File size: 30,168 Bytes
0fe756b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 355,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "dpo_losses": 0.6931471824645996,
      "epoch": 0.0,
      "grad_norm": 1.6018567815095135,
      "learning_rate": 1.3888888888888888e-07,
      "logits/chosen": -2.861618995666504,
      "logits/rejected": -2.8205904960632324,
      "logps/chosen": -271.06011962890625,
      "logps/rejected": -211.1704559326172,
      "loss": 0.6931,
      "positive_losses": 0.0,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/margins_max": 0.0,
      "rewards/margins_min": 0.0,
      "rewards/margins_std": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "dpo_losses": 0.6928361654281616,
      "epoch": 0.03,
      "grad_norm": 14.098492351037597,
      "learning_rate": 1.3888888888888892e-06,
      "logits/chosen": -2.8340628147125244,
      "logits/rejected": -2.7916715145111084,
      "logps/chosen": -324.87408447265625,
      "logps/rejected": -274.8518371582031,
      "loss": 0.6969,
      "positive_losses": 0.03656284138560295,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/chosen": 0.001762597355991602,
      "rewards/margins": 0.0006246823468245566,
      "rewards/margins_max": 0.0034460597671568394,
      "rewards/margins_min": -0.002478615380823612,
      "rewards/margins_std": 0.002669532783329487,
      "rewards/rejected": 0.0011379148345440626,
      "step": 10
    },
    {
      "dpo_losses": 0.6901537775993347,
      "epoch": 0.06,
      "grad_norm": 1.829780676576113,
      "learning_rate": 2.7777777777777783e-06,
      "logits/chosen": -2.7248008251190186,
      "logits/rejected": -2.7065372467041016,
      "logps/chosen": -291.9751892089844,
      "logps/rejected": -214.52914428710938,
      "loss": 0.69,
      "positive_losses": 0.00235748291015625,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/chosen": 0.01850745640695095,
      "rewards/margins": 0.006009287666529417,
      "rewards/margins_max": 0.013369890861213207,
      "rewards/margins_min": -0.0006899007130414248,
      "rewards/margins_std": 0.006301888730376959,
      "rewards/rejected": 0.01249817106872797,
      "step": 20
    },
    {
      "dpo_losses": 0.6790497303009033,
      "epoch": 0.08,
      "grad_norm": 2.096661038575657,
      "learning_rate": 4.166666666666667e-06,
      "logits/chosen": -2.8153939247131348,
      "logits/rejected": -2.7460672855377197,
      "logps/chosen": -298.10052490234375,
      "logps/rejected": -229.7678680419922,
      "loss": 0.677,
      "positive_losses": 0.0,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 0.05605363845825195,
      "rewards/margins": 0.02858993411064148,
      "rewards/margins_max": 0.058357615023851395,
      "rewards/margins_min": 0.004640273749828339,
      "rewards/margins_std": 0.02467900700867176,
      "rewards/rejected": 0.027463700622320175,
      "step": 30
    },
    {
      "dpo_losses": 0.6675597429275513,
      "epoch": 0.11,
      "grad_norm": 1.7320035926217752,
      "learning_rate": 4.998060489154965e-06,
      "logits/chosen": -2.8310070037841797,
      "logits/rejected": -2.751425266265869,
      "logps/chosen": -268.48809814453125,
      "logps/rejected": -222.01107788085938,
      "loss": 0.6662,
      "positive_losses": 0.054492950439453125,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": 0.08996561169624329,
      "rewards/margins": 0.05272960662841797,
      "rewards/margins_max": 0.1101265698671341,
      "rewards/margins_min": 0.003616312053054571,
      "rewards/margins_std": 0.048521898686885834,
      "rewards/rejected": 0.03723599761724472,
      "step": 40
    },
    {
      "dpo_losses": 0.6397972106933594,
      "epoch": 0.14,
      "grad_norm": 9.583890638870626,
      "learning_rate": 4.976275538042932e-06,
      "logits/chosen": -2.7891061305999756,
      "logits/rejected": -2.7175135612487793,
      "logps/chosen": -262.20794677734375,
      "logps/rejected": -231.79653930664062,
      "loss": 0.6446,
      "positive_losses": 0.0,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 0.13362163305282593,
      "rewards/margins": 0.11281381547451019,
      "rewards/margins_max": 0.23626498878002167,
      "rewards/margins_min": 0.022470083087682724,
      "rewards/margins_std": 0.0988926962018013,
      "rewards/rejected": 0.02080780453979969,
      "step": 50
    },
    {
      "dpo_losses": 0.6110584139823914,
      "epoch": 0.17,
      "grad_norm": 2.0747443213986694,
      "learning_rate": 4.93049306999712e-06,
      "logits/chosen": -2.7118520736694336,
      "logits/rejected": -2.6753315925598145,
      "logps/chosen": -296.9767150878906,
      "logps/rejected": -263.8233947753906,
      "loss": 0.628,
      "positive_losses": 0.011554336175322533,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.16662926971912384,
      "rewards/margins": 0.17714819312095642,
      "rewards/margins_max": 0.30765318870544434,
      "rewards/margins_min": 0.05318903177976608,
      "rewards/margins_std": 0.11578011512756348,
      "rewards/rejected": -0.010518952272832394,
      "step": 60
    },
    {
      "dpo_losses": 0.6022371053695679,
      "epoch": 0.2,
      "grad_norm": 1.5871888283763238,
      "learning_rate": 4.861156761634014e-06,
      "logits/chosen": -2.7271430492401123,
      "logits/rejected": -2.6688759326934814,
      "logps/chosen": -303.47613525390625,
      "logps/rejected": -236.2406463623047,
      "loss": 0.6175,
      "positive_losses": 0.19450588524341583,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.19385087490081787,
      "rewards/margins": 0.19984133541584015,
      "rewards/margins_max": 0.4134605824947357,
      "rewards/margins_min": 0.04761160537600517,
      "rewards/margins_std": 0.16880682110786438,
      "rewards/rejected": -0.00599044980481267,
      "step": 70
    },
    {
      "dpo_losses": 0.5768495798110962,
      "epoch": 0.23,
      "grad_norm": 1.804849988880195,
      "learning_rate": 4.7689385491773934e-06,
      "logits/chosen": -2.738285779953003,
      "logits/rejected": -2.684203863143921,
      "logps/chosen": -300.8853454589844,
      "logps/rejected": -292.05633544921875,
      "loss": 0.6017,
      "positive_losses": 0.328561395406723,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.20385125279426575,
      "rewards/margins": 0.26062771677970886,
      "rewards/margins_max": 0.4970013201236725,
      "rewards/margins_min": 0.05170217156410217,
      "rewards/margins_std": 0.2058703452348709,
      "rewards/rejected": -0.056776486337184906,
      "step": 80
    },
    {
      "dpo_losses": 0.5672236084938049,
      "epoch": 0.25,
      "grad_norm": 2.184742961229221,
      "learning_rate": 4.654732116743193e-06,
      "logits/chosen": -2.6370556354522705,
      "logits/rejected": -2.601066827774048,
      "logps/chosen": -252.70535278320312,
      "logps/rejected": -203.89418029785156,
      "loss": 0.5769,
      "positive_losses": 0.07196970283985138,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.2328944206237793,
      "rewards/margins": 0.2819642424583435,
      "rewards/margins_max": 0.514846682548523,
      "rewards/margins_min": 0.09985215216875076,
      "rewards/margins_std": 0.19404996931552887,
      "rewards/rejected": -0.049069829285144806,
      "step": 90
    },
    {
      "dpo_losses": 0.5702880620956421,
      "epoch": 0.28,
      "grad_norm": 2.550586173059517,
      "learning_rate": 4.5196442356717526e-06,
      "logits/chosen": -2.6703598499298096,
      "logits/rejected": -2.6374478340148926,
      "logps/chosen": -264.9583740234375,
      "logps/rejected": -273.49615478515625,
      "loss": 0.6232,
      "positive_losses": 1.2302151918411255,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.16453364491462708,
      "rewards/margins": 0.27762115001678467,
      "rewards/margins_max": 0.5491287708282471,
      "rewards/margins_min": 0.05581303685903549,
      "rewards/margins_std": 0.22483690083026886,
      "rewards/rejected": -0.113087497651577,
      "step": 100
    },
    {
      "epoch": 0.28,
      "eval_dpo_losses": 0.6656126976013184,
      "eval_logits/chosen": -2.67258620262146,
      "eval_logits/rejected": -2.6360833644866943,
      "eval_logps/chosen": -280.30804443359375,
      "eval_logps/rejected": -261.0971984863281,
      "eval_loss": 1.1412982940673828,
      "eval_positive_losses": 4.261031627655029,
      "eval_rewards/accuracies": 0.6230000257492065,
      "eval_rewards/chosen": 0.04285382851958275,
      "eval_rewards/margins": 0.06803657114505768,
      "eval_rewards/margins_max": 0.40864306688308716,
      "eval_rewards/margins_min": -0.22808942198753357,
      "eval_rewards/margins_std": 0.2094314992427826,
      "eval_rewards/rejected": -0.02518274076282978,
      "eval_runtime": 429.2755,
      "eval_samples_per_second": 4.659,
      "eval_steps_per_second": 0.291,
      "step": 100
    },
    {
      "dpo_losses": 0.5097740888595581,
      "epoch": 0.31,
      "grad_norm": 6.336382416368574,
      "learning_rate": 4.364984038837727e-06,
      "logits/chosen": -2.742903709411621,
      "logits/rejected": -2.654869318008423,
      "logps/chosen": -349.24517822265625,
      "logps/rejected": -304.54730224609375,
      "loss": 0.543,
      "positive_losses": 0.44344156980514526,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.25360527634620667,
      "rewards/margins": 0.43474069237709045,
      "rewards/margins_max": 0.7704421281814575,
      "rewards/margins_min": 0.1366521120071411,
      "rewards/margins_std": 0.2834155559539795,
      "rewards/rejected": -0.1811354160308838,
      "step": 110
    },
    {
      "dpo_losses": 0.518837571144104,
      "epoch": 0.34,
      "grad_norm": 2.194144050007341,
      "learning_rate": 4.192250333880045e-06,
      "logits/chosen": -2.7281386852264404,
      "logits/rejected": -2.670868396759033,
      "logps/chosen": -321.75982666015625,
      "logps/rejected": -280.87091064453125,
      "loss": 0.5524,
      "positive_losses": 0.46012669801712036,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.26626402139663696,
      "rewards/margins": 0.4130307137966156,
      "rewards/margins_max": 0.7945607900619507,
      "rewards/margins_min": 0.14706461131572723,
      "rewards/margins_std": 0.2963123917579651,
      "rewards/rejected": -0.14676669239997864,
      "step": 120
    },
    {
      "dpo_losses": 0.4917011260986328,
      "epoch": 0.37,
      "grad_norm": 1.7534787479023215,
      "learning_rate": 4.0031170782990214e-06,
      "logits/chosen": -2.711912155151367,
      "logits/rejected": -2.634033441543579,
      "logps/chosen": -353.554443359375,
      "logps/rejected": -320.6388244628906,
      "loss": 0.5518,
      "positive_losses": 0.8977662920951843,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.2880980372428894,
      "rewards/margins": 0.4901772439479828,
      "rewards/margins_max": 0.8924927711486816,
      "rewards/margins_min": 0.1499636471271515,
      "rewards/margins_std": 0.3346417546272278,
      "rewards/rejected": -0.20207922160625458,
      "step": 130
    },
    {
      "dpo_losses": 0.4866393208503723,
      "epoch": 0.39,
      "grad_norm": 21.27134583914694,
      "learning_rate": 3.7994171571810756e-06,
      "logits/chosen": -2.6895060539245605,
      "logits/rejected": -2.6512811183929443,
      "logps/chosen": -291.05548095703125,
      "logps/rejected": -294.4687805175781,
      "loss": 0.5718,
      "positive_losses": 0.2207096517086029,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.2735855281352997,
      "rewards/margins": 0.5197780132293701,
      "rewards/margins_max": 1.003483772277832,
      "rewards/margins_min": 0.1269286870956421,
      "rewards/margins_std": 0.3979441523551941,
      "rewards/rejected": -0.24619252979755402,
      "step": 140
    },
    {
      "dpo_losses": 0.5046078562736511,
      "epoch": 0.42,
      "grad_norm": 3.3011186957688583,
      "learning_rate": 3.5831246207606597e-06,
      "logits/chosen": -2.6959190368652344,
      "logits/rejected": -2.658679962158203,
      "logps/chosen": -264.2646179199219,
      "logps/rejected": -234.5491180419922,
      "loss": 0.5366,
      "positive_losses": 0.490040123462677,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.24420371651649475,
      "rewards/margins": 0.45886701345443726,
      "rewards/margins_max": 0.8680801391601562,
      "rewards/margins_min": 0.1154303103685379,
      "rewards/margins_std": 0.34930768609046936,
      "rewards/rejected": -0.2146632969379425,
      "step": 150
    },
    {
      "dpo_losses": 0.48088502883911133,
      "epoch": 0.45,
      "grad_norm": 2.135658014816511,
      "learning_rate": 3.3563355539546795e-06,
      "logits/chosen": -2.665548801422119,
      "logits/rejected": -2.6138901710510254,
      "logps/chosen": -274.263427734375,
      "logps/rejected": -260.50518798828125,
      "loss": 0.5724,
      "positive_losses": 0.9731669425964355,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.262834370136261,
      "rewards/margins": 0.5239533185958862,
      "rewards/margins_max": 0.9884392023086548,
      "rewards/margins_min": 0.15575796365737915,
      "rewards/margins_std": 0.3754872977733612,
      "rewards/rejected": -0.26111894845962524,
      "step": 160
    },
    {
      "dpo_losses": 0.4504636824131012,
      "epoch": 0.48,
      "grad_norm": 3.940043763048366,
      "learning_rate": 3.121247763262235e-06,
      "logits/chosen": -2.708754777908325,
      "logits/rejected": -2.657917022705078,
      "logps/chosen": -297.7489013671875,
      "logps/rejected": -327.0563049316406,
      "loss": 0.4813,
      "positive_losses": 0.03098602220416069,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.3187271058559418,
      "rewards/margins": 0.6266334652900696,
      "rewards/margins_max": 1.0517089366912842,
      "rewards/margins_min": 0.17500966787338257,
      "rewards/margins_std": 0.3909396231174469,
      "rewards/rejected": -0.3079063296318054,
      "step": 170
    },
    {
      "dpo_losses": 0.4588772654533386,
      "epoch": 0.51,
      "grad_norm": 8.823245159881209,
      "learning_rate": 2.8801394778833475e-06,
      "logits/chosen": -2.6968963146209717,
      "logits/rejected": -2.6140356063842773,
      "logps/chosen": -305.4325866699219,
      "logps/rejected": -326.99798583984375,
      "loss": 0.5468,
      "positive_losses": 0.8232825994491577,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.2646820843219757,
      "rewards/margins": 0.5928131937980652,
      "rewards/margins_max": 1.0361554622650146,
      "rewards/margins_min": 0.28750157356262207,
      "rewards/margins_std": 0.33570224046707153,
      "rewards/rejected": -0.32813113927841187,
      "step": 180
    },
    {
      "dpo_losses": 0.45539379119873047,
      "epoch": 0.54,
      "grad_norm": 3.517893013000186,
      "learning_rate": 2.6353472714635443e-06,
      "logits/chosen": -2.6537580490112305,
      "logits/rejected": -2.5634191036224365,
      "logps/chosen": -287.6109619140625,
      "logps/rejected": -265.6959228515625,
      "loss": 0.5435,
      "positive_losses": 0.9886103868484497,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.34349915385246277,
      "rewards/margins": 0.6255816221237183,
      "rewards/margins_max": 1.1905597448349,
      "rewards/margins_min": 0.168921560049057,
      "rewards/margins_std": 0.453277051448822,
      "rewards/rejected": -0.2820824980735779,
      "step": 190
    },
    {
      "dpo_losses": 0.44315657019615173,
      "epoch": 0.56,
      "grad_norm": 27.976402148032502,
      "learning_rate": 2.3892434184240536e-06,
      "logits/chosen": -2.7400636672973633,
      "logits/rejected": -2.662397623062134,
      "logps/chosen": -309.39691162109375,
      "logps/rejected": -299.7530212402344,
      "loss": 0.5625,
      "positive_losses": 0.9616166353225708,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.30135902762413025,
      "rewards/margins": 0.6429153084754944,
      "rewards/margins_max": 1.131412148475647,
      "rewards/margins_min": 0.17879006266593933,
      "rewards/margins_std": 0.4260264039039612,
      "rewards/rejected": -0.34155628085136414,
      "step": 200
    },
    {
      "epoch": 0.56,
      "eval_dpo_losses": 0.6469283699989319,
      "eval_logits/chosen": -2.678022623062134,
      "eval_logits/rejected": -2.6380200386047363,
      "eval_logps/chosen": -286.4236145019531,
      "eval_logps/rejected": -272.83990478515625,
      "eval_loss": 1.7185667753219604,
      "eval_positive_losses": 9.667731285095215,
      "eval_rewards/accuracies": 0.6420000195503235,
      "eval_rewards/chosen": -0.018302178010344505,
      "eval_rewards/margins": 0.1243075579404831,
      "eval_rewards/margins_max": 0.6361650228500366,
      "eval_rewards/margins_min": -0.3433184325695038,
      "eval_rewards/margins_std": 0.32774004340171814,
      "eval_rewards/rejected": -0.14260973036289215,
      "eval_runtime": 428.2243,
      "eval_samples_per_second": 4.67,
      "eval_steps_per_second": 0.292,
      "step": 200
    },
    {
      "dpo_losses": 0.4354400634765625,
      "epoch": 0.59,
      "grad_norm": 23.522369776083625,
      "learning_rate": 2.1442129043167877e-06,
      "logits/chosen": -2.6434009075164795,
      "logits/rejected": -2.6138339042663574,
      "logps/chosen": -286.7272033691406,
      "logps/rejected": -291.8896789550781,
      "loss": 0.513,
      "positive_losses": 0.665066123008728,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.33457106351852417,
      "rewards/margins": 0.6892200708389282,
      "rewards/margins_max": 1.2269551753997803,
      "rewards/margins_min": 0.18099449574947357,
      "rewards/margins_std": 0.46556130051612854,
      "rewards/rejected": -0.35464900732040405,
      "step": 210
    },
    {
      "dpo_losses": 0.4387238025665283,
      "epoch": 0.62,
      "grad_norm": 11.92404423048434,
      "learning_rate": 1.9026303129961049e-06,
      "logits/chosen": -2.7612462043762207,
      "logits/rejected": -2.664234161376953,
      "logps/chosen": -319.7461853027344,
      "logps/rejected": -306.0053405761719,
      "loss": 0.5894,
      "positive_losses": 1.1452913284301758,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.33710065484046936,
      "rewards/margins": 0.6538791656494141,
      "rewards/margins_max": 1.1509373188018799,
      "rewards/margins_min": 0.19225715100765228,
      "rewards/margins_std": 0.4403897225856781,
      "rewards/rejected": -0.3167785704135895,
      "step": 220
    },
    {
      "dpo_losses": 0.44511428475379944,
      "epoch": 0.65,
      "grad_norm": 2.419282473127918,
      "learning_rate": 1.66683681459314e-06,
      "logits/chosen": -2.773876428604126,
      "logits/rejected": -2.67607045173645,
      "logps/chosen": -339.04718017578125,
      "logps/rejected": -293.1225891113281,
      "loss": 0.4763,
      "positive_losses": 0.6133368611335754,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.32628515362739563,
      "rewards/margins": 0.6365767121315002,
      "rewards/margins_max": 1.125410795211792,
      "rewards/margins_min": 0.21782192587852478,
      "rewards/margins_std": 0.4051855504512787,
      "rewards/rejected": -0.3102915287017822,
      "step": 230
    },
    {
      "dpo_losses": 0.4544529318809509,
      "epoch": 0.68,
      "grad_norm": 13.447116267552904,
      "learning_rate": 1.4391174773015836e-06,
      "logits/chosen": -2.7197587490081787,
      "logits/rejected": -2.649749279022217,
      "logps/chosen": -302.6105041503906,
      "logps/rejected": -321.8402404785156,
      "loss": 0.692,
      "positive_losses": 2.48455810546875,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 0.22186538577079773,
      "rewards/margins": 0.6085190773010254,
      "rewards/margins_max": 1.1415433883666992,
      "rewards/margins_min": 0.23370866477489471,
      "rewards/margins_std": 0.41311854124069214,
      "rewards/rejected": -0.38665369153022766,
      "step": 240
    },
    {
      "dpo_losses": 0.45861634612083435,
      "epoch": 0.7,
      "grad_norm": 5.111403689556549,
      "learning_rate": 1.2216791228457778e-06,
      "logits/chosen": -2.716823101043701,
      "logits/rejected": -2.640800952911377,
      "logps/chosen": -280.11114501953125,
      "logps/rejected": -281.67138671875,
      "loss": 0.4992,
      "positive_losses": 0.6084854006767273,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.31169968843460083,
      "rewards/margins": 0.6179708242416382,
      "rewards/margins_max": 1.2185614109039307,
      "rewards/margins_min": 0.1615341305732727,
      "rewards/margins_std": 0.4740964472293854,
      "rewards/rejected": -0.30627113580703735,
      "step": 250
    },
    {
      "dpo_losses": 0.4628082811832428,
      "epoch": 0.73,
      "grad_norm": 2.699692592075128,
      "learning_rate": 1.0166289402331391e-06,
      "logits/chosen": -2.7728962898254395,
      "logits/rejected": -2.684753894805908,
      "logps/chosen": -263.36126708984375,
      "logps/rejected": -289.21661376953125,
      "loss": 0.5624,
      "positive_losses": 0.9304378628730774,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.28732261061668396,
      "rewards/margins": 0.5901791453361511,
      "rewards/margins_max": 1.102694034576416,
      "rewards/margins_min": 0.17682021856307983,
      "rewards/margins_std": 0.4229150712490082,
      "rewards/rejected": -0.30285659432411194,
      "step": 260
    },
    {
      "dpo_losses": 0.4588424265384674,
      "epoch": 0.76,
      "grad_norm": 2.4735784513371377,
      "learning_rate": 8.259540650444736e-07,
      "logits/chosen": -2.717153787612915,
      "logits/rejected": -2.662932872772217,
      "logps/chosen": -278.75482177734375,
      "logps/rejected": -291.56866455078125,
      "loss": 0.5853,
      "positive_losses": 0.9098857641220093,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.30360764265060425,
      "rewards/margins": 0.5942984223365784,
      "rewards/margins_max": 1.0322821140289307,
      "rewards/margins_min": 0.21275146305561066,
      "rewards/margins_std": 0.36198341846466064,
      "rewards/rejected": -0.2906908392906189,
      "step": 270
    },
    {
      "dpo_losses": 0.4629085958003998,
      "epoch": 0.79,
      "grad_norm": 13.451546074592132,
      "learning_rate": 6.515023221586722e-07,
      "logits/chosen": -2.6962451934814453,
      "logits/rejected": -2.6575076580047607,
      "logps/chosen": -274.9664001464844,
      "logps/rejected": -304.9722595214844,
      "loss": 0.5625,
      "positive_losses": 1.4465850591659546,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.2849060893058777,
      "rewards/margins": 0.60865718126297,
      "rewards/margins_max": 1.1329301595687866,
      "rewards/margins_min": 0.1755952537059784,
      "rewards/margins_std": 0.4414794445037842,
      "rewards/rejected": -0.3237510919570923,
      "step": 280
    },
    {
      "dpo_losses": 0.47258663177490234,
      "epoch": 0.82,
      "grad_norm": 2.654477953260434,
      "learning_rate": 4.949643185335288e-07,
      "logits/chosen": -2.707307815551758,
      "logits/rejected": -2.652792453765869,
      "logps/chosen": -259.1030578613281,
      "logps/rejected": -292.6324462890625,
      "loss": 0.6149,
      "positive_losses": 1.7202523946762085,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 0.27813172340393066,
      "rewards/margins": 0.5642735958099365,
      "rewards/margins_max": 1.0385398864746094,
      "rewards/margins_min": 0.12702254951000214,
      "rewards/margins_std": 0.4158683717250824,
      "rewards/rejected": -0.28614187240600586,
      "step": 290
    },
    {
      "dpo_losses": 0.4324049949645996,
      "epoch": 0.85,
      "grad_norm": 11.591501845708454,
      "learning_rate": 3.578570595810274e-07,
      "logits/chosen": -2.7821717262268066,
      "logits/rejected": -2.6995315551757812,
      "logps/chosen": -309.7518310546875,
      "logps/rejected": -320.70916748046875,
      "loss": 0.4748,
      "positive_losses": 0.8444260358810425,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3676120638847351,
      "rewards/margins": 0.6803697943687439,
      "rewards/margins_max": 1.199285864830017,
      "rewards/margins_min": 0.21351738274097443,
      "rewards/margins_std": 0.4415613114833832,
      "rewards/rejected": -0.312757670879364,
      "step": 300
    },
    {
      "epoch": 0.85,
      "eval_dpo_losses": 0.6448404788970947,
      "eval_logits/chosen": -2.715327739715576,
      "eval_logits/rejected": -2.6732916831970215,
      "eval_logps/chosen": -284.4527587890625,
      "eval_logps/rejected": -271.32244873046875,
      "eval_loss": 1.6048117876052856,
      "eval_positive_losses": 8.706162452697754,
      "eval_rewards/accuracies": 0.6470000147819519,
      "eval_rewards/chosen": 0.0014067561132833362,
      "eval_rewards/margins": 0.12884218990802765,
      "eval_rewards/margins_max": 0.6374967098236084,
      "eval_rewards/margins_min": -0.34605804085731506,
      "eval_rewards/margins_std": 0.3295030891895294,
      "eval_rewards/rejected": -0.12743544578552246,
      "eval_runtime": 428.2498,
      "eval_samples_per_second": 4.67,
      "eval_steps_per_second": 0.292,
      "step": 300
    },
    {
      "dpo_losses": 0.45941466093063354,
      "epoch": 0.87,
      "grad_norm": 2.6085680781835205,
      "learning_rate": 2.4150924791035037e-07,
      "logits/chosen": -2.774445056915283,
      "logits/rejected": -2.673360824584961,
      "logps/chosen": -267.74237060546875,
      "logps/rejected": -243.88473510742188,
      "loss": 0.5697,
      "positive_losses": 1.3653801679611206,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.30073457956314087,
      "rewards/margins": 0.5973426699638367,
      "rewards/margins_max": 1.1060882806777954,
      "rewards/margins_min": 0.18351522088050842,
      "rewards/margins_std": 0.4086340069770813,
      "rewards/rejected": -0.2966081500053406,
      "step": 310
    },
    {
      "dpo_losses": 0.45310109853744507,
      "epoch": 0.9,
      "grad_norm": 10.060071948421735,
      "learning_rate": 1.4704840690808658e-07,
      "logits/chosen": -2.738978385925293,
      "logits/rejected": -2.680860757827759,
      "logps/chosen": -279.5138854980469,
      "logps/rejected": -293.9893493652344,
      "loss": 0.5692,
      "positive_losses": 1.6892318725585938,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 0.2875928282737732,
      "rewards/margins": 0.6207860708236694,
      "rewards/margins_max": 1.124011754989624,
      "rewards/margins_min": 0.14557920396327972,
      "rewards/margins_std": 0.44626301527023315,
      "rewards/rejected": -0.33319321274757385,
      "step": 320
    },
    {
      "dpo_losses": 0.42673492431640625,
      "epoch": 0.93,
      "grad_norm": 9.476085880429812,
      "learning_rate": 7.538995394063996e-08,
      "logits/chosen": -2.8187005519866943,
      "logits/rejected": -2.7311813831329346,
      "logps/chosen": -318.88360595703125,
      "logps/rejected": -302.66058349609375,
      "loss": 0.5314,
      "positive_losses": 0.5069873929023743,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.35436224937438965,
      "rewards/margins": 0.7115713953971863,
      "rewards/margins_max": 1.23550546169281,
      "rewards/margins_min": 0.2139424830675125,
      "rewards/margins_std": 0.4558965563774109,
      "rewards/rejected": -0.35720914602279663,
      "step": 330
    },
    {
      "dpo_losses": 0.4437997341156006,
      "epoch": 0.96,
      "grad_norm": 2.682118994824555,
      "learning_rate": 2.722832907015971e-08,
      "logits/chosen": -2.6981847286224365,
      "logits/rejected": -2.6440398693084717,
      "logps/chosen": -266.6497802734375,
      "logps/rejected": -282.98199462890625,
      "loss": 0.5024,
      "positive_losses": 0.9627658724784851,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.3319571018218994,
      "rewards/margins": 0.6500804424285889,
      "rewards/margins_max": 1.2494922876358032,
      "rewards/margins_min": 0.25120097398757935,
      "rewards/margins_std": 0.4507668614387512,
      "rewards/rejected": -0.31812337040901184,
      "step": 340
    },
    {
      "dpo_losses": 0.4518283009529114,
      "epoch": 0.99,
      "grad_norm": 5.762126574549782,
      "learning_rate": 3.030265255329623e-09,
      "logits/chosen": -2.6820361614227295,
      "logits/rejected": -2.6376953125,
      "logps/chosen": -285.1527404785156,
      "logps/rejected": -317.6675720214844,
      "loss": 0.5059,
      "positive_losses": 0.9290813207626343,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": 0.2980636656284332,
      "rewards/margins": 0.6102195978164673,
      "rewards/margins_max": 1.0686355829238892,
      "rewards/margins_min": 0.20541608333587646,
      "rewards/margins_std": 0.38572338223457336,
      "rewards/rejected": -0.31215590238571167,
      "step": 350
    },
    {
      "epoch": 1.0,
      "step": 355,
      "total_flos": 0.0,
      "train_loss": 0.5743894765074824,
      "train_runtime": 4311.1014,
      "train_samples_per_second": 1.317,
      "train_steps_per_second": 0.082
    }
  ],
  "logging_steps": 10,
  "max_steps": 355,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}