| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 32, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.04113317007945081, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1767, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.042219935316768153, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.188, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 0.04292721294346761, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1893, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.04056717413915702, | |
| "learning_rate": 5e-05, | |
| "loss": 0.187, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.04330821122698837, | |
| "learning_rate": 4.984280524733107e-05, | |
| "loss": 0.1753, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.0452810422313244, | |
| "learning_rate": 4.937319780454559e-05, | |
| "loss": 0.1875, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 0.039289995019635204, | |
| "learning_rate": 4.8597083257709194e-05, | |
| "loss": 0.1687, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.027002440254188315, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.1663, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 0.02494347654011703, | |
| "learning_rate": 4.6168104980707107e-05, | |
| "loss": 0.1695, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.028920303836423092, | |
| "learning_rate": 4.454578706170075e-05, | |
| "loss": 0.1579, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 0.03943765297659361, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.1568, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.039468128072595324, | |
| "learning_rate": 4.058724504646834e-05, | |
| "loss": 0.1499, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 0.04907120045337006, | |
| "learning_rate": 3.830080191288342e-05, | |
| "loss": 0.1463, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.03536941711454614, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 0.1513, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.023098349626014137, | |
| "learning_rate": 3.3256976548879184e-05, | |
| "loss": 0.1378, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.017512339127206115, | |
| "learning_rate": 3.056302334890786e-05, | |
| "loss": 0.1374, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 0.017368083265511706, | |
| "learning_rate": 2.7799111902582696e-05, | |
| "loss": 0.1505, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.016989863234842244, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1349, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1875, | |
| "grad_norm": 0.01845190519100938, | |
| "learning_rate": 2.2200888097417307e-05, | |
| "loss": 0.1356, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.021100699063456414, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 0.136, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3125, | |
| "grad_norm": 0.018945233559077897, | |
| "learning_rate": 1.6743023451120832e-05, | |
| "loss": 0.1453, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.019023748911866034, | |
| "learning_rate": 1.4152906522061048e-05, | |
| "loss": 0.1297, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 0.01983379100230869, | |
| "learning_rate": 1.1699198087116589e-05, | |
| "loss": 0.1288, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.020004728526161076, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 0.1372, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.018644802377325877, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1264, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.018974115353534597, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 0.1223, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6875, | |
| "grad_norm": 0.02101774905754192, | |
| "learning_rate": 3.831895019292897e-06, | |
| "loss": 0.1242, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.020105790792392823, | |
| "learning_rate": 2.475778302439524e-06, | |
| "loss": 0.1253, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.02049360956396613, | |
| "learning_rate": 1.4029167422908107e-06, | |
| "loss": 0.1437, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.022791100110954515, | |
| "learning_rate": 6.268021954544096e-07, | |
| "loss": 0.1283, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9375, | |
| "grad_norm": 0.018080427643756062, | |
| "learning_rate": 1.571947526689349e-07, | |
| "loss": 0.1283, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.01817465286562608, | |
| "learning_rate": 0.0, | |
| "loss": 0.1267, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 32, | |
| "total_flos": 614948268933120.0, | |
| "train_loss": 0.14902637084014714, | |
| "train_runtime": 1084.9601, | |
| "train_samples_per_second": 0.461, | |
| "train_steps_per_second": 0.029 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 32, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 614948268933120.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |