AlekseyKorshuk commited on
Commit
87b9240
1 Parent(s): f00cb76

huggingartists

Browse files
README.md CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
45
  dataset = load_dataset("huggingartists/eminem")
46
  ```
47
 
48
- [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1tfdfdyn/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Eminem's lyrics.
53
 
54
- Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/2oam1me5) for full transparency and reproducibility.
55
 
56
- At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2oam1me5/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
 
45
  dataset = load_dataset("huggingartists/eminem")
46
  ```
47
 
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3vf8qp69/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Eminem's lyrics.
53
 
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/teyrbiau) for full transparency and reproducibility.
55
 
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/teyrbiau/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
evaluation.txt CHANGED
@@ -1 +1 @@
1
- {"eval_loss": 0.13430377840995789, "eval_runtime": 14.0446, "eval_samples_per_second": 44.43, "eval_steps_per_second": 5.554, "epoch": 2.0}
 
1
+ {"eval_loss": 0.182983860373497, "eval_runtime": 15.8575, "eval_samples_per_second": 41.999, "eval_steps_per_second": 5.297, "epoch": 2.0}
flax_model.msgpack CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0f31ed1a889f71569ec3a441f9ef396888cfb72a5e5321c231c3f7e3422f2e3a
3
  size 497764120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:847f8e5bf0f42177eecf9538fa6414bf9192c4019ebfe0d9814c7a59555b8871
3
  size 497764120
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8304c5fd2768fa01ba9b5712d995a5731bfc92f999564aa85fb2fc766e8d8311
3
  size 995604017
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23ef9e41cf8f2ff08d810e44138a09ba1cf8bbbe1cf06697f92b2baa247225aa
3
  size 995604017
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2036ba89317ecba125c1a82916377c1352593504a5278d40be3161990dfc19f9
3
  size 510396521
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83c763918cad87c921d0ba8ce9d7e909fbdee98f00a5e135e5d23ba48667374d
3
  size 510396521
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:91789756b00c246c75939e963082e15fe38636b3074ce10218a12d23c35888d0
3
  size 14567
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc58ef324cb2c1a967d48b64d114926b825fe7f474d79ba9dc6dcc9008a509ed
3
  size 14567
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:573b86203c7261a17ef6809968ba821e820b1e71725b2471dc35d45266479890
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f6ac60cf933ec0679732f5e8e6c465bca3dbe584300179b279d7fef11183026
3
  size 623
trainer_state.json CHANGED
@@ -1,586 +1,564 @@
1
  {
2
- "best_metric": 0.13430377840995789,
3
- "best_model_checkpoint": "output/eminem/checkpoint-457",
4
  "epoch": 1.0,
5
- "global_step": 457,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.01,
12
- "learning_rate": 0.0001315617372315331,
13
- "loss": 0.3859,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.02,
18
- "learning_rate": 0.0001305819167834288,
19
- "loss": 0.4039,
20
  "step": 10
21
  },
22
  {
23
  "epoch": 0.03,
24
- "learning_rate": 0.0001295279055913259,
25
- "loss": 0.3636,
26
  "step": 15
27
  },
28
  {
29
  "epoch": 0.04,
30
- "learning_rate": 0.00012840096527924066,
31
- "loss": 0.365,
32
  "step": 20
33
  },
34
  {
35
  "epoch": 0.06,
36
- "learning_rate": 0.00012720244476544407,
37
- "loss": 0.4157,
38
  "step": 25
39
  },
40
  {
41
  "epoch": 0.07,
42
- "learning_rate": 0.00012593377864785258,
43
- "loss": 0.3804,
44
  "step": 30
45
  },
46
  {
47
  "epoch": 0.08,
48
- "learning_rate": 0.00012459648548684795,
49
- "loss": 0.3737,
50
  "step": 35
51
  },
52
  {
53
  "epoch": 0.09,
54
- "learning_rate": 0.00012319216598759544,
55
- "loss": 0.4087,
56
  "step": 40
57
  },
58
  {
59
  "epoch": 0.1,
60
- "learning_rate": 0.0001217225010840549,
61
- "loss": 0.3864,
62
  "step": 45
63
  },
64
  {
65
  "epoch": 0.11,
66
- "learning_rate": 0.00012018924992692434,
67
- "loss": 0.3832,
68
  "step": 50
69
  },
70
  {
71
  "epoch": 0.12,
72
- "learning_rate": 0.0001185942477780086,
73
- "loss": 0.3928,
74
  "step": 55
75
  },
76
  {
77
  "epoch": 0.13,
78
- "learning_rate": 0.00011693940381343527,
79
- "loss": 0.3682,
80
  "step": 60
81
  },
82
  {
83
  "epoch": 0.14,
84
- "learning_rate": 0.00011522669883843705,
85
- "loss": 0.3798,
86
  "step": 65
87
  },
88
  {
89
  "epoch": 0.15,
90
- "learning_rate": 0.00011345818291636727,
91
- "loss": 0.3447,
92
  "step": 70
93
  },
94
  {
95
  "epoch": 0.17,
96
- "learning_rate": 0.00011163597291483855,
97
- "loss": 0.3718,
98
  "step": 75
99
  },
100
  {
101
  "epoch": 0.18,
102
- "learning_rate": 0.00010976224997187412,
103
- "loss": 0.4333,
104
  "step": 80
105
  },
106
  {
107
  "epoch": 0.19,
108
- "learning_rate": 0.00010783925688516201,
109
- "loss": 0.43,
110
  "step": 85
111
  },
112
  {
113
  "epoch": 0.2,
114
- "learning_rate": 0.00010586929542745496,
115
- "loss": 0.3813,
116
  "step": 90
117
  },
118
  {
119
  "epoch": 0.21,
120
- "learning_rate": 0.00010385472359144299,
121
- "loss": 0.3779,
122
  "step": 95
123
  },
124
  {
125
  "epoch": 0.22,
126
- "learning_rate": 0.00010179795276726699,
127
- "loss": 0.3846,
128
  "step": 100
129
  },
130
  {
131
  "epoch": 0.23,
132
- "learning_rate": 9.970144485616674e-05,
133
- "loss": 0.3528,
134
  "step": 105
135
  },
136
  {
137
  "epoch": 0.24,
138
- "learning_rate": 9.756770932364287e-05,
139
- "loss": 0.3909,
140
  "step": 110
141
  },
142
  {
143
  "epoch": 0.25,
144
- "learning_rate": 9.53993001956845e-05,
145
- "loss": 0.4177,
146
  "step": 115
147
  },
148
  {
149
- "epoch": 0.26,
150
- "learning_rate": 9.319881300168782e-05,
151
- "loss": 0.4063,
152
  "step": 120
153
  },
154
  {
155
  "epoch": 0.28,
156
- "learning_rate": 9.09688816676409e-05,
157
- "loss": 0.4443,
158
  "step": 125
159
  },
160
  {
161
  "epoch": 0.29,
162
- "learning_rate": 8.871217536342237e-05,
163
- "loss": 0.4231,
164
  "step": 130
165
  },
166
  {
167
  "epoch": 0.3,
168
- "learning_rate": 8.643139530783983e-05,
169
- "loss": 0.3991,
170
  "step": 135
171
  },
172
  {
173
  "epoch": 0.31,
174
- "learning_rate": 8.412927153536512e-05,
175
- "loss": 0.3839,
176
  "step": 140
177
  },
178
  {
179
  "epoch": 0.32,
180
- "learning_rate": 8.180855962833708e-05,
181
- "loss": 0.3923,
182
  "step": 145
183
  },
184
  {
185
  "epoch": 0.33,
186
- "learning_rate": 7.947203741861626e-05,
187
- "loss": 0.4375,
188
  "step": 150
189
  },
190
  {
191
  "epoch": 0.34,
192
- "learning_rate": 7.712250166257713e-05,
193
- "loss": 0.4036,
194
  "step": 155
195
  },
196
  {
197
  "epoch": 0.35,
198
- "learning_rate": 7.476276469347862e-05,
199
- "loss": 0.438,
200
  "step": 160
201
  },
202
  {
203
- "epoch": 0.36,
204
- "learning_rate": 7.239565105515707e-05,
205
- "loss": 0.3947,
206
  "step": 165
207
  },
208
  {
209
- "epoch": 0.37,
210
- "learning_rate": 7.002399412114568e-05,
211
- "loss": 0.3352,
212
  "step": 170
213
  },
214
  {
215
  "epoch": 0.39,
216
- "learning_rate": 6.765063270316318e-05,
217
- "loss": 0.3922,
218
  "step": 175
219
  },
220
  {
221
  "epoch": 0.4,
222
- "learning_rate": 6.527840765317078e-05,
223
- "loss": 0.3592,
224
  "step": 180
225
  },
226
  {
227
  "epoch": 0.41,
228
- "learning_rate": 6.291015846291898e-05,
229
- "loss": 0.4318,
230
  "step": 185
231
  },
232
  {
233
  "epoch": 0.42,
234
- "learning_rate": 6.054871986515355e-05,
235
- "loss": 0.4418,
236
  "step": 190
237
  },
238
  {
239
  "epoch": 0.43,
240
- "learning_rate": 5.819691844052574e-05,
241
- "loss": 0.3699,
242
  "step": 195
243
  },
244
  {
245
  "epoch": 0.44,
246
- "learning_rate": 5.585756923423138e-05,
247
- "loss": 0.3992,
248
  "step": 200
249
  },
250
  {
251
  "epoch": 0.45,
252
- "learning_rate": 5.353347238649785e-05,
253
- "loss": 0.3905,
254
  "step": 205
255
  },
256
  {
257
  "epoch": 0.46,
258
- "learning_rate": 5.122740978086224e-05,
259
- "loss": 0.3767,
260
  "step": 210
261
  },
262
  {
263
- "epoch": 0.47,
264
- "learning_rate": 4.894214171435194e-05,
265
- "loss": 0.3906,
266
  "step": 215
267
  },
268
  {
269
- "epoch": 0.48,
270
- "learning_rate": 4.6680403593463696e-05,
271
- "loss": 0.3188,
272
  "step": 220
273
  },
274
  {
275
  "epoch": 0.5,
276
- "learning_rate": 4.4444902659968225e-05,
277
- "loss": 0.3562,
278
  "step": 225
279
  },
280
  {
281
  "epoch": 0.51,
282
- "learning_rate": 4.22383147503972e-05,
283
- "loss": 0.406,
284
  "step": 230
285
  },
286
  {
287
  "epoch": 0.52,
288
- "learning_rate": 4.006328109315114e-05,
289
- "loss": 0.3462,
290
  "step": 235
291
  },
292
  {
293
  "epoch": 0.53,
294
- "learning_rate": 3.7922405147001485e-05,
295
- "loss": 0.3237,
296
  "step": 240
297
  },
298
  {
299
  "epoch": 0.54,
300
- "learning_rate": 3.581824948484096e-05,
301
- "loss": 0.3355,
302
  "step": 245
303
  },
304
  {
305
  "epoch": 0.55,
306
- "learning_rate": 3.375333272631733e-05,
307
- "loss": 0.3852,
308
  "step": 250
309
  },
310
  {
311
  "epoch": 0.56,
312
- "learning_rate": 3.1730126523147464e-05,
313
- "loss": 0.3729,
314
  "step": 255
315
  },
316
  {
317
- "epoch": 0.57,
318
- "learning_rate": 2.9751052600590164e-05,
319
- "loss": 0.3384,
320
  "step": 260
321
  },
322
  {
323
- "epoch": 0.58,
324
- "learning_rate": 2.7818479858707847e-05,
325
- "loss": 0.3507,
326
  "step": 265
327
  },
328
  {
329
- "epoch": 0.59,
330
- "learning_rate": 2.593472153686425e-05,
331
- "loss": 0.3956,
332
  "step": 270
333
  },
334
  {
335
  "epoch": 0.61,
336
- "learning_rate": 2.4102032444822844e-05,
337
- "loss": 0.3913,
338
  "step": 275
339
  },
340
  {
341
  "epoch": 0.62,
342
- "learning_rate": 2.232260626381796e-05,
343
- "loss": 0.3743,
344
  "step": 280
345
  },
346
  {
347
  "epoch": 0.63,
348
- "learning_rate": 2.0598572920747012e-05,
349
- "loss": 0.4364,
350
  "step": 285
351
  },
352
  {
353
  "epoch": 0.64,
354
- "learning_rate": 1.8931996038735562e-05,
355
- "loss": 0.3568,
356
  "step": 290
357
  },
358
  {
359
  "epoch": 0.65,
360
- "learning_rate": 1.7324870467006735e-05,
361
- "loss": 0.3423,
362
  "step": 295
363
  },
364
  {
365
  "epoch": 0.66,
366
- "learning_rate": 1.577911989311673e-05,
367
- "loss": 0.3864,
368
  "step": 300
369
  },
370
  {
371
  "epoch": 0.67,
372
- "learning_rate": 1.429659454033877e-05,
373
- "loss": 0.4152,
374
  "step": 305
375
  },
376
  {
377
- "epoch": 0.68,
378
- "learning_rate": 1.2879068953004088e-05,
379
- "loss": 0.3684,
380
  "step": 310
381
  },
382
  {
383
- "epoch": 0.69,
384
- "learning_rate": 1.1528239872408715e-05,
385
- "loss": 0.4387,
386
  "step": 315
387
  },
388
  {
389
- "epoch": 0.7,
390
- "learning_rate": 1.0245724205867534e-05,
391
- "loss": 0.4097,
392
  "step": 320
393
  },
394
  {
395
  "epoch": 0.72,
396
- "learning_rate": 9.033057091308445e-06,
397
- "loss": 0.3489,
398
  "step": 325
399
  },
400
  {
401
  "epoch": 0.73,
402
- "learning_rate": 7.891690059763513e-06,
403
- "loss": 0.3714,
404
  "step": 330
405
  },
406
  {
407
  "epoch": 0.74,
408
- "learning_rate": 6.822989297914713e-06,
409
- "loss": 0.3657,
410
  "step": 335
411
  },
412
  {
413
  "epoch": 0.75,
414
- "learning_rate": 5.828234012805588e-06,
415
- "loss": 0.3898,
416
  "step": 340
417
  },
418
  {
419
  "epoch": 0.76,
420
- "learning_rate": 4.90861490066865e-06,
421
- "loss": 0.3966,
422
  "step": 345
423
  },
424
  {
425
  "epoch": 0.77,
426
- "learning_rate": 4.065232721685712e-06,
427
- "loss": 0.3573,
428
  "step": 350
429
  },
430
  {
431
- "epoch": 0.78,
432
- "learning_rate": 3.299096982415909e-06,
433
- "loss": 0.3632,
434
  "step": 355
435
  },
436
  {
437
- "epoch": 0.79,
438
- "learning_rate": 2.6111247274298776e-06,
439
- "loss": 0.37,
440
  "step": 360
441
  },
442
  {
443
- "epoch": 0.8,
444
- "learning_rate": 2.0021394416451063e-06,
445
- "loss": 0.3912,
446
  "step": 365
447
  },
448
  {
449
- "epoch": 0.81,
450
- "learning_rate": 1.472870064624613e-06,
451
- "loss": 0.3434,
452
  "step": 370
453
  },
454
  {
455
  "epoch": 0.83,
456
- "learning_rate": 1.0239501180622683e-06,
457
- "loss": 0.4036,
458
  "step": 375
459
  },
460
  {
461
  "epoch": 0.84,
462
- "learning_rate": 6.559169474696805e-07,
463
- "loss": 0.3656,
464
  "step": 380
465
  },
466
  {
467
  "epoch": 0.85,
468
- "learning_rate": 3.6921107899052446e-07,
469
- "loss": 0.333,
470
  "step": 385
471
  },
472
  {
473
  "epoch": 0.86,
474
- "learning_rate": 1.6417569209981867e-07,
475
- "loss": 0.4543,
476
  "step": 390
477
  },
478
  {
479
  "epoch": 0.87,
480
- "learning_rate": 4.1056208828832076e-08,
481
- "loss": 0.3396,
482
  "step": 395
483
  },
484
  {
485
  "epoch": 0.88,
486
- "learning_rate": 0.0,
487
- "loss": 0.3405,
488
  "step": 400
489
  },
490
  {
491
- "epoch": 0.89,
492
- "learning_rate": 4.105620882877876e-08,
493
- "loss": 0.3422,
494
  "step": 405
495
  },
496
  {
497
- "epoch": 0.9,
498
- "learning_rate": 1.6417569209970443e-07,
499
- "loss": 0.3717,
500
  "step": 410
501
  },
502
  {
503
- "epoch": 0.91,
504
- "learning_rate": 3.692110789903493e-07,
505
- "loss": 0.3561,
506
  "step": 415
507
  },
508
  {
509
  "epoch": 0.93,
510
- "learning_rate": 6.559169474694444e-07,
511
- "loss": 0.4,
512
  "step": 420
513
  },
514
  {
515
  "epoch": 0.94,
516
- "learning_rate": 1.0239501180619788e-06,
517
- "loss": 0.3807,
518
  "step": 425
519
  },
520
  {
521
  "epoch": 0.95,
522
- "learning_rate": 1.4728700646242702e-06,
523
- "loss": 0.3506,
524
  "step": 430
525
  },
526
  {
527
  "epoch": 0.96,
528
- "learning_rate": 2.0021394416451673e-06,
529
- "loss": 0.3554,
530
  "step": 435
531
  },
532
  {
533
  "epoch": 0.97,
534
- "learning_rate": 2.6111247274294206e-06,
535
- "loss": 0.3542,
536
  "step": 440
537
  },
538
  {
539
  "epoch": 0.98,
540
- "learning_rate": 3.299096982415391e-06,
541
- "loss": 0.3452,
542
  "step": 445
543
  },
544
- {
545
- "epoch": 0.99,
546
- "learning_rate": 4.065232721685804e-06,
547
- "loss": 0.3457,
548
- "step": 450
549
- },
550
- {
551
- "epoch": 1.0,
552
- "eval_loss": 0.17592251300811768,
553
- "eval_runtime": 14.827,
554
- "eval_samples_per_second": 43.502,
555
- "eval_steps_per_second": 5.463,
556
- "step": 454
557
- },
558
  {
559
  "epoch": 1.0,
560
- "learning_rate": 4.181858176201967e-07,
561
- "loss": 0.3187,
562
- "step": 455
563
- },
564
- {
565
- "epoch": 1.0,
566
- "eval_loss": 0.15162613987922668,
567
- "eval_runtime": 13.378,
568
- "eval_samples_per_second": 48.139,
569
- "eval_steps_per_second": 6.055,
570
- "step": 455
571
  },
572
  {
573
  "epoch": 1.0,
574
- "eval_loss": 0.13430377840995789,
575
- "eval_runtime": 12.7773,
576
- "eval_samples_per_second": 48.837,
577
- "eval_steps_per_second": 6.105,
578
- "step": 457
579
  }
580
  ],
581
- "max_steps": 914,
582
  "num_train_epochs": 2,
583
- "total_flos": 476466020352000.0,
584
  "trial_name": null,
585
  "trial_params": null
586
  }
 
1
  {
2
+ "best_metric": 0.182983860373497,
3
+ "best_model_checkpoint": "output/eminem/checkpoint-452",
4
  "epoch": 1.0,
5
+ "global_step": 452,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.01,
12
+ "learning_rate": 8.139815125844835e-05,
13
+ "loss": 0.3572,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.02,
18
+ "learning_rate": 7.904875437627615e-05,
19
+ "loss": 0.332,
20
  "step": 10
21
  },
22
  {
23
  "epoch": 0.03,
24
+ "learning_rate": 7.6686739692467e-05,
25
+ "loss": 0.3325,
26
  "step": 15
27
  },
28
  {
29
  "epoch": 0.04,
30
+ "learning_rate": 7.431495955015108e-05,
31
+ "loss": 0.3373,
32
  "step": 20
33
  },
34
  {
35
  "epoch": 0.06,
36
+ "learning_rate": 7.193627808511377e-05,
37
+ "loss": 0.3129,
38
  "step": 25
39
  },
40
  {
41
  "epoch": 0.07,
42
+ "learning_rate": 6.955356776710818e-05,
43
+ "loss": 0.3354,
44
  "step": 30
45
  },
46
  {
47
  "epoch": 0.08,
48
+ "learning_rate": 6.716970593108238e-05,
49
+ "loss": 0.3096,
50
  "step": 35
51
  },
52
  {
53
  "epoch": 0.09,
54
+ "learning_rate": 6.478757130255084e-05,
55
+ "loss": 0.3554,
56
  "step": 40
57
  },
58
  {
59
  "epoch": 0.1,
60
+ "learning_rate": 6.24100405212653e-05,
61
+ "loss": 0.3456,
62
  "step": 45
63
  },
64
  {
65
  "epoch": 0.11,
66
+ "learning_rate": 6.0039984667425774e-05,
67
+ "loss": 0.3283,
68
  "step": 50
69
  },
70
  {
71
  "epoch": 0.12,
72
+ "learning_rate": 5.768026579458568e-05,
73
+ "loss": 0.3464,
74
  "step": 55
75
  },
76
  {
77
  "epoch": 0.13,
78
+ "learning_rate": 5.5333733473478955e-05,
79
+ "loss": 0.3187,
80
  "step": 60
81
  },
82
  {
83
  "epoch": 0.14,
84
+ "learning_rate": 5.3003221350899904e-05,
85
+ "loss": 0.3377,
86
  "step": 65
87
  },
88
  {
89
  "epoch": 0.15,
90
+ "learning_rate": 5.069154372783178e-05,
91
+ "loss": 0.3172,
92
  "step": 70
93
  },
94
  {
95
  "epoch": 0.17,
96
+ "learning_rate": 4.8401492160918125e-05,
97
+ "loss": 0.3366,
98
  "step": 75
99
  },
100
  {
101
  "epoch": 0.18,
102
+ "learning_rate": 4.6135832091409655e-05,
103
+ "loss": 0.3463,
104
  "step": 80
105
  },
106
  {
107
  "epoch": 0.19,
108
+ "learning_rate": 4.389729950565688e-05,
109
+ "loss": 0.3223,
110
  "step": 85
111
  },
112
  {
113
  "epoch": 0.2,
114
+ "learning_rate": 4.168859763115363e-05,
115
+ "loss": 0.3103,
116
  "step": 90
117
  },
118
  {
119
  "epoch": 0.21,
120
+ "learning_rate": 3.9512393672158115e-05,
121
+ "loss": 0.321,
122
  "step": 95
123
  },
124
  {
125
  "epoch": 0.22,
126
+ "learning_rate": 3.737131558879536e-05,
127
+ "loss": 0.341,
128
  "step": 100
129
  },
130
  {
131
  "epoch": 0.23,
132
+ "learning_rate": 3.526794892357027e-05,
133
+ "loss": 0.316,
134
  "step": 105
135
  },
136
  {
137
  "epoch": 0.24,
138
+ "learning_rate": 3.320483367908598e-05,
139
+ "loss": 0.3275,
140
  "step": 110
141
  },
142
  {
143
  "epoch": 0.25,
144
+ "learning_rate": 3.118446125077491e-05,
145
+ "loss": 0.3246,
146
  "step": 115
147
  },
148
  {
149
+ "epoch": 0.27,
150
+ "learning_rate": 2.9209271418308243e-05,
151
+ "loss": 0.3166,
152
  "step": 120
153
  },
154
  {
155
  "epoch": 0.28,
156
+ "learning_rate": 2.7281649399352843e-05,
157
+ "loss": 0.3269,
158
  "step": 125
159
  },
160
  {
161
  "epoch": 0.29,
162
+ "learning_rate": 2.540392296919879e-05,
163
+ "loss": 0.3522,
164
  "step": 130
165
  },
166
  {
167
  "epoch": 0.3,
168
+ "learning_rate": 2.3578359649771312e-05,
169
+ "loss": 0.3318,
170
  "step": 135
171
  },
172
  {
173
  "epoch": 0.31,
174
+ "learning_rate": 2.1807163971388526e-05,
175
+ "loss": 0.3207,
176
  "step": 140
177
  },
178
  {
179
  "epoch": 0.32,
180
+ "learning_rate": 2.009247481060366e-05,
181
+ "loss": 0.3127,
182
  "step": 145
183
  },
184
  {
185
  "epoch": 0.33,
186
+ "learning_rate": 1.843636280731375e-05,
187
+ "loss": 0.2845,
188
  "step": 150
189
  },
190
  {
191
  "epoch": 0.34,
192
+ "learning_rate": 1.6840827864284427e-05,
193
+ "loss": 0.3354,
194
  "step": 155
195
  },
196
  {
197
  "epoch": 0.35,
198
+ "learning_rate": 1.530779673208225e-05,
199
+ "loss": 0.3237,
200
  "step": 160
201
  },
202
  {
203
+ "epoch": 0.37,
204
+ "learning_rate": 1.3839120682352292e-05,
205
+ "loss": 0.3352,
206
  "step": 165
207
  },
208
  {
209
+ "epoch": 0.38,
210
+ "learning_rate": 1.243657327224889e-05,
211
+ "loss": 0.3299,
212
  "step": 170
213
  },
214
  {
215
  "epoch": 0.39,
216
+ "learning_rate": 1.1101848202702286e-05,
217
+ "loss": 0.3208,
218
  "step": 175
219
  },
220
  {
221
  "epoch": 0.4,
222
+ "learning_rate": 9.836557273124208e-06,
223
+ "loss": 0.2988,
224
  "step": 180
225
  },
226
  {
227
  "epoch": 0.41,
228
+ "learning_rate": 8.642228435020553e-06,
229
+ "loss": 0.3097,
230
  "step": 185
231
  },
232
  {
233
  "epoch": 0.42,
234
+ "learning_rate": 7.520303946848552e-06,
235
+ "loss": 0.3237,
236
  "step": 190
237
  },
238
  {
239
  "epoch": 0.43,
240
+ "learning_rate": 6.472138632361784e-06,
241
+ "loss": 0.3487,
242
  "step": 195
243
  },
244
  {
245
  "epoch": 0.44,
246
+ "learning_rate": 5.498998244544231e-06,
247
+ "loss": 0.3417,
248
  "step": 200
249
  },
250
  {
251
  "epoch": 0.45,
252
+ "learning_rate": 4.602057937097127e-06,
253
+ "loss": 0.312,
254
  "step": 205
255
  },
256
  {
257
  "epoch": 0.46,
258
+ "learning_rate": 3.7824008453364313e-06,
259
+ "loss": 0.3209,
260
  "step": 210
261
  },
262
  {
263
+ "epoch": 0.48,
264
+ "learning_rate": 3.041016778213118e-06,
265
+ "loss": 0.3107,
266
  "step": 215
267
  },
268
  {
269
+ "epoch": 0.49,
270
+ "learning_rate": 2.378801023027184e-06,
271
+ "loss": 0.3087,
272
  "step": 220
273
  },
274
  {
275
  "epoch": 0.5,
276
+ "learning_rate": 1.796553264289097e-06,
277
+ "loss": 0.2947,
278
  "step": 225
279
  },
280
  {
281
  "epoch": 0.51,
282
+ "learning_rate": 1.29497661803209e-06,
283
+ "loss": 0.3013,
284
  "step": 230
285
  },
286
  {
287
  "epoch": 0.52,
288
+ "learning_rate": 8.746767827346987e-07,
289
+ "loss": 0.2997,
290
  "step": 235
291
  },
292
  {
293
  "epoch": 0.53,
294
+ "learning_rate": 5.361613078862018e-07,
295
+ "loss": 0.2932,
296
  "step": 240
297
  },
298
  {
299
  "epoch": 0.54,
300
+ "learning_rate": 2.7983898107679704e-07,
301
+ "loss": 0.3453,
302
  "step": 245
303
  },
304
  {
305
  "epoch": 0.55,
306
+ "learning_rate": 1.0601933434688134e-07,
307
+ "loss": 0.3002,
308
  "step": 250
309
  },
310
  {
311
  "epoch": 0.56,
312
+ "learning_rate": 1.4912270403722228e-08,
313
+ "loss": 0.3528,
314
  "step": 255
315
  },
316
  {
317
+ "epoch": 0.58,
318
+ "learning_rate": 6.627809142250918e-09,
319
+ "loss": 0.3299,
320
  "step": 260
321
  },
322
  {
323
+ "epoch": 0.59,
324
+ "learning_rate": 8.117595478739506e-08,
325
+ "loss": 0.3005,
326
  "step": 265
327
  },
328
  {
329
+ "epoch": 0.6,
330
+ "learning_rate": 2.384666838130307e-07,
331
+ "loss": 0.313,
332
  "step": 270
333
  },
334
  {
335
  "epoch": 0.61,
336
+ "learning_rate": 4.783100536523972e-07,
337
+ "loss": 0.2978,
338
  "step": 275
339
  },
340
  {
341
  "epoch": 0.62,
342
+ "learning_rate": 8.004164320734335e-07,
343
+ "loss": 0.3066,
344
  "step": 280
345
  },
346
  {
347
  "epoch": 0.63,
348
+ "learning_rate": 1.2043968469315794e-06,
349
+ "loss": 0.3224,
350
  "step": 285
351
  },
352
  {
353
  "epoch": 0.64,
354
+ "learning_rate": 1.689763455891368e-06,
355
+ "loss": 0.3433,
356
  "step": 290
357
  },
358
  {
359
  "epoch": 0.65,
360
+ "learning_rate": 2.255930135538237e-06,
361
+ "loss": 0.3279,
362
  "step": 295
363
  },
364
  {
365
  "epoch": 0.66,
366
+ "learning_rate": 2.9022131891755135e-06,
367
+ "loss": 0.3128,
368
  "step": 300
369
  },
370
  {
371
  "epoch": 0.67,
372
+ "learning_rate": 3.6278321724456592e-06,
373
+ "loss": 0.3398,
374
  "step": 305
375
  },
376
  {
377
+ "epoch": 0.69,
378
+ "learning_rate": 4.431910835795094e-06,
379
+ "loss": 0.346,
380
  "step": 310
381
  },
382
  {
383
+ "epoch": 0.7,
384
+ "learning_rate": 5.313478182611713e-06,
385
+ "loss": 0.3162,
386
  "step": 315
387
  },
388
  {
389
+ "epoch": 0.71,
390
+ "learning_rate": 6.271469641798839e-06,
391
+ "loss": 0.2976,
392
  "step": 320
393
  },
394
  {
395
  "epoch": 0.72,
396
+ "learning_rate": 7.304728353335969e-06,
397
+ "loss": 0.3161,
398
  "step": 325
399
  },
400
  {
401
  "epoch": 0.73,
402
+ "learning_rate": 8.412006565290002e-06,
403
+ "loss": 0.3494,
404
  "step": 330
405
  },
406
  {
407
  "epoch": 0.74,
408
+ "learning_rate": 9.591967140590498e-06,
409
+ "loss": 0.3082,
410
  "step": 335
411
  },
412
  {
413
  "epoch": 0.75,
414
+ "learning_rate": 1.0843185171737421e-05,
415
+ "loss": 0.3302,
416
  "step": 340
417
  },
418
  {
419
  "epoch": 0.76,
420
+ "learning_rate": 1.2164149701503293e-05,
421
+ "loss": 0.3535,
422
  "step": 345
423
  },
424
  {
425
  "epoch": 0.77,
426
+ "learning_rate": 1.3553265547553905e-05,
427
+ "loss": 0.2955,
428
  "step": 350
429
  },
430
  {
431
+ "epoch": 0.79,
432
+ "learning_rate": 1.500885522876901e-05,
433
+ "loss": 0.3486,
434
  "step": 355
435
  },
436
  {
437
+ "epoch": 0.8,
438
+ "learning_rate": 1.6529160990949952e-05,
439
+ "loss": 0.3031,
440
  "step": 360
441
  },
442
  {
443
+ "epoch": 0.81,
444
+ "learning_rate": 1.8112346929469448e-05,
445
+ "loss": 0.3208,
446
  "step": 365
447
  },
448
  {
449
+ "epoch": 0.82,
450
+ "learning_rate": 1.975650120628388e-05,
451
+ "loss": 0.302,
452
  "step": 370
453
  },
454
  {
455
  "epoch": 0.83,
456
+ "learning_rate": 2.1459638358647496e-05,
457
+ "loss": 0.3412,
458
  "step": 375
459
  },
460
  {
461
  "epoch": 0.84,
462
+ "learning_rate": 2.3219701696742428e-05,
463
+ "loss": 0.3244,
464
  "step": 380
465
  },
466
  {
467
  "epoch": 0.85,
468
+ "learning_rate": 2.5034565787309464e-05,
469
+ "loss": 0.3438,
470
  "step": 385
471
  },
472
  {
473
  "epoch": 0.86,
474
+ "learning_rate": 2.6902039020297848e-05,
475
+ "loss": 0.3173,
476
  "step": 390
477
  },
478
  {
479
  "epoch": 0.87,
480
+ "learning_rate": 2.881986625543576e-05,
481
+ "loss": 0.3234,
482
  "step": 395
483
  },
484
  {
485
  "epoch": 0.88,
486
+ "learning_rate": 3.078573154550556e-05,
487
+ "loss": 0.3361,
488
  "step": 400
489
  },
490
  {
491
+ "epoch": 0.9,
492
+ "learning_rate": 3.279726093305591e-05,
493
+ "loss": 0.3391,
494
  "step": 405
495
  },
496
  {
497
+ "epoch": 0.91,
498
+ "learning_rate": 3.485202531717502e-05,
499
+ "loss": 0.3205,
500
  "step": 410
501
  },
502
  {
503
+ "epoch": 0.92,
504
+ "learning_rate": 3.6947543386838276e-05,
505
+ "loss": 0.3197,
506
  "step": 415
507
  },
508
  {
509
  "epoch": 0.93,
510
+ "learning_rate": 3.9081284617320727e-05,
511
+ "loss": 0.3272,
512
  "step": 420
513
  },
514
  {
515
  "epoch": 0.94,
516
+ "learning_rate": 4.125067232602091e-05,
517
+ "loss": 0.3092,
518
  "step": 425
519
  },
520
  {
521
  "epoch": 0.95,
522
+ "learning_rate": 4.345308678404316e-05,
523
+ "loss": 0.3419,
524
  "step": 430
525
  },
526
  {
527
  "epoch": 0.96,
528
+ "learning_rate": 4.568586837974475e-05,
529
+ "loss": 0.3065,
530
  "step": 435
531
  },
532
  {
533
  "epoch": 0.97,
534
+ "learning_rate": 4.794632083046458e-05,
535
+ "loss": 0.3082,
536
  "step": 440
537
  },
538
  {
539
  "epoch": 0.98,
540
+ "learning_rate": 5.023171443851573e-05,
541
+ "loss": 0.3379,
542
  "step": 445
543
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
544
  {
545
  "epoch": 1.0,
546
+ "learning_rate": 5.2539289387547704e-05,
547
+ "loss": 0.3392,
548
+ "step": 450
 
 
 
 
 
 
 
 
549
  },
550
  {
551
  "epoch": 1.0,
552
+ "eval_loss": 0.182983860373497,
553
+ "eval_runtime": 15.9367,
554
+ "eval_samples_per_second": 41.79,
555
+ "eval_steps_per_second": 5.271,
556
+ "step": 452
557
  }
558
  ],
559
+ "max_steps": 904,
560
  "num_train_epochs": 2,
561
+ "total_flos": 471762763776000.0,
562
  "trial_name": null,
563
  "trial_params": null
564
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3ddd6b3bcc43d3e1b4aaa3a7e199709a9472aec3721534af1679511cccab1a47
3
  size 3311
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:345239b6657277522706cced3d784e83e9d91a342334cf8d8e1c643f51ae5d25
3
  size 3311