lillian039 commited on
Commit
b88ed9a
·
verified ·
1 Parent(s): fbbbc4f

Model save

Browse files
Files changed (5) hide show
  1. README.md +67 -0
  2. all_results.json +9 -0
  3. generation_config.json +12 -0
  4. train_results.json +9 -0
  5. trainer_state.json +1136 -0
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: llama3.1
4
+ base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: transduction-10k-seed50-instruct-fft_lr1e-5_epoch2
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # transduction-10k-seed50-instruct-fft_lr1e-5_epoch2
18
+
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.0693
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 4
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 8
46
+ - gradient_accumulation_steps: 2
47
+ - total_train_batch_size: 128
48
+ - total_eval_batch_size: 32
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.1
52
+ - num_epochs: 2
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:------:|:----:|:---------------:|
58
+ | 0.0681 | 0.9935 | 77 | 0.0811 |
59
+ | 0.0505 | 1.9871 | 154 | 0.0693 |
60
+
61
+
62
+ ### Framework versions
63
+
64
+ - Transformers 4.45.0.dev0
65
+ - Pytorch 2.4.0+cu121
66
+ - Datasets 3.0.1
67
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.9870967741935484,
3
+ "total_flos": 5.361988883508429e+16,
4
+ "train_loss": 0.08745080387437498,
5
+ "train_runtime": 2120.7101,
6
+ "train_samples": 9863,
7
+ "train_samples_per_second": 9.302,
8
+ "train_steps_per_second": 0.073
9
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128000,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 128001,
6
+ 128008,
7
+ 128009
8
+ ],
9
+ "temperature": 0.6,
10
+ "top_p": 0.9,
11
+ "transformers_version": "4.45.0.dev0"
12
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.9870967741935484,
3
+ "total_flos": 5.361988883508429e+16,
4
+ "train_loss": 0.08745080387437498,
5
+ "train_runtime": 2120.7101,
6
+ "train_samples": 9863,
7
+ "train_samples_per_second": 9.302,
8
+ "train_steps_per_second": 0.073
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9870967741935484,
5
+ "eval_steps": 500,
6
+ "global_step": 154,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.012903225806451613,
13
+ "grad_norm": 11.699292902761476,
14
+ "learning_rate": 6.25e-07,
15
+ "loss": 0.3091,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.025806451612903226,
20
+ "grad_norm": 10.2656902023561,
21
+ "learning_rate": 1.25e-06,
22
+ "loss": 0.2644,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.03870967741935484,
27
+ "grad_norm": 10.961480180598883,
28
+ "learning_rate": 1.8750000000000003e-06,
29
+ "loss": 0.3136,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.05161290322580645,
34
+ "grad_norm": 10.621965020575766,
35
+ "learning_rate": 2.5e-06,
36
+ "loss": 0.2649,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.06451612903225806,
41
+ "grad_norm": 7.63251523866155,
42
+ "learning_rate": 3.125e-06,
43
+ "loss": 0.2248,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.07741935483870968,
48
+ "grad_norm": 4.399939556445864,
49
+ "learning_rate": 3.7500000000000005e-06,
50
+ "loss": 0.1872,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.09032258064516129,
55
+ "grad_norm": 3.6018933396960233,
56
+ "learning_rate": 4.3750000000000005e-06,
57
+ "loss": 0.1277,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.1032258064516129,
62
+ "grad_norm": 2.350482184168289,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.1321,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.11612903225806452,
69
+ "grad_norm": 3.6577146283118167,
70
+ "learning_rate": 5.625e-06,
71
+ "loss": 0.1401,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.12903225806451613,
76
+ "grad_norm": 3.095671993174613,
77
+ "learning_rate": 6.25e-06,
78
+ "loss": 0.1247,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.14193548387096774,
83
+ "grad_norm": 1.8930944522483173,
84
+ "learning_rate": 6.875e-06,
85
+ "loss": 0.1243,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.15483870967741936,
90
+ "grad_norm": 3.4213993152717372,
91
+ "learning_rate": 7.500000000000001e-06,
92
+ "loss": 0.1183,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.16774193548387098,
97
+ "grad_norm": 3.0014464259254146,
98
+ "learning_rate": 8.125000000000001e-06,
99
+ "loss": 0.1319,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.18064516129032257,
104
+ "grad_norm": 2.846735403976308,
105
+ "learning_rate": 8.750000000000001e-06,
106
+ "loss": 0.1158,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.1935483870967742,
111
+ "grad_norm": 2.7768857827987614,
112
+ "learning_rate": 9.375000000000001e-06,
113
+ "loss": 0.1053,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.2064516129032258,
118
+ "grad_norm": 2.559286174315675,
119
+ "learning_rate": 1e-05,
120
+ "loss": 0.1351,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.21935483870967742,
125
+ "grad_norm": 2.793791051458797,
126
+ "learning_rate": 9.998704424206747e-06,
127
+ "loss": 0.1532,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.23225806451612904,
132
+ "grad_norm": 2.011717818438971,
133
+ "learning_rate": 9.994818368233639e-06,
134
+ "loss": 0.1041,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.24516129032258063,
139
+ "grad_norm": 2.44195378918142,
140
+ "learning_rate": 9.988343845952697e-06,
141
+ "loss": 0.1074,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.25806451612903225,
146
+ "grad_norm": 1.6874842592386545,
147
+ "learning_rate": 9.979284212657658e-06,
148
+ "loss": 0.1034,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.2709677419354839,
153
+ "grad_norm": 1.3503672327570895,
154
+ "learning_rate": 9.967644163325157e-06,
155
+ "loss": 0.0971,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.2838709677419355,
160
+ "grad_norm": 3.534445687027426,
161
+ "learning_rate": 9.953429730181653e-06,
162
+ "loss": 0.122,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.2967741935483871,
167
+ "grad_norm": 2.731723348296494,
168
+ "learning_rate": 9.93664827957735e-06,
169
+ "loss": 0.1402,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.3096774193548387,
174
+ "grad_norm": 2.516223822228138,
175
+ "learning_rate": 9.917308508168712e-06,
176
+ "loss": 0.1093,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.3225806451612903,
181
+ "grad_norm": 1.7606357719981414,
182
+ "learning_rate": 9.895420438411616e-06,
183
+ "loss": 0.1231,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.33548387096774196,
188
+ "grad_norm": 2.4360379684882654,
189
+ "learning_rate": 9.870995413367397e-06,
190
+ "loss": 0.1113,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.34838709677419355,
195
+ "grad_norm": 1.1166024452341257,
196
+ "learning_rate": 9.844046090824533e-06,
197
+ "loss": 0.0946,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.36129032258064514,
202
+ "grad_norm": 3.095834517141857,
203
+ "learning_rate": 9.814586436738998e-06,
204
+ "loss": 0.1032,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.3741935483870968,
209
+ "grad_norm": 3.200925995404148,
210
+ "learning_rate": 9.782631717996675e-06,
211
+ "loss": 0.1637,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.3870967741935484,
216
+ "grad_norm": 2.7067354639721333,
217
+ "learning_rate": 9.748198494501598e-06,
218
+ "loss": 0.1326,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.4,
223
+ "grad_norm": 1.9657792494045345,
224
+ "learning_rate": 9.711304610594104e-06,
225
+ "loss": 0.0985,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.4129032258064516,
230
+ "grad_norm": 1.2418236122465918,
231
+ "learning_rate": 9.671969185803357e-06,
232
+ "loss": 0.0866,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.4258064516129032,
237
+ "grad_norm": 1.4668414290482827,
238
+ "learning_rate": 9.630212604939026e-06,
239
+ "loss": 0.091,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.43870967741935485,
244
+ "grad_norm": 1.5674054362390097,
245
+ "learning_rate": 9.586056507527266e-06,
246
+ "loss": 0.1166,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.45161290322580644,
251
+ "grad_norm": 1.7105084425031547,
252
+ "learning_rate": 9.539523776596446e-06,
253
+ "loss": 0.121,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.4645161290322581,
258
+ "grad_norm": 1.3111865574883335,
259
+ "learning_rate": 9.490638526818482e-06,
260
+ "loss": 0.0835,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.4774193548387097,
265
+ "grad_norm": 1.0229500712234056,
266
+ "learning_rate": 9.439426092011877e-06,
267
+ "loss": 0.0947,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.49032258064516127,
272
+ "grad_norm": 1.2556959616077457,
273
+ "learning_rate": 9.385913012012972e-06,
274
+ "loss": 0.0911,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.5032258064516129,
279
+ "grad_norm": 1.0982611947906955,
280
+ "learning_rate": 9.330127018922195e-06,
281
+ "loss": 0.105,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.5161290322580645,
286
+ "grad_norm": 1.0216693563567487,
287
+ "learning_rate": 9.272097022732444e-06,
288
+ "loss": 0.0839,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.5290322580645161,
293
+ "grad_norm": 0.887997821512209,
294
+ "learning_rate": 9.211853096347059e-06,
295
+ "loss": 0.0823,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.5419354838709678,
300
+ "grad_norm": 1.5086680290000998,
301
+ "learning_rate": 9.149426459995127e-06,
302
+ "loss": 0.1165,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.5548387096774193,
307
+ "grad_norm": 1.1170270141146377,
308
+ "learning_rate": 9.08484946505221e-06,
309
+ "loss": 0.0964,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.567741935483871,
314
+ "grad_norm": 0.9096654753927038,
315
+ "learning_rate": 9.018155577274891e-06,
316
+ "loss": 0.1003,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.5806451612903226,
321
+ "grad_norm": 1.0160863955008763,
322
+ "learning_rate": 8.949379359457795e-06,
323
+ "loss": 0.0802,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.5935483870967742,
328
+ "grad_norm": 1.0265412727158079,
329
+ "learning_rate": 8.8785564535221e-06,
330
+ "loss": 0.0884,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.6064516129032258,
335
+ "grad_norm": 1.5105953615602765,
336
+ "learning_rate": 8.805723562044825e-06,
337
+ "loss": 0.0855,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.6193548387096774,
342
+ "grad_norm": 1.1851230689340553,
343
+ "learning_rate": 8.730918429238429e-06,
344
+ "loss": 0.0896,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.632258064516129,
349
+ "grad_norm": 0.9547624249115597,
350
+ "learning_rate": 8.65417982139062e-06,
351
+ "loss": 0.0895,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.6451612903225806,
356
+ "grad_norm": 0.8828392166882096,
357
+ "learning_rate": 8.575547506774498e-06,
358
+ "loss": 0.0703,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.6580645161290323,
363
+ "grad_norm": 1.5517765054788453,
364
+ "learning_rate": 8.49506223503941e-06,
365
+ "loss": 0.0982,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.6709677419354839,
370
+ "grad_norm": 0.7610530682010276,
371
+ "learning_rate": 8.412765716093273e-06,
372
+ "loss": 0.09,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.6838709677419355,
377
+ "grad_norm": 1.2971615645647234,
378
+ "learning_rate": 8.328700598487203e-06,
379
+ "loss": 0.0951,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.6967741935483871,
384
+ "grad_norm": 0.6667526127324459,
385
+ "learning_rate": 8.24291044731378e-06,
386
+ "loss": 0.0624,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.7096774193548387,
391
+ "grad_norm": 1.1745372262285316,
392
+ "learning_rate": 8.155439721630265e-06,
393
+ "loss": 0.0955,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.7225806451612903,
398
+ "grad_norm": 0.8393263149118976,
399
+ "learning_rate": 8.066333751418582e-06,
400
+ "loss": 0.0734,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.7354838709677419,
405
+ "grad_norm": 0.9883905258578732,
406
+ "learning_rate": 7.97563871409395e-06,
407
+ "loss": 0.0753,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.7483870967741936,
412
+ "grad_norm": 0.8348230527715342,
413
+ "learning_rate": 7.883401610574338e-06,
414
+ "loss": 0.0764,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.7612903225806451,
419
+ "grad_norm": 1.0240572576125775,
420
+ "learning_rate": 7.789670240923169e-06,
421
+ "loss": 0.1,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.7741935483870968,
426
+ "grad_norm": 1.06880647144967,
427
+ "learning_rate": 7.69449317957788e-06,
428
+ "loss": 0.0917,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.7870967741935484,
433
+ "grad_norm": 0.9745921300249953,
434
+ "learning_rate": 7.597919750177168e-06,
435
+ "loss": 0.0903,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.8,
440
+ "grad_norm": 0.7844807247569731,
441
+ "learning_rate": 7.500000000000001e-06,
442
+ "loss": 0.0673,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.8129032258064516,
447
+ "grad_norm": 1.615686383831043,
448
+ "learning_rate": 7.400784674029579e-06,
449
+ "loss": 0.1024,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.8258064516129032,
454
+ "grad_norm": 0.7238588061260165,
455
+ "learning_rate": 7.300325188655762e-06,
456
+ "loss": 0.0705,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.8387096774193549,
461
+ "grad_norm": 1.3928662222572579,
462
+ "learning_rate": 7.198673605029529e-06,
463
+ "loss": 0.086,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.8516129032258064,
468
+ "grad_norm": 0.9405975350441244,
469
+ "learning_rate": 7.095882602083321e-06,
470
+ "loss": 0.081,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.864516129032258,
475
+ "grad_norm": 0.7204214932073081,
476
+ "learning_rate": 6.9920054492312086e-06,
477
+ "loss": 0.07,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.8774193548387097,
482
+ "grad_norm": 1.0257095309963755,
483
+ "learning_rate": 6.887095978763072e-06,
484
+ "loss": 0.0633,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.8903225806451613,
489
+ "grad_norm": 0.6843552801844485,
490
+ "learning_rate": 6.781208557947085e-06,
491
+ "loss": 0.0535,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.9032258064516129,
496
+ "grad_norm": 0.8369182932506963,
497
+ "learning_rate": 6.674398060854931e-06,
498
+ "loss": 0.0607,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.9161290322580645,
503
+ "grad_norm": 0.9450663518863992,
504
+ "learning_rate": 6.566719839924412e-06,
505
+ "loss": 0.0817,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.9290322580645162,
510
+ "grad_norm": 0.9648342453237144,
511
+ "learning_rate": 6.458229697274125e-06,
512
+ "loss": 0.0784,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.9419354838709677,
517
+ "grad_norm": 0.7226047890347407,
518
+ "learning_rate": 6.348983855785122e-06,
519
+ "loss": 0.0723,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.9548387096774194,
524
+ "grad_norm": 0.9883255400489244,
525
+ "learning_rate": 6.2390389299645e-06,
526
+ "loss": 0.0884,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.967741935483871,
531
+ "grad_norm": 0.8546590385544072,
532
+ "learning_rate": 6.128451896606054e-06,
533
+ "loss": 0.084,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.9806451612903225,
538
+ "grad_norm": 0.9429219413240326,
539
+ "learning_rate": 6.0172800652631706e-06,
540
+ "loss": 0.0809,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.9935483870967742,
545
+ "grad_norm": 0.8587345660767025,
546
+ "learning_rate": 5.905581048549279e-06,
547
+ "loss": 0.0681,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.9935483870967742,
552
+ "eval_loss": 0.08105655014514923,
553
+ "eval_runtime": 39.4157,
554
+ "eval_samples_per_second": 26.36,
555
+ "eval_steps_per_second": 0.837,
556
+ "step": 77
557
+ },
558
+ {
559
+ "epoch": 1.0064516129032257,
560
+ "grad_norm": 1.1624230129889668,
561
+ "learning_rate": 5.793412732281258e-06,
562
+ "loss": 0.0808,
563
+ "step": 78
564
+ },
565
+ {
566
+ "epoch": 1.0193548387096774,
567
+ "grad_norm": 1.0860238876477792,
568
+ "learning_rate": 5.680833245481234e-06,
569
+ "loss": 0.0792,
570
+ "step": 79
571
+ },
572
+ {
573
+ "epoch": 1.032258064516129,
574
+ "grad_norm": 0.8760532041288109,
575
+ "learning_rate": 5.567900930252375e-06,
576
+ "loss": 0.0675,
577
+ "step": 80
578
+ },
579
+ {
580
+ "epoch": 1.0451612903225806,
581
+ "grad_norm": 0.8966395459365959,
582
+ "learning_rate": 5.454674311544236e-06,
583
+ "loss": 0.0682,
584
+ "step": 81
585
+ },
586
+ {
587
+ "epoch": 1.0580645161290323,
588
+ "grad_norm": 1.019077536054118,
589
+ "learning_rate": 5.341212066823356e-06,
590
+ "loss": 0.0799,
591
+ "step": 82
592
+ },
593
+ {
594
+ "epoch": 1.070967741935484,
595
+ "grad_norm": 1.3438302637394202,
596
+ "learning_rate": 5.227572995664819e-06,
597
+ "loss": 0.0882,
598
+ "step": 83
599
+ },
600
+ {
601
+ "epoch": 1.0838709677419356,
602
+ "grad_norm": 0.9235621393269036,
603
+ "learning_rate": 5.113815989280528e-06,
604
+ "loss": 0.0763,
605
+ "step": 84
606
+ },
607
+ {
608
+ "epoch": 1.096774193548387,
609
+ "grad_norm": 0.7779427795576146,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.0773,
612
+ "step": 85
613
+ },
614
+ {
615
+ "epoch": 1.1096774193548387,
616
+ "grad_norm": 1.0066913546499912,
617
+ "learning_rate": 4.886184010719472e-06,
618
+ "loss": 0.0762,
619
+ "step": 86
620
+ },
621
+ {
622
+ "epoch": 1.1225806451612903,
623
+ "grad_norm": 0.849885708644481,
624
+ "learning_rate": 4.772427004335183e-06,
625
+ "loss": 0.085,
626
+ "step": 87
627
+ },
628
+ {
629
+ "epoch": 1.135483870967742,
630
+ "grad_norm": 0.8282232117216065,
631
+ "learning_rate": 4.6587879331766465e-06,
632
+ "loss": 0.0724,
633
+ "step": 88
634
+ },
635
+ {
636
+ "epoch": 1.1483870967741936,
637
+ "grad_norm": 0.6227012177536829,
638
+ "learning_rate": 4.545325688455766e-06,
639
+ "loss": 0.0746,
640
+ "step": 89
641
+ },
642
+ {
643
+ "epoch": 1.1612903225806452,
644
+ "grad_norm": 0.9694041549131266,
645
+ "learning_rate": 4.432099069747625e-06,
646
+ "loss": 0.0876,
647
+ "step": 90
648
+ },
649
+ {
650
+ "epoch": 1.1741935483870969,
651
+ "grad_norm": 1.2344277757510045,
652
+ "learning_rate": 4.319166754518768e-06,
653
+ "loss": 0.0728,
654
+ "step": 91
655
+ },
656
+ {
657
+ "epoch": 1.1870967741935483,
658
+ "grad_norm": 0.5797726042270337,
659
+ "learning_rate": 4.206587267718743e-06,
660
+ "loss": 0.05,
661
+ "step": 92
662
+ },
663
+ {
664
+ "epoch": 1.2,
665
+ "grad_norm": 0.6275337263216433,
666
+ "learning_rate": 4.094418951450721e-06,
667
+ "loss": 0.0458,
668
+ "step": 93
669
+ },
670
+ {
671
+ "epoch": 1.2129032258064516,
672
+ "grad_norm": 0.991136454485937,
673
+ "learning_rate": 3.982719934736832e-06,
674
+ "loss": 0.1011,
675
+ "step": 94
676
+ },
677
+ {
678
+ "epoch": 1.2258064516129032,
679
+ "grad_norm": 0.5693434030485786,
680
+ "learning_rate": 3.871548103393947e-06,
681
+ "loss": 0.0535,
682
+ "step": 95
683
+ },
684
+ {
685
+ "epoch": 1.238709677419355,
686
+ "grad_norm": 0.8648629215753009,
687
+ "learning_rate": 3.7609610700355014e-06,
688
+ "loss": 0.0726,
689
+ "step": 96
690
+ },
691
+ {
692
+ "epoch": 1.2516129032258063,
693
+ "grad_norm": 0.49590292970379035,
694
+ "learning_rate": 3.6510161442148783e-06,
695
+ "loss": 0.0493,
696
+ "step": 97
697
+ },
698
+ {
699
+ "epoch": 1.2645161290322582,
700
+ "grad_norm": 0.5252751061515991,
701
+ "learning_rate": 3.5417703027258752e-06,
702
+ "loss": 0.0499,
703
+ "step": 98
704
+ },
705
+ {
706
+ "epoch": 1.2774193548387096,
707
+ "grad_norm": 0.5165334726304951,
708
+ "learning_rate": 3.4332801600755895e-06,
709
+ "loss": 0.0398,
710
+ "step": 99
711
+ },
712
+ {
713
+ "epoch": 1.2903225806451613,
714
+ "grad_norm": 0.5588113997669555,
715
+ "learning_rate": 3.3256019391450696e-06,
716
+ "loss": 0.0451,
717
+ "step": 100
718
+ },
719
+ {
720
+ "epoch": 1.303225806451613,
721
+ "grad_norm": 0.8000418887806057,
722
+ "learning_rate": 3.2187914420529176e-06,
723
+ "loss": 0.0558,
724
+ "step": 101
725
+ },
726
+ {
727
+ "epoch": 1.3161290322580645,
728
+ "grad_norm": 0.6209463591754386,
729
+ "learning_rate": 3.1129040212369286e-06,
730
+ "loss": 0.0543,
731
+ "step": 102
732
+ },
733
+ {
734
+ "epoch": 1.3290322580645162,
735
+ "grad_norm": 0.5956338266252976,
736
+ "learning_rate": 3.007994550768793e-06,
737
+ "loss": 0.0602,
738
+ "step": 103
739
+ },
740
+ {
741
+ "epoch": 1.3419354838709676,
742
+ "grad_norm": 0.7601669697082293,
743
+ "learning_rate": 2.9041173979166813e-06,
744
+ "loss": 0.061,
745
+ "step": 104
746
+ },
747
+ {
748
+ "epoch": 1.3548387096774195,
749
+ "grad_norm": 0.5346685587339685,
750
+ "learning_rate": 2.8013263949704706e-06,
751
+ "loss": 0.0516,
752
+ "step": 105
753
+ },
754
+ {
755
+ "epoch": 1.367741935483871,
756
+ "grad_norm": 0.6722589800675358,
757
+ "learning_rate": 2.6996748113442397e-06,
758
+ "loss": 0.0667,
759
+ "step": 106
760
+ },
761
+ {
762
+ "epoch": 1.3806451612903226,
763
+ "grad_norm": 0.7382443963105985,
764
+ "learning_rate": 2.599215325970423e-06,
765
+ "loss": 0.0755,
766
+ "step": 107
767
+ },
768
+ {
769
+ "epoch": 1.3935483870967742,
770
+ "grad_norm": 0.717323507281047,
771
+ "learning_rate": 2.5000000000000015e-06,
772
+ "loss": 0.0608,
773
+ "step": 108
774
+ },
775
+ {
776
+ "epoch": 1.4064516129032258,
777
+ "grad_norm": 0.5934322456540527,
778
+ "learning_rate": 2.4020802498228333e-06,
779
+ "loss": 0.0643,
780
+ "step": 109
781
+ },
782
+ {
783
+ "epoch": 1.4193548387096775,
784
+ "grad_norm": 1.0152254575859148,
785
+ "learning_rate": 2.3055068204221226e-06,
786
+ "loss": 0.0926,
787
+ "step": 110
788
+ },
789
+ {
790
+ "epoch": 1.432258064516129,
791
+ "grad_norm": 0.7338463370837035,
792
+ "learning_rate": 2.2103297590768334e-06,
793
+ "loss": 0.0718,
794
+ "step": 111
795
+ },
796
+ {
797
+ "epoch": 1.4451612903225808,
798
+ "grad_norm": 0.6224898983786531,
799
+ "learning_rate": 2.1165983894256647e-06,
800
+ "loss": 0.0722,
801
+ "step": 112
802
+ },
803
+ {
804
+ "epoch": 1.4580645161290322,
805
+ "grad_norm": 0.7111315812802891,
806
+ "learning_rate": 2.0243612859060526e-06,
807
+ "loss": 0.0796,
808
+ "step": 113
809
+ },
810
+ {
811
+ "epoch": 1.4709677419354839,
812
+ "grad_norm": 0.5880677662618246,
813
+ "learning_rate": 1.933666248581418e-06,
814
+ "loss": 0.0612,
815
+ "step": 114
816
+ },
817
+ {
818
+ "epoch": 1.4838709677419355,
819
+ "grad_norm": 0.5688066714876514,
820
+ "learning_rate": 1.8445602783697375e-06,
821
+ "loss": 0.0697,
822
+ "step": 115
823
+ },
824
+ {
825
+ "epoch": 1.4967741935483871,
826
+ "grad_norm": 0.772697127447158,
827
+ "learning_rate": 1.7570895526862202e-06,
828
+ "loss": 0.0863,
829
+ "step": 116
830
+ },
831
+ {
832
+ "epoch": 1.5096774193548388,
833
+ "grad_norm": 0.5188748378473136,
834
+ "learning_rate": 1.6712994015127976e-06,
835
+ "loss": 0.0456,
836
+ "step": 117
837
+ },
838
+ {
839
+ "epoch": 1.5225806451612902,
840
+ "grad_norm": 0.5787690355073418,
841
+ "learning_rate": 1.5872342839067305e-06,
842
+ "loss": 0.0557,
843
+ "step": 118
844
+ },
845
+ {
846
+ "epoch": 1.535483870967742,
847
+ "grad_norm": 1.1554001099512141,
848
+ "learning_rate": 1.5049377649605906e-06,
849
+ "loss": 0.0807,
850
+ "step": 119
851
+ },
852
+ {
853
+ "epoch": 1.5483870967741935,
854
+ "grad_norm": 0.5280610958351776,
855
+ "learning_rate": 1.4244524932255026e-06,
856
+ "loss": 0.0476,
857
+ "step": 120
858
+ },
859
+ {
860
+ "epoch": 1.5612903225806452,
861
+ "grad_norm": 0.6254239981186747,
862
+ "learning_rate": 1.3458201786093795e-06,
863
+ "loss": 0.07,
864
+ "step": 121
865
+ },
866
+ {
867
+ "epoch": 1.5741935483870968,
868
+ "grad_norm": 0.526881850014464,
869
+ "learning_rate": 1.2690815707615727e-06,
870
+ "loss": 0.0412,
871
+ "step": 122
872
+ },
873
+ {
874
+ "epoch": 1.5870967741935482,
875
+ "grad_norm": 0.5947939093981419,
876
+ "learning_rate": 1.194276437955177e-06,
877
+ "loss": 0.0594,
878
+ "step": 123
879
+ },
880
+ {
881
+ "epoch": 1.6,
882
+ "grad_norm": 0.8234862252522671,
883
+ "learning_rate": 1.1214435464779006e-06,
884
+ "loss": 0.0656,
885
+ "step": 124
886
+ },
887
+ {
888
+ "epoch": 1.6129032258064515,
889
+ "grad_norm": 0.6682337549186153,
890
+ "learning_rate": 1.050620640542208e-06,
891
+ "loss": 0.0821,
892
+ "step": 125
893
+ },
894
+ {
895
+ "epoch": 1.6258064516129034,
896
+ "grad_norm": 0.46843664405997654,
897
+ "learning_rate": 9.81844422725109e-07,
898
+ "loss": 0.0538,
899
+ "step": 126
900
+ },
901
+ {
902
+ "epoch": 1.6387096774193548,
903
+ "grad_norm": 0.5846588665419508,
904
+ "learning_rate": 9.151505349477901e-07,
905
+ "loss": 0.0552,
906
+ "step": 127
907
+ },
908
+ {
909
+ "epoch": 1.6516129032258065,
910
+ "grad_norm": 0.6543914516209731,
911
+ "learning_rate": 8.505735400048748e-07,
912
+ "loss": 0.0739,
913
+ "step": 128
914
+ },
915
+ {
916
+ "epoch": 1.664516129032258,
917
+ "grad_norm": 0.6018097015637573,
918
+ "learning_rate": 7.881469036529427e-07,
919
+ "loss": 0.0631,
920
+ "step": 129
921
+ },
922
+ {
923
+ "epoch": 1.6774193548387095,
924
+ "grad_norm": 0.6248249705045374,
925
+ "learning_rate": 7.279029772675572e-07,
926
+ "loss": 0.0678,
927
+ "step": 130
928
+ },
929
+ {
930
+ "epoch": 1.6903225806451614,
931
+ "grad_norm": 0.5074802404785773,
932
+ "learning_rate": 6.698729810778065e-07,
933
+ "loss": 0.0553,
934
+ "step": 131
935
+ },
936
+ {
937
+ "epoch": 1.7032258064516128,
938
+ "grad_norm": 0.6303808241003845,
939
+ "learning_rate": 6.140869879870287e-07,
940
+ "loss": 0.0573,
941
+ "step": 132
942
+ },
943
+ {
944
+ "epoch": 1.7161290322580647,
945
+ "grad_norm": 0.6548357295879033,
946
+ "learning_rate": 5.60573907988124e-07,
947
+ "loss": 0.0718,
948
+ "step": 133
949
+ },
950
+ {
951
+ "epoch": 1.729032258064516,
952
+ "grad_norm": 1.0474619581709672,
953
+ "learning_rate": 5.0936147318152e-07,
954
+ "loss": 0.0684,
955
+ "step": 134
956
+ },
957
+ {
958
+ "epoch": 1.7419354838709677,
959
+ "grad_norm": 0.4233097389816897,
960
+ "learning_rate": 4.604762234035548e-07,
961
+ "loss": 0.0425,
962
+ "step": 135
963
+ },
964
+ {
965
+ "epoch": 1.7548387096774194,
966
+ "grad_norm": 0.5486123769584077,
967
+ "learning_rate": 4.139434924727359e-07,
968
+ "loss": 0.0555,
969
+ "step": 136
970
+ },
971
+ {
972
+ "epoch": 1.7677419354838708,
973
+ "grad_norm": 0.42132757988454256,
974
+ "learning_rate": 3.697873950609737e-07,
975
+ "loss": 0.0454,
976
+ "step": 137
977
+ },
978
+ {
979
+ "epoch": 1.7806451612903227,
980
+ "grad_norm": 0.48857725592632506,
981
+ "learning_rate": 3.2803081419664483e-07,
982
+ "loss": 0.0486,
983
+ "step": 138
984
+ },
985
+ {
986
+ "epoch": 1.793548387096774,
987
+ "grad_norm": 0.6044138979316588,
988
+ "learning_rate": 2.88695389405898e-07,
989
+ "loss": 0.0625,
990
+ "step": 139
991
+ },
992
+ {
993
+ "epoch": 1.8064516129032258,
994
+ "grad_norm": 0.45648144621900827,
995
+ "learning_rate": 2.518015054984041e-07,
996
+ "loss": 0.0485,
997
+ "step": 140
998
+ },
999
+ {
1000
+ "epoch": 1.8193548387096774,
1001
+ "grad_norm": 0.6418678164146125,
1002
+ "learning_rate": 2.1736828200332628e-07,
1003
+ "loss": 0.053,
1004
+ "step": 141
1005
+ },
1006
+ {
1007
+ "epoch": 1.832258064516129,
1008
+ "grad_norm": 0.7770556575132163,
1009
+ "learning_rate": 1.8541356326100436e-07,
1010
+ "loss": 0.0678,
1011
+ "step": 142
1012
+ },
1013
+ {
1014
+ "epoch": 1.8451612903225807,
1015
+ "grad_norm": 0.5056829983103792,
1016
+ "learning_rate": 1.559539091754686e-07,
1017
+ "loss": 0.0513,
1018
+ "step": 143
1019
+ },
1020
+ {
1021
+ "epoch": 1.8580645161290321,
1022
+ "grad_norm": 0.6329718102226061,
1023
+ "learning_rate": 1.2900458663260506e-07,
1024
+ "loss": 0.082,
1025
+ "step": 144
1026
+ },
1027
+ {
1028
+ "epoch": 1.870967741935484,
1029
+ "grad_norm": 0.6625275685465645,
1030
+ "learning_rate": 1.0457956158838545e-07,
1031
+ "loss": 0.0626,
1032
+ "step": 145
1033
+ },
1034
+ {
1035
+ "epoch": 1.8838709677419354,
1036
+ "grad_norm": 0.4523695807369306,
1037
+ "learning_rate": 8.269149183128988e-08,
1038
+ "loss": 0.0457,
1039
+ "step": 146
1040
+ },
1041
+ {
1042
+ "epoch": 1.896774193548387,
1043
+ "grad_norm": 0.6684501205664699,
1044
+ "learning_rate": 6.335172042265192e-08,
1045
+ "loss": 0.0811,
1046
+ "step": 147
1047
+ },
1048
+ {
1049
+ "epoch": 1.9096774193548387,
1050
+ "grad_norm": 0.7933423479927778,
1051
+ "learning_rate": 4.657026981834623e-08,
1052
+ "loss": 0.0675,
1053
+ "step": 148
1054
+ },
1055
+ {
1056
+ "epoch": 1.9225806451612903,
1057
+ "grad_norm": 0.8055874409945138,
1058
+ "learning_rate": 3.235583667484443e-08,
1059
+ "loss": 0.0733,
1060
+ "step": 149
1061
+ },
1062
+ {
1063
+ "epoch": 1.935483870967742,
1064
+ "grad_norm": 0.4268126124624864,
1065
+ "learning_rate": 2.0715787342343586e-08,
1066
+ "loss": 0.0386,
1067
+ "step": 150
1068
+ },
1069
+ {
1070
+ "epoch": 1.9483870967741934,
1071
+ "grad_norm": 0.6755068312271777,
1072
+ "learning_rate": 1.1656154047303691e-08,
1073
+ "loss": 0.0681,
1074
+ "step": 151
1075
+ },
1076
+ {
1077
+ "epoch": 1.9612903225806453,
1078
+ "grad_norm": 0.5936835028163678,
1079
+ "learning_rate": 5.181631766362216e-09,
1080
+ "loss": 0.0675,
1081
+ "step": 152
1082
+ },
1083
+ {
1084
+ "epoch": 1.9741935483870967,
1085
+ "grad_norm": 0.45412737353530935,
1086
+ "learning_rate": 1.2955757932542334e-09,
1087
+ "loss": 0.0523,
1088
+ "step": 153
1089
+ },
1090
+ {
1091
+ "epoch": 1.9870967741935484,
1092
+ "grad_norm": 0.5652252375803395,
1093
+ "learning_rate": 0.0,
1094
+ "loss": 0.0505,
1095
+ "step": 154
1096
+ },
1097
+ {
1098
+ "epoch": 1.9870967741935484,
1099
+ "eval_loss": 0.06934936344623566,
1100
+ "eval_runtime": 38.7654,
1101
+ "eval_samples_per_second": 26.802,
1102
+ "eval_steps_per_second": 0.851,
1103
+ "step": 154
1104
+ },
1105
+ {
1106
+ "epoch": 1.9870967741935484,
1107
+ "step": 154,
1108
+ "total_flos": 5.361988883508429e+16,
1109
+ "train_loss": 0.08745080387437498,
1110
+ "train_runtime": 2120.7101,
1111
+ "train_samples_per_second": 9.302,
1112
+ "train_steps_per_second": 0.073
1113
+ }
1114
+ ],
1115
+ "logging_steps": 1,
1116
+ "max_steps": 154,
1117
+ "num_input_tokens_seen": 0,
1118
+ "num_train_epochs": 2,
1119
+ "save_steps": 500,
1120
+ "stateful_callbacks": {
1121
+ "TrainerControl": {
1122
+ "args": {
1123
+ "should_epoch_stop": false,
1124
+ "should_evaluate": false,
1125
+ "should_log": false,
1126
+ "should_save": true,
1127
+ "should_training_stop": true
1128
+ },
1129
+ "attributes": {}
1130
+ }
1131
+ },
1132
+ "total_flos": 5.361988883508429e+16,
1133
+ "train_batch_size": 8,
1134
+ "trial_name": null,
1135
+ "trial_params": null
1136
+ }