shulijia commited on
Commit
2041cae
·
verified ·
1 Parent(s): 8d6ec75

Training in progress, step 1000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0e9eed224d3bd813c5df4c08c85c0e11044545cfb849478ca85f994c86316646
3
  size 2384234968
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fa45215f37f017ea114f5d41b72d95e062c31d219651ae1208803b3cc29ec1d
3
  size 2384234968
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:552a7fd8dc36826bd7d6cd9f01c77507ab64af35744990ede135cb2acdbe3f53
3
  size 4768662910
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:779732aa59c1ab3aa00d1a1a4bafcd133473f97fb57c9527840c5270b6084f09
3
  size 4768662910
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c062f7f375beded48b5337f5a3f3a5cb38807fa3e85dbf3e294c0ab6b627bfc2
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5edb34d031c0c2b447f3eaadb401a4c1e7e7e6d8c096e28b7092e01a8bd48c92
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e0c8d55f92f35c67a58bda9590da57d8bcab1a0dd313128c3bee737dbccd6b81
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39e70346aa38efc4a7628079f4588786c86aea2dc16a2dfc23ef706d27a775bd
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.25903380391141045,
6
  "eval_steps": 100,
7
- "global_step": 500,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -498,6 +498,496 @@
498
  "eval_samples_per_second": 9.671,
499
  "eval_steps_per_second": 1.212,
500
  "step": 500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
501
  }
502
  ],
503
  "logging_steps": 10,
@@ -517,7 +1007,7 @@
517
  "attributes": {}
518
  }
519
  },
520
- "total_flos": 5412464099328000.0,
521
  "train_batch_size": 1,
522
  "trial_name": null,
523
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.5180676078228209,
6
  "eval_steps": 100,
7
+ "global_step": 1000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
498
  "eval_samples_per_second": 9.671,
499
  "eval_steps_per_second": 1.212,
500
  "step": 500
501
+ },
502
+ {
503
+ "epoch": 0.26421447998963865,
504
+ "grad_norm": 0.9347884654998779,
505
+ "learning_rate": 4.090385722510075e-05,
506
+ "loss": 0.1742,
507
+ "mean_token_accuracy": 0.9687010742723942,
508
+ "num_tokens": 2088960.0,
509
+ "step": 510
510
+ },
511
+ {
512
+ "epoch": 0.26939515606786685,
513
+ "grad_norm": 1.08110511302948,
514
+ "learning_rate": 4.061600460564191e-05,
515
+ "loss": 0.1613,
516
+ "mean_token_accuracy": 0.9706947132945061,
517
+ "num_tokens": 2129920.0,
518
+ "step": 520
519
+ },
520
+ {
521
+ "epoch": 0.27457583214609504,
522
+ "grad_norm": 0.7867632508277893,
523
+ "learning_rate": 4.0328151986183077e-05,
524
+ "loss": 0.1762,
525
+ "mean_token_accuracy": 0.9684197597205639,
526
+ "num_tokens": 2170880.0,
527
+ "step": 530
528
+ },
529
+ {
530
+ "epoch": 0.2797565082243233,
531
+ "grad_norm": 0.8982529044151306,
532
+ "learning_rate": 4.004029936672424e-05,
533
+ "loss": 0.1807,
534
+ "mean_token_accuracy": 0.9678816005587578,
535
+ "num_tokens": 2211840.0,
536
+ "step": 540
537
+ },
538
+ {
539
+ "epoch": 0.2849371843025515,
540
+ "grad_norm": 1.3036047220230103,
541
+ "learning_rate": 3.9752446747265404e-05,
542
+ "loss": 0.1548,
543
+ "mean_token_accuracy": 0.971722112596035,
544
+ "num_tokens": 2252800.0,
545
+ "step": 550
546
+ },
547
+ {
548
+ "epoch": 0.2901178603807797,
549
+ "grad_norm": 0.9593939781188965,
550
+ "learning_rate": 3.9464594127806564e-05,
551
+ "loss": 0.169,
552
+ "mean_token_accuracy": 0.9703033223748208,
553
+ "num_tokens": 2293760.0,
554
+ "step": 560
555
+ },
556
+ {
557
+ "epoch": 0.2952985364590079,
558
+ "grad_norm": 1.2803624868392944,
559
+ "learning_rate": 3.9176741508347724e-05,
560
+ "loss": 0.1616,
561
+ "mean_token_accuracy": 0.9705234833061696,
562
+ "num_tokens": 2334720.0,
563
+ "step": 570
564
+ },
565
+ {
566
+ "epoch": 0.3004792125372361,
567
+ "grad_norm": 0.7705172896385193,
568
+ "learning_rate": 3.888888888888889e-05,
569
+ "loss": 0.1667,
570
+ "mean_token_accuracy": 0.9702054738998414,
571
+ "num_tokens": 2375680.0,
572
+ "step": 580
573
+ },
574
+ {
575
+ "epoch": 0.30565988861546434,
576
+ "grad_norm": 1.0682554244995117,
577
+ "learning_rate": 3.860103626943005e-05,
578
+ "loss": 0.1813,
579
+ "mean_token_accuracy": 0.9671722054481506,
580
+ "num_tokens": 2416640.0,
581
+ "step": 590
582
+ },
583
+ {
584
+ "epoch": 0.31084056469369253,
585
+ "grad_norm": 0.9426740407943726,
586
+ "learning_rate": 3.831318364997122e-05,
587
+ "loss": 0.1711,
588
+ "step": 600
589
+ },
590
+ {
591
+ "epoch": 0.31084056469369253,
592
+ "eval_loss": 0.1666339486837387,
593
+ "eval_mean_token_accuracy": 0.9699312761772511,
594
+ "eval_num_tokens": 2457600.0,
595
+ "eval_runtime": 177.424,
596
+ "eval_samples_per_second": 9.672,
597
+ "eval_steps_per_second": 1.212,
598
+ "step": 600
599
+ },
600
+ {
601
+ "epoch": 0.31602124077192073,
602
+ "grad_norm": 2.044517755508423,
603
+ "learning_rate": 3.802533103051238e-05,
604
+ "loss": 0.1784,
605
+ "mean_token_accuracy": 0.9679060637950897,
606
+ "num_tokens": 2498560.0,
607
+ "step": 610
608
+ },
609
+ {
610
+ "epoch": 0.3212019168501489,
611
+ "grad_norm": 0.7277724146842957,
612
+ "learning_rate": 3.7737478411053545e-05,
613
+ "loss": 0.1609,
614
+ "mean_token_accuracy": 0.9709148697555066,
615
+ "num_tokens": 2539520.0,
616
+ "step": 620
617
+ },
618
+ {
619
+ "epoch": 0.3263825929283772,
620
+ "grad_norm": 0.849914014339447,
621
+ "learning_rate": 3.7449625791594705e-05,
622
+ "loss": 0.1724,
623
+ "mean_token_accuracy": 0.9685909919440746,
624
+ "num_tokens": 2580480.0,
625
+ "step": 630
626
+ },
627
+ {
628
+ "epoch": 0.3315632690066054,
629
+ "grad_norm": 0.7287548780441284,
630
+ "learning_rate": 3.7161773172135865e-05,
631
+ "loss": 0.1599,
632
+ "mean_token_accuracy": 0.9709393352270126,
633
+ "num_tokens": 2621440.0,
634
+ "step": 640
635
+ },
636
+ {
637
+ "epoch": 0.3367439450848336,
638
+ "grad_norm": 0.8491587042808533,
639
+ "learning_rate": 3.687392055267703e-05,
640
+ "loss": 0.1798,
641
+ "mean_token_accuracy": 0.9670988254249095,
642
+ "num_tokens": 2662400.0,
643
+ "step": 650
644
+ },
645
+ {
646
+ "epoch": 0.34192462116306177,
647
+ "grad_norm": 0.9557061791419983,
648
+ "learning_rate": 3.658606793321819e-05,
649
+ "loss": 0.1634,
650
+ "mean_token_accuracy": 0.9707680970430375,
651
+ "num_tokens": 2703360.0,
652
+ "step": 660
653
+ },
654
+ {
655
+ "epoch": 0.34710529724128997,
656
+ "grad_norm": 0.902422308921814,
657
+ "learning_rate": 3.629821531375936e-05,
658
+ "loss": 0.1582,
659
+ "mean_token_accuracy": 0.9710371777415275,
660
+ "num_tokens": 2744320.0,
661
+ "step": 670
662
+ },
663
+ {
664
+ "epoch": 0.3522859733195182,
665
+ "grad_norm": 0.7796424627304077,
666
+ "learning_rate": 3.601036269430052e-05,
667
+ "loss": 0.167,
668
+ "mean_token_accuracy": 0.9700097791850567,
669
+ "num_tokens": 2785280.0,
670
+ "step": 680
671
+ },
672
+ {
673
+ "epoch": 0.3574666493977464,
674
+ "grad_norm": 0.7159249186515808,
675
+ "learning_rate": 3.572251007484168e-05,
676
+ "loss": 0.1636,
677
+ "mean_token_accuracy": 0.971477497369051,
678
+ "num_tokens": 2826240.0,
679
+ "step": 690
680
+ },
681
+ {
682
+ "epoch": 0.3626473254759746,
683
+ "grad_norm": 0.8235995769500732,
684
+ "learning_rate": 3.5434657455382846e-05,
685
+ "loss": 0.1782,
686
+ "step": 700
687
+ },
688
+ {
689
+ "epoch": 0.3626473254759746,
690
+ "eval_loss": 0.16461288928985596,
691
+ "eval_mean_token_accuracy": 0.9701497241508129,
692
+ "eval_num_tokens": 2867200.0,
693
+ "eval_runtime": 177.4307,
694
+ "eval_samples_per_second": 9.671,
695
+ "eval_steps_per_second": 1.212,
696
+ "step": 700
697
+ },
698
+ {
699
+ "epoch": 0.3678280015542028,
700
+ "grad_norm": 0.8182228803634644,
701
+ "learning_rate": 3.5146804835924006e-05,
702
+ "loss": 0.1719,
703
+ "mean_token_accuracy": 0.9677715227007866,
704
+ "num_tokens": 2908160.0,
705
+ "step": 710
706
+ },
707
+ {
708
+ "epoch": 0.373008677632431,
709
+ "grad_norm": 0.6779491901397705,
710
+ "learning_rate": 3.485895221646517e-05,
711
+ "loss": 0.1565,
712
+ "mean_token_accuracy": 0.9719422653317451,
713
+ "num_tokens": 2949120.0,
714
+ "step": 720
715
+ },
716
+ {
717
+ "epoch": 0.37818935371065926,
718
+ "grad_norm": 0.8998910784721375,
719
+ "learning_rate": 3.4571099597006333e-05,
720
+ "loss": 0.1682,
721
+ "mean_token_accuracy": 0.9695205435156822,
722
+ "num_tokens": 2990080.0,
723
+ "step": 730
724
+ },
725
+ {
726
+ "epoch": 0.38337002978888746,
727
+ "grad_norm": 0.994229257106781,
728
+ "learning_rate": 3.4283246977547494e-05,
729
+ "loss": 0.151,
730
+ "mean_token_accuracy": 0.9732632078230381,
731
+ "num_tokens": 3031040.0,
732
+ "step": 740
733
+ },
734
+ {
735
+ "epoch": 0.38855070586711565,
736
+ "grad_norm": 0.7867749929428101,
737
+ "learning_rate": 3.399539435808866e-05,
738
+ "loss": 0.1688,
739
+ "mean_token_accuracy": 0.9701565541327,
740
+ "num_tokens": 3072000.0,
741
+ "step": 750
742
+ },
743
+ {
744
+ "epoch": 0.39373138194534385,
745
+ "grad_norm": 0.8778353929519653,
746
+ "learning_rate": 3.370754173862982e-05,
747
+ "loss": 0.169,
748
+ "mean_token_accuracy": 0.9686888411641121,
749
+ "num_tokens": 3112960.0,
750
+ "step": 760
751
+ },
752
+ {
753
+ "epoch": 0.3989120580235721,
754
+ "grad_norm": 0.8085846304893494,
755
+ "learning_rate": 3.341968911917099e-05,
756
+ "loss": 0.155,
757
+ "mean_token_accuracy": 0.9710127174854278,
758
+ "num_tokens": 3153920.0,
759
+ "step": 770
760
+ },
761
+ {
762
+ "epoch": 0.4040927341018003,
763
+ "grad_norm": 0.9327249526977539,
764
+ "learning_rate": 3.313183649971215e-05,
765
+ "loss": 0.1571,
766
+ "mean_token_accuracy": 0.9713307224214077,
767
+ "num_tokens": 3194880.0,
768
+ "step": 780
769
+ },
770
+ {
771
+ "epoch": 0.4092734101800285,
772
+ "grad_norm": 0.7871725559234619,
773
+ "learning_rate": 3.284398388025331e-05,
774
+ "loss": 0.1626,
775
+ "mean_token_accuracy": 0.9693493120372295,
776
+ "num_tokens": 3235840.0,
777
+ "step": 790
778
+ },
779
+ {
780
+ "epoch": 0.4144540862582567,
781
+ "grad_norm": 0.7657867670059204,
782
+ "learning_rate": 3.2556131260794475e-05,
783
+ "loss": 0.1475,
784
+ "step": 800
785
+ },
786
+ {
787
+ "epoch": 0.4144540862582567,
788
+ "eval_loss": 0.1627623736858368,
789
+ "eval_mean_token_accuracy": 0.9704717109369677,
790
+ "eval_num_tokens": 3276800.0,
791
+ "eval_runtime": 177.485,
792
+ "eval_samples_per_second": 9.668,
793
+ "eval_steps_per_second": 1.211,
794
+ "step": 800
795
+ },
796
+ {
797
+ "epoch": 0.4196347623364849,
798
+ "grad_norm": 0.8354265093803406,
799
+ "learning_rate": 3.2268278641335635e-05,
800
+ "loss": 0.1541,
801
+ "mean_token_accuracy": 0.9727005835622549,
802
+ "num_tokens": 3317760.0,
803
+ "step": 810
804
+ },
805
+ {
806
+ "epoch": 0.42481543841471314,
807
+ "grad_norm": 0.9434787631034851,
808
+ "learning_rate": 3.19804260218768e-05,
809
+ "loss": 0.1692,
810
+ "mean_token_accuracy": 0.9691780798137188,
811
+ "num_tokens": 3358720.0,
812
+ "step": 820
813
+ },
814
+ {
815
+ "epoch": 0.42999611449294134,
816
+ "grad_norm": 0.9330864548683167,
817
+ "learning_rate": 3.169257340241796e-05,
818
+ "loss": 0.1655,
819
+ "mean_token_accuracy": 0.9689579218626022,
820
+ "num_tokens": 3399680.0,
821
+ "step": 830
822
+ },
823
+ {
824
+ "epoch": 0.43517679057116954,
825
+ "grad_norm": 1.0121676921844482,
826
+ "learning_rate": 3.140472078295912e-05,
827
+ "loss": 0.1462,
828
+ "mean_token_accuracy": 0.9737524449825287,
829
+ "num_tokens": 3440640.0,
830
+ "step": 840
831
+ },
832
+ {
833
+ "epoch": 0.44035746664939773,
834
+ "grad_norm": 0.8801805973052979,
835
+ "learning_rate": 3.111686816350029e-05,
836
+ "loss": 0.1625,
837
+ "mean_token_accuracy": 0.9710616409778595,
838
+ "num_tokens": 3481600.0,
839
+ "step": 850
840
+ },
841
+ {
842
+ "epoch": 0.44553814272762593,
843
+ "grad_norm": 1.5192477703094482,
844
+ "learning_rate": 3.082901554404145e-05,
845
+ "loss": 0.1529,
846
+ "mean_token_accuracy": 0.9727984294295311,
847
+ "num_tokens": 3522560.0,
848
+ "step": 860
849
+ },
850
+ {
851
+ "epoch": 0.4507188188058542,
852
+ "grad_norm": 0.9691808223724365,
853
+ "learning_rate": 3.0541162924582616e-05,
854
+ "loss": 0.1564,
855
+ "mean_token_accuracy": 0.9713551864027977,
856
+ "num_tokens": 3563520.0,
857
+ "step": 870
858
+ },
859
+ {
860
+ "epoch": 0.4558994948840824,
861
+ "grad_norm": 0.9524751901626587,
862
+ "learning_rate": 3.0253310305123776e-05,
863
+ "loss": 0.1515,
864
+ "mean_token_accuracy": 0.9722113460302353,
865
+ "num_tokens": 3604480.0,
866
+ "step": 880
867
+ },
868
+ {
869
+ "epoch": 0.4610801709623106,
870
+ "grad_norm": 0.6413014531135559,
871
+ "learning_rate": 2.996545768566494e-05,
872
+ "loss": 0.1521,
873
+ "mean_token_accuracy": 0.9713551841676236,
874
+ "num_tokens": 3645440.0,
875
+ "step": 890
876
+ },
877
+ {
878
+ "epoch": 0.4662608470405388,
879
+ "grad_norm": 0.7614185214042664,
880
+ "learning_rate": 2.9677605066206103e-05,
881
+ "loss": 0.154,
882
+ "step": 900
883
+ },
884
+ {
885
+ "epoch": 0.4662608470405388,
886
+ "eval_loss": 0.16088420152664185,
887
+ "eval_mean_token_accuracy": 0.9706810574198879,
888
+ "eval_num_tokens": 3686400.0,
889
+ "eval_runtime": 177.4178,
890
+ "eval_samples_per_second": 9.672,
891
+ "eval_steps_per_second": 1.212,
892
+ "step": 900
893
+ },
894
+ {
895
+ "epoch": 0.471441523118767,
896
+ "grad_norm": 0.9191246032714844,
897
+ "learning_rate": 2.9389752446747267e-05,
898
+ "loss": 0.1726,
899
+ "mean_token_accuracy": 0.970560172200203,
900
+ "num_tokens": 3727360.0,
901
+ "step": 910
902
+ },
903
+ {
904
+ "epoch": 0.4766221991969952,
905
+ "grad_norm": 0.7410148978233337,
906
+ "learning_rate": 2.910189982728843e-05,
907
+ "loss": 0.1652,
908
+ "mean_token_accuracy": 0.9696673147380352,
909
+ "num_tokens": 3768320.0,
910
+ "step": 920
911
+ },
912
+ {
913
+ "epoch": 0.4818028752752234,
914
+ "grad_norm": 1.0268230438232422,
915
+ "learning_rate": 2.881404720782959e-05,
916
+ "loss": 0.1623,
917
+ "mean_token_accuracy": 0.9706213265657425,
918
+ "num_tokens": 3809280.0,
919
+ "step": 930
920
+ },
921
+ {
922
+ "epoch": 0.4869835513534516,
923
+ "grad_norm": 0.6969717144966125,
924
+ "learning_rate": 2.8526194588370757e-05,
925
+ "loss": 0.1566,
926
+ "mean_token_accuracy": 0.9719667293131351,
927
+ "num_tokens": 3850240.0,
928
+ "step": 940
929
+ },
930
+ {
931
+ "epoch": 0.4921642274316798,
932
+ "grad_norm": 0.7350455522537231,
933
+ "learning_rate": 2.8238341968911917e-05,
934
+ "loss": 0.1759,
935
+ "mean_token_accuracy": 0.9677592933177948,
936
+ "num_tokens": 3891200.0,
937
+ "step": 950
938
+ },
939
+ {
940
+ "epoch": 0.49734490350990807,
941
+ "grad_norm": 1.0810635089874268,
942
+ "learning_rate": 2.795048934945308e-05,
943
+ "loss": 0.147,
944
+ "mean_token_accuracy": 0.9727250464260578,
945
+ "num_tokens": 3932160.0,
946
+ "step": 960
947
+ },
948
+ {
949
+ "epoch": 0.5025255795881363,
950
+ "grad_norm": 1.0493009090423584,
951
+ "learning_rate": 2.7662636729994245e-05,
952
+ "loss": 0.142,
953
+ "mean_token_accuracy": 0.9734099790453911,
954
+ "num_tokens": 3973120.0,
955
+ "step": 970
956
+ },
957
+ {
958
+ "epoch": 0.5077062556663645,
959
+ "grad_norm": 0.788886308670044,
960
+ "learning_rate": 2.7374784110535408e-05,
961
+ "loss": 0.1547,
962
+ "mean_token_accuracy": 0.9717221103608609,
963
+ "num_tokens": 4014080.0,
964
+ "step": 980
965
+ },
966
+ {
967
+ "epoch": 0.5128869317445927,
968
+ "grad_norm": 0.8576117753982544,
969
+ "learning_rate": 2.708693149107657e-05,
970
+ "loss": 0.1543,
971
+ "mean_token_accuracy": 0.9718933448195457,
972
+ "num_tokens": 4055040.0,
973
+ "step": 990
974
+ },
975
+ {
976
+ "epoch": 0.5180676078228209,
977
+ "grad_norm": 0.7536800503730774,
978
+ "learning_rate": 2.6799078871617732e-05,
979
+ "loss": 0.1548,
980
+ "step": 1000
981
+ },
982
+ {
983
+ "epoch": 0.5180676078228209,
984
+ "eval_loss": 0.15957099199295044,
985
+ "eval_mean_token_accuracy": 0.9708847165107727,
986
+ "eval_num_tokens": 4096000.0,
987
+ "eval_runtime": 177.5545,
988
+ "eval_samples_per_second": 9.665,
989
+ "eval_steps_per_second": 1.211,
990
+ "step": 1000
991
  }
992
  ],
993
  "logging_steps": 10,
 
1007
  "attributes": {}
1008
  }
1009
  },
1010
+ "total_flos": 1.0824928198656e+16,
1011
  "train_batch_size": 1,
1012
  "trial_name": null,
1013
  "trial_params": null