486, "grad_norm": 0.7718070181049002, "learning_rate": 1.0073627575565629e-06, "loss": 0.5693, "mean_token_accuracy": 0.8100354075431824, "num_tokens": 323865434.0, "step": 7443 }, { "epoch": 0.9838752312979117, "grad_norm": 0.7540394980442272, "learning_rate": 1.0072445139952053e-06, "loss": 0.5476, "mean_token_accuracy": 0.8170552849769592, "num_tokens": 323930970.0, "step": 7444 }, { "epoch": 0.9840074015331747, "grad_norm": 0.7781439762783305, "learning_rate": 1.0071272268476887e-06, "loss": 0.5605, "mean_token_accuracy": 0.8132248520851135, "num_tokens": 323996506.0, "step": 7445 }, { "epoch": 0.9841395717684377, "grad_norm": 0.7606521617079162, "learning_rate": 1.0070108961389805e-06, "loss": 0.563, "mean_token_accuracy": 0.8132859468460083, "num_tokens": 324062042.0, "step": 7446 }, { "epoch": 0.9842717420037007, "grad_norm": 0.7722046435101585, "learning_rate": 1.006895521893846e-06, "loss": 0.5915, "mean_token_accuracy": 0.8038853406906128, "num_tokens": 324127578.0, "step": 7447 }, { "epoch": 0.9844039122389638, "grad_norm": 0.8624713188320556, "learning_rate": 1.0067811041368455e-06, "loss": 0.5836, "mean_token_accuracy": 0.8079904913902283, "num_tokens": 324193114.0, "step": 7448 }, { "epoch": 0.9845360824742269, "grad_norm": 0.7868485407544017, "learning_rate": 1.0066676428923365e-06, "loss": 0.5853, "mean_token_accuracy": 0.8040532469749451, "num_tokens": 324258650.0, "step": 7449 }, { "epoch": 0.9846682527094899, "grad_norm": 0.7556059425852159, "learning_rate": 1.006555138184472e-06, "loss": 0.5492, "mean_token_accuracy": 0.8177420496940613, "num_tokens": 324324186.0, "step": 7450 }, { "epoch": 0.9848004229447529, "grad_norm": 0.7317357355057614, "learning_rate": 1.0064435900372025e-06, "loss": 0.5252, "mean_token_accuracy": 0.8233579397201538, "num_tokens": 324389722.0, "step": 7451 }, { "epoch": 0.9849325931800159, "grad_norm": 0.8009103486078728, "learning_rate": 1.0063329984742735e-06, "loss": 0.5735, "mean_token_accuracy": 0.8094860315322876, "num_tokens": 324455258.0, "step": 7452 }, { "epoch": 0.9850647634152789, "grad_norm": 0.7933462824657633, "learning_rate": 1.0062233635192286e-06, "loss": 0.5742, "mean_token_accuracy": 0.8089061379432678, "num_tokens": 324520794.0, "step": 7453 }, { "epoch": 0.985196933650542, "grad_norm": 0.7472376870290559, "learning_rate": 1.0061146851954063e-06, "loss": 0.5264, "mean_token_accuracy": 0.8222897052764893, "num_tokens": 324586330.0, "step": 7454 }, { "epoch": 0.985329103885805, "grad_norm": 0.783634167466546, "learning_rate": 1.0060069635259413e-06, "loss": 0.5548, "mean_token_accuracy": 0.8151782155036926, "num_tokens": 324651866.0, "step": 7455 }, { "epoch": 0.985461274121068, "grad_norm": 0.7788741159025527, "learning_rate": 1.0059001985337668e-06, "loss": 0.5154, "mean_token_accuracy": 0.827325701713562, "num_tokens": 324717402.0, "step": 7456 }, { "epoch": 0.985593444356331, "grad_norm": 0.7525583360613858, "learning_rate": 1.0057943902416093e-06, "loss": 0.5307, "mean_token_accuracy": 0.8246245980262756, "num_tokens": 324782938.0, "step": 7457 }, { "epoch": 0.985725614591594, "grad_norm": 0.8520012788826345, "learning_rate": 1.0056895386719946e-06, "loss": 0.5754, "mean_token_accuracy": 0.807731032371521, "num_tokens": 324848474.0, "step": 7458 }, { "epoch": 0.985857784826857, "grad_norm": 0.9028696235594964, "learning_rate": 1.0055856438472422e-06, "loss": 0.6607, "mean_token_accuracy": 0.7799108624458313, "num_tokens": 324914010.0, "step": 7459 }, { "epoch": 0.9859899550621201, "grad_norm": 0.776198318219249, "learning_rate": 1.0054827057894694e-06, "loss": 0.5419, "mean_token_accuracy": 0.8192528486251831, "num_tokens": 324979546.0, "step": 7460 }, { "epoch": 0.9861221252973831, "grad_norm": 0.7676229767853711, "learning_rate": 1.0053807245205904e-06, "loss": 0.529, "mean_token_accuracy": 0.8230832815170288, "num_tokens": 325045082.0, "step": 7461 }, { "epoch": 0.9862542955326461, "grad_norm": 0.7999773449873342, "learning_rate": 1.0052797000623134e-06, "loss": 0.5596, "mean_token_accuracy": 0.8129196763038635, "num_tokens": 325110618.0, "step": 7462 }, { "epoch": 0.9863864657679091, "grad_norm": 0.8487824504708735, "learning_rate": 1.0051796324361463e-06, "loss": 0.6004, "mean_token_accuracy": 0.7991850972175598, "num_tokens": 325176154.0, "step": 7463 }, { "epoch": 0.9865186360031721, "grad_norm": 0.8330151833443221, "learning_rate": 1.0050805216633893e-06, "loss": 0.5835, "mean_token_accuracy": 0.8055793046951294, "num_tokens": 325241690.0, "step": 7464 }, { "epoch": 0.9866508062384352, "grad_norm": 0.839449608161749, "learning_rate": 1.0049823677651426e-06, "loss": 0.5945, "mean_token_accuracy": 0.8017336130142212, "num_tokens": 325307226.0, "step": 7465 }, { "epoch": 0.9867829764736982, "grad_norm": 0.9131088534022356, "learning_rate": 1.004885170762301e-06, "loss": 0.6148, "mean_token_accuracy": 0.7963008284568787, "num_tokens": 325372762.0, "step": 7466 }, { "epoch": 0.9869151467089612, "grad_norm": 0.797349720479484, "learning_rate": 1.0047889306755545e-06, "loss": 0.5961, "mean_token_accuracy": 0.8042058348655701, "num_tokens": 325438298.0, "step": 7467 }, { "epoch": 0.9870473169442242, "grad_norm": 0.8004688051120843, "learning_rate": 1.0046936475253921e-06, "loss": 0.5839, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 325503834.0, "step": 7468 }, { "epoch": 0.9871794871794872, "grad_norm": 0.7757868954287298, "learning_rate": 1.004599321332097e-06, "loss": 0.5747, "mean_token_accuracy": 0.8088908791542053, "num_tokens": 325569370.0, "step": 7469 }, { "epoch": 0.9873116574147502, "grad_norm": 0.7817654214192876, "learning_rate": 1.0045059521157484e-06, "loss": 0.5462, "mean_token_accuracy": 0.8201837539672852, "num_tokens": 325634906.0, "step": 7470 }, { "epoch": 0.9874438276500133, "grad_norm": 0.8306022448335302, "learning_rate": 1.0044135398962241e-06, "loss": 0.5758, "mean_token_accuracy": 0.8084940910339355, "num_tokens": 325700442.0, "step": 7471 }, { "epoch": 0.9875759978852763, "grad_norm": 0.8158210142630151, "learning_rate": 1.0043220846931956e-06, "loss": 0.5733, "mean_token_accuracy": 0.8072884678840637, "num_tokens": 325765978.0, "step": 7472 }, { "epoch": 0.9877081681205393, "grad_norm": 0.8186964561695762, "learning_rate": 1.0042315865261323e-06, "loss": 0.5667, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 325831514.0, "step": 7473 }, { "epoch": 0.9878403383558023, "grad_norm": 0.861672391589617, "learning_rate": 1.0041420454142997e-06, "loss": 0.6276, "mean_token_accuracy": 0.7917225956916809, "num_tokens": 325897050.0, "step": 7474 }, { "epoch": 0.9879725085910653, "grad_norm": 0.7500499292759357, "learning_rate": 1.0040534613767589e-06, "loss": 0.4886, "mean_token_accuracy": 0.8347424268722534, "num_tokens": 325962586.0, "step": 7475 }, { "epoch": 0.9881046788263284, "grad_norm": 0.7781959704377319, "learning_rate": 1.0039658344323673e-06, "loss": 0.5582, "mean_token_accuracy": 0.8141710162162781, "num_tokens": 326028122.0, "step": 7476 }, { "epoch": 0.9882368490615914, "grad_norm": 0.8206143079822582, "learning_rate": 1.003879164599779e-06, "loss": 0.5835, "mean_token_accuracy": 0.8053045868873596, "num_tokens": 326093658.0, "step": 7477 }, { "epoch": 0.9883690192968544, "grad_norm": 0.8059839482223967, "learning_rate": 1.0037934518974444e-06, "loss": 0.5278, "mean_token_accuracy": 0.8230832815170288, "num_tokens": 326159194.0, "step": 7478 }, { "epoch": 0.9885011895321174, "grad_norm": 0.7307899608680346, "learning_rate": 1.0037086963436102e-06, "loss": 0.5056, "mean_token_accuracy": 0.8315224051475525, "num_tokens": 326224730.0, "step": 7479 }, { "epoch": 0.9886333597673804, "grad_norm": 0.803703464418621, "learning_rate": 1.0036248979563184e-06, "loss": 0.589, "mean_token_accuracy": 0.8038700819015503, "num_tokens": 326290266.0, "step": 7480 }, { "epoch": 0.9887655300026434, "grad_norm": 0.9048857531113742, "learning_rate": 1.0035420567534086e-06, "loss": 0.639, "mean_token_accuracy": 0.7865645289421082, "num_tokens": 326355802.0, "step": 7481 }, { "epoch": 0.9888977002379065, "grad_norm": 0.7195989690379259, "learning_rate": 1.003460172752515e-06, "loss": 0.5474, "mean_token_accuracy": 0.818596601486206, "num_tokens": 326421338.0, "step": 7482 }, { "epoch": 0.9890298704731695, "grad_norm": 0.8708218152452586, "learning_rate": 1.0033792459710701e-06, "loss": 0.6005, "mean_token_accuracy": 0.8008332252502441, "num_tokens": 326486874.0, "step": 7483 }, { "epoch": 0.9891620407084325, "grad_norm": 0.823493406916416, "learning_rate": 1.0032992764263014e-06, "loss": 0.5862, "mean_token_accuracy": 0.8042821288108826, "num_tokens": 326552410.0, "step": 7484 }, { "epoch": 0.9892942109436955, "grad_norm": 0.8348252132823442, "learning_rate": 1.003220264135232e-06, "loss": 0.6209, "mean_token_accuracy": 0.7907917499542236, "num_tokens": 326617946.0, "step": 7485 }, { "epoch": 0.9894263811789585, "grad_norm": 0.8542744347045491, "learning_rate": 1.003142209114682e-06, "loss": 0.5736, "mean_token_accuracy": 0.8045263290405273, "num_tokens": 326683482.0, "step": 7486 }, { "epoch": 0.9895585514142216, "grad_norm": 0.7377104433360273, "learning_rate": 1.0030651113812683e-06, "loss": 0.5493, "mean_token_accuracy": 0.8158039450645447, "num_tokens": 326749018.0, "step": 7487 }, { "epoch": 0.9896907216494846, "grad_norm": 0.7825057786057624, "learning_rate": 1.0029889709514035e-06, "loss": 0.5545, "mean_token_accuracy": 0.8159870505332947, "num_tokens": 326814554.0, "step": 7488 }, { "epoch": 0.9898228918847476, "grad_norm": 0.75877647698748, "learning_rate": 1.0029137878412959e-06, "loss": 0.5389, "mean_token_accuracy": 0.8192223310470581, "num_tokens": 326880090.0, "step": 7489 }, { "epoch": 0.9899550621200106, "grad_norm": 0.8002244365874476, "learning_rate": 1.00283956206695e-06, "loss": 0.5751, "mean_token_accuracy": 0.8054877519607544, "num_tokens": 326945626.0, "step": 7490 }, { "epoch": 0.9900872323552736, "grad_norm": 0.8208088015770476, "learning_rate": 1.0027662936441677e-06, "loss": 0.5966, "mean_token_accuracy": 0.8038243055343628, "num_tokens": 327011162.0, "step": 7491 }, { "epoch": 0.9902194025905366, "grad_norm": 0.7973440631405133, "learning_rate": 1.0026939825885457e-06, "loss": 0.5289, "mean_token_accuracy": 0.8211146593093872, "num_tokens": 327076698.0, "step": 7492 }, { "epoch": 0.9903515728257997, "grad_norm": 0.7805906591125252, "learning_rate": 1.0026226289154782e-06, "loss": 0.5628, "mean_token_accuracy": 0.8114393949508667, "num_tokens": 327142234.0, "step": 7493 }, { "epoch": 0.9904837430610627, "grad_norm": 0.7928008046415608, "learning_rate": 1.0025522326401546e-06, "loss": 0.5575, "mean_token_accuracy": 0.8132248520851135, "num_tokens": 327207770.0, "step": 7494 }, { "epoch": 0.9906159132963257, "grad_norm": 0.8252156734404471, "learning_rate": 1.0024827937775607e-06, "loss": 0.5924, "mean_token_accuracy": 0.8001770377159119, "num_tokens": 327273306.0, "step": 7495 }, { "epoch": 0.9907480835315887, "grad_norm": 0.8565802615683288, "learning_rate": 1.002414312342478e-06, "loss": 0.5857, "mean_token_accuracy": 0.8053503632545471, "num_tokens": 327338842.0, "step": 7496 }, { "epoch": 0.9908802537668517, "grad_norm": 0.7589527070329312, "learning_rate": 1.002346788349486e-06, "loss": 0.5975, "mean_token_accuracy": 0.8010010719299316, "num_tokens": 327404378.0, "step": 7497 }, { "epoch": 0.9910124240021148, "grad_norm": 0.825285173293102, "learning_rate": 1.0022802218129586e-06, "loss": 0.5872, "mean_token_accuracy": 0.803213894367218, "num_tokens": 327469914.0, "step": 7498 }, { "epoch": 0.9911445942373778, "grad_norm": 0.8019071305943933, "learning_rate": 1.0022146127470659e-06, "loss": 0.5884, "mean_token_accuracy": 0.8032596707344055, "num_tokens": 327535450.0, "step": 7499 }, { "epoch": 0.9912767644726408, "grad_norm": 0.7762702393142433, "learning_rate": 1.0021499611657753e-06, "loss": 0.5177, "mean_token_accuracy": 0.827264666557312, "num_tokens": 327600986.0, "step": 7500 }, { "epoch": 0.9914089347079038, "grad_norm": 0.8255814039636664, "learning_rate": 1.00208626708285e-06, "loss": 0.6366, "mean_token_accuracy": 0.7854962944984436, "num_tokens": 327666522.0, "step": 7501 }, { "epoch": 0.9915411049431668, "grad_norm": 0.8146525140131685, "learning_rate": 1.0020235305118482e-06, "loss": 0.5487, "mean_token_accuracy": 0.8148730397224426, "num_tokens": 327732058.0, "step": 7502 }, { "epoch": 0.9916732751784298, "grad_norm": 0.8029959679830432, "learning_rate": 1.0019617514661262e-06, "loss": 0.5622, "mean_token_accuracy": 0.8126907348632812, "num_tokens": 327797594.0, "step": 7503 }, { "epoch": 0.9918054454136929, "grad_norm": 0.8331156904442197, "learning_rate": 1.0019009299588345e-06, "loss": 0.6205, "mean_token_accuracy": 0.7954004406929016, "num_tokens": 327863130.0, "step": 7504 }, { "epoch": 0.9919376156489559, "grad_norm": 0.8280242046919091, "learning_rate": 1.0018410660029216e-06, "loss": 0.5664, "mean_token_accuracy": 0.8142015337944031, "num_tokens": 327928666.0, "step": 7505 }, { "epoch": 0.9920697858842189, "grad_norm": 0.7611749443741423, "learning_rate": 1.0017821596111314e-06, "loss": 0.5254, "mean_token_accuracy": 0.8260438442230225, "num_tokens": 327994202.0, "step": 7506 }, { "epoch": 0.9922019561194819, "grad_norm": 0.8126950994021803, "learning_rate": 1.0017242107960028e-06, "loss": 0.5651, "mean_token_accuracy": 0.8114088773727417, "num_tokens": 328059738.0, "step": 7507 }, { "epoch": 0.9923341263547449, "grad_norm": 0.8967144917871935, "learning_rate": 1.0016672195698724e-06, "loss": 0.6127, "mean_token_accuracy": 0.7955835461616516, "num_tokens": 328125274.0, "step": 7508 }, { "epoch": 0.992466296590008, "grad_norm": 0.7963915116771735, "learning_rate": 1.001611185944873e-06, "loss": 0.593, "mean_token_accuracy": 0.8024813532829285, "num_tokens": 328190810.0, "step": 7509 }, { "epoch": 0.992598466825271, "grad_norm": 0.7693230258327102, "learning_rate": 1.0015561099329318e-06, "loss": 0.5784, "mean_token_accuracy": 0.8059150576591492, "num_tokens": 328256346.0, "step": 7510 }, { "epoch": 0.992730637060534, "grad_norm": 0.7568047993293335, "learning_rate": 1.0015019915457749e-06, "loss": 0.5502, "mean_token_accuracy": 0.8155292272567749, "num_tokens": 328321882.0, "step": 7511 }, { "epoch": 0.992862807295797, "grad_norm": 0.8221296365197529, "learning_rate": 1.0014488307949213e-06, "loss": 0.6216, "mean_token_accuracy": 0.7941642999649048, "num_tokens": 328387418.0, "step": 7512 }, { "epoch": 0.99299497753106, "grad_norm": 0.801076759177788, "learning_rate": 1.0013966276916899e-06, "loss": 0.5727, "mean_token_accuracy": 0.8116835355758667, "num_tokens": 328452954.0, "step": 7513 }, { "epoch": 0.993127147766323, "grad_norm": 0.7586771320813769, "learning_rate": 1.0013453822471915e-06, "loss": 0.522, "mean_token_accuracy": 0.827203631401062, "num_tokens": 328518490.0, "step": 7514 }, { "epoch": 0.9932593180015861, "grad_norm": 0.7882611000322176, "learning_rate": 1.0012950944723355e-06, "loss": 0.5263, "mean_token_accuracy": 0.8251434564590454, "num_tokens": 328584026.0, "step": 7515 }, { "epoch": 0.9933914882368491, "grad_norm": 0.774819910484788, "learning_rate": 1.0012457643778288e-06, "loss": 0.5235, "mean_token_accuracy": 0.8256623148918152, "num_tokens": 328649562.0, "step": 7516 }, { "epoch": 0.9935236584721121, "grad_norm": 0.7982911940020694, "learning_rate": 1.0011973919741711e-06, "loss": 0.5865, "mean_token_accuracy": 0.806464433670044, "num_tokens": 328715098.0, "step": 7517 }, { "epoch": 0.9936558287073751, "grad_norm": 0.8196498370059717, "learning_rate": 1.0011499772716604e-06, "loss": 0.5819, "mean_token_accuracy": 0.8078836798667908, "num_tokens": 328780634.0, "step": 7518 }, { "epoch": 0.9937879989426381, "grad_norm": 0.7597777424828713, "learning_rate": 1.0011035202803905e-06, "loss": 0.5631, "mean_token_accuracy": 0.8122634887695312, "num_tokens": 328846170.0, "step": 7519 }, { "epoch": 0.9939201691779012, "grad_norm": 0.8375841140123327, "learning_rate": 1.0010580210102511e-06, "loss": 0.5722, "mean_token_accuracy": 0.8089671730995178, "num_tokens": 328911706.0, "step": 7520 }, { "epoch": 0.9940523394131642, "grad_norm": 0.7647924946619945, "learning_rate": 1.0010134794709271e-06, "loss": 0.5691, "mean_token_accuracy": 0.8109815716743469, "num_tokens": 328977242.0, "step": 7521 }, { "epoch": 0.9941845096484272, "grad_norm": 0.7574387700232109, "learning_rate": 1.0009698956719019e-06, "loss": 0.5459, "mean_token_accuracy": 0.8191765546798706, "num_tokens": 329042778.0, "step": 7522 }, { "epoch": 0.9943166798836902, "grad_norm": 0.8201984725089956, "learning_rate": 1.0009272696224531e-06, "loss": 0.58, "mean_token_accuracy": 0.8051825165748596, "num_tokens": 329108314.0, "step": 7523 }, { "epoch": 0.9944488501189532, "grad_norm": 0.8357053050436605, "learning_rate": 1.0008856013316545e-06, "loss": 0.6255, "mean_token_accuracy": 0.7899676561355591, "num_tokens": 329173850.0, "step": 7524 }, { "epoch": 0.9945810203542162, "grad_norm": 0.8112413931033001, "learning_rate": 1.0008448908083765e-06, "loss": 0.5915, "mean_token_accuracy": 0.8006653785705566, "num_tokens": 329239386.0, "step": 7525 }, { "epoch": 0.9947131905894793, "grad_norm": 0.8059223428459412, "learning_rate": 1.0008051380612854e-06, "loss": 0.5559, "mean_token_accuracy": 0.8142778873443604, "num_tokens": 329304922.0, "step": 7526 }, { "epoch": 0.9948453608247423, "grad_norm": 0.8436367995034788, "learning_rate": 1.0007663430988443e-06, "loss": 0.5698, "mean_token_accuracy": 0.808814525604248, "num_tokens": 329370458.0, "step": 7527 }, { "epoch": 0.9949775310600053, "grad_norm": 0.8762693798740134, "learning_rate": 1.0007285059293118e-06, "loss": 0.5869, "mean_token_accuracy": 0.8029391765594482, "num_tokens": 329435994.0, "step": 7528 }, { "epoch": 0.9951097012952683, "grad_norm": 0.8241062741339843, "learning_rate": 1.0006916265607419e-06, "loss": 0.6169, "mean_token_accuracy": 0.7963618636131287, "num_tokens": 329501530.0, "step": 7529 }, { "epoch": 0.9952418715305313, "grad_norm": 0.7902574319545217, "learning_rate": 1.000655705000986e-06, "loss": 0.5762, "mean_token_accuracy": 0.8077768087387085, "num_tokens": 329567066.0, "step": 7530 }, { "epoch": 0.9953740417657944, "grad_norm": 0.7649091827862938, "learning_rate": 1.0006207412576913e-06, "loss": 0.5642, "mean_token_accuracy": 0.8122329115867615, "num_tokens": 329632602.0, "step": 7531 }, { "epoch": 0.9955062120010574, "grad_norm": 0.8324651072646676, "learning_rate": 1.0005867353383e-06, "loss": 0.563, "mean_token_accuracy": 0.8115614652633667, "num_tokens": 329698138.0, "step": 7532 }, { "epoch": 0.9956383822363204, "grad_norm": 0.8408682729825473, "learning_rate": 1.000553687250052e-06, "loss": 0.6012, "mean_token_accuracy": 0.8007264137268066, "num_tokens": 329763674.0, "step": 7533 }, { "epoch": 0.9957705524715834, "grad_norm": 0.8391966664535103, "learning_rate": 1.0005215969999827e-06, "loss": 0.6192, "mean_token_accuracy": 0.7939964532852173, "num_tokens": 329829210.0, "step": 7534 }, { "epoch": 0.9959027227068464, "grad_norm": 0.8247844097987423, "learning_rate": 1.0004904645949224e-06, "loss": 0.5937, "mean_token_accuracy": 0.802405059337616, "num_tokens": 329894746.0, "step": 7535 }, { "epoch": 0.9960348929421095, "grad_norm": 0.7951091585234616, "learning_rate": 1.0004602900414995e-06, "loss": 0.6178, "mean_token_accuracy": 0.7933250069618225, "num_tokens": 329960282.0, "step": 7536 }, { "epoch": 0.9961670631773725, "grad_norm": 0.8077305257001686, "learning_rate": 1.0004310733461374e-06, "loss": 0.5802, "mean_token_accuracy": 0.8048315048217773, "num_tokens": 330025818.0, "step": 7537 }, { "epoch": 0.9962992334126355, "grad_norm": 0.8300092591807735, "learning_rate": 1.0004028145150554e-06, "loss": 0.572, "mean_token_accuracy": 0.8107831478118896, "num_tokens": 330091354.0, "step": 7538 }, { "epoch": 0.9964314036478985, "grad_norm": 0.7760246141150845, "learning_rate": 1.0003755135542691e-06, "loss": 0.5682, "mean_token_accuracy": 0.8107984662055969, "num_tokens": 330156890.0, "step": 7539 }, { "epoch": 0.9965635738831615, "grad_norm": 0.8515056937964701, "learning_rate": 1.000349170469591e-06, "loss": 0.6009, "mean_token_accuracy": 0.8026645183563232, "num_tokens": 330222426.0, "step": 7540 }, { "epoch": 0.9966957441184245, "grad_norm": 0.7526955789572701, "learning_rate": 1.0003237852666284e-06, "loss": 0.5272, "mean_token_accuracy": 0.8238767981529236, "num_tokens": 330287962.0, "step": 7541 }, { "epoch": 0.9968279143536876, "grad_norm": 0.7777333864874454, "learning_rate": 1.0002993579507852e-06, "loss": 0.5439, "mean_token_accuracy": 0.8178030848503113, "num_tokens": 330353498.0, "step": 7542 }, { "epoch": 0.9969600845889506, "grad_norm": 0.7185254422086322, "learning_rate": 1.000275888527262e-06, "loss": 0.5474, "mean_token_accuracy": 0.8172537088394165, "num_tokens": 330419034.0, "step": 7543 }, { "epoch": 0.9970922548242136, "grad_norm": 0.8089390728637225, "learning_rate": 1.0002533770010545e-06, "loss": 0.5625, "mean_token_accuracy": 0.8129501938819885, "num_tokens": 330484570.0, "step": 7544 }, { "epoch": 0.9972244250594766, "grad_norm": 0.7339071618290348, "learning_rate": 1.0002318233769556e-06, "loss": 0.5066, "mean_token_accuracy": 0.83162921667099, "num_tokens": 330550106.0, "step": 7545 }, { "epoch": 0.9973565952947396, "grad_norm": 0.7842428769781453, "learning_rate": 1.0002112276595525e-06, "loss": 0.5981, "mean_token_accuracy": 0.7996429204940796, "num_tokens": 330615642.0, "step": 7546 }, { "epoch": 0.9974887655300027, "grad_norm": 0.8178515686438265, "learning_rate": 1.0001915898532305e-06, "loss": 0.5668, "mean_token_accuracy": 0.8102032542228699, "num_tokens": 330681178.0, "step": 7547 }, { "epoch": 0.9976209357652657, "grad_norm": 0.7374322074182146, "learning_rate": 1.0001729099621693e-06, "loss": 0.538, "mean_token_accuracy": 0.8211146593093872, "num_tokens": 330746714.0, "step": 7548 }, { "epoch": 0.9977531060005287, "grad_norm": 0.910862847552687, "learning_rate": 1.0001551879903464e-06, "loss": 0.6263, "mean_token_accuracy": 0.7895098328590393, "num_tokens": 330812250.0, "step": 7549 }, { "epoch": 0.9978852762357917, "grad_norm": 0.7452000499805466, "learning_rate": 1.0001384239415338e-06, "loss": 0.5787, "mean_token_accuracy": 0.8087077140808105, "num_tokens": 330877786.0, "step": 7550 }, { "epoch": 0.9980174464710547, "grad_norm": 0.7981190723757702, "learning_rate": 1.0001226178193008e-06, "loss": 0.5252, "mean_token_accuracy": 0.8255249857902527, "num_tokens": 330943322.0, "step": 7551 }, { "epoch": 0.9981496167063177, "grad_norm": 0.7995280939001643, "learning_rate": 1.0001077696270116e-06, "loss": 0.5823, "mean_token_accuracy": 0.8069222569465637, "num_tokens": 331008858.0, "step": 7552 }, { "epoch": 0.9982817869415808, "grad_norm": 0.7774497793512188, "learning_rate": 1.0000938793678267e-06, "loss": 0.5415, "mean_token_accuracy": 0.8177420496940613, "num_tokens": 331074394.0, "step": 7553 }, { "epoch": 0.9984139571768438, "grad_norm": 0.7458645867370769, "learning_rate": 1.0000809470447043e-06, "loss": 0.5413, "mean_token_accuracy": 0.818444013595581, "num_tokens": 331139930.0, "step": 7554 }, { "epoch": 0.9985461274121068, "grad_norm": 0.8720539397836408, "learning_rate": 1.0000689726603967e-06, "loss": 0.6529, "mean_token_accuracy": 0.783924400806427, "num_tokens": 331205466.0, "step": 7555 }, { "epoch": 0.9986782976473698, "grad_norm": 0.8016954507274141, "learning_rate": 1.0000579562174528e-06, "loss": 0.5619, "mean_token_accuracy": 0.8106305599212646, "num_tokens": 331271002.0, "step": 7556 }, { "epoch": 0.9988104678826328, "grad_norm": 0.7985364895193362, "learning_rate": 1.0000478977182178e-06, "loss": 0.5466, "mean_token_accuracy": 0.817268967628479, "num_tokens": 331336538.0, "step": 7557 }, { "epoch": 0.9989426381178959, "grad_norm": 0.8361171922946252, "learning_rate": 1.0000387971648334e-06, "loss": 0.6263, "mean_token_accuracy": 0.7907611727714539, "num_tokens": 331402074.0, "step": 7558 }, { "epoch": 0.9990748083531589, "grad_norm": 0.8166200103494413, "learning_rate": 1.0000306545592368e-06, "loss": 0.5671, "mean_token_accuracy": 0.8097912073135376, "num_tokens": 331467610.0, "step": 7559 }, { "epoch": 0.9992069785884219, "grad_norm": 0.8142399334111575, "learning_rate": 1.000023469903161e-06, "loss": 0.6184, "mean_token_accuracy": 0.7934165596961975, "num_tokens": 331533146.0, "step": 7560 }, { "epoch": 0.9993391488236849, "grad_norm": 0.8197592729819727, "learning_rate": 1.0000172431981355e-06, "loss": 0.5746, "mean_token_accuracy": 0.807395339012146, "num_tokens": 331598682.0, "step": 7561 }, { "epoch": 0.9994713190589479, "grad_norm": 0.7797927646545476, "learning_rate": 1.0000119744454864e-06, "loss": 0.5215, "mean_token_accuracy": 0.8274173140525818, "num_tokens": 331664218.0, "step": 7562 }, { "epoch": 0.999603489294211, "grad_norm": 0.7574390451719882, "learning_rate": 1.0000076636463352e-06, "loss": 0.5426, "mean_token_accuracy": 0.8193291425704956, "num_tokens": 331729754.0, "step": 7563 }, { "epoch": 0.999735659529474, "grad_norm": 0.7804488810589181, "learning_rate": 1.0000043108015989e-06, "loss": 0.5742, "mean_token_accuracy": 0.8108289837837219, "num_tokens": 331795290.0, "step": 7564 }, { "epoch": 0.999867829764737, "grad_norm": 0.7823763271207023, "learning_rate": 1.0000019159119917e-06, "loss": 0.5652, "mean_token_accuracy": 0.8093792200088501, "num_tokens": 331860826.0, "step": 7565 }, { "epoch": 1.0, "grad_norm": 0.7465338978368082, "learning_rate": 1.0000004789780234e-06, "loss": 0.5061, "mean_token_accuracy": 0.8289738893508911, "num_tokens": 331926362.0, "step": 7566 }, { "epoch": 1.0, "step": 7566, "total_flos": 376152465051648.0, "train_loss": 0.392165879506993, "train_runtime": 99618.8993, "train_samples_per_second": 0.608, "train_steps_per_second": 0.076 } ], "logging_steps": 1, "max_steps": 7566, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 376152465051648.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }