Hugging Face's logo

AmberYifan
/
Llama-2-7b-sft-spin-10k

Text Generation
Transformers
Safetensors
llama
Generated from Trainer
trl
dpo
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
AmberYifan commited on
Commit
5808646
·
verified ·
1 Parent(s): 15392b6

Training in progress, epoch 2, checkpoint

Browse files
Files changed (18) hide show
  1. last-checkpoint/global_step468/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
  2. last-checkpoint/global_step468/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
  3. last-checkpoint/global_step468/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
  4. last-checkpoint/global_step468/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
  5. last-checkpoint/global_step468/zero_pp_rank_0_mp_rank_00_model_states.pt +3 -0
  6. last-checkpoint/global_step468/zero_pp_rank_1_mp_rank_00_model_states.pt +3 -0
  7. last-checkpoint/global_step468/zero_pp_rank_2_mp_rank_00_model_states.pt +3 -0
  8. last-checkpoint/global_step468/zero_pp_rank_3_mp_rank_00_model_states.pt +3 -0
  9. last-checkpoint/latest +1 -1
  10. last-checkpoint/model-00001-of-00003.safetensors +1 -1
  11. last-checkpoint/model-00002-of-00003.safetensors +1 -1
  12. last-checkpoint/model-00003-of-00003.safetensors +1 -1
  13. last-checkpoint/rng_state_0.pth +1 -1
  14. last-checkpoint/rng_state_1.pth +1 -1
  15. last-checkpoint/rng_state_2.pth +1 -1
  16. last-checkpoint/rng_state_3.pth +1 -1
  17. last-checkpoint/scheduler.pt +1 -1
  18. last-checkpoint/trainer_state.json +244 -3
last-checkpoint/global_step468/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c0b378fc8ef0d4735b0c3525dba790872fc51d572c89d6b2d2a58eefb687b98
3
+ size 13476835648
last-checkpoint/global_step468/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4afa88b175bbbb23fa8b1612843aad10843fd3054464e6ebd4372ae4eeaa05ea
3
+ size 13476835648
last-checkpoint/global_step468/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ccf212e00810995f4b376941dc46420d796610e5dc9fddf8ef52531c62bb3a3
3
+ size 13476835648
last-checkpoint/global_step468/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40786568d9322386f0e6b3a74a22916bef0f82e63efe0d4fbc40ac55e1fda9fd
3
+ size 13476835648
last-checkpoint/global_step468/zero_pp_rank_0_mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4613cee34efe0bc385ba9cb4c5ca6454d19829b4c9f5911cf2fd48960e5cf646
3
+ size 150693
last-checkpoint/global_step468/zero_pp_rank_1_mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffbce7bfdfe05e0b0f31fffa3177c7612d34b54d0c6094cd60b0935318aaaf42
3
+ size 150693
last-checkpoint/global_step468/zero_pp_rank_2_mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:272ef9949809a500bf11df67bcddafc78bfb2b58d62d74de61590bf470209104
3
+ size 150693
last-checkpoint/global_step468/zero_pp_rank_3_mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cec4d477c9dd7bcc2b833fa9dce73de9b30cbc1b1e70679eb9c8c4cc92d5c608
3
+ size 150693
last-checkpoint/latest CHANGED
@@ -1 +1 @@
1
- global_step313
 
1
+ global_step468
last-checkpoint/model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c339c890ec08f3e5e58f5ad06ebae0dbd6ddb23ad121d561e63e9bac188bfc9b
3
  size 4938985352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a93733cee85d786b6798854e1a95b040a48817ac943434a2bae7205a6a8bd49e
3
  size 4938985352
last-checkpoint/model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:76dbefc709489db092eb6a498172ab21f68b4a5a5dff5455f75484d6faa34133
3
  size 4947390880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91c58676920f5c166cbdbc4f92b57b8ee0d959fe2b8b99e4ee8bc4f6907b01a4
3
  size 4947390880
last-checkpoint/model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a533572b108506e6e59af9b867786193b44716117ef6ea8ec975a5d6308541b8
3
  size 3590488816
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5eb6d35fb63e02478b5b415b42a0f05b5ba1906bef9e8818e84761aa37366643
3
  size 3590488816
last-checkpoint/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:50ca27603f7c4a4d9a980223e1a102922ac31561acfa7da4f28042e4e8b3528b
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85a44fde7a8ca235658f035c0812d344f55779640520411739b17e214177e846
3
  size 15024
last-checkpoint/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c57f658c3d9b6de611da50021823095fc4067036fb9a4b34655cc2e34fbf5c2b
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2caad91201b0df46e56af9eecef567ed2037350c38fdc81837724a4727d0b050
3
  size 15024
last-checkpoint/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:109da2c363a1e09432fa2ea2b7dceb6f67d1f91975069fc669e71dd536d654d8
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73e988587366c84fd1b4fc5931610543c6da84cf501cd98c78745fd102dd5024
3
  size 15024
last-checkpoint/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bda65e490958bfe3fe01157ed131eb2354f8fc19536842d27c998ecf29d48991
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55af3bb04c01d99df727583b8b7a09241b4e47d58ff1b9cf5532d3fa0d878b52
3
  size 15024
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:afa7143d1f3515855bc2dc683c033423673c7ec610b3bb70a98e9e265d312b47
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe28dd76f2f7f400d1189fda0915b1785eeec35348cbceb412ac262d936f961c
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 2.0,
5
  "eval_steps": 500,
6
- "global_step": 313,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -519,6 +519,247 @@
519
  "eval_samples_per_second": 17.305,
520
  "eval_steps_per_second": 0.606,
521
  "step": 313
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
522
  }
523
  ],
524
  "logging_steps": 10,
@@ -533,7 +774,7 @@
533
  "should_evaluate": false,
534
  "should_log": false,
535
  "should_save": true,
536
- "should_training_stop": false
537
  },
538
  "attributes": {}
539
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 2.9904153354632586,
5
  "eval_steps": 500,
6
+ "global_step": 468,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
519
  "eval_samples_per_second": 17.305,
520
  "eval_steps_per_second": 0.606,
521
  "step": 313
522
+ },
523
+ {
524
+ "epoch": 2.0447284345047922,
525
+ "grad_norm": 0.01361688937176726,
526
+ "learning_rate": 1.7577197149643706e-07,
527
+ "logits/chosen": -1.578125,
528
+ "logits/rejected": -1.53125,
529
+ "logps/chosen": -157.0,
530
+ "logps/rejected": -270.0,
531
+ "loss": 0.0,
532
+ "rewards/accuracies": 1.0,
533
+ "rewards/chosen": -1.203125,
534
+ "rewards/margins": 12.6875,
535
+ "rewards/rejected": -13.9375,
536
+ "step": 320
537
+ },
538
+ {
539
+ "epoch": 2.108626198083067,
540
+ "grad_norm": 0.05503143614977478,
541
+ "learning_rate": 1.6389548693586697e-07,
542
+ "logits/chosen": -1.5859375,
543
+ "logits/rejected": -1.5,
544
+ "logps/chosen": -162.0,
545
+ "logps/rejected": -266.0,
546
+ "loss": 0.0002,
547
+ "rewards/accuracies": 1.0,
548
+ "rewards/chosen": -1.1796875,
549
+ "rewards/margins": 12.375,
550
+ "rewards/rejected": -13.5625,
551
+ "step": 330
552
+ },
553
+ {
554
+ "epoch": 2.1725239616613417,
555
+ "grad_norm": 0.010269985783571838,
556
+ "learning_rate": 1.520190023752969e-07,
557
+ "logits/chosen": -1.53125,
558
+ "logits/rejected": -1.5234375,
559
+ "logps/chosen": -152.0,
560
+ "logps/rejected": -268.0,
561
+ "loss": 0.0001,
562
+ "rewards/accuracies": 1.0,
563
+ "rewards/chosen": -1.078125,
564
+ "rewards/margins": 12.75,
565
+ "rewards/rejected": -13.8125,
566
+ "step": 340
567
+ },
568
+ {
569
+ "epoch": 2.236421725239617,
570
+ "grad_norm": 0.0683919347457657,
571
+ "learning_rate": 1.4014251781472683e-07,
572
+ "logits/chosen": -1.5625,
573
+ "logits/rejected": -1.4921875,
574
+ "logps/chosen": -156.0,
575
+ "logps/rejected": -266.0,
576
+ "loss": 0.0001,
577
+ "rewards/accuracies": 1.0,
578
+ "rewards/chosen": -1.09375,
579
+ "rewards/margins": 12.75,
580
+ "rewards/rejected": -13.8125,
581
+ "step": 350
582
+ },
583
+ {
584
+ "epoch": 2.3003194888178915,
585
+ "grad_norm": 0.015461126488895941,
586
+ "learning_rate": 1.2826603325415677e-07,
587
+ "logits/chosen": -1.5546875,
588
+ "logits/rejected": -1.546875,
589
+ "logps/chosen": -146.0,
590
+ "logps/rejected": -266.0,
591
+ "loss": 0.0001,
592
+ "rewards/accuracies": 1.0,
593
+ "rewards/chosen": -1.1015625,
594
+ "rewards/margins": 12.625,
595
+ "rewards/rejected": -13.75,
596
+ "step": 360
597
+ },
598
+ {
599
+ "epoch": 2.364217252396166,
600
+ "grad_norm": 0.004666909345169454,
601
+ "learning_rate": 1.163895486935867e-07,
602
+ "logits/chosen": -1.5390625,
603
+ "logits/rejected": -1.4921875,
604
+ "logps/chosen": -160.0,
605
+ "logps/rejected": -270.0,
606
+ "loss": 0.0002,
607
+ "rewards/accuracies": 1.0,
608
+ "rewards/chosen": -1.1328125,
609
+ "rewards/margins": 12.8125,
610
+ "rewards/rejected": -13.9375,
611
+ "step": 370
612
+ },
613
+ {
614
+ "epoch": 2.428115015974441,
615
+ "grad_norm": 0.021768817881518163,
616
+ "learning_rate": 1.0451306413301662e-07,
617
+ "logits/chosen": -1.5859375,
618
+ "logits/rejected": -1.5078125,
619
+ "logps/chosen": -157.0,
620
+ "logps/rejected": -266.0,
621
+ "loss": 0.0001,
622
+ "rewards/accuracies": 1.0,
623
+ "rewards/chosen": -1.1484375,
624
+ "rewards/margins": 12.8125,
625
+ "rewards/rejected": -14.0,
626
+ "step": 380
627
+ },
628
+ {
629
+ "epoch": 2.4920127795527156,
630
+ "grad_norm": 0.006712968496189408,
631
+ "learning_rate": 9.263657957244655e-08,
632
+ "logits/chosen": -1.578125,
633
+ "logits/rejected": -1.4765625,
634
+ "logps/chosen": -150.0,
635
+ "logps/rejected": -272.0,
636
+ "loss": 0.0002,
637
+ "rewards/accuracies": 1.0,
638
+ "rewards/chosen": -1.046875,
639
+ "rewards/margins": 13.125,
640
+ "rewards/rejected": -14.125,
641
+ "step": 390
642
+ },
643
+ {
644
+ "epoch": 2.5559105431309903,
645
+ "grad_norm": 0.03157363956222121,
646
+ "learning_rate": 8.076009501187649e-08,
647
+ "logits/chosen": -1.578125,
648
+ "logits/rejected": -1.515625,
649
+ "logps/chosen": -160.0,
650
+ "logps/rejected": -268.0,
651
+ "loss": 0.0001,
652
+ "rewards/accuracies": 1.0,
653
+ "rewards/chosen": -1.0625,
654
+ "rewards/margins": 13.0625,
655
+ "rewards/rejected": -14.125,
656
+ "step": 400
657
+ },
658
+ {
659
+ "epoch": 2.619808306709265,
660
+ "grad_norm": 0.012508580641767728,
661
+ "learning_rate": 6.88836104513064e-08,
662
+ "logits/chosen": -1.5390625,
663
+ "logits/rejected": -1.5,
664
+ "logps/chosen": -151.0,
665
+ "logps/rejected": -268.0,
666
+ "loss": 0.0001,
667
+ "rewards/accuracies": 1.0,
668
+ "rewards/chosen": -1.03125,
669
+ "rewards/margins": 12.9375,
670
+ "rewards/rejected": -13.9375,
671
+ "step": 410
672
+ },
673
+ {
674
+ "epoch": 2.68370607028754,
675
+ "grad_norm": 0.014858628369684351,
676
+ "learning_rate": 5.700712589073634e-08,
677
+ "logits/chosen": -1.5546875,
678
+ "logits/rejected": -1.515625,
679
+ "logps/chosen": -159.0,
680
+ "logps/rejected": -268.0,
681
+ "loss": 0.0001,
682
+ "rewards/accuracies": 1.0,
683
+ "rewards/chosen": -1.171875,
684
+ "rewards/margins": 12.8125,
685
+ "rewards/rejected": -14.0,
686
+ "step": 420
687
+ },
688
+ {
689
+ "epoch": 2.747603833865815,
690
+ "grad_norm": 0.00119033851952325,
691
+ "learning_rate": 4.5130641330166267e-08,
692
+ "logits/chosen": -1.59375,
693
+ "logits/rejected": -1.46875,
694
+ "logps/chosen": -165.0,
695
+ "logps/rejected": -270.0,
696
+ "loss": 0.0001,
697
+ "rewards/accuracies": 1.0,
698
+ "rewards/chosen": -1.0703125,
699
+ "rewards/margins": 13.25,
700
+ "rewards/rejected": -14.3125,
701
+ "step": 430
702
+ },
703
+ {
704
+ "epoch": 2.8115015974440896,
705
+ "grad_norm": 0.01000781639917085,
706
+ "learning_rate": 3.32541567695962e-08,
707
+ "logits/chosen": -1.5390625,
708
+ "logits/rejected": -1.5234375,
709
+ "logps/chosen": -142.0,
710
+ "logps/rejected": -270.0,
711
+ "loss": 0.0001,
712
+ "rewards/accuracies": 1.0,
713
+ "rewards/chosen": -1.0234375,
714
+ "rewards/margins": 13.375,
715
+ "rewards/rejected": -14.375,
716
+ "step": 440
717
+ },
718
+ {
719
+ "epoch": 2.8753993610223643,
720
+ "grad_norm": 0.032333252949445446,
721
+ "learning_rate": 2.1377672209026125e-08,
722
+ "logits/chosen": -1.5546875,
723
+ "logits/rejected": -1.5078125,
724
+ "logps/chosen": -159.0,
725
+ "logps/rejected": -270.0,
726
+ "loss": 0.0,
727
+ "rewards/accuracies": 1.0,
728
+ "rewards/chosen": -1.078125,
729
+ "rewards/margins": 13.3125,
730
+ "rewards/rejected": -14.375,
731
+ "step": 450
732
+ },
733
+ {
734
+ "epoch": 2.939297124600639,
735
+ "grad_norm": 0.0060308968638626,
736
+ "learning_rate": 9.501187648456057e-09,
737
+ "logits/chosen": -1.578125,
738
+ "logits/rejected": -1.4921875,
739
+ "logps/chosen": -153.0,
740
+ "logps/rejected": -270.0,
741
+ "loss": 0.0001,
742
+ "rewards/accuracies": 1.0,
743
+ "rewards/chosen": -1.0390625,
744
+ "rewards/margins": 13.3125,
745
+ "rewards/rejected": -14.375,
746
+ "step": 460
747
+ },
748
+ {
749
+ "epoch": 2.9904153354632586,
750
+ "eval_logits/chosen": -1.5078125,
751
+ "eval_logits/rejected": -1.5234375,
752
+ "eval_logps/chosen": -150.0,
753
+ "eval_logps/rejected": -270.0,
754
+ "eval_loss": 1.3302266779646743e-05,
755
+ "eval_rewards/accuracies": 1.0,
756
+ "eval_rewards/chosen": -1.09375,
757
+ "eval_rewards/margins": 12.8125,
758
+ "eval_rewards/rejected": -13.9375,
759
+ "eval_runtime": 14.4196,
760
+ "eval_samples_per_second": 13.87,
761
+ "eval_steps_per_second": 0.485,
762
+ "step": 468
763
  }
764
  ],
765
  "logging_steps": 10,
 
774
  "should_evaluate": false,
775
  "should_log": false,
776
  "should_save": true,
777
+ "should_training_stop": true
778
  },
779
  "attributes": {}
780
  }