lewtun HF staff commited on
Commit
8104d49
1 Parent(s): be993f0

Model save

Browse files
README.md CHANGED
@@ -2,13 +2,11 @@
2
  license: apache-2.0
3
  base_model: mistralai/Mistral-7B-v0.1
4
  tags:
5
- - alignment-handbook
6
- - generated_from_trainer
7
  - trl
8
  - sft
9
  - generated_from_trainer
10
  datasets:
11
- - HuggingFaceH4/ultrachat_200k
12
  model-index:
13
  - name: zephyr-7b-sft-full
14
  results: []
@@ -19,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  # zephyr-7b-sft-full
21
 
22
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the HuggingFaceH4/ultrachat_200k dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.9354
25
 
26
  ## Model description
27
 
@@ -57,7 +55,7 @@ The following hyperparameters were used during training:
57
 
58
  | Training Loss | Epoch | Step | Validation Loss |
59
  |:-------------:|:-----:|:----:|:---------------:|
60
- | 0.9077 | 1.0 | 1090 | 0.9354 |
61
 
62
 
63
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: mistralai/Mistral-7B-v0.1
4
  tags:
 
 
5
  - trl
6
  - sft
7
  - generated_from_trainer
8
  datasets:
9
+ - generator
10
  model-index:
11
  - name: zephyr-7b-sft-full
12
  results: []
 
17
 
18
  # zephyr-7b-sft-full
19
 
20
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.9353
23
 
24
  ## Model description
25
 
 
55
 
56
  | Training Loss | Epoch | Step | Validation Loss |
57
  |:-------------:|:-----:|:----:|:---------------:|
58
+ | 0.9075 | 1.0 | 1090 | 0.9353 |
59
 
60
 
61
  ### Framework versions
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
  "epoch": 1.0,
3
- "eval_loss": 0.9353834986686707,
4
- "eval_runtime": 143.2861,
5
  "eval_samples": 23110,
6
- "eval_samples_per_second": 107.694,
7
- "eval_steps_per_second": 1.689,
8
- "train_loss": 0.9776670060026537,
9
- "train_runtime": 5614.3798,
10
  "train_samples": 207865,
11
- "train_samples_per_second": 24.834,
12
  "train_steps_per_second": 0.194
13
  }
 
1
  {
2
  "epoch": 1.0,
3
+ "eval_loss": 0.9353461265563965,
4
+ "eval_runtime": 141.7596,
5
  "eval_samples": 23110,
6
+ "eval_samples_per_second": 108.853,
7
+ "eval_steps_per_second": 1.707,
8
+ "train_loss": 0.9776473476252424,
9
+ "train_runtime": 5608.2856,
10
  "train_samples": 207865,
11
+ "train_samples_per_second": 24.861,
12
  "train_steps_per_second": 0.194
13
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 1.0,
3
- "eval_loss": 0.9353834986686707,
4
- "eval_runtime": 143.2861,
5
  "eval_samples": 23110,
6
- "eval_samples_per_second": 107.694,
7
- "eval_steps_per_second": 1.689
8
  }
 
1
  {
2
  "epoch": 1.0,
3
+ "eval_loss": 0.9353461265563965,
4
+ "eval_runtime": 141.7596,
5
  "eval_samples": 23110,
6
+ "eval_samples_per_second": 108.853,
7
+ "eval_steps_per_second": 1.707
8
  }
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8f0995a437e42c9046c3964bb7d858c34866d6b4f1de4ef12f98129888ee3fc
3
  size 4943162336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab3121e8d66828f8888b456267d5ce7e87063138a04ddde287016df065852277
3
  size 4943162336
model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3b789d79949a4f5de8001f9e01a619675b6d2675041e49fb912ebe6c8e7d5b52
3
  size 4999819336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fa2d53c86d1b508a7440b1a6aa57652ac00fe7307f67b3b184cfbe2e84d8d55
3
  size 4999819336
model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4d4b611a06358f2462ff5cc487cd28bd2de498fde53ba13e219b972869ebe1d5
3
  size 4540516344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5a120540ca1ffa8b3b7e66aa3a45760092305697b9e0b23a2399ab56b3f0771
3
  size 4540516344
runs/Jan10_01-40-06_ip-26-0-168-34/events.out.tfevents.1704850836.ip-26-0-168-34.1133327.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fd358ecb98e337e7bdb0fadcbeedbec16960ffebf7e300cd81091cb11328b54b
3
- size 35928
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd7f25fbf29b87da1eb52c6961c9817d4df7207797ebcaa5e3a8d23380bce5d5
3
+ size 39379
runs/Jan10_01-40-06_ip-26-0-168-34/events.out.tfevents.1704856586.ip-26-0-168-34.1133327.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98419bcde11bdd300eba86cae7c9590c8fa1aa6a585ea4be1e4772dafbb18fc6
3
+ size 359
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 1.0,
3
- "train_loss": 0.9776670060026537,
4
- "train_runtime": 5614.3798,
5
  "train_samples": 207865,
6
- "train_samples_per_second": 24.834,
7
  "train_steps_per_second": 0.194
8
  }
 
1
  {
2
  "epoch": 1.0,
3
+ "train_loss": 0.9776473476252424,
4
+ "train_runtime": 5608.2856,
5
  "train_samples": 207865,
6
+ "train_samples_per_second": 24.861,
7
  "train_steps_per_second": 0.194
8
  }
trainer_state.json CHANGED
@@ -23,7 +23,7 @@
23
  {
24
  "epoch": 0.01,
25
  "learning_rate": 1.8348623853211011e-06,
26
- "loss": 1.3918,
27
  "step": 10
28
  },
29
  {
@@ -41,13 +41,13 @@
41
  {
42
  "epoch": 0.02,
43
  "learning_rate": 4.587155963302753e-06,
44
- "loss": 1.1173,
45
  "step": 25
46
  },
47
  {
48
  "epoch": 0.03,
49
  "learning_rate": 5.504587155963303e-06,
50
- "loss": 1.0705,
51
  "step": 30
52
  },
53
  {
@@ -59,7 +59,7 @@
59
  {
60
  "epoch": 0.04,
61
  "learning_rate": 7.3394495412844045e-06,
62
- "loss": 1.0521,
63
  "step": 40
64
  },
65
  {
@@ -71,49 +71,49 @@
71
  {
72
  "epoch": 0.05,
73
  "learning_rate": 9.174311926605506e-06,
74
- "loss": 1.0309,
75
  "step": 50
76
  },
77
  {
78
  "epoch": 0.05,
79
  "learning_rate": 1.0091743119266055e-05,
80
- "loss": 1.0092,
81
  "step": 55
82
  },
83
  {
84
  "epoch": 0.06,
85
  "learning_rate": 1.1009174311926607e-05,
86
- "loss": 1.0359,
87
  "step": 60
88
  },
89
  {
90
  "epoch": 0.06,
91
  "learning_rate": 1.1926605504587156e-05,
92
- "loss": 1.0081,
93
  "step": 65
94
  },
95
  {
96
  "epoch": 0.06,
97
  "learning_rate": 1.2844036697247708e-05,
98
- "loss": 1.0002,
99
  "step": 70
100
  },
101
  {
102
  "epoch": 0.07,
103
  "learning_rate": 1.3761467889908258e-05,
104
- "loss": 1.0003,
105
  "step": 75
106
  },
107
  {
108
  "epoch": 0.07,
109
  "learning_rate": 1.4678899082568809e-05,
110
- "loss": 0.992,
111
  "step": 80
112
  },
113
  {
114
  "epoch": 0.08,
115
  "learning_rate": 1.559633027522936e-05,
116
- "loss": 1.0151,
117
  "step": 85
118
  },
119
  {
@@ -131,7 +131,7 @@
131
  {
132
  "epoch": 0.09,
133
  "learning_rate": 1.834862385321101e-05,
134
- "loss": 1.0162,
135
  "step": 100
136
  },
137
  {
@@ -143,13 +143,13 @@
143
  {
144
  "epoch": 0.1,
145
  "learning_rate": 1.999994872196626e-05,
146
- "loss": 1.0054,
147
  "step": 110
148
  },
149
  {
150
  "epoch": 0.11,
151
  "learning_rate": 1.9998154046002822e-05,
152
- "loss": 1.014,
153
  "step": 115
154
  },
155
  {
@@ -161,37 +161,37 @@
161
  {
162
  "epoch": 0.11,
163
  "learning_rate": 1.9986875683942535e-05,
164
- "loss": 1.0201,
165
  "step": 125
166
  },
167
  {
168
  "epoch": 0.12,
169
  "learning_rate": 1.9977394889447526e-05,
170
- "loss": 1.0236,
171
  "step": 130
172
  },
173
  {
174
  "epoch": 0.12,
175
  "learning_rate": 1.9965356041462954e-05,
176
- "loss": 1.0303,
177
  "step": 135
178
  },
179
  {
180
  "epoch": 0.13,
181
  "learning_rate": 1.9950762226567783e-05,
182
- "loss": 1.0282,
183
  "step": 140
184
  },
185
  {
186
  "epoch": 0.13,
187
  "learning_rate": 1.9933617186395917e-05,
188
- "loss": 1.0044,
189
  "step": 145
190
  },
191
  {
192
  "epoch": 0.14,
193
  "learning_rate": 1.9913925316676946e-05,
194
- "loss": 1.0025,
195
  "step": 150
196
  },
197
  {
@@ -209,97 +209,97 @@
209
  {
210
  "epoch": 0.15,
211
  "learning_rate": 1.9839622474129595e-05,
212
- "loss": 1.0161,
213
  "step": 165
214
  },
215
  {
216
  "epoch": 0.16,
217
  "learning_rate": 1.9809800282473014e-05,
218
- "loss": 0.9981,
219
  "step": 170
220
  },
221
  {
222
  "epoch": 0.16,
223
  "learning_rate": 1.977746300605507e-05,
224
- "loss": 1.0223,
225
  "step": 175
226
  },
227
  {
228
  "epoch": 0.17,
229
  "learning_rate": 1.9742618935665478e-05,
230
- "loss": 1.0088,
231
  "step": 180
232
  },
233
  {
234
  "epoch": 0.17,
235
  "learning_rate": 1.9705277004798072e-05,
236
- "loss": 0.9986,
237
  "step": 185
238
  },
239
  {
240
  "epoch": 0.17,
241
  "learning_rate": 1.9665446787360444e-05,
242
- "loss": 1.0107,
243
  "step": 190
244
  },
245
  {
246
  "epoch": 0.18,
247
  "learning_rate": 1.9623138495219292e-05,
248
- "loss": 1.0088,
249
  "step": 195
250
  },
251
  {
252
  "epoch": 0.18,
253
  "learning_rate": 1.957836297558229e-05,
254
- "loss": 1.0274,
255
  "step": 200
256
  },
257
  {
258
  "epoch": 0.19,
259
  "learning_rate": 1.9531131708217005e-05,
260
- "loss": 0.9892,
261
  "step": 205
262
  },
263
  {
264
  "epoch": 0.19,
265
  "learning_rate": 1.948145680250766e-05,
266
- "loss": 1.0117,
267
  "step": 210
268
  },
269
  {
270
  "epoch": 0.2,
271
  "learning_rate": 1.9429350994350483e-05,
272
- "loss": 0.9933,
273
  "step": 215
274
  },
275
  {
276
  "epoch": 0.2,
277
  "learning_rate": 1.93748276428884e-05,
278
- "loss": 1.0232,
279
  "step": 220
280
  },
281
  {
282
  "epoch": 0.21,
283
  "learning_rate": 1.931790072708596e-05,
284
- "loss": 1.0189,
285
  "step": 225
286
  },
287
  {
288
  "epoch": 0.21,
289
  "learning_rate": 1.9258584842145342e-05,
290
- "loss": 1.0018,
291
  "step": 230
292
  },
293
  {
294
  "epoch": 0.22,
295
  "learning_rate": 1.9196895195764363e-05,
296
- "loss": 0.9957,
297
  "step": 235
298
  },
299
  {
300
  "epoch": 0.22,
301
  "learning_rate": 1.913284760423745e-05,
302
- "loss": 1.0213,
303
  "step": 240
304
  },
305
  {
@@ -311,13 +311,13 @@
311
  {
312
  "epoch": 0.23,
313
  "learning_rate": 1.8997744869421248e-05,
314
- "loss": 1.0037,
315
  "step": 250
316
  },
317
  {
318
  "epoch": 0.23,
319
  "learning_rate": 1.8926724364434447e-05,
320
- "loss": 0.9953,
321
  "step": 255
322
  },
323
  {
@@ -329,7 +329,7 @@
329
  {
330
  "epoch": 0.24,
331
  "learning_rate": 1.8777836117563894e-05,
332
- "loss": 1.0198,
333
  "step": 265
334
  },
335
  {
@@ -341,13 +341,13 @@
341
  {
342
  "epoch": 0.25,
343
  "learning_rate": 1.861994642880105e-05,
344
- "loss": 1.0035,
345
  "step": 275
346
  },
347
  {
348
  "epoch": 0.26,
349
  "learning_rate": 1.8537676285033886e-05,
350
- "loss": 1.0062,
351
  "step": 280
352
  },
353
  {
@@ -359,13 +359,13 @@
359
  {
360
  "epoch": 0.27,
361
  "learning_rate": 1.8366590857437182e-05,
362
- "loss": 1.015,
363
  "step": 290
364
  },
365
  {
366
  "epoch": 0.27,
367
  "learning_rate": 1.8277819437329577e-05,
368
- "loss": 0.9928,
369
  "step": 295
370
  },
371
  {
@@ -377,61 +377,61 @@
377
  {
378
  "epoch": 0.28,
379
  "learning_rate": 1.809393297692334e-05,
380
- "loss": 0.9891,
381
  "step": 305
382
  },
383
  {
384
  "epoch": 0.28,
385
  "learning_rate": 1.799886508233829e-05,
386
- "loss": 0.9866,
387
  "step": 310
388
  },
389
  {
390
  "epoch": 0.29,
391
  "learning_rate": 1.790174639944997e-05,
392
- "loss": 0.9891,
393
  "step": 315
394
  },
395
  {
396
  "epoch": 0.29,
397
  "learning_rate": 1.780260182802314e-05,
398
- "loss": 0.998,
399
  "step": 320
400
  },
401
  {
402
  "epoch": 0.3,
403
  "learning_rate": 1.7701456787229805e-05,
404
- "loss": 0.9875,
405
  "step": 325
406
  },
407
  {
408
  "epoch": 0.3,
409
  "learning_rate": 1.7598337209132142e-05,
410
- "loss": 1.01,
411
  "step": 330
412
  },
413
  {
414
  "epoch": 0.31,
415
  "learning_rate": 1.7493269532033882e-05,
416
- "loss": 0.9939,
417
  "step": 335
418
  },
419
  {
420
  "epoch": 0.31,
421
  "learning_rate": 1.738628069370195e-05,
422
- "loss": 0.9979,
423
  "step": 340
424
  },
425
  {
426
  "epoch": 0.32,
427
  "learning_rate": 1.7277398124460022e-05,
428
- "loss": 1.0069,
429
  "step": 345
430
  },
431
  {
432
  "epoch": 0.32,
433
  "learning_rate": 1.71666497401558e-05,
434
- "loss": 0.9926,
435
  "step": 350
436
  },
437
  {
@@ -443,7 +443,7 @@
443
  {
444
  "epoch": 0.33,
445
  "learning_rate": 1.6939669574305565e-05,
446
- "loss": 1.0071,
447
  "step": 360
448
  },
449
  {
@@ -455,25 +455,25 @@
455
  {
456
  "epoch": 0.34,
457
  "learning_rate": 1.6705572958388576e-05,
458
- "loss": 0.999,
459
  "step": 370
460
  },
461
  {
462
  "epoch": 0.34,
463
  "learning_rate": 1.6585930722009602e-05,
464
- "loss": 0.9953,
465
  "step": 375
466
  },
467
  {
468
  "epoch": 0.35,
469
  "learning_rate": 1.6464599952375998e-05,
470
- "loss": 0.9973,
471
  "step": 380
472
  },
473
  {
474
  "epoch": 0.35,
475
  "learning_rate": 1.63416117568662e-05,
476
- "loss": 0.9813,
477
  "step": 385
478
  },
479
  {
@@ -485,31 +485,31 @@
485
  {
486
  "epoch": 0.36,
487
  "learning_rate": 1.6090789634342278e-05,
488
- "loss": 1.003,
489
  "step": 395
490
  },
491
  {
492
  "epoch": 0.37,
493
  "learning_rate": 1.5963020014335437e-05,
494
- "loss": 0.9932,
495
  "step": 400
496
  },
497
  {
498
  "epoch": 0.37,
499
  "learning_rate": 1.583372156597961e-05,
500
- "loss": 0.9754,
501
  "step": 405
502
  },
503
  {
504
  "epoch": 0.38,
505
  "learning_rate": 1.570292743944583e-05,
506
- "loss": 1.0097,
507
  "step": 410
508
  },
509
  {
510
  "epoch": 0.38,
511
  "learning_rate": 1.557067116837444e-05,
512
- "loss": 0.9678,
513
  "step": 415
514
  },
515
  {
@@ -521,43 +521,43 @@
521
  {
522
  "epoch": 0.39,
523
  "learning_rate": 1.530190819284555e-05,
524
- "loss": 1.0128,
525
  "step": 425
526
  },
527
  {
528
  "epoch": 0.39,
529
  "learning_rate": 1.5165470395159314e-05,
530
- "loss": 0.9882,
531
  "step": 430
532
  },
533
  {
534
  "epoch": 0.4,
535
  "learning_rate": 1.5027708248811331e-05,
536
- "loss": 0.9888,
537
  "step": 435
538
  },
539
  {
540
  "epoch": 0.4,
541
  "learning_rate": 1.4888657073937077e-05,
542
- "loss": 0.979,
543
  "step": 440
544
  },
545
  {
546
  "epoch": 0.41,
547
  "learning_rate": 1.4748352521159492e-05,
548
- "loss": 0.9767,
549
  "step": 445
550
  },
551
  {
552
  "epoch": 0.41,
553
  "learning_rate": 1.4606830562448692e-05,
554
- "loss": 0.9875,
555
  "step": 450
556
  },
557
  {
558
  "epoch": 0.42,
559
  "learning_rate": 1.4464127481899312e-05,
560
- "loss": 0.9999,
561
  "step": 455
562
  },
563
  {
@@ -581,13 +581,13 @@
581
  {
582
  "epoch": 0.44,
583
  "learning_rate": 1.3882240024460928e-05,
584
- "loss": 0.9861,
585
  "step": 475
586
  },
587
  {
588
  "epoch": 0.44,
589
  "learning_rate": 1.3734185865027061e-05,
590
- "loss": 0.9833,
591
  "step": 480
592
  },
593
  {
@@ -599,31 +599,31 @@
599
  {
600
  "epoch": 0.45,
601
  "learning_rate": 1.3435243583752294e-05,
602
- "loss": 0.9752,
603
  "step": 490
604
  },
605
  {
606
  "epoch": 0.45,
607
  "learning_rate": 1.3284432106201233e-05,
608
- "loss": 0.9796,
609
  "step": 495
610
  },
611
  {
612
  "epoch": 0.46,
613
  "learning_rate": 1.313277854982062e-05,
614
- "loss": 0.9584,
615
  "step": 500
616
  },
617
  {
618
  "epoch": 0.46,
619
  "learning_rate": 1.2980321796293838e-05,
620
- "loss": 0.9884,
621
  "step": 505
622
  },
623
  {
624
  "epoch": 0.47,
625
  "learning_rate": 1.2827100933231904e-05,
626
- "loss": 0.9695,
627
  "step": 510
628
  },
629
  {
@@ -641,7 +641,7 @@
641
  {
642
  "epoch": 0.48,
643
  "learning_rate": 1.2363247441059775e-05,
644
- "loss": 0.971,
645
  "step": 525
646
  },
647
  {
@@ -659,7 +659,7 @@
659
  {
660
  "epoch": 0.5,
661
  "learning_rate": 1.189394177207125e-05,
662
- "loss": 0.9825,
663
  "step": 540
664
  },
665
  {
@@ -671,13 +671,13 @@
671
  {
672
  "epoch": 0.5,
673
  "learning_rate": 1.1578576573543541e-05,
674
- "loss": 0.9612,
675
  "step": 550
676
  },
677
  {
678
  "epoch": 0.51,
679
  "learning_rate": 1.1420266647205232e-05,
680
- "loss": 0.9692,
681
  "step": 555
682
  },
683
  {
@@ -689,13 +689,13 @@
689
  {
690
  "epoch": 0.52,
691
  "learning_rate": 1.1102595071354471e-05,
692
- "loss": 0.9446,
693
  "step": 565
694
  },
695
  {
696
  "epoch": 0.52,
697
  "learning_rate": 1.0943314868040365e-05,
698
- "loss": 0.9582,
699
  "step": 570
700
  },
701
  {
@@ -713,13 +713,13 @@
713
  {
714
  "epoch": 0.54,
715
  "learning_rate": 1.0464186795625481e-05,
716
- "loss": 0.9636,
717
  "step": 585
718
  },
719
  {
720
  "epoch": 0.54,
721
  "learning_rate": 1.0304184775393642e-05,
722
- "loss": 0.9571,
723
  "step": 590
724
  },
725
  {
@@ -737,25 +737,25 @@
737
  {
738
  "epoch": 0.56,
739
  "learning_rate": 9.823874962247565e-06,
740
- "loss": 0.9651,
741
  "step": 605
742
  },
743
  {
744
  "epoch": 0.56,
745
  "learning_rate": 9.663807268427197e-06,
746
- "loss": 0.9511,
747
  "step": 610
748
  },
749
  {
750
  "epoch": 0.56,
751
  "learning_rate": 9.503825769350016e-06,
752
- "loss": 0.9731,
753
  "step": 615
754
  },
755
  {
756
  "epoch": 0.57,
757
  "learning_rate": 9.343971481858246e-06,
758
- "loss": 0.9487,
759
  "step": 620
760
  },
761
  {
@@ -767,7 +767,7 @@
767
  {
768
  "epoch": 0.58,
769
  "learning_rate": 9.024808435416435e-06,
770
- "loss": 0.9431,
771
  "step": 630
772
  },
773
  {
@@ -779,19 +779,19 @@
779
  {
780
  "epoch": 0.59,
781
  "learning_rate": 8.706645422477739e-06,
782
- "loss": 0.9618,
783
  "step": 640
784
  },
785
  {
786
  "epoch": 0.59,
787
  "learning_rate": 8.548040936496989e-06,
788
- "loss": 0.9452,
789
  "step": 645
790
  },
791
  {
792
  "epoch": 0.6,
793
  "learning_rate": 8.389808710909881e-06,
794
- "loss": 0.9739,
795
  "step": 650
796
  },
797
  {
@@ -809,25 +809,25 @@
809
  {
810
  "epoch": 0.61,
811
  "learning_rate": 7.917750740470116e-06,
812
- "loss": 0.9695,
813
  "step": 665
814
  },
815
  {
816
  "epoch": 0.61,
817
  "learning_rate": 7.761412129735853e-06,
818
- "loss": 0.9626,
819
  "step": 670
820
  },
821
  {
822
  "epoch": 0.62,
823
  "learning_rate": 7.605647459150961e-06,
824
- "loss": 0.9482,
825
  "step": 675
826
  },
827
  {
828
  "epoch": 0.62,
829
  "learning_rate": 7.4504966644264775e-06,
830
- "loss": 0.9482,
831
  "step": 680
832
  },
833
  {
@@ -839,7 +839,7 @@
839
  {
840
  "epoch": 0.63,
841
  "learning_rate": 7.142195648261747e-06,
842
- "loss": 0.9646,
843
  "step": 690
844
  },
845
  {
@@ -851,31 +851,31 @@
851
  {
852
  "epoch": 0.64,
853
  "learning_rate": 6.83682523588902e-06,
854
- "loss": 0.9436,
855
  "step": 700
856
  },
857
  {
858
  "epoch": 0.65,
859
  "learning_rate": 6.685336991505122e-06,
860
- "loss": 0.9576,
861
  "step": 705
862
  },
863
  {
864
  "epoch": 0.65,
865
  "learning_rate": 6.5346985766979384e-06,
866
- "loss": 0.9497,
867
  "step": 710
868
  },
869
  {
870
  "epoch": 0.66,
871
  "learning_rate": 6.384948612883872e-06,
872
- "loss": 0.9474,
873
  "step": 715
874
  },
875
  {
876
  "epoch": 0.66,
877
  "learning_rate": 6.2361254936939e-06,
878
- "loss": 0.9382,
879
  "step": 720
880
  },
881
  {
@@ -893,31 +893,31 @@
893
  {
894
  "epoch": 0.67,
895
  "learning_rate": 5.79559751711138e-06,
896
- "loss": 0.9349,
897
  "step": 735
898
  },
899
  {
900
  "epoch": 0.68,
901
  "learning_rate": 5.650860813791786e-06,
902
- "loss": 0.9444,
903
  "step": 740
904
  },
905
  {
906
  "epoch": 0.68,
907
  "learning_rate": 5.507239164130501e-06,
908
- "loss": 0.9522,
909
  "step": 745
910
  },
911
  {
912
  "epoch": 0.69,
913
  "learning_rate": 5.364769390551225e-06,
914
- "loss": 0.9507,
915
  "step": 750
916
  },
917
  {
918
  "epoch": 0.69,
919
  "learning_rate": 5.223488020154028e-06,
920
- "loss": 0.9461,
921
  "step": 755
922
  },
923
  {
@@ -935,13 +935,13 @@
935
  {
936
  "epoch": 0.71,
937
  "learning_rate": 4.807134973085036e-06,
938
- "loss": 0.9437,
939
  "step": 770
940
  },
941
  {
942
  "epoch": 0.71,
943
  "learning_rate": 4.670966253826027e-06,
944
- "loss": 0.9515,
945
  "step": 775
946
  },
947
  {
@@ -953,25 +953,25 @@
953
  {
954
  "epoch": 0.72,
955
  "learning_rate": 4.402762228128531e-06,
956
- "loss": 0.9412,
957
  "step": 785
958
  },
959
  {
960
  "epoch": 0.72,
961
  "learning_rate": 4.270795685155001e-06,
962
- "loss": 0.9489,
963
  "step": 790
964
  },
965
  {
966
  "epoch": 0.73,
967
  "learning_rate": 4.140298023713416e-06,
968
- "loss": 0.9292,
969
  "step": 795
970
  },
971
  {
972
  "epoch": 0.73,
973
  "learning_rate": 4.0113027014349374e-06,
974
- "loss": 0.944,
975
  "step": 800
976
  },
977
  {
@@ -983,49 +983,49 @@
983
  {
984
  "epoch": 0.74,
985
  "learning_rate": 3.757950970527249e-06,
986
- "loss": 0.9365,
987
  "step": 810
988
  },
989
  {
990
  "epoch": 0.75,
991
  "learning_rate": 3.633659517458736e-06,
992
- "loss": 0.9312,
993
  "step": 815
994
  },
995
  {
996
  "epoch": 0.75,
997
  "learning_rate": 3.511000298021098e-06,
998
- "loss": 0.95,
999
  "step": 820
1000
  },
1001
  {
1002
  "epoch": 0.76,
1003
  "learning_rate": 3.39000476018726e-06,
1004
- "loss": 0.944,
1005
  "step": 825
1006
  },
1007
  {
1008
  "epoch": 0.76,
1009
  "learning_rate": 3.2707039253872796e-06,
1010
- "loss": 0.9354,
1011
  "step": 830
1012
  },
1013
  {
1014
  "epoch": 0.77,
1015
  "learning_rate": 3.153128380554941e-06,
1016
- "loss": 0.9418,
1017
  "step": 835
1018
  },
1019
  {
1020
  "epoch": 0.77,
1021
  "learning_rate": 3.037308270285709e-06,
1022
- "loss": 0.9204,
1023
  "step": 840
1024
  },
1025
  {
1026
  "epoch": 0.78,
1027
  "learning_rate": 2.923273289108115e-06,
1028
- "loss": 0.9462,
1029
  "step": 845
1030
  },
1031
  {
@@ -1037,7 +1037,7 @@
1037
  {
1038
  "epoch": 0.78,
1039
  "learning_rate": 2.700675196245288e-06,
1040
- "loss": 0.9466,
1041
  "step": 855
1042
  },
1043
  {
@@ -1049,55 +1049,55 @@
1049
  {
1050
  "epoch": 0.79,
1051
  "learning_rate": 2.485562370502279e-06,
1052
- "loss": 0.9283,
1053
  "step": 865
1054
  },
1055
  {
1056
  "epoch": 0.8,
1057
  "learning_rate": 2.3808821740669608e-06,
1058
- "loss": 0.9294,
1059
  "step": 870
1060
  },
1061
  {
1062
  "epoch": 0.8,
1063
  "learning_rate": 2.2781554044688015e-06,
1064
- "loss": 0.9404,
1065
  "step": 875
1066
  },
1067
  {
1068
  "epoch": 0.81,
1069
  "learning_rate": 2.1774083993013715e-06,
1070
- "loss": 0.9363,
1071
  "step": 880
1072
  },
1073
  {
1074
  "epoch": 0.81,
1075
  "learning_rate": 2.0786669885765044e-06,
1076
- "loss": 0.9345,
1077
  "step": 885
1078
  },
1079
  {
1080
  "epoch": 0.82,
1081
  "learning_rate": 1.981956488101898e-06,
1082
- "loss": 0.9244,
1083
  "step": 890
1084
  },
1085
  {
1086
  "epoch": 0.82,
1087
  "learning_rate": 1.8873016929904942e-06,
1088
- "loss": 0.9336,
1089
  "step": 895
1090
  },
1091
  {
1092
  "epoch": 0.83,
1093
  "learning_rate": 1.7947268713034128e-06,
1094
- "loss": 0.9481,
1095
  "step": 900
1096
  },
1097
  {
1098
  "epoch": 0.83,
1099
  "learning_rate": 1.704255757827963e-06,
1100
- "loss": 0.935,
1101
  "step": 905
1102
  },
1103
  {
@@ -1109,13 +1109,13 @@
1109
  {
1110
  "epoch": 0.84,
1111
  "learning_rate": 1.529716891919074e-06,
1112
- "loss": 0.9213,
1113
  "step": 915
1114
  },
1115
  {
1116
  "epoch": 0.84,
1117
  "learning_rate": 1.4456938886170413e-06,
1118
- "loss": 0.9381,
1119
  "step": 920
1120
  },
1121
  {
@@ -1127,19 +1127,19 @@
1127
  {
1128
  "epoch": 0.85,
1129
  "learning_rate": 1.2842484469453365e-06,
1130
- "loss": 0.9158,
1131
  "step": 930
1132
  },
1133
  {
1134
  "epoch": 0.86,
1135
  "learning_rate": 1.2068674007506787e-06,
1136
- "loss": 0.9288,
1137
  "step": 935
1138
  },
1139
  {
1140
  "epoch": 0.86,
1141
  "learning_rate": 1.1317407810650372e-06,
1142
- "loss": 0.9213,
1143
  "step": 940
1144
  },
1145
  {
@@ -1163,37 +1163,37 @@
1163
  {
1164
  "epoch": 0.88,
1165
  "learning_rate": 8.541550213737171e-07,
1166
- "loss": 0.924,
1167
  "step": 960
1168
  },
1169
  {
1170
  "epoch": 0.89,
1171
  "learning_rate": 7.905777244947954e-07,
1172
- "loss": 0.9192,
1173
  "step": 965
1174
  },
1175
  {
1176
  "epoch": 0.89,
1177
  "learning_rate": 7.293615845160196e-07,
1178
- "loss": 0.9125,
1179
  "step": 970
1180
  },
1181
  {
1182
  "epoch": 0.89,
1183
  "learning_rate": 6.705222963319191e-07,
1184
- "loss": 0.9282,
1185
  "step": 975
1186
  },
1187
  {
1188
  "epoch": 0.9,
1189
  "learning_rate": 6.140749454480932e-07,
1190
- "loss": 0.9384,
1191
  "step": 980
1192
  },
1193
  {
1194
  "epoch": 0.9,
1195
  "learning_rate": 5.600340041135133e-07,
1196
- "loss": 0.9259,
1197
  "step": 985
1198
  },
1199
  {
@@ -1211,25 +1211,25 @@
1211
  {
1212
  "epoch": 0.92,
1213
  "learning_rate": 4.124850842338779e-07,
1214
- "loss": 0.9252,
1215
  "step": 1000
1216
  },
1217
  {
1218
  "epoch": 0.92,
1219
  "learning_rate": 3.6820211191520127e-07,
1220
- "loss": 0.9231,
1221
  "step": 1005
1222
  },
1223
  {
1224
  "epoch": 0.93,
1225
  "learning_rate": 3.263885872300343e-07,
1226
- "loss": 0.9087,
1227
  "step": 1010
1228
  },
1229
  {
1230
  "epoch": 0.93,
1231
  "learning_rate": 2.870552305351382e-07,
1232
- "loss": 0.9291,
1233
  "step": 1015
1234
  },
1235
  {
@@ -1247,7 +1247,7 @@
1247
  {
1248
  "epoch": 0.94,
1249
  "learning_rate": 1.840338184455881e-07,
1250
- "loss": 0.9148,
1251
  "step": 1030
1252
  },
1253
  {
@@ -1265,43 +1265,43 @@
1265
  {
1266
  "epoch": 0.96,
1267
  "learning_rate": 1.0365852555447642e-07,
1268
- "loss": 0.9407,
1269
  "step": 1045
1270
  },
1271
  {
1272
  "epoch": 0.96,
1273
  "learning_rate": 8.19327959602012e-08,
1274
- "loss": 0.9346,
1275
  "step": 1050
1276
  },
1277
  {
1278
  "epoch": 0.97,
1279
  "learning_rate": 6.274990913221035e-08,
1280
- "loss": 0.9198,
1281
  "step": 1055
1282
  },
1283
  {
1284
  "epoch": 0.97,
1285
  "learning_rate": 4.6114783273213395e-08,
1286
- "loss": 0.9214,
1287
  "step": 1060
1288
  },
1289
  {
1290
  "epoch": 0.98,
1291
  "learning_rate": 3.203168337845508e-08,
1292
- "loss": 0.9125,
1293
  "step": 1065
1294
  },
1295
  {
1296
  "epoch": 0.98,
1297
  "learning_rate": 2.05042201422323e-08,
1298
- "loss": 0.9268,
1299
  "step": 1070
1300
  },
1301
  {
1302
  "epoch": 0.99,
1303
  "learning_rate": 1.1535349032167908e-08,
1304
- "loss": 0.9249,
1305
  "step": 1075
1306
  },
1307
  {
@@ -1313,30 +1313,30 @@
1313
  {
1314
  "epoch": 1.0,
1315
  "learning_rate": 1.2819245493955746e-09,
1316
- "loss": 0.9161,
1317
  "step": 1085
1318
  },
1319
  {
1320
  "epoch": 1.0,
1321
  "learning_rate": 0.0,
1322
- "loss": 0.9077,
1323
  "step": 1090
1324
  },
1325
  {
1326
  "epoch": 1.0,
1327
- "eval_loss": 0.9353834986686707,
1328
- "eval_runtime": 144.309,
1329
- "eval_samples_per_second": 106.93,
1330
- "eval_steps_per_second": 1.677,
1331
  "step": 1090
1332
  },
1333
  {
1334
  "epoch": 1.0,
1335
  "step": 1090,
1336
  "total_flos": 456447649382400.0,
1337
- "train_loss": 0.9776670060026537,
1338
- "train_runtime": 5614.3798,
1339
- "train_samples_per_second": 24.834,
1340
  "train_steps_per_second": 0.194
1341
  }
1342
  ],
 
23
  {
24
  "epoch": 0.01,
25
  "learning_rate": 1.8348623853211011e-06,
26
+ "loss": 1.3919,
27
  "step": 10
28
  },
29
  {
 
41
  {
42
  "epoch": 0.02,
43
  "learning_rate": 4.587155963302753e-06,
44
+ "loss": 1.1172,
45
  "step": 25
46
  },
47
  {
48
  "epoch": 0.03,
49
  "learning_rate": 5.504587155963303e-06,
50
+ "loss": 1.0704,
51
  "step": 30
52
  },
53
  {
 
59
  {
60
  "epoch": 0.04,
61
  "learning_rate": 7.3394495412844045e-06,
62
+ "loss": 1.052,
63
  "step": 40
64
  },
65
  {
 
71
  {
72
  "epoch": 0.05,
73
  "learning_rate": 9.174311926605506e-06,
74
+ "loss": 1.0308,
75
  "step": 50
76
  },
77
  {
78
  "epoch": 0.05,
79
  "learning_rate": 1.0091743119266055e-05,
80
+ "loss": 1.0091,
81
  "step": 55
82
  },
83
  {
84
  "epoch": 0.06,
85
  "learning_rate": 1.1009174311926607e-05,
86
+ "loss": 1.0356,
87
  "step": 60
88
  },
89
  {
90
  "epoch": 0.06,
91
  "learning_rate": 1.1926605504587156e-05,
92
+ "loss": 1.0079,
93
  "step": 65
94
  },
95
  {
96
  "epoch": 0.06,
97
  "learning_rate": 1.2844036697247708e-05,
98
+ "loss": 1.0001,
99
  "step": 70
100
  },
101
  {
102
  "epoch": 0.07,
103
  "learning_rate": 1.3761467889908258e-05,
104
+ "loss": 1.0001,
105
  "step": 75
106
  },
107
  {
108
  "epoch": 0.07,
109
  "learning_rate": 1.4678899082568809e-05,
110
+ "loss": 0.9918,
111
  "step": 80
112
  },
113
  {
114
  "epoch": 0.08,
115
  "learning_rate": 1.559633027522936e-05,
116
+ "loss": 1.0148,
117
  "step": 85
118
  },
119
  {
 
131
  {
132
  "epoch": 0.09,
133
  "learning_rate": 1.834862385321101e-05,
134
+ "loss": 1.0163,
135
  "step": 100
136
  },
137
  {
 
143
  {
144
  "epoch": 0.1,
145
  "learning_rate": 1.999994872196626e-05,
146
+ "loss": 1.0056,
147
  "step": 110
148
  },
149
  {
150
  "epoch": 0.11,
151
  "learning_rate": 1.9998154046002822e-05,
152
+ "loss": 1.0141,
153
  "step": 115
154
  },
155
  {
 
161
  {
162
  "epoch": 0.11,
163
  "learning_rate": 1.9986875683942535e-05,
164
+ "loss": 1.02,
165
  "step": 125
166
  },
167
  {
168
  "epoch": 0.12,
169
  "learning_rate": 1.9977394889447526e-05,
170
+ "loss": 1.0237,
171
  "step": 130
172
  },
173
  {
174
  "epoch": 0.12,
175
  "learning_rate": 1.9965356041462954e-05,
176
+ "loss": 1.0304,
177
  "step": 135
178
  },
179
  {
180
  "epoch": 0.13,
181
  "learning_rate": 1.9950762226567783e-05,
182
+ "loss": 1.0284,
183
  "step": 140
184
  },
185
  {
186
  "epoch": 0.13,
187
  "learning_rate": 1.9933617186395917e-05,
188
+ "loss": 1.0046,
189
  "step": 145
190
  },
191
  {
192
  "epoch": 0.14,
193
  "learning_rate": 1.9913925316676946e-05,
194
+ "loss": 1.0026,
195
  "step": 150
196
  },
197
  {
 
209
  {
210
  "epoch": 0.15,
211
  "learning_rate": 1.9839622474129595e-05,
212
+ "loss": 1.016,
213
  "step": 165
214
  },
215
  {
216
  "epoch": 0.16,
217
  "learning_rate": 1.9809800282473014e-05,
218
+ "loss": 0.9979,
219
  "step": 170
220
  },
221
  {
222
  "epoch": 0.16,
223
  "learning_rate": 1.977746300605507e-05,
224
+ "loss": 1.0219,
225
  "step": 175
226
  },
227
  {
228
  "epoch": 0.17,
229
  "learning_rate": 1.9742618935665478e-05,
230
+ "loss": 1.0086,
231
  "step": 180
232
  },
233
  {
234
  "epoch": 0.17,
235
  "learning_rate": 1.9705277004798072e-05,
236
+ "loss": 0.9983,
237
  "step": 185
238
  },
239
  {
240
  "epoch": 0.17,
241
  "learning_rate": 1.9665446787360444e-05,
242
+ "loss": 1.0106,
243
  "step": 190
244
  },
245
  {
246
  "epoch": 0.18,
247
  "learning_rate": 1.9623138495219292e-05,
248
+ "loss": 1.0084,
249
  "step": 195
250
  },
251
  {
252
  "epoch": 0.18,
253
  "learning_rate": 1.957836297558229e-05,
254
+ "loss": 1.0272,
255
  "step": 200
256
  },
257
  {
258
  "epoch": 0.19,
259
  "learning_rate": 1.9531131708217005e-05,
260
+ "loss": 0.989,
261
  "step": 205
262
  },
263
  {
264
  "epoch": 0.19,
265
  "learning_rate": 1.948145680250766e-05,
266
+ "loss": 1.0114,
267
  "step": 210
268
  },
269
  {
270
  "epoch": 0.2,
271
  "learning_rate": 1.9429350994350483e-05,
272
+ "loss": 0.993,
273
  "step": 215
274
  },
275
  {
276
  "epoch": 0.2,
277
  "learning_rate": 1.93748276428884e-05,
278
+ "loss": 1.0229,
279
  "step": 220
280
  },
281
  {
282
  "epoch": 0.21,
283
  "learning_rate": 1.931790072708596e-05,
284
+ "loss": 1.0191,
285
  "step": 225
286
  },
287
  {
288
  "epoch": 0.21,
289
  "learning_rate": 1.9258584842145342e-05,
290
+ "loss": 1.0016,
291
  "step": 230
292
  },
293
  {
294
  "epoch": 0.22,
295
  "learning_rate": 1.9196895195764363e-05,
296
+ "loss": 0.9955,
297
  "step": 235
298
  },
299
  {
300
  "epoch": 0.22,
301
  "learning_rate": 1.913284760423745e-05,
302
+ "loss": 1.0212,
303
  "step": 240
304
  },
305
  {
 
311
  {
312
  "epoch": 0.23,
313
  "learning_rate": 1.8997744869421248e-05,
314
+ "loss": 1.0036,
315
  "step": 250
316
  },
317
  {
318
  "epoch": 0.23,
319
  "learning_rate": 1.8926724364434447e-05,
320
+ "loss": 0.9951,
321
  "step": 255
322
  },
323
  {
 
329
  {
330
  "epoch": 0.24,
331
  "learning_rate": 1.8777836117563894e-05,
332
+ "loss": 1.0197,
333
  "step": 265
334
  },
335
  {
 
341
  {
342
  "epoch": 0.25,
343
  "learning_rate": 1.861994642880105e-05,
344
+ "loss": 1.0036,
345
  "step": 275
346
  },
347
  {
348
  "epoch": 0.26,
349
  "learning_rate": 1.8537676285033886e-05,
350
+ "loss": 1.0064,
351
  "step": 280
352
  },
353
  {
 
359
  {
360
  "epoch": 0.27,
361
  "learning_rate": 1.8366590857437182e-05,
362
+ "loss": 1.0154,
363
  "step": 290
364
  },
365
  {
366
  "epoch": 0.27,
367
  "learning_rate": 1.8277819437329577e-05,
368
+ "loss": 0.993,
369
  "step": 295
370
  },
371
  {
 
377
  {
378
  "epoch": 0.28,
379
  "learning_rate": 1.809393297692334e-05,
380
+ "loss": 0.9892,
381
  "step": 305
382
  },
383
  {
384
  "epoch": 0.28,
385
  "learning_rate": 1.799886508233829e-05,
386
+ "loss": 0.9869,
387
  "step": 310
388
  },
389
  {
390
  "epoch": 0.29,
391
  "learning_rate": 1.790174639944997e-05,
392
+ "loss": 0.9894,
393
  "step": 315
394
  },
395
  {
396
  "epoch": 0.29,
397
  "learning_rate": 1.780260182802314e-05,
398
+ "loss": 0.9984,
399
  "step": 320
400
  },
401
  {
402
  "epoch": 0.3,
403
  "learning_rate": 1.7701456787229805e-05,
404
+ "loss": 0.9878,
405
  "step": 325
406
  },
407
  {
408
  "epoch": 0.3,
409
  "learning_rate": 1.7598337209132142e-05,
410
+ "loss": 1.0103,
411
  "step": 330
412
  },
413
  {
414
  "epoch": 0.31,
415
  "learning_rate": 1.7493269532033882e-05,
416
+ "loss": 0.994,
417
  "step": 335
418
  },
419
  {
420
  "epoch": 0.31,
421
  "learning_rate": 1.738628069370195e-05,
422
+ "loss": 0.9978,
423
  "step": 340
424
  },
425
  {
426
  "epoch": 0.32,
427
  "learning_rate": 1.7277398124460022e-05,
428
+ "loss": 1.0071,
429
  "step": 345
430
  },
431
  {
432
  "epoch": 0.32,
433
  "learning_rate": 1.71666497401558e-05,
434
+ "loss": 0.9925,
435
  "step": 350
436
  },
437
  {
 
443
  {
444
  "epoch": 0.33,
445
  "learning_rate": 1.6939669574305565e-05,
446
+ "loss": 1.0069,
447
  "step": 360
448
  },
449
  {
 
455
  {
456
  "epoch": 0.34,
457
  "learning_rate": 1.6705572958388576e-05,
458
+ "loss": 0.9989,
459
  "step": 370
460
  },
461
  {
462
  "epoch": 0.34,
463
  "learning_rate": 1.6585930722009602e-05,
464
+ "loss": 0.9951,
465
  "step": 375
466
  },
467
  {
468
  "epoch": 0.35,
469
  "learning_rate": 1.6464599952375998e-05,
470
+ "loss": 0.9971,
471
  "step": 380
472
  },
473
  {
474
  "epoch": 0.35,
475
  "learning_rate": 1.63416117568662e-05,
476
+ "loss": 0.9811,
477
  "step": 385
478
  },
479
  {
 
485
  {
486
  "epoch": 0.36,
487
  "learning_rate": 1.6090789634342278e-05,
488
+ "loss": 1.0029,
489
  "step": 395
490
  },
491
  {
492
  "epoch": 0.37,
493
  "learning_rate": 1.5963020014335437e-05,
494
+ "loss": 0.9934,
495
  "step": 400
496
  },
497
  {
498
  "epoch": 0.37,
499
  "learning_rate": 1.583372156597961e-05,
500
+ "loss": 0.9753,
501
  "step": 405
502
  },
503
  {
504
  "epoch": 0.38,
505
  "learning_rate": 1.570292743944583e-05,
506
+ "loss": 1.009,
507
  "step": 410
508
  },
509
  {
510
  "epoch": 0.38,
511
  "learning_rate": 1.557067116837444e-05,
512
+ "loss": 0.968,
513
  "step": 415
514
  },
515
  {
 
521
  {
522
  "epoch": 0.39,
523
  "learning_rate": 1.530190819284555e-05,
524
+ "loss": 1.0131,
525
  "step": 425
526
  },
527
  {
528
  "epoch": 0.39,
529
  "learning_rate": 1.5165470395159314e-05,
530
+ "loss": 0.9881,
531
  "step": 430
532
  },
533
  {
534
  "epoch": 0.4,
535
  "learning_rate": 1.5027708248811331e-05,
536
+ "loss": 0.9887,
537
  "step": 435
538
  },
539
  {
540
  "epoch": 0.4,
541
  "learning_rate": 1.4888657073937077e-05,
542
+ "loss": 0.9793,
543
  "step": 440
544
  },
545
  {
546
  "epoch": 0.41,
547
  "learning_rate": 1.4748352521159492e-05,
548
+ "loss": 0.9766,
549
  "step": 445
550
  },
551
  {
552
  "epoch": 0.41,
553
  "learning_rate": 1.4606830562448692e-05,
554
+ "loss": 0.9877,
555
  "step": 450
556
  },
557
  {
558
  "epoch": 0.42,
559
  "learning_rate": 1.4464127481899312e-05,
560
+ "loss": 0.9998,
561
  "step": 455
562
  },
563
  {
 
581
  {
582
  "epoch": 0.44,
583
  "learning_rate": 1.3882240024460928e-05,
584
+ "loss": 0.986,
585
  "step": 475
586
  },
587
  {
588
  "epoch": 0.44,
589
  "learning_rate": 1.3734185865027061e-05,
590
+ "loss": 0.9834,
591
  "step": 480
592
  },
593
  {
 
599
  {
600
  "epoch": 0.45,
601
  "learning_rate": 1.3435243583752294e-05,
602
+ "loss": 0.9749,
603
  "step": 490
604
  },
605
  {
606
  "epoch": 0.45,
607
  "learning_rate": 1.3284432106201233e-05,
608
+ "loss": 0.9799,
609
  "step": 495
610
  },
611
  {
612
  "epoch": 0.46,
613
  "learning_rate": 1.313277854982062e-05,
614
+ "loss": 0.9585,
615
  "step": 500
616
  },
617
  {
618
  "epoch": 0.46,
619
  "learning_rate": 1.2980321796293838e-05,
620
+ "loss": 0.9886,
621
  "step": 505
622
  },
623
  {
624
  "epoch": 0.47,
625
  "learning_rate": 1.2827100933231904e-05,
626
+ "loss": 0.9694,
627
  "step": 510
628
  },
629
  {
 
641
  {
642
  "epoch": 0.48,
643
  "learning_rate": 1.2363247441059775e-05,
644
+ "loss": 0.9711,
645
  "step": 525
646
  },
647
  {
 
659
  {
660
  "epoch": 0.5,
661
  "learning_rate": 1.189394177207125e-05,
662
+ "loss": 0.9826,
663
  "step": 540
664
  },
665
  {
 
671
  {
672
  "epoch": 0.5,
673
  "learning_rate": 1.1578576573543541e-05,
674
+ "loss": 0.9614,
675
  "step": 550
676
  },
677
  {
678
  "epoch": 0.51,
679
  "learning_rate": 1.1420266647205232e-05,
680
+ "loss": 0.9689,
681
  "step": 555
682
  },
683
  {
 
689
  {
690
  "epoch": 0.52,
691
  "learning_rate": 1.1102595071354471e-05,
692
+ "loss": 0.9447,
693
  "step": 565
694
  },
695
  {
696
  "epoch": 0.52,
697
  "learning_rate": 1.0943314868040365e-05,
698
+ "loss": 0.9584,
699
  "step": 570
700
  },
701
  {
 
713
  {
714
  "epoch": 0.54,
715
  "learning_rate": 1.0464186795625481e-05,
716
+ "loss": 0.9637,
717
  "step": 585
718
  },
719
  {
720
  "epoch": 0.54,
721
  "learning_rate": 1.0304184775393642e-05,
722
+ "loss": 0.9573,
723
  "step": 590
724
  },
725
  {
 
737
  {
738
  "epoch": 0.56,
739
  "learning_rate": 9.823874962247565e-06,
740
+ "loss": 0.9652,
741
  "step": 605
742
  },
743
  {
744
  "epoch": 0.56,
745
  "learning_rate": 9.663807268427197e-06,
746
+ "loss": 0.951,
747
  "step": 610
748
  },
749
  {
750
  "epoch": 0.56,
751
  "learning_rate": 9.503825769350016e-06,
752
+ "loss": 0.9735,
753
  "step": 615
754
  },
755
  {
756
  "epoch": 0.57,
757
  "learning_rate": 9.343971481858246e-06,
758
+ "loss": 0.9491,
759
  "step": 620
760
  },
761
  {
 
767
  {
768
  "epoch": 0.58,
769
  "learning_rate": 9.024808435416435e-06,
770
+ "loss": 0.9434,
771
  "step": 630
772
  },
773
  {
 
779
  {
780
  "epoch": 0.59,
781
  "learning_rate": 8.706645422477739e-06,
782
+ "loss": 0.9619,
783
  "step": 640
784
  },
785
  {
786
  "epoch": 0.59,
787
  "learning_rate": 8.548040936496989e-06,
788
+ "loss": 0.9455,
789
  "step": 645
790
  },
791
  {
792
  "epoch": 0.6,
793
  "learning_rate": 8.389808710909881e-06,
794
+ "loss": 0.9741,
795
  "step": 650
796
  },
797
  {
 
809
  {
810
  "epoch": 0.61,
811
  "learning_rate": 7.917750740470116e-06,
812
+ "loss": 0.9696,
813
  "step": 665
814
  },
815
  {
816
  "epoch": 0.61,
817
  "learning_rate": 7.761412129735853e-06,
818
+ "loss": 0.9628,
819
  "step": 670
820
  },
821
  {
822
  "epoch": 0.62,
823
  "learning_rate": 7.605647459150961e-06,
824
+ "loss": 0.9485,
825
  "step": 675
826
  },
827
  {
828
  "epoch": 0.62,
829
  "learning_rate": 7.4504966644264775e-06,
830
+ "loss": 0.9485,
831
  "step": 680
832
  },
833
  {
 
839
  {
840
  "epoch": 0.63,
841
  "learning_rate": 7.142195648261747e-06,
842
+ "loss": 0.9648,
843
  "step": 690
844
  },
845
  {
 
851
  {
852
  "epoch": 0.64,
853
  "learning_rate": 6.83682523588902e-06,
854
+ "loss": 0.944,
855
  "step": 700
856
  },
857
  {
858
  "epoch": 0.65,
859
  "learning_rate": 6.685336991505122e-06,
860
+ "loss": 0.9575,
861
  "step": 705
862
  },
863
  {
864
  "epoch": 0.65,
865
  "learning_rate": 6.5346985766979384e-06,
866
+ "loss": 0.9498,
867
  "step": 710
868
  },
869
  {
870
  "epoch": 0.66,
871
  "learning_rate": 6.384948612883872e-06,
872
+ "loss": 0.9475,
873
  "step": 715
874
  },
875
  {
876
  "epoch": 0.66,
877
  "learning_rate": 6.2361254936939e-06,
878
+ "loss": 0.9385,
879
  "step": 720
880
  },
881
  {
 
893
  {
894
  "epoch": 0.67,
895
  "learning_rate": 5.79559751711138e-06,
896
+ "loss": 0.9352,
897
  "step": 735
898
  },
899
  {
900
  "epoch": 0.68,
901
  "learning_rate": 5.650860813791786e-06,
902
+ "loss": 0.9446,
903
  "step": 740
904
  },
905
  {
906
  "epoch": 0.68,
907
  "learning_rate": 5.507239164130501e-06,
908
+ "loss": 0.9523,
909
  "step": 745
910
  },
911
  {
912
  "epoch": 0.69,
913
  "learning_rate": 5.364769390551225e-06,
914
+ "loss": 0.9508,
915
  "step": 750
916
  },
917
  {
918
  "epoch": 0.69,
919
  "learning_rate": 5.223488020154028e-06,
920
+ "loss": 0.946,
921
  "step": 755
922
  },
923
  {
 
935
  {
936
  "epoch": 0.71,
937
  "learning_rate": 4.807134973085036e-06,
938
+ "loss": 0.9438,
939
  "step": 770
940
  },
941
  {
942
  "epoch": 0.71,
943
  "learning_rate": 4.670966253826027e-06,
944
+ "loss": 0.9502,
945
  "step": 775
946
  },
947
  {
 
953
  {
954
  "epoch": 0.72,
955
  "learning_rate": 4.402762228128531e-06,
956
+ "loss": 0.9408,
957
  "step": 785
958
  },
959
  {
960
  "epoch": 0.72,
961
  "learning_rate": 4.270795685155001e-06,
962
+ "loss": 0.9482,
963
  "step": 790
964
  },
965
  {
966
  "epoch": 0.73,
967
  "learning_rate": 4.140298023713416e-06,
968
+ "loss": 0.9288,
969
  "step": 795
970
  },
971
  {
972
  "epoch": 0.73,
973
  "learning_rate": 4.0113027014349374e-06,
974
+ "loss": 0.9438,
975
  "step": 800
976
  },
977
  {
 
983
  {
984
  "epoch": 0.74,
985
  "learning_rate": 3.757950970527249e-06,
986
+ "loss": 0.9366,
987
  "step": 810
988
  },
989
  {
990
  "epoch": 0.75,
991
  "learning_rate": 3.633659517458736e-06,
992
+ "loss": 0.931,
993
  "step": 815
994
  },
995
  {
996
  "epoch": 0.75,
997
  "learning_rate": 3.511000298021098e-06,
998
+ "loss": 0.9497,
999
  "step": 820
1000
  },
1001
  {
1002
  "epoch": 0.76,
1003
  "learning_rate": 3.39000476018726e-06,
1004
+ "loss": 0.9438,
1005
  "step": 825
1006
  },
1007
  {
1008
  "epoch": 0.76,
1009
  "learning_rate": 3.2707039253872796e-06,
1010
+ "loss": 0.9353,
1011
  "step": 830
1012
  },
1013
  {
1014
  "epoch": 0.77,
1015
  "learning_rate": 3.153128380554941e-06,
1016
+ "loss": 0.9419,
1017
  "step": 835
1018
  },
1019
  {
1020
  "epoch": 0.77,
1021
  "learning_rate": 3.037308270285709e-06,
1022
+ "loss": 0.9201,
1023
  "step": 840
1024
  },
1025
  {
1026
  "epoch": 0.78,
1027
  "learning_rate": 2.923273289108115e-06,
1028
+ "loss": 0.9461,
1029
  "step": 845
1030
  },
1031
  {
 
1037
  {
1038
  "epoch": 0.78,
1039
  "learning_rate": 2.700675196245288e-06,
1040
+ "loss": 0.9465,
1041
  "step": 855
1042
  },
1043
  {
 
1049
  {
1050
  "epoch": 0.79,
1051
  "learning_rate": 2.485562370502279e-06,
1052
+ "loss": 0.9282,
1053
  "step": 865
1054
  },
1055
  {
1056
  "epoch": 0.8,
1057
  "learning_rate": 2.3808821740669608e-06,
1058
+ "loss": 0.9293,
1059
  "step": 870
1060
  },
1061
  {
1062
  "epoch": 0.8,
1063
  "learning_rate": 2.2781554044688015e-06,
1064
+ "loss": 0.9402,
1065
  "step": 875
1066
  },
1067
  {
1068
  "epoch": 0.81,
1069
  "learning_rate": 2.1774083993013715e-06,
1070
+ "loss": 0.9362,
1071
  "step": 880
1072
  },
1073
  {
1074
  "epoch": 0.81,
1075
  "learning_rate": 2.0786669885765044e-06,
1076
+ "loss": 0.9342,
1077
  "step": 885
1078
  },
1079
  {
1080
  "epoch": 0.82,
1081
  "learning_rate": 1.981956488101898e-06,
1082
+ "loss": 0.9245,
1083
  "step": 890
1084
  },
1085
  {
1086
  "epoch": 0.82,
1087
  "learning_rate": 1.8873016929904942e-06,
1088
+ "loss": 0.9338,
1089
  "step": 895
1090
  },
1091
  {
1092
  "epoch": 0.83,
1093
  "learning_rate": 1.7947268713034128e-06,
1094
+ "loss": 0.9483,
1095
  "step": 900
1096
  },
1097
  {
1098
  "epoch": 0.83,
1099
  "learning_rate": 1.704255757827963e-06,
1100
+ "loss": 0.9351,
1101
  "step": 905
1102
  },
1103
  {
 
1109
  {
1110
  "epoch": 0.84,
1111
  "learning_rate": 1.529716891919074e-06,
1112
+ "loss": 0.9212,
1113
  "step": 915
1114
  },
1115
  {
1116
  "epoch": 0.84,
1117
  "learning_rate": 1.4456938886170413e-06,
1118
+ "loss": 0.9379,
1119
  "step": 920
1120
  },
1121
  {
 
1127
  {
1128
  "epoch": 0.85,
1129
  "learning_rate": 1.2842484469453365e-06,
1130
+ "loss": 0.9156,
1131
  "step": 930
1132
  },
1133
  {
1134
  "epoch": 0.86,
1135
  "learning_rate": 1.2068674007506787e-06,
1136
+ "loss": 0.9285,
1137
  "step": 935
1138
  },
1139
  {
1140
  "epoch": 0.86,
1141
  "learning_rate": 1.1317407810650372e-06,
1142
+ "loss": 0.9214,
1143
  "step": 940
1144
  },
1145
  {
 
1163
  {
1164
  "epoch": 0.88,
1165
  "learning_rate": 8.541550213737171e-07,
1166
+ "loss": 0.9241,
1167
  "step": 960
1168
  },
1169
  {
1170
  "epoch": 0.89,
1171
  "learning_rate": 7.905777244947954e-07,
1172
+ "loss": 0.9194,
1173
  "step": 965
1174
  },
1175
  {
1176
  "epoch": 0.89,
1177
  "learning_rate": 7.293615845160196e-07,
1178
+ "loss": 0.9124,
1179
  "step": 970
1180
  },
1181
  {
1182
  "epoch": 0.89,
1183
  "learning_rate": 6.705222963319191e-07,
1184
+ "loss": 0.9281,
1185
  "step": 975
1186
  },
1187
  {
1188
  "epoch": 0.9,
1189
  "learning_rate": 6.140749454480932e-07,
1190
+ "loss": 0.9382,
1191
  "step": 980
1192
  },
1193
  {
1194
  "epoch": 0.9,
1195
  "learning_rate": 5.600340041135133e-07,
1196
+ "loss": 0.9257,
1197
  "step": 985
1198
  },
1199
  {
 
1211
  {
1212
  "epoch": 0.92,
1213
  "learning_rate": 4.124850842338779e-07,
1214
+ "loss": 0.9249,
1215
  "step": 1000
1216
  },
1217
  {
1218
  "epoch": 0.92,
1219
  "learning_rate": 3.6820211191520127e-07,
1220
+ "loss": 0.9233,
1221
  "step": 1005
1222
  },
1223
  {
1224
  "epoch": 0.93,
1225
  "learning_rate": 3.263885872300343e-07,
1226
+ "loss": 0.9086,
1227
  "step": 1010
1228
  },
1229
  {
1230
  "epoch": 0.93,
1231
  "learning_rate": 2.870552305351382e-07,
1232
+ "loss": 0.929,
1233
  "step": 1015
1234
  },
1235
  {
 
1247
  {
1248
  "epoch": 0.94,
1249
  "learning_rate": 1.840338184455881e-07,
1250
+ "loss": 0.9144,
1251
  "step": 1030
1252
  },
1253
  {
 
1265
  {
1266
  "epoch": 0.96,
1267
  "learning_rate": 1.0365852555447642e-07,
1268
+ "loss": 0.9406,
1269
  "step": 1045
1270
  },
1271
  {
1272
  "epoch": 0.96,
1273
  "learning_rate": 8.19327959602012e-08,
1274
+ "loss": 0.9343,
1275
  "step": 1050
1276
  },
1277
  {
1278
  "epoch": 0.97,
1279
  "learning_rate": 6.274990913221035e-08,
1280
+ "loss": 0.92,
1281
  "step": 1055
1282
  },
1283
  {
1284
  "epoch": 0.97,
1285
  "learning_rate": 4.6114783273213395e-08,
1286
+ "loss": 0.9212,
1287
  "step": 1060
1288
  },
1289
  {
1290
  "epoch": 0.98,
1291
  "learning_rate": 3.203168337845508e-08,
1292
+ "loss": 0.9127,
1293
  "step": 1065
1294
  },
1295
  {
1296
  "epoch": 0.98,
1297
  "learning_rate": 2.05042201422323e-08,
1298
+ "loss": 0.9267,
1299
  "step": 1070
1300
  },
1301
  {
1302
  "epoch": 0.99,
1303
  "learning_rate": 1.1535349032167908e-08,
1304
+ "loss": 0.9248,
1305
  "step": 1075
1306
  },
1307
  {
 
1313
  {
1314
  "epoch": 1.0,
1315
  "learning_rate": 1.2819245493955746e-09,
1316
+ "loss": 0.9159,
1317
  "step": 1085
1318
  },
1319
  {
1320
  "epoch": 1.0,
1321
  "learning_rate": 0.0,
1322
+ "loss": 0.9075,
1323
  "step": 1090
1324
  },
1325
  {
1326
  "epoch": 1.0,
1327
+ "eval_loss": 0.9353461265563965,
1328
+ "eval_runtime": 142.6393,
1329
+ "eval_samples_per_second": 108.182,
1330
+ "eval_steps_per_second": 1.697,
1331
  "step": 1090
1332
  },
1333
  {
1334
  "epoch": 1.0,
1335
  "step": 1090,
1336
  "total_flos": 456447649382400.0,
1337
+ "train_loss": 0.9776473476252424,
1338
+ "train_runtime": 5608.2856,
1339
+ "train_samples_per_second": 24.861,
1340
  "train_steps_per_second": 0.194
1341
  }
1342
  ],