TomPei commited on
Commit
ba010c6
·
verified ·
1 Parent(s): 1b1ad6e

Delete trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +0 -1221
trainer_state.json DELETED
@@ -1,1221 +0,0 @@
1
- {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 1.3071895424836601,
5
- "eval_steps": 500,
6
- "global_step": 200,
7
- "is_hyper_param_search": false,
8
- "is_local_process_zero": true,
9
- "is_world_process_zero": true,
10
- "log_history": [
11
- {
12
- "epoch": 0.01,
13
- "learning_rate": 0.0,
14
- "loss": 0.5341,
15
- "step": 1
16
- },
17
- {
18
- "epoch": 0.01,
19
- "learning_rate": 4.075900941810124e-06,
20
- "loss": 0.4805,
21
- "step": 2
22
- },
23
- {
24
- "epoch": 0.02,
25
- "learning_rate": 6.46015014942309e-06,
26
- "loss": 0.4839,
27
- "step": 3
28
- },
29
- {
30
- "epoch": 0.03,
31
- "learning_rate": 8.151801883620247e-06,
32
- "loss": 0.4227,
33
- "step": 4
34
- },
35
- {
36
- "epoch": 0.03,
37
- "learning_rate": 9.463948908766788e-06,
38
- "loss": 0.4289,
39
- "step": 5
40
- },
41
- {
42
- "epoch": 0.04,
43
- "learning_rate": 1.0536051091233212e-05,
44
- "loss": 0.3633,
45
- "step": 6
46
- },
47
- {
48
- "epoch": 0.05,
49
- "learning_rate": 1.1442500570809876e-05,
50
- "loss": 0.3749,
51
- "step": 7
52
- },
53
- {
54
- "epoch": 0.05,
55
- "learning_rate": 1.222770282543037e-05,
56
- "loss": 0.3379,
57
- "step": 8
58
- },
59
- {
60
- "epoch": 0.06,
61
- "learning_rate": 1.292030029884618e-05,
62
- "loss": 0.3637,
63
- "step": 9
64
- },
65
- {
66
- "epoch": 0.07,
67
- "learning_rate": 1.3539849850576912e-05,
68
- "loss": 0.3716,
69
- "step": 10
70
- },
71
- {
72
- "epoch": 0.07,
73
- "learning_rate": 1.4100300592531481e-05,
74
- "loss": 0.3544,
75
- "step": 11
76
- },
77
- {
78
- "epoch": 0.08,
79
- "learning_rate": 1.4611952033043337e-05,
80
- "loss": 0.3463,
81
- "step": 12
82
- },
83
- {
84
- "epoch": 0.08,
85
- "learning_rate": 1.5082625732282867e-05,
86
- "loss": 0.3121,
87
- "step": 13
88
- },
89
- {
90
- "epoch": 0.09,
91
- "learning_rate": 1.551840151262e-05,
92
- "loss": 0.3364,
93
- "step": 14
94
- },
95
- {
96
- "epoch": 0.1,
97
- "learning_rate": 1.5924099058189875e-05,
98
- "loss": 0.3259,
99
- "step": 15
100
- },
101
- {
102
- "epoch": 0.1,
103
- "learning_rate": 1.6303603767240495e-05,
104
- "loss": 0.3348,
105
- "step": 16
106
- },
107
- {
108
- "epoch": 0.11,
109
- "learning_rate": 1.6660093644266146e-05,
110
- "loss": 0.3125,
111
- "step": 17
112
- },
113
- {
114
- "epoch": 0.12,
115
- "learning_rate": 1.6996201240656302e-05,
116
- "loss": 0.3377,
117
- "step": 18
118
- },
119
- {
120
- "epoch": 0.12,
121
- "learning_rate": 1.7314131752785847e-05,
122
- "loss": 0.3086,
123
- "step": 19
124
- },
125
- {
126
- "epoch": 0.13,
127
- "learning_rate": 1.7615750792387035e-05,
128
- "loss": 0.3012,
129
- "step": 20
130
- },
131
- {
132
- "epoch": 0.14,
133
- "learning_rate": 1.7902650720232966e-05,
134
- "loss": 0.3212,
135
- "step": 21
136
- },
137
- {
138
- "epoch": 0.14,
139
- "learning_rate": 1.8176201534341607e-05,
140
- "loss": 0.3179,
141
- "step": 22
142
- },
143
- {
144
- "epoch": 0.15,
145
- "learning_rate": 1.8437590437029225e-05,
146
- "loss": 0.3196,
147
- "step": 23
148
- },
149
- {
150
- "epoch": 0.16,
151
- "learning_rate": 1.868785297485346e-05,
152
- "loss": 0.3175,
153
- "step": 24
154
- },
155
- {
156
- "epoch": 0.16,
157
- "learning_rate": 1.8927897817533575e-05,
158
- "loss": 0.3312,
159
- "step": 25
160
- },
161
- {
162
- "epoch": 0.17,
163
- "learning_rate": 1.915852667409299e-05,
164
- "loss": 0.2915,
165
- "step": 26
166
- },
167
- {
168
- "epoch": 0.18,
169
- "learning_rate": 1.9380450448269272e-05,
170
- "loss": 0.3208,
171
- "step": 27
172
- },
173
- {
174
- "epoch": 0.18,
175
- "learning_rate": 1.9594302454430122e-05,
176
- "loss": 0.2906,
177
- "step": 28
178
- },
179
- {
180
- "epoch": 0.19,
181
- "learning_rate": 1.9800649313336155e-05,
182
- "loss": 0.2998,
183
- "step": 29
184
- },
185
- {
186
- "epoch": 0.2,
187
- "learning_rate": 2e-05,
188
- "loss": 0.3057,
189
- "step": 30
190
- },
191
- {
192
- "epoch": 0.2,
193
- "learning_rate": 2e-05,
194
- "loss": 0.2943,
195
- "step": 31
196
- },
197
- {
198
- "epoch": 0.21,
199
- "learning_rate": 2e-05,
200
- "loss": 0.3099,
201
- "step": 32
202
- },
203
- {
204
- "epoch": 0.22,
205
- "learning_rate": 2e-05,
206
- "loss": 0.2935,
207
- "step": 33
208
- },
209
- {
210
- "epoch": 0.22,
211
- "learning_rate": 2e-05,
212
- "loss": 0.2903,
213
- "step": 34
214
- },
215
- {
216
- "epoch": 0.23,
217
- "learning_rate": 2e-05,
218
- "loss": 0.2782,
219
- "step": 35
220
- },
221
- {
222
- "epoch": 0.24,
223
- "learning_rate": 2e-05,
224
- "loss": 0.3099,
225
- "step": 36
226
- },
227
- {
228
- "epoch": 0.24,
229
- "learning_rate": 2e-05,
230
- "loss": 0.2966,
231
- "step": 37
232
- },
233
- {
234
- "epoch": 0.25,
235
- "learning_rate": 2e-05,
236
- "loss": 0.2939,
237
- "step": 38
238
- },
239
- {
240
- "epoch": 0.25,
241
- "learning_rate": 2e-05,
242
- "loss": 0.2949,
243
- "step": 39
244
- },
245
- {
246
- "epoch": 0.26,
247
- "learning_rate": 2e-05,
248
- "loss": 0.2842,
249
- "step": 40
250
- },
251
- {
252
- "epoch": 0.27,
253
- "learning_rate": 2e-05,
254
- "loss": 0.279,
255
- "step": 41
256
- },
257
- {
258
- "epoch": 0.27,
259
- "learning_rate": 2e-05,
260
- "loss": 0.3078,
261
- "step": 42
262
- },
263
- {
264
- "epoch": 0.28,
265
- "learning_rate": 2e-05,
266
- "loss": 0.3019,
267
- "step": 43
268
- },
269
- {
270
- "epoch": 0.29,
271
- "learning_rate": 2e-05,
272
- "loss": 0.3208,
273
- "step": 44
274
- },
275
- {
276
- "epoch": 0.29,
277
- "learning_rate": 2e-05,
278
- "loss": 0.312,
279
- "step": 45
280
- },
281
- {
282
- "epoch": 0.3,
283
- "learning_rate": 2e-05,
284
- "loss": 0.2994,
285
- "step": 46
286
- },
287
- {
288
- "epoch": 0.31,
289
- "learning_rate": 2e-05,
290
- "loss": 0.3032,
291
- "step": 47
292
- },
293
- {
294
- "epoch": 0.31,
295
- "learning_rate": 2e-05,
296
- "loss": 0.3176,
297
- "step": 48
298
- },
299
- {
300
- "epoch": 0.32,
301
- "learning_rate": 2e-05,
302
- "loss": 0.3116,
303
- "step": 49
304
- },
305
- {
306
- "epoch": 0.33,
307
- "learning_rate": 2e-05,
308
- "loss": 0.3035,
309
- "step": 50
310
- },
311
- {
312
- "epoch": 0.33,
313
- "learning_rate": 2e-05,
314
- "loss": 0.3081,
315
- "step": 51
316
- },
317
- {
318
- "epoch": 0.34,
319
- "learning_rate": 2e-05,
320
- "loss": 0.2975,
321
- "step": 52
322
- },
323
- {
324
- "epoch": 0.35,
325
- "learning_rate": 2e-05,
326
- "loss": 0.2893,
327
- "step": 53
328
- },
329
- {
330
- "epoch": 0.35,
331
- "learning_rate": 2e-05,
332
- "loss": 0.2919,
333
- "step": 54
334
- },
335
- {
336
- "epoch": 0.36,
337
- "learning_rate": 2e-05,
338
- "loss": 0.2922,
339
- "step": 55
340
- },
341
- {
342
- "epoch": 0.37,
343
- "learning_rate": 2e-05,
344
- "loss": 0.3126,
345
- "step": 56
346
- },
347
- {
348
- "epoch": 0.37,
349
- "learning_rate": 2e-05,
350
- "loss": 0.2758,
351
- "step": 57
352
- },
353
- {
354
- "epoch": 0.38,
355
- "learning_rate": 2e-05,
356
- "loss": 0.2856,
357
- "step": 58
358
- },
359
- {
360
- "epoch": 0.39,
361
- "learning_rate": 2e-05,
362
- "loss": 0.2936,
363
- "step": 59
364
- },
365
- {
366
- "epoch": 0.39,
367
- "learning_rate": 2e-05,
368
- "loss": 0.2778,
369
- "step": 60
370
- },
371
- {
372
- "epoch": 0.4,
373
- "learning_rate": 2e-05,
374
- "loss": 0.2777,
375
- "step": 61
376
- },
377
- {
378
- "epoch": 0.41,
379
- "learning_rate": 2e-05,
380
- "loss": 0.2757,
381
- "step": 62
382
- },
383
- {
384
- "epoch": 0.41,
385
- "learning_rate": 2e-05,
386
- "loss": 0.2961,
387
- "step": 63
388
- },
389
- {
390
- "epoch": 0.42,
391
- "learning_rate": 2e-05,
392
- "loss": 0.2864,
393
- "step": 64
394
- },
395
- {
396
- "epoch": 0.42,
397
- "learning_rate": 2e-05,
398
- "loss": 0.2746,
399
- "step": 65
400
- },
401
- {
402
- "epoch": 0.43,
403
- "learning_rate": 2e-05,
404
- "loss": 0.2824,
405
- "step": 66
406
- },
407
- {
408
- "epoch": 0.44,
409
- "learning_rate": 2e-05,
410
- "loss": 0.2719,
411
- "step": 67
412
- },
413
- {
414
- "epoch": 0.44,
415
- "learning_rate": 2e-05,
416
- "loss": 0.2786,
417
- "step": 68
418
- },
419
- {
420
- "epoch": 0.45,
421
- "learning_rate": 2e-05,
422
- "loss": 0.2739,
423
- "step": 69
424
- },
425
- {
426
- "epoch": 0.46,
427
- "learning_rate": 2e-05,
428
- "loss": 0.2846,
429
- "step": 70
430
- },
431
- {
432
- "epoch": 0.46,
433
- "learning_rate": 2e-05,
434
- "loss": 0.2762,
435
- "step": 71
436
- },
437
- {
438
- "epoch": 0.47,
439
- "learning_rate": 2e-05,
440
- "loss": 0.2834,
441
- "step": 72
442
- },
443
- {
444
- "epoch": 0.48,
445
- "learning_rate": 2e-05,
446
- "loss": 0.2827,
447
- "step": 73
448
- },
449
- {
450
- "epoch": 0.48,
451
- "learning_rate": 2e-05,
452
- "loss": 0.2697,
453
- "step": 74
454
- },
455
- {
456
- "epoch": 0.49,
457
- "learning_rate": 2e-05,
458
- "loss": 0.2873,
459
- "step": 75
460
- },
461
- {
462
- "epoch": 0.5,
463
- "learning_rate": 2e-05,
464
- "loss": 0.2596,
465
- "step": 76
466
- },
467
- {
468
- "epoch": 0.5,
469
- "learning_rate": 2e-05,
470
- "loss": 0.2791,
471
- "step": 77
472
- },
473
- {
474
- "epoch": 0.51,
475
- "learning_rate": 2e-05,
476
- "loss": 0.2683,
477
- "step": 78
478
- },
479
- {
480
- "epoch": 0.52,
481
- "learning_rate": 2e-05,
482
- "loss": 0.2771,
483
- "step": 79
484
- },
485
- {
486
- "epoch": 0.52,
487
- "learning_rate": 2e-05,
488
- "loss": 0.2836,
489
- "step": 80
490
- },
491
- {
492
- "epoch": 0.53,
493
- "learning_rate": 2e-05,
494
- "loss": 0.2656,
495
- "step": 81
496
- },
497
- {
498
- "epoch": 0.54,
499
- "learning_rate": 2e-05,
500
- "loss": 0.2832,
501
- "step": 82
502
- },
503
- {
504
- "epoch": 0.54,
505
- "learning_rate": 2e-05,
506
- "loss": 0.2572,
507
- "step": 83
508
- },
509
- {
510
- "epoch": 0.55,
511
- "learning_rate": 2e-05,
512
- "loss": 0.2691,
513
- "step": 84
514
- },
515
- {
516
- "epoch": 0.56,
517
- "learning_rate": 2e-05,
518
- "loss": 0.2811,
519
- "step": 85
520
- },
521
- {
522
- "epoch": 0.56,
523
- "learning_rate": 2e-05,
524
- "loss": 0.2757,
525
- "step": 86
526
- },
527
- {
528
- "epoch": 0.57,
529
- "learning_rate": 2e-05,
530
- "loss": 0.2752,
531
- "step": 87
532
- },
533
- {
534
- "epoch": 0.58,
535
- "learning_rate": 2e-05,
536
- "loss": 0.275,
537
- "step": 88
538
- },
539
- {
540
- "epoch": 0.58,
541
- "learning_rate": 2e-05,
542
- "loss": 0.2871,
543
- "step": 89
544
- },
545
- {
546
- "epoch": 0.59,
547
- "learning_rate": 2e-05,
548
- "loss": 0.2601,
549
- "step": 90
550
- },
551
- {
552
- "epoch": 0.59,
553
- "learning_rate": 2e-05,
554
- "loss": 0.2832,
555
- "step": 91
556
- },
557
- {
558
- "epoch": 0.6,
559
- "learning_rate": 2e-05,
560
- "loss": 0.2717,
561
- "step": 92
562
- },
563
- {
564
- "epoch": 0.61,
565
- "learning_rate": 2e-05,
566
- "loss": 0.2845,
567
- "step": 93
568
- },
569
- {
570
- "epoch": 0.61,
571
- "learning_rate": 2e-05,
572
- "loss": 0.2825,
573
- "step": 94
574
- },
575
- {
576
- "epoch": 0.62,
577
- "learning_rate": 2e-05,
578
- "loss": 0.2886,
579
- "step": 95
580
- },
581
- {
582
- "epoch": 0.63,
583
- "learning_rate": 2e-05,
584
- "loss": 0.2713,
585
- "step": 96
586
- },
587
- {
588
- "epoch": 0.63,
589
- "learning_rate": 2e-05,
590
- "loss": 0.2771,
591
- "step": 97
592
- },
593
- {
594
- "epoch": 0.64,
595
- "learning_rate": 2e-05,
596
- "loss": 0.2577,
597
- "step": 98
598
- },
599
- {
600
- "epoch": 0.65,
601
- "learning_rate": 2e-05,
602
- "loss": 0.2648,
603
- "step": 99
604
- },
605
- {
606
- "epoch": 0.65,
607
- "learning_rate": 2e-05,
608
- "loss": 0.2794,
609
- "step": 100
610
- },
611
- {
612
- "epoch": 0.66,
613
- "learning_rate": 2e-05,
614
- "loss": 0.2868,
615
- "step": 101
616
- },
617
- {
618
- "epoch": 0.67,
619
- "learning_rate": 2e-05,
620
- "loss": 0.2822,
621
- "step": 102
622
- },
623
- {
624
- "epoch": 0.67,
625
- "learning_rate": 2e-05,
626
- "loss": 0.2802,
627
- "step": 103
628
- },
629
- {
630
- "epoch": 0.68,
631
- "learning_rate": 2e-05,
632
- "loss": 0.2764,
633
- "step": 104
634
- },
635
- {
636
- "epoch": 0.69,
637
- "learning_rate": 2e-05,
638
- "loss": 0.2716,
639
- "step": 105
640
- },
641
- {
642
- "epoch": 0.69,
643
- "learning_rate": 2e-05,
644
- "loss": 0.2679,
645
- "step": 106
646
- },
647
- {
648
- "epoch": 0.7,
649
- "learning_rate": 2e-05,
650
- "loss": 0.2825,
651
- "step": 107
652
- },
653
- {
654
- "epoch": 0.71,
655
- "learning_rate": 2e-05,
656
- "loss": 0.2852,
657
- "step": 108
658
- },
659
- {
660
- "epoch": 0.71,
661
- "learning_rate": 2e-05,
662
- "loss": 0.277,
663
- "step": 109
664
- },
665
- {
666
- "epoch": 0.72,
667
- "learning_rate": 2e-05,
668
- "loss": 0.2666,
669
- "step": 110
670
- },
671
- {
672
- "epoch": 0.73,
673
- "learning_rate": 2e-05,
674
- "loss": 0.2815,
675
- "step": 111
676
- },
677
- {
678
- "epoch": 0.73,
679
- "learning_rate": 2e-05,
680
- "loss": 0.2722,
681
- "step": 112
682
- },
683
- {
684
- "epoch": 0.74,
685
- "learning_rate": 2e-05,
686
- "loss": 0.2648,
687
- "step": 113
688
- },
689
- {
690
- "epoch": 0.75,
691
- "learning_rate": 2e-05,
692
- "loss": 0.2826,
693
- "step": 114
694
- },
695
- {
696
- "epoch": 0.75,
697
- "learning_rate": 2e-05,
698
- "loss": 0.2667,
699
- "step": 115
700
- },
701
- {
702
- "epoch": 0.76,
703
- "learning_rate": 2e-05,
704
- "loss": 0.2724,
705
- "step": 116
706
- },
707
- {
708
- "epoch": 0.76,
709
- "learning_rate": 2e-05,
710
- "loss": 0.2757,
711
- "step": 117
712
- },
713
- {
714
- "epoch": 0.77,
715
- "learning_rate": 2e-05,
716
- "loss": 0.2748,
717
- "step": 118
718
- },
719
- {
720
- "epoch": 0.78,
721
- "learning_rate": 2e-05,
722
- "loss": 0.2708,
723
- "step": 119
724
- },
725
- {
726
- "epoch": 0.78,
727
- "learning_rate": 2e-05,
728
- "loss": 0.2774,
729
- "step": 120
730
- },
731
- {
732
- "epoch": 0.79,
733
- "learning_rate": 2e-05,
734
- "loss": 0.2737,
735
- "step": 121
736
- },
737
- {
738
- "epoch": 0.8,
739
- "learning_rate": 2e-05,
740
- "loss": 0.2603,
741
- "step": 122
742
- },
743
- {
744
- "epoch": 0.8,
745
- "learning_rate": 2e-05,
746
- "loss": 0.2792,
747
- "step": 123
748
- },
749
- {
750
- "epoch": 0.81,
751
- "learning_rate": 2e-05,
752
- "loss": 0.2788,
753
- "step": 124
754
- },
755
- {
756
- "epoch": 0.82,
757
- "learning_rate": 2e-05,
758
- "loss": 0.2803,
759
- "step": 125
760
- },
761
- {
762
- "epoch": 0.82,
763
- "learning_rate": 2e-05,
764
- "loss": 0.2862,
765
- "step": 126
766
- },
767
- {
768
- "epoch": 0.83,
769
- "learning_rate": 2e-05,
770
- "loss": 0.2892,
771
- "step": 127
772
- },
773
- {
774
- "epoch": 0.84,
775
- "learning_rate": 2e-05,
776
- "loss": 0.2794,
777
- "step": 128
778
- },
779
- {
780
- "epoch": 0.84,
781
- "learning_rate": 2e-05,
782
- "loss": 0.2721,
783
- "step": 129
784
- },
785
- {
786
- "epoch": 0.85,
787
- "learning_rate": 2e-05,
788
- "loss": 0.2714,
789
- "step": 130
790
- },
791
- {
792
- "epoch": 0.86,
793
- "learning_rate": 2e-05,
794
- "loss": 0.2693,
795
- "step": 131
796
- },
797
- {
798
- "epoch": 0.86,
799
- "learning_rate": 2e-05,
800
- "loss": 0.2851,
801
- "step": 132
802
- },
803
- {
804
- "epoch": 0.87,
805
- "learning_rate": 2e-05,
806
- "loss": 0.2804,
807
- "step": 133
808
- },
809
- {
810
- "epoch": 0.88,
811
- "learning_rate": 2e-05,
812
- "loss": 0.2772,
813
- "step": 134
814
- },
815
- {
816
- "epoch": 0.88,
817
- "learning_rate": 2e-05,
818
- "loss": 0.271,
819
- "step": 135
820
- },
821
- {
822
- "epoch": 0.89,
823
- "learning_rate": 2e-05,
824
- "loss": 0.2752,
825
- "step": 136
826
- },
827
- {
828
- "epoch": 0.9,
829
- "learning_rate": 2e-05,
830
- "loss": 0.2527,
831
- "step": 137
832
- },
833
- {
834
- "epoch": 0.9,
835
- "learning_rate": 2e-05,
836
- "loss": 0.2707,
837
- "step": 138
838
- },
839
- {
840
- "epoch": 0.91,
841
- "learning_rate": 2e-05,
842
- "loss": 0.2579,
843
- "step": 139
844
- },
845
- {
846
- "epoch": 0.92,
847
- "learning_rate": 2e-05,
848
- "loss": 0.2707,
849
- "step": 140
850
- },
851
- {
852
- "epoch": 0.92,
853
- "learning_rate": 2e-05,
854
- "loss": 0.2584,
855
- "step": 141
856
- },
857
- {
858
- "epoch": 0.93,
859
- "learning_rate": 2e-05,
860
- "loss": 0.2652,
861
- "step": 142
862
- },
863
- {
864
- "epoch": 0.93,
865
- "learning_rate": 2e-05,
866
- "loss": 0.269,
867
- "step": 143
868
- },
869
- {
870
- "epoch": 0.94,
871
- "learning_rate": 2e-05,
872
- "loss": 0.2666,
873
- "step": 144
874
- },
875
- {
876
- "epoch": 0.95,
877
- "learning_rate": 2e-05,
878
- "loss": 0.2685,
879
- "step": 145
880
- },
881
- {
882
- "epoch": 0.95,
883
- "learning_rate": 2e-05,
884
- "loss": 0.2818,
885
- "step": 146
886
- },
887
- {
888
- "epoch": 0.96,
889
- "learning_rate": 2e-05,
890
- "loss": 0.2583,
891
- "step": 147
892
- },
893
- {
894
- "epoch": 0.97,
895
- "learning_rate": 2e-05,
896
- "loss": 0.2834,
897
- "step": 148
898
- },
899
- {
900
- "epoch": 0.97,
901
- "learning_rate": 2e-05,
902
- "loss": 0.2691,
903
- "step": 149
904
- },
905
- {
906
- "epoch": 0.98,
907
- "learning_rate": 2e-05,
908
- "loss": 0.2617,
909
- "step": 150
910
- },
911
- {
912
- "epoch": 0.99,
913
- "learning_rate": 2e-05,
914
- "loss": 0.2772,
915
- "step": 151
916
- },
917
- {
918
- "epoch": 0.99,
919
- "learning_rate": 2e-05,
920
- "loss": 0.2785,
921
- "step": 152
922
- },
923
- {
924
- "epoch": 1.0,
925
- "learning_rate": 2e-05,
926
- "loss": 0.2714,
927
- "step": 153
928
- },
929
- {
930
- "epoch": 1.01,
931
- "learning_rate": 2e-05,
932
- "loss": 0.215,
933
- "step": 154
934
- },
935
- {
936
- "epoch": 1.01,
937
- "learning_rate": 2e-05,
938
- "loss": 0.2257,
939
- "step": 155
940
- },
941
- {
942
- "epoch": 1.02,
943
- "learning_rate": 2e-05,
944
- "loss": 0.2122,
945
- "step": 156
946
- },
947
- {
948
- "epoch": 1.03,
949
- "learning_rate": 2e-05,
950
- "loss": 0.2267,
951
- "step": 157
952
- },
953
- {
954
- "epoch": 1.03,
955
- "learning_rate": 2e-05,
956
- "loss": 0.2159,
957
- "step": 158
958
- },
959
- {
960
- "epoch": 1.04,
961
- "learning_rate": 2e-05,
962
- "loss": 0.2114,
963
- "step": 159
964
- },
965
- {
966
- "epoch": 1.05,
967
- "learning_rate": 2e-05,
968
- "loss": 0.2124,
969
- "step": 160
970
- },
971
- {
972
- "epoch": 1.05,
973
- "learning_rate": 2e-05,
974
- "loss": 0.2085,
975
- "step": 161
976
- },
977
- {
978
- "epoch": 1.06,
979
- "learning_rate": 2e-05,
980
- "loss": 0.221,
981
- "step": 162
982
- },
983
- {
984
- "epoch": 1.07,
985
- "learning_rate": 2e-05,
986
- "loss": 0.2177,
987
- "step": 163
988
- },
989
- {
990
- "epoch": 1.07,
991
- "learning_rate": 2e-05,
992
- "loss": 0.2226,
993
- "step": 164
994
- },
995
- {
996
- "epoch": 1.08,
997
- "learning_rate": 2e-05,
998
- "loss": 0.2198,
999
- "step": 165
1000
- },
1001
- {
1002
- "epoch": 1.08,
1003
- "learning_rate": 2e-05,
1004
- "loss": 0.2179,
1005
- "step": 166
1006
- },
1007
- {
1008
- "epoch": 1.09,
1009
- "learning_rate": 2e-05,
1010
- "loss": 0.215,
1011
- "step": 167
1012
- },
1013
- {
1014
- "epoch": 1.1,
1015
- "learning_rate": 2e-05,
1016
- "loss": 0.2223,
1017
- "step": 168
1018
- },
1019
- {
1020
- "epoch": 1.1,
1021
- "learning_rate": 2e-05,
1022
- "loss": 0.2168,
1023
- "step": 169
1024
- },
1025
- {
1026
- "epoch": 1.11,
1027
- "learning_rate": 2e-05,
1028
- "loss": 0.2081,
1029
- "step": 170
1030
- },
1031
- {
1032
- "epoch": 1.12,
1033
- "learning_rate": 2e-05,
1034
- "loss": 0.2101,
1035
- "step": 171
1036
- },
1037
- {
1038
- "epoch": 1.12,
1039
- "learning_rate": 2e-05,
1040
- "loss": 0.2091,
1041
- "step": 172
1042
- },
1043
- {
1044
- "epoch": 1.13,
1045
- "learning_rate": 2e-05,
1046
- "loss": 0.2459,
1047
- "step": 173
1048
- },
1049
- {
1050
- "epoch": 1.14,
1051
- "learning_rate": 2e-05,
1052
- "loss": 0.2135,
1053
- "step": 174
1054
- },
1055
- {
1056
- "epoch": 1.14,
1057
- "learning_rate": 2e-05,
1058
- "loss": 0.2182,
1059
- "step": 175
1060
- },
1061
- {
1062
- "epoch": 1.15,
1063
- "learning_rate": 2e-05,
1064
- "loss": 0.2165,
1065
- "step": 176
1066
- },
1067
- {
1068
- "epoch": 1.16,
1069
- "learning_rate": 2e-05,
1070
- "loss": 0.2239,
1071
- "step": 177
1072
- },
1073
- {
1074
- "epoch": 1.16,
1075
- "learning_rate": 2e-05,
1076
- "loss": 0.2184,
1077
- "step": 178
1078
- },
1079
- {
1080
- "epoch": 1.17,
1081
- "learning_rate": 2e-05,
1082
- "loss": 0.2272,
1083
- "step": 179
1084
- },
1085
- {
1086
- "epoch": 1.18,
1087
- "learning_rate": 2e-05,
1088
- "loss": 0.2066,
1089
- "step": 180
1090
- },
1091
- {
1092
- "epoch": 1.18,
1093
- "learning_rate": 2e-05,
1094
- "loss": 0.2123,
1095
- "step": 181
1096
- },
1097
- {
1098
- "epoch": 1.19,
1099
- "learning_rate": 2e-05,
1100
- "loss": 0.2059,
1101
- "step": 182
1102
- },
1103
- {
1104
- "epoch": 1.2,
1105
- "learning_rate": 2e-05,
1106
- "loss": 0.2192,
1107
- "step": 183
1108
- },
1109
- {
1110
- "epoch": 1.2,
1111
- "learning_rate": 2e-05,
1112
- "loss": 0.2102,
1113
- "step": 184
1114
- },
1115
- {
1116
- "epoch": 1.21,
1117
- "learning_rate": 2e-05,
1118
- "loss": 0.2192,
1119
- "step": 185
1120
- },
1121
- {
1122
- "epoch": 1.22,
1123
- "learning_rate": 2e-05,
1124
- "loss": 0.2156,
1125
- "step": 186
1126
- },
1127
- {
1128
- "epoch": 1.22,
1129
- "learning_rate": 2e-05,
1130
- "loss": 0.22,
1131
- "step": 187
1132
- },
1133
- {
1134
- "epoch": 1.23,
1135
- "learning_rate": 2e-05,
1136
- "loss": 0.2154,
1137
- "step": 188
1138
- },
1139
- {
1140
- "epoch": 1.24,
1141
- "learning_rate": 2e-05,
1142
- "loss": 0.2045,
1143
- "step": 189
1144
- },
1145
- {
1146
- "epoch": 1.24,
1147
- "learning_rate": 2e-05,
1148
- "loss": 0.2153,
1149
- "step": 190
1150
- },
1151
- {
1152
- "epoch": 1.25,
1153
- "learning_rate": 2e-05,
1154
- "loss": 0.2242,
1155
- "step": 191
1156
- },
1157
- {
1158
- "epoch": 1.25,
1159
- "learning_rate": 2e-05,
1160
- "loss": 0.2119,
1161
- "step": 192
1162
- },
1163
- {
1164
- "epoch": 1.26,
1165
- "learning_rate": 2e-05,
1166
- "loss": 0.2163,
1167
- "step": 193
1168
- },
1169
- {
1170
- "epoch": 1.27,
1171
- "learning_rate": 2e-05,
1172
- "loss": 0.2172,
1173
- "step": 194
1174
- },
1175
- {
1176
- "epoch": 1.27,
1177
- "learning_rate": 2e-05,
1178
- "loss": 0.2002,
1179
- "step": 195
1180
- },
1181
- {
1182
- "epoch": 1.28,
1183
- "learning_rate": 2e-05,
1184
- "loss": 0.2138,
1185
- "step": 196
1186
- },
1187
- {
1188
- "epoch": 1.29,
1189
- "learning_rate": 2e-05,
1190
- "loss": 0.2247,
1191
- "step": 197
1192
- },
1193
- {
1194
- "epoch": 1.29,
1195
- "learning_rate": 2e-05,
1196
- "loss": 0.2208,
1197
- "step": 198
1198
- },
1199
- {
1200
- "epoch": 1.3,
1201
- "learning_rate": 2e-05,
1202
- "loss": 0.2188,
1203
- "step": 199
1204
- },
1205
- {
1206
- "epoch": 1.31,
1207
- "learning_rate": 2e-05,
1208
- "loss": 0.2129,
1209
- "step": 200
1210
- }
1211
- ],
1212
- "logging_steps": 1.0,
1213
- "max_steps": 306,
1214
- "num_input_tokens_seen": 0,
1215
- "num_train_epochs": 2,
1216
- "save_steps": 100,
1217
- "total_flos": 5.340257905864081e+18,
1218
- "train_batch_size": 16,
1219
- "trial_name": null,
1220
- "trial_params": null
1221
- }