nthakur commited on
Commit
cb8e87f
·
verified ·
1 Parent(s): 89adaf9

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: mistralai/Mistral-7B-Instruct-v0.2
9
+ model-index:
10
+ - name: Mistral-7B-Instruct-v0.2-mirage-gpt-4o-sft-instruct-mistral
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
+ # Mistral-7B-Instruct-v0.2-mirage-gpt-4o-sft-instruct-mistral
18
+
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.2326
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: 0.0002
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 4
46
+ - total_train_batch_size: 32
47
+ - total_eval_batch_size: 32
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: cosine
50
+ - lr_scheduler_warmup_ratio: 0.1
51
+ - num_epochs: 1
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:------:|:----:|:---------------:|
57
+ | 0.3189 | 0.2535 | 200 | 0.2828 |
58
+ | 0.2906 | 0.5070 | 400 | 0.2465 |
59
+ | 0.2525 | 0.7605 | 600 | 0.2326 |
60
+
61
+
62
+ ### Framework versions
63
+
64
+ - PEFT 0.7.1
65
+ - Transformers 4.41.2
66
+ - Pytorch 2.3.0+cu121
67
+ - Datasets 2.20.0
68
+ - Tokenizers 0.19.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4aa95dcfcff5fe5e6f48c9ba4063ae680a68e6efc4ece2ed248e9c53cfa4d5ab
3
  size 83946192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:421842c06b301217070ca85feb1cc9a24bb21739151f1c272bc36fd434a3c933
3
  size 83946192
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 1.0371979095834624e+16,
4
+ "train_loss": 0.2726454405050314,
5
+ "train_runtime": 20804.2809,
6
+ "train_samples": 25235,
7
+ "train_samples_per_second": 1.213,
8
+ "train_steps_per_second": 0.038
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 1.0371979095834624e+16,
4
+ "train_loss": 0.2726454405050314,
5
+ "train_runtime": 20804.2809,
6
+ "train_samples": 25235,
7
+ "train_samples_per_second": 1.213,
8
+ "train_steps_per_second": 0.038
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1172 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 200,
6
+ "global_step": 789,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0012674271229404308,
13
+ "grad_norm": 0.8615043469450224,
14
+ "learning_rate": 2.531645569620253e-06,
15
+ "loss": 0.479,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0063371356147021544,
20
+ "grad_norm": 1.2064683000577734,
21
+ "learning_rate": 1.2658227848101267e-05,
22
+ "loss": 0.5395,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.012674271229404309,
27
+ "grad_norm": 1.0410453983480228,
28
+ "learning_rate": 2.5316455696202533e-05,
29
+ "loss": 0.6061,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.019011406844106463,
34
+ "grad_norm": 0.6487558463156515,
35
+ "learning_rate": 3.79746835443038e-05,
36
+ "loss": 0.5462,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.025348542458808618,
41
+ "grad_norm": 0.5370587969238408,
42
+ "learning_rate": 5.0632911392405066e-05,
43
+ "loss": 0.4583,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.031685678073510776,
48
+ "grad_norm": 0.5034416625324662,
49
+ "learning_rate": 6.329113924050633e-05,
50
+ "loss": 0.4883,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.03802281368821293,
55
+ "grad_norm": 0.39983399371170963,
56
+ "learning_rate": 7.59493670886076e-05,
57
+ "loss": 0.3405,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.044359949302915085,
62
+ "grad_norm": 0.43348705894690165,
63
+ "learning_rate": 8.860759493670887e-05,
64
+ "loss": 0.4164,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.050697084917617236,
69
+ "grad_norm": 0.38455150629020124,
70
+ "learning_rate": 0.00010126582278481013,
71
+ "loss": 0.3655,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.057034220532319393,
76
+ "grad_norm": 0.4620346635793738,
77
+ "learning_rate": 0.0001139240506329114,
78
+ "loss": 0.321,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.06337135614702155,
83
+ "grad_norm": 0.5487179791839946,
84
+ "learning_rate": 0.00012658227848101267,
85
+ "loss": 0.3917,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.0697084917617237,
90
+ "grad_norm": 0.4322988318215336,
91
+ "learning_rate": 0.00013924050632911395,
92
+ "loss": 0.2954,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.07604562737642585,
97
+ "grad_norm": 0.45659465975830865,
98
+ "learning_rate": 0.0001518987341772152,
99
+ "loss": 0.3362,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.08238276299112801,
104
+ "grad_norm": 0.5012029391973714,
105
+ "learning_rate": 0.00016455696202531648,
106
+ "loss": 0.3201,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.08871989860583017,
111
+ "grad_norm": 0.4257121777036391,
112
+ "learning_rate": 0.00017721518987341773,
113
+ "loss": 0.3103,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.09505703422053231,
118
+ "grad_norm": 0.49930786967179713,
119
+ "learning_rate": 0.00018987341772151899,
120
+ "loss": 0.3895,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.10139416983523447,
125
+ "grad_norm": 0.3987283812690162,
126
+ "learning_rate": 0.00019999902106840922,
127
+ "loss": 0.2607,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.10773130544993663,
132
+ "grad_norm": 0.4941124105408692,
133
+ "learning_rate": 0.00019996476047513454,
134
+ "loss": 0.3509,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.11406844106463879,
139
+ "grad_norm": 0.35089312674321954,
140
+ "learning_rate": 0.00019988157246677513,
141
+ "loss": 0.323,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.12040557667934093,
146
+ "grad_norm": 0.4238705680350401,
147
+ "learning_rate": 0.00019974949775942134,
148
+ "loss": 0.2834,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.1267427122940431,
153
+ "grad_norm": 0.45602469539148643,
154
+ "learning_rate": 0.00019956860099659346,
155
+ "loss": 0.3492,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.13307984790874525,
160
+ "grad_norm": 0.40420721250262537,
161
+ "learning_rate": 0.00019933897071760235,
162
+ "loss": 0.2865,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.1394169835234474,
167
+ "grad_norm": 0.47242229343887376,
168
+ "learning_rate": 0.00019906071931421413,
169
+ "loss": 0.3047,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.14575411913814956,
174
+ "grad_norm": 0.3662230634874705,
175
+ "learning_rate": 0.00019873398297564037,
176
+ "loss": 0.3183,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.1520912547528517,
181
+ "grad_norm": 0.3786550636833906,
182
+ "learning_rate": 0.00019835892162188066,
183
+ "loss": 0.2903,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.15842839036755388,
188
+ "grad_norm": 0.42556611928998644,
189
+ "learning_rate": 0.00019793571882545047,
190
+ "loss": 0.3298,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.16476552598225602,
195
+ "grad_norm": 0.3477579859642712,
196
+ "learning_rate": 0.0001974645817215322,
197
+ "loss": 0.2811,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.17110266159695817,
202
+ "grad_norm": 0.338149840344514,
203
+ "learning_rate": 0.0001969457409065933,
204
+ "loss": 0.3218,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.17743979721166034,
209
+ "grad_norm": 0.31604418234996284,
210
+ "learning_rate": 0.0001963794503255219,
211
+ "loss": 0.3181,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.18377693282636248,
216
+ "grad_norm": 0.37941148877648684,
217
+ "learning_rate": 0.00019576598714733431,
218
+ "loss": 0.2708,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.19011406844106463,
223
+ "grad_norm": 0.41662892297732984,
224
+ "learning_rate": 0.00019510565162951537,
225
+ "loss": 0.3133,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.1964512040557668,
230
+ "grad_norm": 0.33440472234808516,
231
+ "learning_rate": 0.0001943987669710586,
232
+ "loss": 0.3027,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.20278833967046894,
237
+ "grad_norm": 0.3858834779514916,
238
+ "learning_rate": 0.0001936456791542776,
239
+ "loss": 0.3148,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.20912547528517111,
244
+ "grad_norm": 0.3588666651678752,
245
+ "learning_rate": 0.000192846756775466,
246
+ "loss": 0.3038,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.21546261089987326,
251
+ "grad_norm": 0.3923595420574365,
252
+ "learning_rate": 0.00019200239086448933,
253
+ "loss": 0.251,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.2217997465145754,
258
+ "grad_norm": 0.3720411933468231,
259
+ "learning_rate": 0.0001911129946933968,
260
+ "loss": 0.3066,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.22813688212927757,
265
+ "grad_norm": 0.32287305381502635,
266
+ "learning_rate": 0.00019017900357414669,
267
+ "loss": 0.3021,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.23447401774397972,
272
+ "grad_norm": 0.3574163723783811,
273
+ "learning_rate": 0.00018920087464554427,
274
+ "loss": 0.3156,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.24081115335868186,
279
+ "grad_norm": 0.4152681195493882,
280
+ "learning_rate": 0.0001881790866494969,
281
+ "loss": 0.2839,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.24714828897338403,
286
+ "grad_norm": 0.36834582922377895,
287
+ "learning_rate": 0.00018711413969669526,
288
+ "loss": 0.2945,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.2534854245880862,
293
+ "grad_norm": 0.39981519235640095,
294
+ "learning_rate": 0.00018600655502183612,
295
+ "loss": 0.3189,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.2534854245880862,
300
+ "eval_loss": 0.28277337551116943,
301
+ "eval_runtime": 1084.4496,
302
+ "eval_samples_per_second": 3.689,
303
+ "eval_steps_per_second": 0.115,
304
+ "step": 200
305
+ },
306
+ {
307
+ "epoch": 0.2598225602027883,
308
+ "grad_norm": 0.31219177577589374,
309
+ "learning_rate": 0.0001848568747285054,
310
+ "loss": 0.2299,
311
+ "step": 205
312
+ },
313
+ {
314
+ "epoch": 0.2661596958174905,
315
+ "grad_norm": 0.37017559919032633,
316
+ "learning_rate": 0.00018366566152384773,
317
+ "loss": 0.2753,
318
+ "step": 210
319
+ },
320
+ {
321
+ "epoch": 0.27249683143219267,
322
+ "grad_norm": 0.29894743373031823,
323
+ "learning_rate": 0.00018243349844315117,
324
+ "loss": 0.2734,
325
+ "step": 215
326
+ },
327
+ {
328
+ "epoch": 0.2788339670468948,
329
+ "grad_norm": 0.3227139696142255,
330
+ "learning_rate": 0.00018116098856448253,
331
+ "loss": 0.2595,
332
+ "step": 220
333
+ },
334
+ {
335
+ "epoch": 0.28517110266159695,
336
+ "grad_norm": 0.3780707564332847,
337
+ "learning_rate": 0.00017984875471351302,
338
+ "loss": 0.3107,
339
+ "step": 225
340
+ },
341
+ {
342
+ "epoch": 0.2915082382762991,
343
+ "grad_norm": 0.31519655631233245,
344
+ "learning_rate": 0.00017849743915867807,
345
+ "loss": 0.2336,
346
+ "step": 230
347
+ },
348
+ {
349
+ "epoch": 0.29784537389100124,
350
+ "grad_norm": 0.35840275346083816,
351
+ "learning_rate": 0.00017710770329682144,
352
+ "loss": 0.2939,
353
+ "step": 235
354
+ },
355
+ {
356
+ "epoch": 0.3041825095057034,
357
+ "grad_norm": 0.2971942370747859,
358
+ "learning_rate": 0.0001756802273294766,
359
+ "loss": 0.2959,
360
+ "step": 240
361
+ },
362
+ {
363
+ "epoch": 0.3105196451204056,
364
+ "grad_norm": 0.3247193570102783,
365
+ "learning_rate": 0.0001742157099299445,
366
+ "loss": 0.2373,
367
+ "step": 245
368
+ },
369
+ {
370
+ "epoch": 0.31685678073510776,
371
+ "grad_norm": 0.38491700907647547,
372
+ "learning_rate": 0.00017271486790133023,
373
+ "loss": 0.3272,
374
+ "step": 250
375
+ },
376
+ {
377
+ "epoch": 0.3231939163498099,
378
+ "grad_norm": 0.3075011053275653,
379
+ "learning_rate": 0.00017117843582570608,
380
+ "loss": 0.2377,
381
+ "step": 255
382
+ },
383
+ {
384
+ "epoch": 0.32953105196451205,
385
+ "grad_norm": 0.31453423737207853,
386
+ "learning_rate": 0.00016960716570457292,
387
+ "loss": 0.2842,
388
+ "step": 260
389
+ },
390
+ {
391
+ "epoch": 0.3358681875792142,
392
+ "grad_norm": 0.24817557882025235,
393
+ "learning_rate": 0.00016800182659079568,
394
+ "loss": 0.2721,
395
+ "step": 265
396
+ },
397
+ {
398
+ "epoch": 0.34220532319391633,
399
+ "grad_norm": 0.3377039172873109,
400
+ "learning_rate": 0.00016636320421219278,
401
+ "loss": 0.2463,
402
+ "step": 270
403
+ },
404
+ {
405
+ "epoch": 0.3485424588086185,
406
+ "grad_norm": 0.3661471684768712,
407
+ "learning_rate": 0.00016469210058696446,
408
+ "loss": 0.3025,
409
+ "step": 275
410
+ },
411
+ {
412
+ "epoch": 0.3548795944233207,
413
+ "grad_norm": 0.27351558488185734,
414
+ "learning_rate": 0.0001629893336311477,
415
+ "loss": 0.2335,
416
+ "step": 280
417
+ },
418
+ {
419
+ "epoch": 0.3612167300380228,
420
+ "grad_norm": 0.4358770290773004,
421
+ "learning_rate": 0.00016125573675828983,
422
+ "loss": 0.279,
423
+ "step": 285
424
+ },
425
+ {
426
+ "epoch": 0.36755386565272496,
427
+ "grad_norm": 0.2774172437400498,
428
+ "learning_rate": 0.00015949215847153717,
429
+ "loss": 0.2581,
430
+ "step": 290
431
+ },
432
+ {
433
+ "epoch": 0.37389100126742714,
434
+ "grad_norm": 0.28991353506835355,
435
+ "learning_rate": 0.00015769946194833817,
436
+ "loss": 0.2314,
437
+ "step": 295
438
+ },
439
+ {
440
+ "epoch": 0.38022813688212925,
441
+ "grad_norm": 0.4117767724226869,
442
+ "learning_rate": 0.00015587852461796376,
443
+ "loss": 0.3218,
444
+ "step": 300
445
+ },
446
+ {
447
+ "epoch": 0.3865652724968314,
448
+ "grad_norm": 0.31038917537934363,
449
+ "learning_rate": 0.00015403023773205286,
450
+ "loss": 0.2315,
451
+ "step": 305
452
+ },
453
+ {
454
+ "epoch": 0.3929024081115336,
455
+ "grad_norm": 0.3358971163136089,
456
+ "learning_rate": 0.00015215550592839218,
457
+ "loss": 0.302,
458
+ "step": 310
459
+ },
460
+ {
461
+ "epoch": 0.39923954372623577,
462
+ "grad_norm": 0.24064351622215968,
463
+ "learning_rate": 0.00015025524678814427,
464
+ "loss": 0.2795,
465
+ "step": 315
466
+ },
467
+ {
468
+ "epoch": 0.4055766793409379,
469
+ "grad_norm": 0.32648334764287845,
470
+ "learning_rate": 0.00014833039038674047,
471
+ "loss": 0.2398,
472
+ "step": 320
473
+ },
474
+ {
475
+ "epoch": 0.41191381495564006,
476
+ "grad_norm": 0.4100611479861272,
477
+ "learning_rate": 0.0001463818788386588,
478
+ "loss": 0.2849,
479
+ "step": 325
480
+ },
481
+ {
482
+ "epoch": 0.41825095057034223,
483
+ "grad_norm": 0.333871597957756,
484
+ "learning_rate": 0.00014441066583630906,
485
+ "loss": 0.2297,
486
+ "step": 330
487
+ },
488
+ {
489
+ "epoch": 0.42458808618504434,
490
+ "grad_norm": 0.34174204454819196,
491
+ "learning_rate": 0.00014241771618325123,
492
+ "loss": 0.2704,
493
+ "step": 335
494
+ },
495
+ {
496
+ "epoch": 0.4309252217997465,
497
+ "grad_norm": 0.2539128471983557,
498
+ "learning_rate": 0.00014040400532197583,
499
+ "loss": 0.2579,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 0.4372623574144487,
504
+ "grad_norm": 0.29762852904975656,
505
+ "learning_rate": 0.0001383705188564767,
506
+ "loss": 0.255,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 0.4435994930291508,
511
+ "grad_norm": 0.3477101390224288,
512
+ "learning_rate": 0.00013631825206985063,
513
+ "loss": 0.2841,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 0.449936628643853,
518
+ "grad_norm": 0.2807172517529508,
519
+ "learning_rate": 0.0001342482094371591,
520
+ "loss": 0.2112,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 0.45627376425855515,
525
+ "grad_norm": 0.3838095185531826,
526
+ "learning_rate": 0.00013216140413379167,
527
+ "loss": 0.282,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 0.46261089987325726,
532
+ "grad_norm": 0.2421625693621495,
533
+ "learning_rate": 0.00013005885753957048,
534
+ "loss": 0.2473,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 0.46894803548795944,
539
+ "grad_norm": 0.32700829134754156,
540
+ "learning_rate": 0.0001279415987388395,
541
+ "loss": 0.2174,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 0.4752851711026616,
546
+ "grad_norm": 0.3713274528328548,
547
+ "learning_rate": 0.0001258106640167826,
548
+ "loss": 0.283,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 0.4816223067173637,
553
+ "grad_norm": 0.27730345855582755,
554
+ "learning_rate": 0.0001236670963522172,
555
+ "loss": 0.2021,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 0.4879594423320659,
560
+ "grad_norm": 0.34446917830376517,
561
+ "learning_rate": 0.00012151194490711178,
562
+ "loss": 0.2561,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 0.49429657794676807,
567
+ "grad_norm": 0.2522933072667094,
568
+ "learning_rate": 0.00011934626451307726,
569
+ "loss": 0.2433,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 0.5006337135614702,
574
+ "grad_norm": 0.3469006061112617,
575
+ "learning_rate": 0.00011717111515508319,
576
+ "loss": 0.2266,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 0.5069708491761724,
581
+ "grad_norm": 0.3778944814729631,
582
+ "learning_rate": 0.00011498756145265144,
583
+ "loss": 0.2906,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 0.5069708491761724,
588
+ "eval_loss": 0.2465018332004547,
589
+ "eval_runtime": 1086.0248,
590
+ "eval_samples_per_second": 3.683,
591
+ "eval_steps_per_second": 0.115,
592
+ "step": 400
593
+ },
594
+ {
595
+ "epoch": 0.5133079847908745,
596
+ "grad_norm": 0.2906640770941992,
597
+ "learning_rate": 0.00011279667213878205,
598
+ "loss": 0.2329,
599
+ "step": 405
600
+ },
601
+ {
602
+ "epoch": 0.5196451204055766,
603
+ "grad_norm": 0.3562675808467966,
604
+ "learning_rate": 0.00011059951953686535,
605
+ "loss": 0.2727,
606
+ "step": 410
607
+ },
608
+ {
609
+ "epoch": 0.5259822560202788,
610
+ "grad_norm": 0.23100289329465973,
611
+ "learning_rate": 0.00010839717903583684,
612
+ "loss": 0.2559,
613
+ "step": 415
614
+ },
615
+ {
616
+ "epoch": 0.532319391634981,
617
+ "grad_norm": 0.31979110117969145,
618
+ "learning_rate": 0.00010619072856383181,
619
+ "loss": 0.2413,
620
+ "step": 420
621
+ },
622
+ {
623
+ "epoch": 0.5386565272496832,
624
+ "grad_norm": 0.3131449805956835,
625
+ "learning_rate": 0.00010398124806059701,
626
+ "loss": 0.2807,
627
+ "step": 425
628
+ },
629
+ {
630
+ "epoch": 0.5449936628643853,
631
+ "grad_norm": 0.30855801225213353,
632
+ "learning_rate": 0.00010176981894891768,
633
+ "loss": 0.1961,
634
+ "step": 430
635
+ },
636
+ {
637
+ "epoch": 0.5513307984790875,
638
+ "grad_norm": 0.3422494582211193,
639
+ "learning_rate": 9.955752360531896e-05,
640
+ "loss": 0.2805,
641
+ "step": 435
642
+ },
643
+ {
644
+ "epoch": 0.5576679340937896,
645
+ "grad_norm": 0.26688868743798966,
646
+ "learning_rate": 9.734544483030026e-05,
647
+ "loss": 0.2597,
648
+ "step": 440
649
+ },
650
+ {
651
+ "epoch": 0.5640050697084917,
652
+ "grad_norm": 0.3161252521303147,
653
+ "learning_rate": 9.513466531836221e-05,
654
+ "loss": 0.2166,
655
+ "step": 445
656
+ },
657
+ {
658
+ "epoch": 0.5703422053231939,
659
+ "grad_norm": 0.3401785018630494,
660
+ "learning_rate": 9.292626712808556e-05,
661
+ "loss": 0.2827,
662
+ "step": 450
663
+ },
664
+ {
665
+ "epoch": 0.5766793409378961,
666
+ "grad_norm": 0.3036061299387351,
667
+ "learning_rate": 9.072133115252112e-05,
668
+ "loss": 0.2254,
669
+ "step": 455
670
+ },
671
+ {
672
+ "epoch": 0.5830164765525983,
673
+ "grad_norm": 0.35415945685790207,
674
+ "learning_rate": 8.85209365901505e-05,
675
+ "loss": 0.2692,
676
+ "step": 460
677
+ },
678
+ {
679
+ "epoch": 0.5893536121673004,
680
+ "grad_norm": 0.25401069755470806,
681
+ "learning_rate": 8.632616041667577e-05,
682
+ "loss": 0.2539,
683
+ "step": 465
684
+ },
685
+ {
686
+ "epoch": 0.5956907477820025,
687
+ "grad_norm": 0.31003086790259704,
688
+ "learning_rate": 8.41380768578976e-05,
689
+ "loss": 0.2271,
690
+ "step": 470
691
+ },
692
+ {
693
+ "epoch": 0.6020278833967047,
694
+ "grad_norm": 0.3237135373224526,
695
+ "learning_rate": 8.195775686393897e-05,
696
+ "loss": 0.2792,
697
+ "step": 475
698
+ },
699
+ {
700
+ "epoch": 0.6083650190114068,
701
+ "grad_norm": 0.2952913468902984,
702
+ "learning_rate": 7.978626758507217e-05,
703
+ "loss": 0.2033,
704
+ "step": 480
705
+ },
706
+ {
707
+ "epoch": 0.614702154626109,
708
+ "grad_norm": 0.32514176968318775,
709
+ "learning_rate": 7.762467184940574e-05,
710
+ "loss": 0.2632,
711
+ "step": 485
712
+ },
713
+ {
714
+ "epoch": 0.6210392902408112,
715
+ "grad_norm": 0.2704128994845209,
716
+ "learning_rate": 7.547402764268689e-05,
717
+ "loss": 0.2657,
718
+ "step": 490
719
+ },
720
+ {
721
+ "epoch": 0.6273764258555133,
722
+ "grad_norm": 0.3419677722251191,
723
+ "learning_rate": 7.333538759047389e-05,
724
+ "loss": 0.2432,
725
+ "step": 495
726
+ },
727
+ {
728
+ "epoch": 0.6337135614702155,
729
+ "grad_norm": 0.3336718483492428,
730
+ "learning_rate": 7.120979844293201e-05,
731
+ "loss": 0.2789,
732
+ "step": 500
733
+ },
734
+ {
735
+ "epoch": 0.6400506970849176,
736
+ "grad_norm": 0.28861366582406006,
737
+ "learning_rate": 6.909830056250527e-05,
738
+ "loss": 0.2185,
739
+ "step": 505
740
+ },
741
+ {
742
+ "epoch": 0.6463878326996197,
743
+ "grad_norm": 0.3292639675337998,
744
+ "learning_rate": 6.700192741471447e-05,
745
+ "loss": 0.2622,
746
+ "step": 510
747
+ },
748
+ {
749
+ "epoch": 0.6527249683143219,
750
+ "grad_norm": 0.2539285743910528,
751
+ "learning_rate": 6.4921705062331e-05,
752
+ "loss": 0.2442,
753
+ "step": 515
754
+ },
755
+ {
756
+ "epoch": 0.6590621039290241,
757
+ "grad_norm": 0.3067269034698268,
758
+ "learning_rate": 6.285865166317386e-05,
759
+ "loss": 0.2143,
760
+ "step": 520
761
+ },
762
+ {
763
+ "epoch": 0.6653992395437263,
764
+ "grad_norm": 0.3245132022152399,
765
+ "learning_rate": 6.081377697177576e-05,
766
+ "loss": 0.2719,
767
+ "step": 525
768
+ },
769
+ {
770
+ "epoch": 0.6717363751584284,
771
+ "grad_norm": 0.27285602978719187,
772
+ "learning_rate": 5.8788081845162246e-05,
773
+ "loss": 0.205,
774
+ "step": 530
775
+ },
776
+ {
777
+ "epoch": 0.6780735107731305,
778
+ "grad_norm": 0.33017438859174614,
779
+ "learning_rate": 5.678255775298542e-05,
780
+ "loss": 0.2595,
781
+ "step": 535
782
+ },
783
+ {
784
+ "epoch": 0.6844106463878327,
785
+ "grad_norm": 0.22042843479196494,
786
+ "learning_rate": 5.479818629225259e-05,
787
+ "loss": 0.2508,
788
+ "step": 540
789
+ },
790
+ {
791
+ "epoch": 0.6907477820025348,
792
+ "grad_norm": 0.3262775579247169,
793
+ "learning_rate": 5.2835938706886966e-05,
794
+ "loss": 0.2252,
795
+ "step": 545
796
+ },
797
+ {
798
+ "epoch": 0.697084917617237,
799
+ "grad_norm": 0.3604718135367501,
800
+ "learning_rate": 5.0896775412355434e-05,
801
+ "loss": 0.2703,
802
+ "step": 550
803
+ },
804
+ {
805
+ "epoch": 0.7034220532319392,
806
+ "grad_norm": 0.2849330234510489,
807
+ "learning_rate": 4.89816455255966e-05,
808
+ "loss": 0.1977,
809
+ "step": 555
810
+ },
811
+ {
812
+ "epoch": 0.7097591888466414,
813
+ "grad_norm": 0.2992590502661103,
814
+ "learning_rate": 4.7091486400478604e-05,
815
+ "loss": 0.252,
816
+ "step": 560
817
+ },
818
+ {
819
+ "epoch": 0.7160963244613435,
820
+ "grad_norm": 0.2424976560859796,
821
+ "learning_rate": 4.5227223169014456e-05,
822
+ "loss": 0.2465,
823
+ "step": 565
824
+ },
825
+ {
826
+ "epoch": 0.7224334600760456,
827
+ "grad_norm": 0.35965307208087977,
828
+ "learning_rate": 4.338976828855939e-05,
829
+ "loss": 0.2249,
830
+ "step": 570
831
+ },
832
+ {
833
+ "epoch": 0.7287705956907478,
834
+ "grad_norm": 0.3741830397170476,
835
+ "learning_rate": 4.1580021095211486e-05,
836
+ "loss": 0.284,
837
+ "step": 575
838
+ },
839
+ {
840
+ "epoch": 0.7351077313054499,
841
+ "grad_norm": 0.2903803822883755,
842
+ "learning_rate": 3.9798867363634814e-05,
843
+ "loss": 0.2123,
844
+ "step": 580
845
+ },
846
+ {
847
+ "epoch": 0.7414448669201521,
848
+ "grad_norm": 0.31710020807356026,
849
+ "learning_rate": 3.804717887351991e-05,
850
+ "loss": 0.2605,
851
+ "step": 585
852
+ },
853
+ {
854
+ "epoch": 0.7477820025348543,
855
+ "grad_norm": 0.24733722916292933,
856
+ "learning_rate": 3.632581298289427e-05,
857
+ "loss": 0.2289,
858
+ "step": 590
859
+ },
860
+ {
861
+ "epoch": 0.7541191381495564,
862
+ "grad_norm": 0.3001878742603873,
863
+ "learning_rate": 3.4635612208491194e-05,
864
+ "loss": 0.2322,
865
+ "step": 595
866
+ },
867
+ {
868
+ "epoch": 0.7604562737642585,
869
+ "grad_norm": 0.3250883573895976,
870
+ "learning_rate": 3.2977403813382926e-05,
871
+ "loss": 0.2525,
872
+ "step": 600
873
+ },
874
+ {
875
+ "epoch": 0.7604562737642585,
876
+ "eval_loss": 0.2325838953256607,
877
+ "eval_runtime": 1086.1174,
878
+ "eval_samples_per_second": 3.683,
879
+ "eval_steps_per_second": 0.115,
880
+ "step": 600
881
+ },
882
+ {
883
+ "epoch": 0.7667934093789607,
884
+ "grad_norm": 0.3246331902911399,
885
+ "learning_rate": 3.135199940207947e-05,
886
+ "loss": 0.2206,
887
+ "step": 605
888
+ },
889
+ {
890
+ "epoch": 0.7731305449936628,
891
+ "grad_norm": 0.3364965552008111,
892
+ "learning_rate": 2.976019452329153e-05,
893
+ "loss": 0.2566,
894
+ "step": 610
895
+ },
896
+ {
897
+ "epoch": 0.779467680608365,
898
+ "grad_norm": 0.22924356326780843,
899
+ "learning_rate": 2.8202768280551894e-05,
900
+ "loss": 0.2448,
901
+ "step": 615
902
+ },
903
+ {
904
+ "epoch": 0.7858048162230672,
905
+ "grad_norm": 0.3207071242715804,
906
+ "learning_rate": 2.6680482950885777e-05,
907
+ "loss": 0.2269,
908
+ "step": 620
909
+ },
910
+ {
911
+ "epoch": 0.7921419518377694,
912
+ "grad_norm": 0.37606836614121164,
913
+ "learning_rate": 2.5194083611716935e-05,
914
+ "loss": 0.27,
915
+ "step": 625
916
+ },
917
+ {
918
+ "epoch": 0.7984790874524715,
919
+ "grad_norm": 0.2830616844795413,
920
+ "learning_rate": 2.374429777619205e-05,
921
+ "loss": 0.1848,
922
+ "step": 630
923
+ },
924
+ {
925
+ "epoch": 0.8048162230671736,
926
+ "grad_norm": 0.3123281734582231,
927
+ "learning_rate": 2.2331835037101823e-05,
928
+ "loss": 0.2562,
929
+ "step": 635
930
+ },
931
+ {
932
+ "epoch": 0.8111533586818758,
933
+ "grad_norm": 0.27281906149266577,
934
+ "learning_rate": 2.0957386719573224e-05,
935
+ "loss": 0.2417,
936
+ "step": 640
937
+ },
938
+ {
939
+ "epoch": 0.8174904942965779,
940
+ "grad_norm": 0.3463842809980822,
941
+ "learning_rate": 1.962162554270267e-05,
942
+ "loss": 0.2108,
943
+ "step": 645
944
+ },
945
+ {
946
+ "epoch": 0.8238276299112801,
947
+ "grad_norm": 0.35511851586585724,
948
+ "learning_rate": 1.83252052902961e-05,
949
+ "loss": 0.2789,
950
+ "step": 650
951
+ },
952
+ {
953
+ "epoch": 0.8301647655259823,
954
+ "grad_norm": 0.2720460231999165,
955
+ "learning_rate": 1.7068760490876422e-05,
956
+ "loss": 0.1886,
957
+ "step": 655
958
+ },
959
+ {
960
+ "epoch": 0.8365019011406845,
961
+ "grad_norm": 0.323785370919597,
962
+ "learning_rate": 1.5852906107115893e-05,
963
+ "loss": 0.2617,
964
+ "step": 660
965
+ },
966
+ {
967
+ "epoch": 0.8428390367553865,
968
+ "grad_norm": 0.23954436901187928,
969
+ "learning_rate": 1.4678237234844649e-05,
970
+ "loss": 0.2501,
971
+ "step": 665
972
+ },
973
+ {
974
+ "epoch": 0.8491761723700887,
975
+ "grad_norm": 0.3169133202131331,
976
+ "learning_rate": 1.354532881178301e-05,
977
+ "loss": 0.1988,
978
+ "step": 670
979
+ },
980
+ {
981
+ "epoch": 0.8555133079847909,
982
+ "grad_norm": 0.3516256902668012,
983
+ "learning_rate": 1.2454735336140167e-05,
984
+ "loss": 0.2765,
985
+ "step": 675
986
+ },
987
+ {
988
+ "epoch": 0.861850443599493,
989
+ "grad_norm": 0.2923347791144997,
990
+ "learning_rate": 1.1406990595216971e-05,
991
+ "loss": 0.2065,
992
+ "step": 680
993
+ },
994
+ {
995
+ "epoch": 0.8681875792141952,
996
+ "grad_norm": 0.30823497046755327,
997
+ "learning_rate": 1.0402607404145449e-05,
998
+ "loss": 0.2623,
999
+ "step": 685
1000
+ },
1001
+ {
1002
+ "epoch": 0.8745247148288974,
1003
+ "grad_norm": 0.25292131222665615,
1004
+ "learning_rate": 9.442077354893198e-06,
1005
+ "loss": 0.2461,
1006
+ "step": 690
1007
+ },
1008
+ {
1009
+ "epoch": 0.8808618504435995,
1010
+ "grad_norm": 0.32414756933713984,
1011
+ "learning_rate": 8.525870575655392e-06,
1012
+ "loss": 0.217,
1013
+ "step": 695
1014
+ },
1015
+ {
1016
+ "epoch": 0.8871989860583016,
1017
+ "grad_norm": 0.3423100582401461,
1018
+ "learning_rate": 7.654435500752055e-06,
1019
+ "loss": 0.2675,
1020
+ "step": 700
1021
+ },
1022
+ {
1023
+ "epoch": 0.8935361216730038,
1024
+ "grad_norm": 0.29544160874192754,
1025
+ "learning_rate": 6.828198651143425e-06,
1026
+ "loss": 0.1927,
1027
+ "step": 705
1028
+ },
1029
+ {
1030
+ "epoch": 0.899873257287706,
1031
+ "grad_norm": 0.32477510117980407,
1032
+ "learning_rate": 6.047564425670749e-06,
1033
+ "loss": 0.2613,
1034
+ "step": 710
1035
+ },
1036
+ {
1037
+ "epoch": 0.9062103929024081,
1038
+ "grad_norm": 0.23676655623304554,
1039
+ "learning_rate": 5.312914903124566e-06,
1040
+ "loss": 0.2414,
1041
+ "step": 715
1042
+ },
1043
+ {
1044
+ "epoch": 0.9125475285171103,
1045
+ "grad_norm": 0.2967989522178623,
1046
+ "learning_rate": 4.624609655237544e-06,
1047
+ "loss": 0.2002,
1048
+ "step": 720
1049
+ },
1050
+ {
1051
+ "epoch": 0.9188846641318125,
1052
+ "grad_norm": 0.340334151463555,
1053
+ "learning_rate": 3.982985570693354e-06,
1054
+ "loss": 0.2764,
1055
+ "step": 725
1056
+ },
1057
+ {
1058
+ "epoch": 0.9252217997465145,
1059
+ "grad_norm": 0.2776495387493925,
1060
+ "learning_rate": 3.388356690237582e-06,
1061
+ "loss": 0.1859,
1062
+ "step": 730
1063
+ },
1064
+ {
1065
+ "epoch": 0.9315589353612167,
1066
+ "grad_norm": 0.37073108669893173,
1067
+ "learning_rate": 2.84101405297158e-06,
1068
+ "loss": 0.2378,
1069
+ "step": 735
1070
+ },
1071
+ {
1072
+ "epoch": 0.9378960709759189,
1073
+ "grad_norm": 0.2570912693772944,
1074
+ "learning_rate": 2.341225553904336e-06,
1075
+ "loss": 0.2398,
1076
+ "step": 740
1077
+ },
1078
+ {
1079
+ "epoch": 0.944233206590621,
1080
+ "grad_norm": 0.3306131510091784,
1081
+ "learning_rate": 1.8892358128322018e-06,
1082
+ "loss": 0.2092,
1083
+ "step": 745
1084
+ },
1085
+ {
1086
+ "epoch": 0.9505703422053232,
1087
+ "grad_norm": 0.33888565708232804,
1088
+ "learning_rate": 1.4852660546105234e-06,
1089
+ "loss": 0.2873,
1090
+ "step": 750
1091
+ },
1092
+ {
1093
+ "epoch": 0.9569074778200254,
1094
+ "grad_norm": 0.2906867586955816,
1095
+ "learning_rate": 1.1295140008758864e-06,
1096
+ "loss": 0.2079,
1097
+ "step": 755
1098
+ },
1099
+ {
1100
+ "epoch": 0.9632446134347274,
1101
+ "grad_norm": 0.34402820600782946,
1102
+ "learning_rate": 8.221537732719275e-07,
1103
+ "loss": 0.2648,
1104
+ "step": 760
1105
+ },
1106
+ {
1107
+ "epoch": 0.9695817490494296,
1108
+ "grad_norm": 0.26017431588284967,
1109
+ "learning_rate": 5.633358082260954e-07,
1110
+ "loss": 0.2481,
1111
+ "step": 765
1112
+ },
1113
+ {
1114
+ "epoch": 0.9759188846641318,
1115
+ "grad_norm": 0.3339069885340444,
1116
+ "learning_rate": 3.5318678331904833e-07,
1117
+ "loss": 0.2237,
1118
+ "step": 770
1119
+ },
1120
+ {
1121
+ "epoch": 0.982256020278834,
1122
+ "grad_norm": 0.34754105161997045,
1123
+ "learning_rate": 1.9180955528270706e-07,
1124
+ "loss": 0.27,
1125
+ "step": 775
1126
+ },
1127
+ {
1128
+ "epoch": 0.9885931558935361,
1129
+ "grad_norm": 0.3246226295094946,
1130
+ "learning_rate": 7.928310965742425e-08,
1131
+ "loss": 0.2226,
1132
+ "step": 780
1133
+ },
1134
+ {
1135
+ "epoch": 0.9949302915082383,
1136
+ "grad_norm": 0.31594487456225406,
1137
+ "learning_rate": 1.5662522132742218e-08,
1138
+ "loss": 0.2444,
1139
+ "step": 785
1140
+ },
1141
+ {
1142
+ "epoch": 1.0,
1143
+ "step": 789,
1144
+ "total_flos": 1.0371979095834624e+16,
1145
+ "train_loss": 0.2726454405050314,
1146
+ "train_runtime": 20804.2809,
1147
+ "train_samples_per_second": 1.213,
1148
+ "train_steps_per_second": 0.038
1149
+ }
1150
+ ],
1151
+ "logging_steps": 5,
1152
+ "max_steps": 789,
1153
+ "num_input_tokens_seen": 0,
1154
+ "num_train_epochs": 1,
1155
+ "save_steps": 200,
1156
+ "stateful_callbacks": {
1157
+ "TrainerControl": {
1158
+ "args": {
1159
+ "should_epoch_stop": false,
1160
+ "should_evaluate": false,
1161
+ "should_log": false,
1162
+ "should_save": true,
1163
+ "should_training_stop": false
1164
+ },
1165
+ "attributes": {}
1166
+ }
1167
+ },
1168
+ "total_flos": 1.0371979095834624e+16,
1169
+ "train_batch_size": 8,
1170
+ "trial_name": null,
1171
+ "trial_params": null
1172
+ }