lyan62 commited on
Commit
a8d0e00
1 Parent(s): e739f85

Training in progress, step 5000

Browse files
config.json CHANGED
@@ -20,7 +20,7 @@
20
  "mask_ratio": 0.25,
21
  "model_type": "pixel",
22
  "norm_pix_input": true,
23
- "norm_pix_loss": true,
24
  "num_attention_heads": 6,
25
  "num_channels": 3,
26
  "num_hidden_layers": 12,
 
20
  "mask_ratio": 0.25,
21
  "model_type": "pixel",
22
  "norm_pix_input": true,
23
+ "norm_pix_loss": false,
24
  "num_attention_heads": 6,
25
  "num_channels": 3,
26
  "num_hidden_layers": 12,
last-checkpoint/config.json CHANGED
@@ -20,7 +20,7 @@
20
  "mask_ratio": 0.25,
21
  "model_type": "pixel",
22
  "norm_pix_input": true,
23
- "norm_pix_loss": true,
24
  "num_attention_heads": 6,
25
  "num_channels": 3,
26
  "num_hidden_layers": 12,
 
20
  "mask_ratio": 0.25,
21
  "model_type": "pixel",
22
  "norm_pix_input": true,
23
+ "norm_pix_loss": false,
24
  "num_attention_heads": 6,
25
  "num_channels": 3,
26
  "num_hidden_layers": 12,
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6cb3d0e61b5d108ce8144603307306d33ec57a740054d047b674d88a24d4047f
3
- size 202194449
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd08e118d8510f2cb4a9c2891ba7cb95c1953052e571d833019de6ce82573790
3
+ size 202193937
last-checkpoint/pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:01986e29e7b00ff5ed5012297e50124030c090218d489d0fb41425e1a13097b1
3
  size 102501541
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e31313ef25632d40f04a4d20bebc02ef847b472d4200e8e9085e3e9835ffafaa
3
  size 102501541
last-checkpoint/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6c0b6dc93078eb284ca60353df2fed8bb86e0d867e2ab46b3e3f4b2755de65b8
3
  size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eb26b59330b615a98e78d6fa6f00cb6141e029593d57c287ed1c46236d2e017
3
  size 14503
last-checkpoint/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6c0b6dc93078eb284ca60353df2fed8bb86e0d867e2ab46b3e3f4b2755de65b8
3
  size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eb26b59330b615a98e78d6fa6f00cb6141e029593d57c287ed1c46236d2e017
3
  size 14503
last-checkpoint/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6c0b6dc93078eb284ca60353df2fed8bb86e0d867e2ab46b3e3f4b2755de65b8
3
  size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eb26b59330b615a98e78d6fa6f00cb6141e029593d57c287ed1c46236d2e017
3
  size 14503
last-checkpoint/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6c0b6dc93078eb284ca60353df2fed8bb86e0d867e2ab46b3e3f4b2755de65b8
3
  size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eb26b59330b615a98e78d6fa6f00cb6141e029593d57c287ed1c46236d2e017
3
  size 14503
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72ec509fe27f6bd7a85a63791b7907f44298546cc560f9361537aced29997e23
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e42b82e9522a85b7c37d23355628f9314507631e34b534ae0e598c498d2271b9
3
  size 623
last-checkpoint/trainer_state.json CHANGED
@@ -1,1516 +1,76 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.91490942514649,
5
- "global_step": 125000,
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": 2.9999999999999997e-06,
13
- "loss": 0.7239,
14
  "step": 500
15
  },
16
  {
17
  "epoch": 0.02,
18
- "learning_rate": 5.999999999999999e-06,
19
- "loss": 0.2594,
20
  "step": 1000
21
  },
22
  {
23
  "epoch": 0.02,
24
- "learning_rate": 8.999999999999999e-06,
25
- "loss": 0.0798,
26
  "step": 1500
27
  },
28
  {
29
  "epoch": 0.03,
30
- "learning_rate": 1.1999999999999999e-05,
31
- "loss": 0.0611,
32
  "step": 2000
33
  },
34
  {
35
  "epoch": 0.04,
36
- "learning_rate": 1.4999999999999999e-05,
37
- "loss": 0.0603,
38
  "step": 2500
39
  },
40
  {
41
  "epoch": 0.05,
42
- "learning_rate": 1.7999999999999997e-05,
43
- "loss": 0.06,
44
  "step": 3000
45
  },
46
  {
47
  "epoch": 0.05,
48
- "learning_rate": 2.1e-05,
49
- "loss": 0.0597,
50
  "step": 3500
51
  },
52
  {
53
  "epoch": 0.06,
54
- "learning_rate": 2.3999999999999997e-05,
55
- "loss": 0.0596,
56
  "step": 4000
57
  },
58
  {
59
  "epoch": 0.07,
60
- "learning_rate": 2.6999999999999996e-05,
61
- "loss": 0.0595,
62
  "step": 4500
63
  },
64
  {
65
  "epoch": 0.08,
66
- "learning_rate": 2.9999999999999997e-05,
67
- "loss": 0.0594,
68
  "step": 5000
69
- },
70
- {
71
- "epoch": 0.08,
72
- "learning_rate": 3.2999999999999996e-05,
73
- "loss": 0.0593,
74
- "step": 5500
75
- },
76
- {
77
- "epoch": 0.09,
78
- "learning_rate": 3.5999999999999994e-05,
79
- "loss": 0.0592,
80
- "step": 6000
81
- },
82
- {
83
- "epoch": 0.1,
84
- "learning_rate": 3.9e-05,
85
- "loss": 0.0591,
86
- "step": 6500
87
- },
88
- {
89
- "epoch": 0.11,
90
- "learning_rate": 4.2e-05,
91
- "loss": 0.0589,
92
- "step": 7000
93
- },
94
- {
95
- "epoch": 0.11,
96
- "learning_rate": 4.4999999999999996e-05,
97
- "loss": 0.0585,
98
- "step": 7500
99
- },
100
- {
101
- "epoch": 0.12,
102
- "learning_rate": 4.7999999999999994e-05,
103
- "loss": 0.0582,
104
- "step": 8000
105
- },
106
- {
107
- "epoch": 0.13,
108
- "learning_rate": 5.1e-05,
109
- "loss": 0.0581,
110
- "step": 8500
111
- },
112
- {
113
- "epoch": 0.14,
114
- "learning_rate": 5.399999999999999e-05,
115
- "loss": 0.058,
116
- "step": 9000
117
- },
118
- {
119
- "epoch": 0.15,
120
- "learning_rate": 5.6999999999999996e-05,
121
- "loss": 0.058,
122
- "step": 9500
123
- },
124
- {
125
- "epoch": 0.15,
126
- "learning_rate": 5.9999999999999995e-05,
127
- "loss": 0.058,
128
- "step": 10000
129
- },
130
- {
131
- "epoch": 0.16,
132
- "learning_rate": 6.299999999999999e-05,
133
- "loss": 0.0579,
134
- "step": 10500
135
- },
136
- {
137
- "epoch": 0.17,
138
- "learning_rate": 6.599999999999999e-05,
139
- "loss": 0.0579,
140
- "step": 11000
141
- },
142
- {
143
- "epoch": 0.18,
144
- "learning_rate": 6.9e-05,
145
- "loss": 0.0579,
146
- "step": 11500
147
- },
148
- {
149
- "epoch": 0.18,
150
- "learning_rate": 7.199999999999999e-05,
151
- "loss": 0.0579,
152
- "step": 12000
153
- },
154
- {
155
- "epoch": 0.19,
156
- "learning_rate": 7.5e-05,
157
- "loss": 0.0579,
158
- "step": 12500
159
- },
160
- {
161
- "epoch": 0.2,
162
- "learning_rate": 7.8e-05,
163
- "loss": 0.0578,
164
- "step": 13000
165
- },
166
- {
167
- "epoch": 0.21,
168
- "learning_rate": 8.1e-05,
169
- "loss": 0.0574,
170
- "step": 13500
171
- },
172
- {
173
- "epoch": 0.21,
174
- "learning_rate": 8.4e-05,
175
- "loss": 0.056,
176
- "step": 14000
177
- },
178
- {
179
- "epoch": 0.22,
180
- "learning_rate": 8.699999999999999e-05,
181
- "loss": 0.0543,
182
- "step": 14500
183
- },
184
- {
185
- "epoch": 0.23,
186
- "learning_rate": 8.999999999999999e-05,
187
- "loss": 0.052,
188
- "step": 15000
189
- },
190
- {
191
- "epoch": 0.24,
192
- "learning_rate": 9.3e-05,
193
- "loss": 0.0496,
194
- "step": 15500
195
- },
196
- {
197
- "epoch": 0.25,
198
- "learning_rate": 9.599999999999999e-05,
199
- "loss": 0.0483,
200
- "step": 16000
201
- },
202
- {
203
- "epoch": 0.25,
204
- "learning_rate": 9.9e-05,
205
- "loss": 0.0474,
206
- "step": 16500
207
- },
208
- {
209
- "epoch": 0.26,
210
- "learning_rate": 0.000102,
211
- "loss": 0.0467,
212
- "step": 17000
213
- },
214
- {
215
- "epoch": 0.27,
216
- "learning_rate": 0.00010499999999999999,
217
- "loss": 0.0462,
218
- "step": 17500
219
- },
220
- {
221
- "epoch": 0.28,
222
- "learning_rate": 0.00010799999999999998,
223
- "loss": 0.0458,
224
- "step": 18000
225
- },
226
- {
227
- "epoch": 0.28,
228
- "learning_rate": 0.00011099999999999999,
229
- "loss": 0.0454,
230
- "step": 18500
231
- },
232
- {
233
- "epoch": 0.29,
234
- "learning_rate": 0.00011399999999999999,
235
- "loss": 0.0451,
236
- "step": 19000
237
- },
238
- {
239
- "epoch": 0.3,
240
- "learning_rate": 0.000117,
241
- "loss": 0.0448,
242
- "step": 19500
243
- },
244
- {
245
- "epoch": 0.31,
246
- "learning_rate": 0.00011999999999999999,
247
- "loss": 0.0446,
248
- "step": 20000
249
- },
250
- {
251
- "epoch": 0.31,
252
- "learning_rate": 0.00012299999999999998,
253
- "loss": 0.0443,
254
- "step": 20500
255
- },
256
- {
257
- "epoch": 0.32,
258
- "learning_rate": 0.00012599999999999997,
259
- "loss": 0.0441,
260
- "step": 21000
261
- },
262
- {
263
- "epoch": 0.33,
264
- "learning_rate": 0.000129,
265
- "loss": 0.0439,
266
- "step": 21500
267
- },
268
- {
269
- "epoch": 0.34,
270
- "learning_rate": 0.00013199999999999998,
271
- "loss": 0.0437,
272
- "step": 22000
273
- },
274
- {
275
- "epoch": 0.34,
276
- "learning_rate": 0.000135,
277
- "loss": 0.0436,
278
- "step": 22500
279
- },
280
- {
281
- "epoch": 0.35,
282
- "learning_rate": 0.000138,
283
- "loss": 0.0434,
284
- "step": 23000
285
- },
286
- {
287
- "epoch": 0.36,
288
- "learning_rate": 0.00014099999999999998,
289
- "loss": 0.0432,
290
- "step": 23500
291
- },
292
- {
293
- "epoch": 0.37,
294
- "learning_rate": 0.00014399999999999998,
295
- "loss": 0.0431,
296
- "step": 24000
297
- },
298
- {
299
- "epoch": 0.38,
300
- "learning_rate": 0.000147,
301
- "loss": 0.0429,
302
- "step": 24500
303
- },
304
- {
305
- "epoch": 0.38,
306
- "learning_rate": 0.00015,
307
- "loss": 0.0424,
308
- "step": 25000
309
- },
310
- {
311
- "epoch": 0.39,
312
- "learning_rate": 0.00014999960357580777,
313
- "loss": 0.0422,
314
- "step": 25500
315
- },
316
- {
317
- "epoch": 0.4,
318
- "learning_rate": 0.0001499984143075663,
319
- "loss": 0.042,
320
- "step": 26000
321
- },
322
- {
323
- "epoch": 0.41,
324
- "learning_rate": 0.0001499964322082812,
325
- "loss": 0.0419,
326
- "step": 26500
327
- },
328
- {
329
- "epoch": 0.41,
330
- "learning_rate": 0.00014999365729962845,
331
- "loss": 0.0417,
332
- "step": 27000
333
- },
334
- {
335
- "epoch": 0.42,
336
- "learning_rate": 0.000149990089611954,
337
- "loss": 0.0415,
338
- "step": 27500
339
- },
340
- {
341
- "epoch": 0.43,
342
- "learning_rate": 0.00014998572918427345,
343
- "loss": 0.0414,
344
- "step": 28000
345
- },
346
- {
347
- "epoch": 0.44,
348
- "learning_rate": 0.00014998057606427183,
349
- "loss": 0.0412,
350
- "step": 28500
351
- },
352
- {
353
- "epoch": 0.44,
354
- "learning_rate": 0.00014997463030830277,
355
- "loss": 0.041,
356
- "step": 29000
357
- },
358
- {
359
- "epoch": 0.45,
360
- "learning_rate": 0.0001499678919813881,
361
- "loss": 0.0408,
362
- "step": 29500
363
- },
364
- {
365
- "epoch": 0.46,
366
- "learning_rate": 0.00014996036115721712,
367
- "loss": 0.0406,
368
- "step": 30000
369
- },
370
- {
371
- "epoch": 0.47,
372
- "learning_rate": 0.0001499520379181457,
373
- "loss": 0.0405,
374
- "step": 30500
375
- },
376
- {
377
- "epoch": 0.47,
378
- "learning_rate": 0.00014994292235519547,
379
- "loss": 0.0403,
380
- "step": 31000
381
- },
382
- {
383
- "epoch": 0.48,
384
- "learning_rate": 0.0001499330145680527,
385
- "loss": 0.0402,
386
- "step": 31500
387
- },
388
- {
389
- "epoch": 0.49,
390
- "learning_rate": 0.00014992231466506745,
391
- "loss": 0.04,
392
- "step": 32000
393
- },
394
- {
395
- "epoch": 0.5,
396
- "learning_rate": 0.0001499108227632521,
397
- "loss": 0.0399,
398
- "step": 32500
399
- },
400
- {
401
- "epoch": 0.51,
402
- "learning_rate": 0.00014989853898828023,
403
- "loss": 0.0397,
404
- "step": 33000
405
- },
406
- {
407
- "epoch": 0.51,
408
- "learning_rate": 0.00014988546347448526,
409
- "loss": 0.0396,
410
- "step": 33500
411
- },
412
- {
413
- "epoch": 0.52,
414
- "learning_rate": 0.0001498715963648589,
415
- "loss": 0.0395,
416
- "step": 34000
417
- },
418
- {
419
- "epoch": 0.53,
420
- "learning_rate": 0.00014985693781104968,
421
- "loss": 0.0394,
422
- "step": 34500
423
- },
424
- {
425
- "epoch": 0.54,
426
- "learning_rate": 0.00014984148797336113,
427
- "loss": 0.0393,
428
- "step": 35000
429
- },
430
- {
431
- "epoch": 0.54,
432
- "learning_rate": 0.00014982524702075026,
433
- "loss": 0.0391,
434
- "step": 35500
435
- },
436
- {
437
- "epoch": 0.55,
438
- "learning_rate": 0.00014980821513082548,
439
- "loss": 0.039,
440
- "step": 36000
441
- },
442
- {
443
- "epoch": 0.56,
444
- "learning_rate": 0.00014979039248984487,
445
- "loss": 0.0389,
446
- "step": 36500
447
- },
448
- {
449
- "epoch": 0.57,
450
- "learning_rate": 0.0001497717792927139,
451
- "loss": 0.0388,
452
- "step": 37000
453
- },
454
- {
455
- "epoch": 0.57,
456
- "learning_rate": 0.00014975237574298356,
457
- "loss": 0.0387,
458
- "step": 37500
459
- },
460
- {
461
- "epoch": 0.58,
462
- "learning_rate": 0.00014973218205284794,
463
- "loss": 0.0386,
464
- "step": 38000
465
- },
466
- {
467
- "epoch": 0.59,
468
- "learning_rate": 0.000149711198443142,
469
- "loss": 0.0385,
470
- "step": 38500
471
- },
472
- {
473
- "epoch": 0.6,
474
- "learning_rate": 0.00014968942514333915,
475
- "loss": 0.0384,
476
- "step": 39000
477
- },
478
- {
479
- "epoch": 0.61,
480
- "learning_rate": 0.00014966686239154873,
481
- "loss": 0.0383,
482
- "step": 39500
483
- },
484
- {
485
- "epoch": 0.61,
486
- "learning_rate": 0.00014964351043451332,
487
- "loss": 0.0383,
488
- "step": 40000
489
- },
490
- {
491
- "epoch": 0.62,
492
- "learning_rate": 0.00014961936952760622,
493
- "loss": 0.0382,
494
- "step": 40500
495
- },
496
- {
497
- "epoch": 0.63,
498
- "learning_rate": 0.0001495944399348285,
499
- "loss": 0.0381,
500
- "step": 41000
501
- },
502
- {
503
- "epoch": 0.64,
504
- "learning_rate": 0.00014956872192880622,
505
- "loss": 0.038,
506
- "step": 41500
507
- },
508
- {
509
- "epoch": 0.64,
510
- "learning_rate": 0.00014954221579078732,
511
- "loss": 0.0379,
512
- "step": 42000
513
- },
514
- {
515
- "epoch": 0.65,
516
- "learning_rate": 0.00014951492181063876,
517
- "loss": 0.0378,
518
- "step": 42500
519
- },
520
- {
521
- "epoch": 0.66,
522
- "learning_rate": 0.00014948684028684306,
523
- "loss": 0.0377,
524
- "step": 43000
525
- },
526
- {
527
- "epoch": 0.67,
528
- "learning_rate": 0.00014945797152649532,
529
- "loss": 0.0376,
530
- "step": 43500
531
- },
532
- {
533
- "epoch": 0.67,
534
- "learning_rate": 0.00014942831584529963,
535
- "loss": 0.0374,
536
- "step": 44000
537
- },
538
- {
539
- "epoch": 0.68,
540
- "learning_rate": 0.0001493978735675658,
541
- "loss": 0.0372,
542
- "step": 44500
543
- },
544
- {
545
- "epoch": 0.69,
546
- "learning_rate": 0.00014936664502620569,
547
- "loss": 0.0371,
548
- "step": 45000
549
- },
550
- {
551
- "epoch": 0.7,
552
- "learning_rate": 0.00014933463056272963,
553
- "loss": 0.037,
554
- "step": 45500
555
- },
556
- {
557
- "epoch": 0.7,
558
- "learning_rate": 0.00014930183052724267,
559
- "loss": 0.0369,
560
- "step": 46000
561
- },
562
- {
563
- "epoch": 0.71,
564
- "learning_rate": 0.00014926824527844071,
565
- "loss": 0.0368,
566
- "step": 46500
567
- },
568
- {
569
- "epoch": 0.72,
570
- "learning_rate": 0.00014923387518360668,
571
- "loss": 0.0367,
572
- "step": 47000
573
- },
574
- {
575
- "epoch": 0.73,
576
- "learning_rate": 0.00014919872061860644,
577
- "loss": 0.0365,
578
- "step": 47500
579
- },
580
- {
581
- "epoch": 0.74,
582
- "learning_rate": 0.00014916278196788467,
583
- "loss": 0.0364,
584
- "step": 48000
585
- },
586
- {
587
- "epoch": 0.74,
588
- "learning_rate": 0.0001491260596244607,
589
- "loss": 0.0363,
590
- "step": 48500
591
- },
592
- {
593
- "epoch": 0.75,
594
- "learning_rate": 0.00014908855398992416,
595
- "loss": 0.0362,
596
- "step": 49000
597
- },
598
- {
599
- "epoch": 0.76,
600
- "learning_rate": 0.00014905026547443068,
601
- "loss": 0.0361,
602
- "step": 49500
603
- },
604
- {
605
- "epoch": 0.77,
606
- "learning_rate": 0.00014901119449669737,
607
- "loss": 0.036,
608
- "step": 50000
609
- },
610
- {
611
- "epoch": 0.77,
612
- "learning_rate": 0.0001489713414839981,
613
- "loss": 0.0359,
614
- "step": 50500
615
- },
616
- {
617
- "epoch": 0.78,
618
- "learning_rate": 0.00014893070687215914,
619
- "loss": 0.0358,
620
- "step": 51000
621
- },
622
- {
623
- "epoch": 0.79,
624
- "learning_rate": 0.000148889291105554,
625
- "loss": 0.0357,
626
- "step": 51500
627
- },
628
- {
629
- "epoch": 0.8,
630
- "learning_rate": 0.00014884709463709893,
631
- "loss": 0.0356,
632
- "step": 52000
633
- },
634
- {
635
- "epoch": 0.8,
636
- "learning_rate": 0.0001488041179282477,
637
- "loss": 0.0355,
638
- "step": 52500
639
- },
640
- {
641
- "epoch": 0.81,
642
- "learning_rate": 0.00014876036144898677,
643
- "loss": 0.0354,
644
- "step": 53000
645
- },
646
- {
647
- "epoch": 0.82,
648
- "learning_rate": 0.00014871582567782993,
649
- "loss": 0.0353,
650
- "step": 53500
651
- },
652
- {
653
- "epoch": 0.83,
654
- "learning_rate": 0.00014867051110181327,
655
- "loss": 0.0352,
656
- "step": 54000
657
- },
658
- {
659
- "epoch": 0.83,
660
- "learning_rate": 0.00014862441821648968,
661
- "loss": 0.0352,
662
- "step": 54500
663
- },
664
- {
665
- "epoch": 0.84,
666
- "learning_rate": 0.00014857754752592366,
667
- "loss": 0.0351,
668
- "step": 55000
669
- },
670
- {
671
- "epoch": 0.85,
672
- "learning_rate": 0.0001485298995426855,
673
- "loss": 0.035,
674
- "step": 55500
675
- },
676
- {
677
- "epoch": 0.86,
678
- "learning_rate": 0.00014848147478784598,
679
- "loss": 0.0349,
680
- "step": 56000
681
- },
682
- {
683
- "epoch": 0.87,
684
- "learning_rate": 0.00014843227379097038,
685
- "loss": 0.0348,
686
- "step": 56500
687
- },
688
- {
689
- "epoch": 0.87,
690
- "learning_rate": 0.00014838229709011297,
691
- "loss": 0.0348,
692
- "step": 57000
693
- },
694
- {
695
- "epoch": 0.88,
696
- "learning_rate": 0.00014833154523181095,
697
- "loss": 0.0347,
698
- "step": 57500
699
- },
700
- {
701
- "epoch": 0.89,
702
- "learning_rate": 0.00014828001877107845,
703
- "loss": 0.0346,
704
- "step": 58000
705
- },
706
- {
707
- "epoch": 0.9,
708
- "learning_rate": 0.00014822771827140065,
709
- "loss": 0.0345,
710
- "step": 58500
711
- },
712
- {
713
- "epoch": 0.9,
714
- "learning_rate": 0.00014817464430472743,
715
- "loss": 0.0345,
716
- "step": 59000
717
- },
718
- {
719
- "epoch": 0.91,
720
- "learning_rate": 0.00014812079745146715,
721
- "loss": 0.0344,
722
- "step": 59500
723
- },
724
- {
725
- "epoch": 0.92,
726
- "learning_rate": 0.00014806617830048042,
727
- "loss": 0.0343,
728
- "step": 60000
729
- },
730
- {
731
- "epoch": 0.93,
732
- "learning_rate": 0.00014801078744907346,
733
- "loss": 0.0343,
734
- "step": 60500
735
- },
736
- {
737
- "epoch": 0.93,
738
- "learning_rate": 0.00014795462550299182,
739
- "loss": 0.0342,
740
- "step": 61000
741
- },
742
- {
743
- "epoch": 0.94,
744
- "learning_rate": 0.0001478976930764135,
745
- "loss": 0.0341,
746
- "step": 61500
747
- },
748
- {
749
- "epoch": 0.95,
750
- "learning_rate": 0.0001478399907919424,
751
- "loss": 0.0341,
752
- "step": 62000
753
- },
754
- {
755
- "epoch": 0.96,
756
- "learning_rate": 0.00014778151928060144,
757
- "loss": 0.034,
758
- "step": 62500
759
- },
760
- {
761
- "epoch": 0.97,
762
- "learning_rate": 0.00014772227918182575,
763
- "loss": 0.034,
764
- "step": 63000
765
- },
766
- {
767
- "epoch": 0.97,
768
- "learning_rate": 0.00014766227114345552,
769
- "loss": 0.0339,
770
- "step": 63500
771
- },
772
- {
773
- "epoch": 0.98,
774
- "learning_rate": 0.00014760149582172905,
775
- "loss": 0.0338,
776
- "step": 64000
777
- },
778
- {
779
- "epoch": 0.99,
780
- "learning_rate": 0.00014753995388127553,
781
- "loss": 0.0338,
782
- "step": 64500
783
- },
784
- {
785
- "epoch": 1.0,
786
- "learning_rate": 0.00014747764599510777,
787
- "loss": 0.0337,
788
- "step": 65000
789
- },
790
- {
791
- "epoch": 1.0,
792
- "learning_rate": 0.0001474145728446148,
793
- "loss": 0.0337,
794
- "step": 65500
795
- },
796
- {
797
- "epoch": 1.01,
798
- "learning_rate": 0.00014735073511955453,
799
- "loss": 0.0336,
800
- "step": 66000
801
- },
802
- {
803
- "epoch": 1.02,
804
- "learning_rate": 0.00014728613351804604,
805
- "loss": 0.0336,
806
- "step": 66500
807
- },
808
- {
809
- "epoch": 1.03,
810
- "learning_rate": 0.00014722076874656214,
811
- "loss": 0.0335,
812
- "step": 67000
813
- },
814
- {
815
- "epoch": 1.03,
816
- "learning_rate": 0.00014715464151992148,
817
- "loss": 0.0335,
818
- "step": 67500
819
- },
820
- {
821
- "epoch": 1.04,
822
- "learning_rate": 0.0001470877525612808,
823
- "loss": 0.0334,
824
- "step": 68000
825
- },
826
- {
827
- "epoch": 1.05,
828
- "learning_rate": 0.00014702010260212708,
829
- "loss": 0.0334,
830
- "step": 68500
831
- },
832
- {
833
- "epoch": 1.06,
834
- "learning_rate": 0.0001469516923822694,
835
- "loss": 0.0333,
836
- "step": 69000
837
- },
838
- {
839
- "epoch": 1.06,
840
- "learning_rate": 0.00014688252264983097,
841
- "loss": 0.0333,
842
- "step": 69500
843
- },
844
- {
845
- "epoch": 1.07,
846
- "learning_rate": 0.00014681259416124092,
847
- "loss": 0.0332,
848
- "step": 70000
849
- },
850
- {
851
- "epoch": 1.08,
852
- "learning_rate": 0.00014674190768122595,
853
- "loss": 0.0332,
854
- "step": 70500
855
- },
856
- {
857
- "epoch": 1.09,
858
- "learning_rate": 0.00014667046398280213,
859
- "loss": 0.0331,
860
- "step": 71000
861
- },
862
- {
863
- "epoch": 1.1,
864
- "learning_rate": 0.0001465982638472663,
865
- "loss": 0.0331,
866
- "step": 71500
867
- },
868
- {
869
- "epoch": 1.1,
870
- "learning_rate": 0.00014652530806418755,
871
- "loss": 0.033,
872
- "step": 72000
873
- },
874
- {
875
- "epoch": 1.11,
876
- "learning_rate": 0.00014645159743139862,
877
- "loss": 0.033,
878
- "step": 72500
879
- },
880
- {
881
- "epoch": 1.12,
882
- "learning_rate": 0.0001463771327549872,
883
- "loss": 0.033,
884
- "step": 73000
885
- },
886
- {
887
- "epoch": 1.13,
888
- "learning_rate": 0.00014630191484928709,
889
- "loss": 0.0329,
890
- "step": 73500
891
- },
892
- {
893
- "epoch": 1.13,
894
- "learning_rate": 0.00014622594453686923,
895
- "loss": 0.0329,
896
- "step": 74000
897
- },
898
- {
899
- "epoch": 1.14,
900
- "learning_rate": 0.00014614922264853276,
901
- "loss": 0.0328,
902
- "step": 74500
903
- },
904
- {
905
- "epoch": 1.15,
906
- "learning_rate": 0.000146071750023296,
907
- "loss": 0.0328,
908
- "step": 75000
909
- },
910
- {
911
- "epoch": 1.16,
912
- "learning_rate": 0.00014599352750838715,
913
- "loss": 0.0327,
914
- "step": 75500
915
- },
916
- {
917
- "epoch": 1.16,
918
- "learning_rate": 0.0001459145559592351,
919
- "loss": 0.0327,
920
- "step": 76000
921
- },
922
- {
923
- "epoch": 1.17,
924
- "learning_rate": 0.00014583483623946004,
925
- "loss": 0.0326,
926
- "step": 76500
927
- },
928
- {
929
- "epoch": 1.18,
930
- "learning_rate": 0.00014575436922086412,
931
- "loss": 0.0326,
932
- "step": 77000
933
- },
934
- {
935
- "epoch": 1.19,
936
- "learning_rate": 0.00014567315578342167,
937
- "loss": 0.0325,
938
- "step": 77500
939
- },
940
- {
941
- "epoch": 1.19,
942
- "learning_rate": 0.0001455911968152699,
943
- "loss": 0.0325,
944
- "step": 78000
945
- },
946
- {
947
- "epoch": 1.2,
948
- "learning_rate": 0.00014550849321269896,
949
- "loss": 0.0324,
950
- "step": 78500
951
- },
952
- {
953
- "epoch": 1.21,
954
- "learning_rate": 0.0001454250458801421,
955
- "loss": 0.0323,
956
- "step": 79000
957
- },
958
- {
959
- "epoch": 1.22,
960
- "learning_rate": 0.00014534085573016613,
961
- "loss": 0.0323,
962
- "step": 79500
963
- },
964
- {
965
- "epoch": 1.23,
966
- "learning_rate": 0.0001452559236834609,
967
- "loss": 0.0323,
968
- "step": 80000
969
- },
970
- {
971
- "epoch": 1.23,
972
- "learning_rate": 0.00014517025066882974,
973
- "loss": 0.0322,
974
- "step": 80500
975
- },
976
- {
977
- "epoch": 1.24,
978
- "learning_rate": 0.00014508383762317902,
979
- "loss": 0.0321,
980
- "step": 81000
981
- },
982
- {
983
- "epoch": 1.25,
984
- "learning_rate": 0.00014499668549150792,
985
- "loss": 0.0321,
986
- "step": 81500
987
- },
988
- {
989
- "epoch": 1.26,
990
- "learning_rate": 0.0001449087952268982,
991
- "loss": 0.0321,
992
- "step": 82000
993
- },
994
- {
995
- "epoch": 1.26,
996
- "learning_rate": 0.00014482016779050375,
997
- "loss": 0.032,
998
- "step": 82500
999
- },
1000
- {
1001
- "epoch": 1.27,
1002
- "learning_rate": 0.00014473080415153999,
1003
- "loss": 0.032,
1004
- "step": 83000
1005
- },
1006
- {
1007
- "epoch": 1.28,
1008
- "learning_rate": 0.00014464070528727332,
1009
- "loss": 0.0319,
1010
- "step": 83500
1011
- },
1012
- {
1013
- "epoch": 1.29,
1014
- "learning_rate": 0.00014454987218301055,
1015
- "loss": 0.0319,
1016
- "step": 84000
1017
- },
1018
- {
1019
- "epoch": 1.29,
1020
- "learning_rate": 0.00014445830583208793,
1021
- "loss": 0.0318,
1022
- "step": 84500
1023
- },
1024
- {
1025
- "epoch": 1.3,
1026
- "learning_rate": 0.00014436600723586037,
1027
- "loss": 0.0318,
1028
- "step": 85000
1029
- },
1030
- {
1031
- "epoch": 1.31,
1032
- "learning_rate": 0.0001442729774036906,
1033
- "loss": 0.0317,
1034
- "step": 85500
1035
- },
1036
- {
1037
- "epoch": 1.32,
1038
- "learning_rate": 0.0001441792173529379,
1039
- "loss": 0.0317,
1040
- "step": 86000
1041
- },
1042
- {
1043
- "epoch": 1.33,
1044
- "learning_rate": 0.00014408472810894718,
1045
- "loss": 0.0316,
1046
- "step": 86500
1047
- },
1048
- {
1049
- "epoch": 1.33,
1050
- "learning_rate": 0.00014398951070503766,
1051
- "loss": 0.0315,
1052
- "step": 87000
1053
- },
1054
- {
1055
- "epoch": 1.34,
1056
- "learning_rate": 0.00014389356618249165,
1057
- "loss": 0.0314,
1058
- "step": 87500
1059
- },
1060
- {
1061
- "epoch": 1.35,
1062
- "learning_rate": 0.00014379689559054304,
1063
- "loss": 0.0312,
1064
- "step": 88000
1065
- },
1066
- {
1067
- "epoch": 1.36,
1068
- "learning_rate": 0.000143699499986366,
1069
- "loss": 0.0311,
1070
- "step": 88500
1071
- },
1072
- {
1073
- "epoch": 1.36,
1074
- "learning_rate": 0.00014360138043506318,
1075
- "loss": 0.031,
1076
- "step": 89000
1077
- },
1078
- {
1079
- "epoch": 1.37,
1080
- "learning_rate": 0.0001435025380096543,
1081
- "loss": 0.031,
1082
- "step": 89500
1083
- },
1084
- {
1085
- "epoch": 1.38,
1086
- "learning_rate": 0.00014340297379106427,
1087
- "loss": 0.031,
1088
- "step": 90000
1089
- },
1090
- {
1091
- "epoch": 1.39,
1092
- "learning_rate": 0.00014330268886811147,
1093
- "loss": 0.0309,
1094
- "step": 90500
1095
- },
1096
- {
1097
- "epoch": 1.39,
1098
- "learning_rate": 0.00014320168433749572,
1099
- "loss": 0.0308,
1100
- "step": 91000
1101
- },
1102
- {
1103
- "epoch": 1.4,
1104
- "learning_rate": 0.0001430999613037864,
1105
- "loss": 0.0308,
1106
- "step": 91500
1107
- },
1108
- {
1109
- "epoch": 1.41,
1110
- "learning_rate": 0.0001429975208794103,
1111
- "loss": 0.0308,
1112
- "step": 92000
1113
- },
1114
- {
1115
- "epoch": 1.42,
1116
- "learning_rate": 0.00014289436418463952,
1117
- "loss": 0.0307,
1118
- "step": 92500
1119
- },
1120
- {
1121
- "epoch": 1.42,
1122
- "learning_rate": 0.0001427904923475791,
1123
- "loss": 0.0307,
1124
- "step": 93000
1125
- },
1126
- {
1127
- "epoch": 1.43,
1128
- "learning_rate": 0.00014268590650415482,
1129
- "loss": 0.0306,
1130
- "step": 93500
1131
- },
1132
- {
1133
- "epoch": 1.44,
1134
- "learning_rate": 0.0001425806077981007,
1135
- "loss": 0.0306,
1136
- "step": 94000
1137
- },
1138
- {
1139
- "epoch": 1.45,
1140
- "learning_rate": 0.0001424745973809466,
1141
- "loss": 0.0305,
1142
- "step": 94500
1143
- },
1144
- {
1145
- "epoch": 1.46,
1146
- "learning_rate": 0.00014236787641200535,
1147
- "loss": 0.0305,
1148
- "step": 95000
1149
- },
1150
- {
1151
- "epoch": 1.46,
1152
- "learning_rate": 0.0001422604460583604,
1153
- "loss": 0.0304,
1154
- "step": 95500
1155
- },
1156
- {
1157
- "epoch": 1.47,
1158
- "learning_rate": 0.00014215230749485292,
1159
- "loss": 0.0304,
1160
- "step": 96000
1161
- },
1162
- {
1163
- "epoch": 1.48,
1164
- "learning_rate": 0.00014204346190406888,
1165
- "loss": 0.0303,
1166
- "step": 96500
1167
- },
1168
- {
1169
- "epoch": 1.49,
1170
- "learning_rate": 0.00014193391047632623,
1171
- "loss": 0.0303,
1172
- "step": 97000
1173
- },
1174
- {
1175
- "epoch": 1.49,
1176
- "learning_rate": 0.00014182365440966184,
1177
- "loss": 0.0303,
1178
- "step": 97500
1179
- },
1180
- {
1181
- "epoch": 1.5,
1182
- "learning_rate": 0.00014171269490981838,
1183
- "loss": 0.0302,
1184
- "step": 98000
1185
- },
1186
- {
1187
- "epoch": 1.51,
1188
- "learning_rate": 0.0001416010331902312,
1189
- "loss": 0.0302,
1190
- "step": 98500
1191
- },
1192
- {
1193
- "epoch": 1.52,
1194
- "learning_rate": 0.00014148867047201493,
1195
- "loss": 0.0302,
1196
- "step": 99000
1197
- },
1198
- {
1199
- "epoch": 1.52,
1200
- "learning_rate": 0.00014137560798395025,
1201
- "loss": 0.0301,
1202
- "step": 99500
1203
- },
1204
- {
1205
- "epoch": 1.53,
1206
- "learning_rate": 0.00014126184696247043,
1207
- "loss": 0.0301,
1208
- "step": 100000
1209
- },
1210
- {
1211
- "epoch": 1.54,
1212
- "learning_rate": 0.0001411473886516478,
1213
- "loss": 0.03,
1214
- "step": 100500
1215
- },
1216
- {
1217
- "epoch": 1.55,
1218
- "learning_rate": 0.0001410322343031801,
1219
- "loss": 0.03,
1220
- "step": 101000
1221
- },
1222
- {
1223
- "epoch": 1.55,
1224
- "learning_rate": 0.00014091638517637681,
1225
- "loss": 0.03,
1226
- "step": 101500
1227
- },
1228
- {
1229
- "epoch": 1.56,
1230
- "learning_rate": 0.00014079984253814544,
1231
- "loss": 0.03,
1232
- "step": 102000
1233
- },
1234
- {
1235
- "epoch": 1.57,
1236
- "learning_rate": 0.00014068260766297757,
1237
- "loss": 0.0299,
1238
- "step": 102500
1239
- },
1240
- {
1241
- "epoch": 1.58,
1242
- "learning_rate": 0.00014056468183293511,
1243
- "loss": 0.0299,
1244
- "step": 103000
1245
- },
1246
- {
1247
- "epoch": 1.59,
1248
- "learning_rate": 0.00014044606633763592,
1249
- "loss": 0.0298,
1250
- "step": 103500
1251
- },
1252
- {
1253
- "epoch": 1.59,
1254
- "learning_rate": 0.00014032676247424016,
1255
- "loss": 0.0298,
1256
- "step": 104000
1257
- },
1258
- {
1259
- "epoch": 1.6,
1260
- "learning_rate": 0.00014020677154743567,
1261
- "loss": 0.0298,
1262
- "step": 104500
1263
- },
1264
- {
1265
- "epoch": 1.61,
1266
- "learning_rate": 0.0001400860948694241,
1267
- "loss": 0.0298,
1268
- "step": 105000
1269
- },
1270
- {
1271
- "epoch": 1.62,
1272
- "learning_rate": 0.0001399647337599062,
1273
- "loss": 0.0297,
1274
- "step": 105500
1275
- },
1276
- {
1277
- "epoch": 1.62,
1278
- "learning_rate": 0.00013984268954606762,
1279
- "loss": 0.0297,
1280
- "step": 106000
1281
- },
1282
- {
1283
- "epoch": 1.63,
1284
- "learning_rate": 0.00013971996356256435,
1285
- "loss": 0.0297,
1286
- "step": 106500
1287
- },
1288
- {
1289
- "epoch": 1.64,
1290
- "learning_rate": 0.0001395965571515081,
1291
- "loss": 0.0296,
1292
- "step": 107000
1293
- },
1294
- {
1295
- "epoch": 1.65,
1296
- "learning_rate": 0.00013947247166245158,
1297
- "loss": 0.0296,
1298
- "step": 107500
1299
- },
1300
- {
1301
- "epoch": 1.65,
1302
- "learning_rate": 0.0001393477084523738,
1303
- "loss": 0.0296,
1304
- "step": 108000
1305
- },
1306
- {
1307
- "epoch": 1.66,
1308
- "learning_rate": 0.0001392222688856653,
1309
- "loss": 0.0296,
1310
- "step": 108500
1311
- },
1312
- {
1313
- "epoch": 1.67,
1314
- "learning_rate": 0.00013909615433411305,
1315
- "loss": 0.0295,
1316
- "step": 109000
1317
- },
1318
- {
1319
- "epoch": 1.68,
1320
- "learning_rate": 0.00013896936617688554,
1321
- "loss": 0.0295,
1322
- "step": 109500
1323
- },
1324
- {
1325
- "epoch": 1.69,
1326
- "learning_rate": 0.0001388419058005178,
1327
- "loss": 0.0294,
1328
- "step": 110000
1329
- },
1330
- {
1331
- "epoch": 1.69,
1332
- "learning_rate": 0.0001387137745988961,
1333
- "loss": 0.0294,
1334
- "step": 110500
1335
- },
1336
- {
1337
- "epoch": 1.7,
1338
- "learning_rate": 0.00013858497397324266,
1339
- "loss": 0.0294,
1340
- "step": 111000
1341
- },
1342
- {
1343
- "epoch": 1.71,
1344
- "learning_rate": 0.00013845550533210052,
1345
- "loss": 0.0293,
1346
- "step": 111500
1347
- },
1348
- {
1349
- "epoch": 1.72,
1350
- "learning_rate": 0.00013832537009131804,
1351
- "loss": 0.0293,
1352
- "step": 112000
1353
- },
1354
- {
1355
- "epoch": 1.72,
1356
- "learning_rate": 0.0001381945696740333,
1357
- "loss": 0.0293,
1358
- "step": 112500
1359
- },
1360
- {
1361
- "epoch": 1.73,
1362
- "learning_rate": 0.00013806310551065876,
1363
- "loss": 0.0293,
1364
- "step": 113000
1365
- },
1366
- {
1367
- "epoch": 1.74,
1368
- "learning_rate": 0.00013793097903886542,
1369
- "loss": 0.0293,
1370
- "step": 113500
1371
- },
1372
- {
1373
- "epoch": 1.75,
1374
- "learning_rate": 0.00013779819170356718,
1375
- "loss": 0.0292,
1376
- "step": 114000
1377
- },
1378
- {
1379
- "epoch": 1.75,
1380
- "learning_rate": 0.00013766474495690513,
1381
- "loss": 0.0292,
1382
- "step": 114500
1383
- },
1384
- {
1385
- "epoch": 1.76,
1386
- "learning_rate": 0.00013753064025823143,
1387
- "loss": 0.0291,
1388
- "step": 115000
1389
- },
1390
- {
1391
- "epoch": 1.77,
1392
- "learning_rate": 0.00013739587907409366,
1393
- "loss": 0.0291,
1394
- "step": 115500
1395
- },
1396
- {
1397
- "epoch": 1.78,
1398
- "learning_rate": 0.0001372604628782185,
1399
- "loss": 0.0291,
1400
- "step": 116000
1401
- },
1402
- {
1403
- "epoch": 1.78,
1404
- "learning_rate": 0.00013712439315149579,
1405
- "loss": 0.0291,
1406
- "step": 116500
1407
- },
1408
- {
1409
- "epoch": 1.79,
1410
- "learning_rate": 0.0001369876713819622,
1411
- "loss": 0.029,
1412
- "step": 117000
1413
- },
1414
- {
1415
- "epoch": 1.8,
1416
- "learning_rate": 0.0001368502990647852,
1417
- "loss": 0.0291,
1418
- "step": 117500
1419
- },
1420
- {
1421
- "epoch": 1.81,
1422
- "learning_rate": 0.00013671227770224636,
1423
- "loss": 0.029,
1424
- "step": 118000
1425
- },
1426
- {
1427
- "epoch": 1.82,
1428
- "learning_rate": 0.0001365736088037252,
1429
- "loss": 0.029,
1430
- "step": 118500
1431
- },
1432
- {
1433
- "epoch": 1.82,
1434
- "learning_rate": 0.00013643429388568264,
1435
- "loss": 0.029,
1436
- "step": 119000
1437
- },
1438
- {
1439
- "epoch": 1.83,
1440
- "learning_rate": 0.0001362943344716442,
1441
- "loss": 0.0289,
1442
- "step": 119500
1443
- },
1444
- {
1445
- "epoch": 1.84,
1446
- "learning_rate": 0.0001361537320921837,
1447
- "loss": 0.0289,
1448
- "step": 120000
1449
- },
1450
- {
1451
- "epoch": 1.85,
1452
- "learning_rate": 0.00013601248828490616,
1453
- "loss": 0.0289,
1454
- "step": 120500
1455
- },
1456
- {
1457
- "epoch": 1.85,
1458
- "learning_rate": 0.0001358706045944313,
1459
- "loss": 0.0289,
1460
- "step": 121000
1461
- },
1462
- {
1463
- "epoch": 1.86,
1464
- "learning_rate": 0.00013572808257237637,
1465
- "loss": 0.0289,
1466
- "step": 121500
1467
- },
1468
- {
1469
- "epoch": 1.87,
1470
- "learning_rate": 0.0001355849237773394,
1471
- "loss": 0.0288,
1472
- "step": 122000
1473
- },
1474
- {
1475
- "epoch": 1.88,
1476
- "learning_rate": 0.00013544112977488204,
1477
- "loss": 0.0288,
1478
- "step": 122500
1479
- },
1480
- {
1481
- "epoch": 1.88,
1482
- "learning_rate": 0.00013529670213751244,
1483
- "loss": 0.0288,
1484
- "step": 123000
1485
- },
1486
- {
1487
- "epoch": 1.89,
1488
- "learning_rate": 0.00013515164244466812,
1489
- "loss": 0.0287,
1490
- "step": 123500
1491
- },
1492
- {
1493
- "epoch": 1.9,
1494
- "learning_rate": 0.00013500595228269863,
1495
- "loss": 0.0287,
1496
- "step": 124000
1497
- },
1498
- {
1499
- "epoch": 1.91,
1500
- "learning_rate": 0.00013485963324484825,
1501
- "loss": 0.0287,
1502
- "step": 124500
1503
- },
1504
- {
1505
- "epoch": 1.91,
1506
- "learning_rate": 0.00013471268693123853,
1507
- "loss": 0.0287,
1508
- "step": 125000
1509
  }
1510
  ],
1511
  "max_steps": 500000,
1512
  "num_train_epochs": 8,
1513
- "total_flos": 1.996802604193089e+21,
1514
  "trial_name": null,
1515
  "trial_params": null
1516
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.07659607062157711,
5
+ "global_step": 5000,
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": 3.0000000000000004e-07,
13
+ "loss": 0.8817,
14
  "step": 500
15
  },
16
  {
17
  "epoch": 0.02,
18
+ "learning_rate": 6.000000000000001e-07,
19
+ "loss": 0.7274,
20
  "step": 1000
21
  },
22
  {
23
  "epoch": 0.02,
24
+ "learning_rate": 9e-07,
25
+ "loss": 0.5446,
26
  "step": 1500
27
  },
28
  {
29
  "epoch": 0.03,
30
+ "learning_rate": 1.2000000000000002e-06,
31
+ "loss": 0.3974,
32
  "step": 2000
33
  },
34
  {
35
  "epoch": 0.04,
36
+ "learning_rate": 1.5e-06,
37
+ "loss": 0.2731,
38
  "step": 2500
39
  },
40
  {
41
  "epoch": 0.05,
42
+ "learning_rate": 1.8e-06,
43
+ "loss": 0.1759,
44
  "step": 3000
45
  },
46
  {
47
  "epoch": 0.05,
48
+ "learning_rate": 2.1000000000000002e-06,
49
+ "loss": 0.112,
50
  "step": 3500
51
  },
52
  {
53
  "epoch": 0.06,
54
+ "learning_rate": 2.4000000000000003e-06,
55
+ "loss": 0.0785,
56
  "step": 4000
57
  },
58
  {
59
  "epoch": 0.07,
60
+ "learning_rate": 2.7e-06,
61
+ "loss": 0.0652,
62
  "step": 4500
63
  },
64
  {
65
  "epoch": 0.08,
66
+ "learning_rate": 3e-06,
67
+ "loss": 0.0614,
68
  "step": 5000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
69
  }
70
  ],
71
  "max_steps": 500000,
72
  "num_train_epochs": 8,
73
+ "total_flos": 7.9871784693625e+19,
74
  "trial_name": null,
75
  "trial_params": null
76
  }
last-checkpoint/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:05fc0c551198ca4907e70bb4b0044b0f6f08d72a20ab8c2f1ed25d998371c8a9
3
  size 3247
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab366a33e91138d5d6c5aa5a082b4d18493ea68b9488d5d44ed4bcabf1978754
3
  size 3247
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:01986e29e7b00ff5ed5012297e50124030c090218d489d0fb41425e1a13097b1
3
  size 102501541
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e31313ef25632d40f04a4d20bebc02ef847b472d4200e8e9085e3e9835ffafaa
3
  size 102501541
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:05fc0c551198ca4907e70bb4b0044b0f6f08d72a20ab8c2f1ed25d998371c8a9
3
  size 3247
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab366a33e91138d5d6c5aa5a082b4d18493ea68b9488d5d44ed4bcabf1978754
3
  size 3247