NicholasCorrado commited on
Commit
d62f00a
1 Parent(s): 9a67f54

Model save

Browse files
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: mistralai/Mistral-7B-v0.3
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: mistral-7b-ift
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # mistral-7b-ift
20
+
21
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.0288
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 5e-07
43
+ - train_batch_size: 8
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 4
49
+ - total_train_batch_size: 256
50
+ - total_eval_batch_size: 64
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:-----:|:----:|:---------------:|
60
+ | 0.9524 | 1.0 | 461 | 1.0288 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - Transformers 4.44.1
66
+ - Pytorch 2.1.2+cu121
67
+ - Datasets 2.21.0
68
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 192943352709120.0,
4
+ "train_loss": 1.0081440616326323,
5
+ "train_runtime": 5423.7345,
6
+ "train_samples": 77486,
7
+ "train_samples_per_second": 21.752,
8
+ "train_steps_per_second": 0.085
9
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.44.1"
6
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 192943352709120.0,
4
+ "train_loss": 1.0081440616326323,
5
+ "train_runtime": 5423.7345,
6
+ "train_samples": 77486,
7
+ "train_samples_per_second": 21.752,
8
+ "train_steps_per_second": 0.085
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,701 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 461,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0021691973969631237,
13
+ "grad_norm": 11.479755487865775,
14
+ "learning_rate": 1.0638297872340425e-08,
15
+ "loss": 1.204,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.010845986984815618,
20
+ "grad_norm": 11.719854924785,
21
+ "learning_rate": 5.3191489361702123e-08,
22
+ "loss": 1.2705,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.021691973969631236,
27
+ "grad_norm": 12.5013105505897,
28
+ "learning_rate": 1.0638297872340425e-07,
29
+ "loss": 1.2569,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.03253796095444685,
34
+ "grad_norm": 10.664645875250606,
35
+ "learning_rate": 1.5957446808510638e-07,
36
+ "loss": 1.2504,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.04338394793926247,
41
+ "grad_norm": 9.933034533041308,
42
+ "learning_rate": 2.127659574468085e-07,
43
+ "loss": 1.2132,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.05422993492407809,
48
+ "grad_norm": 5.664415659701916,
49
+ "learning_rate": 2.659574468085106e-07,
50
+ "loss": 1.1718,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.0650759219088937,
55
+ "grad_norm": 4.799955074967526,
56
+ "learning_rate": 3.1914893617021275e-07,
57
+ "loss": 1.1492,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.07592190889370933,
62
+ "grad_norm": 3.5056771066313868,
63
+ "learning_rate": 3.7234042553191484e-07,
64
+ "loss": 1.1193,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.08676789587852494,
69
+ "grad_norm": 2.758357067802773,
70
+ "learning_rate": 4.25531914893617e-07,
71
+ "loss": 1.0995,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.09761388286334056,
76
+ "grad_norm": 2.512596535973654,
77
+ "learning_rate": 4.787234042553192e-07,
78
+ "loss": 1.0789,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.10845986984815618,
83
+ "grad_norm": 2.4219547091489244,
84
+ "learning_rate": 4.999352212103373e-07,
85
+ "loss": 1.0741,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.1193058568329718,
90
+ "grad_norm": 2.359757061351952,
91
+ "learning_rate": 4.995394723941067e-07,
92
+ "loss": 1.0585,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.1301518438177874,
97
+ "grad_norm": 2.4328040984593247,
98
+ "learning_rate": 4.98784531935359e-07,
99
+ "loss": 1.0568,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.14099783080260303,
104
+ "grad_norm": 2.3052389483909543,
105
+ "learning_rate": 4.976714865090826e-07,
106
+ "loss": 1.0638,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.15184381778741865,
111
+ "grad_norm": 2.198699363060523,
112
+ "learning_rate": 4.96201938253052e-07,
113
+ "loss": 1.0349,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.16268980477223427,
118
+ "grad_norm": 2.138037840029951,
119
+ "learning_rate": 4.943780024616802e-07,
120
+ "loss": 1.034,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.1735357917570499,
125
+ "grad_norm": 2.1989759782977076,
126
+ "learning_rate": 4.922023045412265e-07,
127
+ "loss": 1.0295,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.1843817787418655,
132
+ "grad_norm": 2.2941860079446235,
133
+ "learning_rate": 4.896779762307389e-07,
134
+ "loss": 1.0262,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.19522776572668113,
139
+ "grad_norm": 2.1481967771932773,
140
+ "learning_rate": 4.868086510941716e-07,
141
+ "loss": 1.0154,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.20607375271149675,
146
+ "grad_norm": 2.160671871794234,
147
+ "learning_rate": 4.835984592901677e-07,
148
+ "loss": 1.0468,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.21691973969631237,
153
+ "grad_norm": 2.1550419124695783,
154
+ "learning_rate": 4.800520216270341e-07,
155
+ "loss": 1.0336,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.227765726681128,
160
+ "grad_norm": 2.2253199590666157,
161
+ "learning_rate": 4.7617444291146555e-07,
162
+ "loss": 1.0224,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.2386117136659436,
167
+ "grad_norm": 2.183801883640947,
168
+ "learning_rate": 4.7197130460059377e-07,
169
+ "loss": 1.0139,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.24945770065075923,
174
+ "grad_norm": 2.177485956300299,
175
+ "learning_rate": 4.6744865676793666e-07,
176
+ "loss": 1.0053,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.2603036876355748,
181
+ "grad_norm": 2.1256298478006856,
182
+ "learning_rate": 4.6261300939481274e-07,
183
+ "loss": 0.9914,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.27114967462039047,
188
+ "grad_norm": 2.1075218318516886,
189
+ "learning_rate": 4.574713229997563e-07,
190
+ "loss": 1.0006,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.28199566160520606,
195
+ "grad_norm": 2.1304778495096457,
196
+ "learning_rate": 4.520309986194201e-07,
197
+ "loss": 1.0338,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.2928416485900217,
202
+ "grad_norm": 2.1650488920006046,
203
+ "learning_rate": 4.462998671553897e-07,
204
+ "loss": 1.0017,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.3036876355748373,
209
+ "grad_norm": 2.214789820606799,
210
+ "learning_rate": 4.4028617810224115e-07,
211
+ "loss": 1.0101,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.31453362255965295,
216
+ "grad_norm": 2.157917041478876,
217
+ "learning_rate": 4.3399858767306927e-07,
218
+ "loss": 0.989,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.32537960954446854,
223
+ "grad_norm": 2.092011368725107,
224
+ "learning_rate": 4.2744614633957723e-07,
225
+ "loss": 1.0115,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.3362255965292842,
230
+ "grad_norm": 2.219786897391391,
231
+ "learning_rate": 4.206382858046635e-07,
232
+ "loss": 0.9853,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.3470715835140998,
237
+ "grad_norm": 2.0979071528204756,
238
+ "learning_rate": 4.135848054262578e-07,
239
+ "loss": 1.0163,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.3579175704989154,
244
+ "grad_norm": 2.1698498942112288,
245
+ "learning_rate": 4.062958581119472e-07,
246
+ "loss": 0.9865,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.368763557483731,
251
+ "grad_norm": 2.132812386449239,
252
+ "learning_rate": 3.9878193570469743e-07,
253
+ "loss": 0.9871,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.3796095444685466,
258
+ "grad_norm": 2.337071820270912,
259
+ "learning_rate": 3.91053853880703e-07,
260
+ "loss": 0.984,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.39045553145336226,
265
+ "grad_norm": 2.1129666341045263,
266
+ "learning_rate": 3.831227365811074e-07,
267
+ "loss": 0.9943,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.40130151843817785,
272
+ "grad_norm": 2.121223339662348,
273
+ "learning_rate": 3.75e-07,
274
+ "loss": 0.9918,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.4121475054229935,
279
+ "grad_norm": 2.285212865590998,
280
+ "learning_rate": 3.6669733615173965e-07,
281
+ "loss": 0.9843,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.4229934924078091,
286
+ "grad_norm": 2.123810421477716,
287
+ "learning_rate": 3.5822669604125684e-07,
288
+ "loss": 0.9781,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.43383947939262474,
293
+ "grad_norm": 2.1627048249065997,
294
+ "learning_rate": 3.4960027246156036e-07,
295
+ "loss": 0.9781,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.44468546637744033,
300
+ "grad_norm": 2.137516375815561,
301
+ "learning_rate": 3.408304824432103e-07,
302
+ "loss": 0.9867,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.455531453362256,
307
+ "grad_norm": 2.058882247220635,
308
+ "learning_rate": 3.319299493810187e-07,
309
+ "loss": 0.9776,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.46637744034707157,
314
+ "grad_norm": 2.1433762495027864,
315
+ "learning_rate": 3.229114848637062e-07,
316
+ "loss": 0.9879,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.4772234273318872,
321
+ "grad_norm": 2.1877299182133734,
322
+ "learning_rate": 3.13788070232669e-07,
323
+ "loss": 0.9812,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.4880694143167028,
328
+ "grad_norm": 2.0613909390774525,
329
+ "learning_rate": 3.0457283789640036e-07,
330
+ "loss": 1.0005,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.49891540130151846,
335
+ "grad_norm": 2.137287960677187,
336
+ "learning_rate": 2.9527905242746395e-07,
337
+ "loss": 0.9994,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.5097613882863341,
342
+ "grad_norm": 2.062311669373047,
343
+ "learning_rate": 2.85920091469227e-07,
344
+ "loss": 0.9806,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.5206073752711496,
349
+ "grad_norm": 2.1009325069210054,
350
+ "learning_rate": 2.765094264798387e-07,
351
+ "loss": 0.9905,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.5314533622559653,
356
+ "grad_norm": 2.252996820494139,
357
+ "learning_rate": 2.6706060334116775e-07,
358
+ "loss": 0.9833,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.5422993492407809,
363
+ "grad_norm": 2.063111058570344,
364
+ "learning_rate": 2.575872228606156e-07,
365
+ "loss": 0.9466,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.5531453362255966,
370
+ "grad_norm": 2.1028294210091794,
371
+ "learning_rate": 2.4810292119386674e-07,
372
+ "loss": 0.9556,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.5639913232104121,
377
+ "grad_norm": 2.121237281068362,
378
+ "learning_rate": 2.3862135021675915e-07,
379
+ "loss": 0.9717,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.5748373101952278,
384
+ "grad_norm": 2.2205226946736905,
385
+ "learning_rate": 2.2915615787452664e-07,
386
+ "loss": 0.9638,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.5856832971800434,
391
+ "grad_norm": 2.0940361008010333,
392
+ "learning_rate": 2.1972096853669903e-07,
393
+ "loss": 0.9671,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.596529284164859,
398
+ "grad_norm": 2.0586787374748576,
399
+ "learning_rate": 2.1032936338593717e-07,
400
+ "loss": 0.9773,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.6073752711496746,
405
+ "grad_norm": 2.072790057342988,
406
+ "learning_rate": 2.0099486086903294e-07,
407
+ "loss": 0.9684,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.6182212581344902,
412
+ "grad_norm": 2.1778824831003147,
413
+ "learning_rate": 1.9173089723821087e-07,
414
+ "loss": 0.9667,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.6290672451193059,
419
+ "grad_norm": 2.115203742538656,
420
+ "learning_rate": 1.825508072107439e-07,
421
+ "loss": 0.9919,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.6399132321041214,
426
+ "grad_norm": 2.9734190506035536,
427
+ "learning_rate": 1.7346780477471897e-07,
428
+ "loss": 0.9741,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.6507592190889371,
433
+ "grad_norm": 2.109405285224148,
434
+ "learning_rate": 1.6449496416858282e-07,
435
+ "loss": 0.9815,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.6616052060737527,
440
+ "grad_norm": 2.0612586919012355,
441
+ "learning_rate": 1.5564520106184643e-07,
442
+ "loss": 0.9637,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.6724511930585684,
447
+ "grad_norm": 2.126282772097656,
448
+ "learning_rate": 1.4693125396403562e-07,
449
+ "loss": 0.9576,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.6832971800433839,
454
+ "grad_norm": 2.151977422363772,
455
+ "learning_rate": 1.3836566588865e-07,
456
+ "loss": 0.9871,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.6941431670281996,
461
+ "grad_norm": 2.10699457923422,
462
+ "learning_rate": 1.2996076629852112e-07,
463
+ "loss": 0.9775,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.7049891540130152,
468
+ "grad_norm": 2.2306202203282366,
469
+ "learning_rate": 1.2172865335856064e-07,
470
+ "loss": 0.9869,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.7158351409978309,
475
+ "grad_norm": 2.1309989319293696,
476
+ "learning_rate": 1.1368117652144185e-07,
477
+ "loss": 0.9871,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.7266811279826464,
482
+ "grad_norm": 2.1871677586329175,
483
+ "learning_rate": 1.0582991947128323e-07,
484
+ "loss": 0.983,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.737527114967462,
489
+ "grad_norm": 2.089837928751529,
490
+ "learning_rate": 9.818618344988258e-08,
491
+ "loss": 0.9817,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.7483731019522777,
496
+ "grad_norm": 2.0529724920050905,
497
+ "learning_rate": 9.076097098950541e-08,
498
+ "loss": 0.9581,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.7592190889370932,
503
+ "grad_norm": 2.077217246150324,
504
+ "learning_rate": 8.356497007563986e-08,
505
+ "loss": 0.9845,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.7700650759219089,
510
+ "grad_norm": 2.1176827367409876,
511
+ "learning_rate": 7.660853876251683e-08,
512
+ "loss": 0.9625,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.7809110629067245,
517
+ "grad_norm": 2.1409566690457433,
518
+ "learning_rate": 6.990169026353867e-08,
519
+ "loss": 0.9999,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.7917570498915402,
524
+ "grad_norm": 2.0254717410055703,
525
+ "learning_rate": 6.345407853807863e-08,
526
+ "loss": 0.9642,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.8026030368763557,
531
+ "grad_norm": 2.140899328755212,
532
+ "learning_rate": 5.727498439539602e-08,
533
+ "loss": 0.9717,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.8134490238611713,
538
+ "grad_norm": 2.0666287819389404,
539
+ "learning_rate": 5.13733021356714e-08,
540
+ "loss": 0.9697,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.824295010845987,
545
+ "grad_norm": 2.0741290971837127,
546
+ "learning_rate": 4.57575267473895e-08,
547
+ "loss": 0.9641,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.8351409978308026,
552
+ "grad_norm": 2.108990519804528,
553
+ "learning_rate": 4.043574167949892e-08,
554
+ "loss": 0.9902,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.8459869848156182,
559
+ "grad_norm": 2.094620559269544,
560
+ "learning_rate": 3.541560720594869e-08,
561
+ "loss": 0.9926,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.8568329718004338,
566
+ "grad_norm": 2.0594257251140586,
567
+ "learning_rate": 3.0704349399351435e-08,
568
+ "loss": 0.9614,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.8676789587852495,
573
+ "grad_norm": 2.1291510379356753,
574
+ "learning_rate": 2.6308749729643058e-08,
575
+ "loss": 0.9651,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.8785249457700651,
580
+ "grad_norm": 2.1035620640567734,
581
+ "learning_rate": 2.2235135302712092e-08,
582
+ "loss": 0.9952,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.8893709327548807,
587
+ "grad_norm": 2.0424482935650614,
588
+ "learning_rate": 1.8489369753048682e-08,
589
+ "loss": 0.9631,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.9002169197396963,
594
+ "grad_norm": 2.0939056533766847,
595
+ "learning_rate": 1.507684480352292e-08,
596
+ "loss": 0.9747,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.911062906724512,
601
+ "grad_norm": 4.429046210547233,
602
+ "learning_rate": 1.2002472504440807e-08,
603
+ "loss": 0.9843,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.9219088937093276,
608
+ "grad_norm": 2.1047314550346297,
609
+ "learning_rate": 9.270678163050217e-09,
610
+ "loss": 0.9738,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.9327548806941431,
615
+ "grad_norm": 2.0998363541587888,
616
+ "learning_rate": 6.885393973673298e-09,
617
+ "loss": 0.9509,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.9436008676789588,
622
+ "grad_norm": 2.138035512894866,
623
+ "learning_rate": 4.850053357634693e-09,
624
+ "loss": 0.9799,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.9544468546637744,
629
+ "grad_norm": 2.053994471901985,
630
+ "learning_rate": 3.1675860211325954e-09,
631
+ "loss": 0.9762,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.96529284164859,
636
+ "grad_norm": 2.054984790757136,
637
+ "learning_rate": 1.840413738166402e-09,
638
+ "loss": 0.9499,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.9761388286334056,
643
+ "grad_norm": 2.0710147404144097,
644
+ "learning_rate": 8.704468645914787e-10,
645
+ "loss": 0.9692,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.9869848156182213,
650
+ "grad_norm": 2.067619347023652,
651
+ "learning_rate": 2.5908158831811077e-10,
652
+ "loss": 0.9994,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.9978308026030369,
657
+ "grad_norm": 2.095822455094207,
658
+ "learning_rate": 7.197919613455284e-12,
659
+ "loss": 0.9524,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 1.0,
664
+ "eval_loss": 1.0288481712341309,
665
+ "eval_runtime": 3.4782,
666
+ "eval_samples_per_second": 74.465,
667
+ "eval_steps_per_second": 1.438,
668
+ "step": 461
669
+ },
670
+ {
671
+ "epoch": 1.0,
672
+ "step": 461,
673
+ "total_flos": 192943352709120.0,
674
+ "train_loss": 1.0081440616326323,
675
+ "train_runtime": 5423.7345,
676
+ "train_samples_per_second": 21.752,
677
+ "train_steps_per_second": 0.085
678
+ }
679
+ ],
680
+ "logging_steps": 5,
681
+ "max_steps": 461,
682
+ "num_input_tokens_seen": 0,
683
+ "num_train_epochs": 1,
684
+ "save_steps": 100,
685
+ "stateful_callbacks": {
686
+ "TrainerControl": {
687
+ "args": {
688
+ "should_epoch_stop": false,
689
+ "should_evaluate": false,
690
+ "should_log": false,
691
+ "should_save": true,
692
+ "should_training_stop": true
693
+ },
694
+ "attributes": {}
695
+ }
696
+ },
697
+ "total_flos": 192943352709120.0,
698
+ "train_batch_size": 8,
699
+ "trial_name": null,
700
+ "trial_params": null
701
+ }