ruwwww commited on
Commit
8d42ebd
·
verified ·
1 Parent(s): 462dc3b

qwen25_7b: checkpoint-134

Browse files
models/rosy-thesis-sft-v3-qwen25_7b/last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:669ec53f2d60c3b57ab1f80e81af47a7cb8f29e5ea9bd6bb6f5dc719be56f137
3
  size 161533192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf93a7dece8c4e6c6a0853e09728dcfaa9d3dc90efec21cf6a72475efeaf99f6
3
  size 161533192
models/rosy-thesis-sft-v3-qwen25_7b/last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cb08787a8f98f2016378d2e05e4915006f42f9c8aab490b617e9998f4c42ae7a
3
  size 82469637
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50b77839717562c2fdd5dbf61a9b9b861b1186d6f84fd7ce520181089cc68258
3
  size 82469637
models/rosy-thesis-sft-v3-qwen25_7b/last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3887a744cff2da51e333b51dddf3452bdb6cab201cc1726e92f6c47aec21f81
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f8bd0e6441d2a592c690db3f1c99f957068ef14cafac284e2f5340c0718c363
3
  size 1465
models/rosy-thesis-sft-v3-qwen25_7b/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": 0.32858707557502737,
6
  "eval_steps": 500,
7
- "global_step": 75,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -533,6 +533,419 @@
533
  "learning_rate": 4.9847856983226996e-05,
534
  "loss": 0.4938378632068634,
535
  "step": 75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
536
  }
537
  ],
538
  "logging_steps": 1,
@@ -547,12 +960,12 @@
547
  "should_evaluate": false,
548
  "should_log": false,
549
  "should_save": true,
550
- "should_training_stop": false
551
  },
552
  "attributes": {}
553
  }
554
  },
555
- "total_flos": 4.870668147100385e+17,
556
  "train_batch_size": 1,
557
  "trial_name": null,
558
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.5870755750273823,
6
  "eval_steps": 500,
7
+ "global_step": 134,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
533
  "learning_rate": 4.9847856983226996e-05,
534
  "loss": 0.4938378632068634,
535
  "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.3329682365826944,
539
+ "grad_norm": 0.18689356744289398,
540
+ "learning_rate": 4.9839963190492576e-05,
541
+ "loss": 0.5809458494186401,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.3373493975903614,
546
+ "grad_norm": 0.14314883947372437,
547
+ "learning_rate": 4.9831870420040126e-05,
548
+ "loss": 0.4784907102584839,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.3417305585980285,
553
+ "grad_norm": 0.17245735228061676,
554
+ "learning_rate": 4.982357873669589e-05,
555
+ "loss": 0.5192238688468933,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.34611171960569553,
560
+ "grad_norm": 0.1744268536567688,
561
+ "learning_rate": 4.981508820687943e-05,
562
+ "loss": 0.5425693988800049,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.35049288061336253,
567
+ "grad_norm": 0.1737440526485443,
568
+ "learning_rate": 4.9806398898603206e-05,
569
+ "loss": 0.4831830859184265,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.3548740416210296,
574
+ "grad_norm": 0.16995087265968323,
575
+ "learning_rate": 4.979751088147192e-05,
576
+ "loss": 0.3716176748275757,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.3592552026286966,
581
+ "grad_norm": 0.1777695119380951,
582
+ "learning_rate": 4.978842422668204e-05,
583
+ "loss": 0.41289517283439636,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.36363636363636365,
588
+ "grad_norm": 0.16704976558685303,
589
+ "learning_rate": 4.9779139007021184e-05,
590
+ "loss": 0.4531249701976776,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.36801752464403065,
595
+ "grad_norm": 0.17657746374607086,
596
+ "learning_rate": 4.9769655296867554e-05,
597
+ "loss": 0.48118069767951965,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.3723986856516977,
602
+ "grad_norm": 0.18034295737743378,
603
+ "learning_rate": 4.9759973172189334e-05,
604
+ "loss": 0.5912143588066101,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.37677984665936476,
609
+ "grad_norm": 0.15948718786239624,
610
+ "learning_rate": 4.975009271054409e-05,
611
+ "loss": 0.6708583235740662,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.38116100766703176,
616
+ "grad_norm": 0.1701880693435669,
617
+ "learning_rate": 4.974001399107816e-05,
618
+ "loss": 0.6332480907440186,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.3855421686746988,
623
+ "grad_norm": 0.19453497231006622,
624
+ "learning_rate": 4.972973709452597e-05,
625
+ "loss": 0.5207229852676392,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.3899233296823658,
630
+ "grad_norm": 0.16726645827293396,
631
+ "learning_rate": 4.971926210320944e-05,
632
+ "loss": 0.47982877492904663,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.39430449069003287,
637
+ "grad_norm": 0.23486323654651642,
638
+ "learning_rate": 4.9708589101037306e-05,
639
+ "loss": 0.5936954021453857,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.39868565169769987,
644
+ "grad_norm": 0.17078223824501038,
645
+ "learning_rate": 4.969771817350445e-05,
646
+ "loss": 0.44712382555007935,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.40306681270536693,
651
+ "grad_norm": 0.1948413848876953,
652
+ "learning_rate": 4.968664940769121e-05,
653
+ "loss": 0.5840800404548645,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.40744797371303393,
658
+ "grad_norm": 0.1884104311466217,
659
+ "learning_rate": 4.967538289226267e-05,
660
+ "loss": 0.4561323821544647,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.411829134720701,
665
+ "grad_norm": 0.1873539388179779,
666
+ "learning_rate": 4.9663918717467996e-05,
667
+ "loss": 0.5569716691970825,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.41621029572836804,
672
+ "grad_norm": 0.16940157115459442,
673
+ "learning_rate": 4.965225697513965e-05,
674
+ "loss": 0.44654473662376404,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.42059145673603504,
679
+ "grad_norm": 0.21278893947601318,
680
+ "learning_rate": 4.9640397758692715e-05,
681
+ "loss": 0.48756664991378784,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.4249726177437021,
686
+ "grad_norm": 0.19782011210918427,
687
+ "learning_rate": 4.962834116312409e-05,
688
+ "loss": 0.5450175404548645,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.4293537787513691,
693
+ "grad_norm": 0.20583702623844147,
694
+ "learning_rate": 4.961608728501178e-05,
695
+ "loss": 0.6042219996452332,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.43373493975903615,
700
+ "grad_norm": 0.22986851632595062,
701
+ "learning_rate": 4.960363622251409e-05,
702
+ "loss": 0.5261135697364807,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.43811610076670315,
707
+ "grad_norm": 0.1990327686071396,
708
+ "learning_rate": 4.959098807536885e-05,
709
+ "loss": 0.7777824401855469,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.4424972617743702,
714
+ "grad_norm": 0.19824311137199402,
715
+ "learning_rate": 4.957814294489261e-05,
716
+ "loss": 0.6562072038650513,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.44687842278203727,
721
+ "grad_norm": 0.17608404159545898,
722
+ "learning_rate": 4.956510093397983e-05,
723
+ "loss": 0.427765816450119,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.45125958378970427,
728
+ "grad_norm": 0.19360114634037018,
729
+ "learning_rate": 4.955186214710208e-05,
730
+ "loss": 0.5328750014305115,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.4556407447973713,
735
+ "grad_norm": 0.23907044529914856,
736
+ "learning_rate": 4.953842669030716e-05,
737
+ "loss": 0.5417872071266174,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.4600219058050383,
742
+ "grad_norm": 0.19629034399986267,
743
+ "learning_rate": 4.952479467121827e-05,
744
+ "loss": 0.46531593799591064,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.4644030668127054,
749
+ "grad_norm": 0.2029380202293396,
750
+ "learning_rate": 4.9510966199033174e-05,
751
+ "loss": 0.5872620344161987,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.4687842278203724,
756
+ "grad_norm": 0.22634181380271912,
757
+ "learning_rate": 4.949694138452327e-05,
758
+ "loss": 0.5499263405799866,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.47316538882803943,
763
+ "grad_norm": 0.19832900166511536,
764
+ "learning_rate": 4.948272034003275e-05,
765
+ "loss": 0.43778613209724426,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.47754654983570644,
770
+ "grad_norm": 0.1316746473312378,
771
+ "learning_rate": 4.9468303179477706e-05,
772
+ "loss": 0.2693359851837158,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.4819277108433735,
777
+ "grad_norm": 0.2278345823287964,
778
+ "learning_rate": 4.9453690018345144e-05,
779
+ "loss": 0.5739112496376038,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.48630887185104055,
784
+ "grad_norm": 0.17547903954982758,
785
+ "learning_rate": 4.943888097369216e-05,
786
+ "loss": 0.5327550172805786,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.49069003285870755,
791
+ "grad_norm": 0.20869563519954681,
792
+ "learning_rate": 4.942387616414492e-05,
793
+ "loss": 0.5430151224136353,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.4950711938663746,
798
+ "grad_norm": 0.20007291436195374,
799
+ "learning_rate": 4.940867570989777e-05,
800
+ "loss": 0.531854510307312,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.4994523548740416,
805
+ "grad_norm": 0.2701137363910675,
806
+ "learning_rate": 4.939327973271221e-05,
807
+ "loss": 0.5614831447601318,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.5038335158817087,
812
+ "grad_norm": 0.1753612905740738,
813
+ "learning_rate": 4.9377688355915994e-05,
814
+ "loss": 0.5305143594741821,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.5082146768893757,
819
+ "grad_norm": 0.18635676801204681,
820
+ "learning_rate": 4.936190170440208e-05,
821
+ "loss": 0.4858599901199341,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.5125958378970427,
826
+ "grad_norm": 0.2185445874929428,
827
+ "learning_rate": 4.934591990462766e-05,
828
+ "loss": 0.4726346433162689,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.5169769989047097,
833
+ "grad_norm": 0.23962336778640747,
834
+ "learning_rate": 4.932974308461311e-05,
835
+ "loss": 0.418442964553833,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.5213581599123768,
840
+ "grad_norm": 0.2367740273475647,
841
+ "learning_rate": 4.931337137394105e-05,
842
+ "loss": 0.5662670731544495,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.5257393209200438,
847
+ "grad_norm": 0.24327610433101654,
848
+ "learning_rate": 4.92968049037552e-05,
849
+ "loss": 0.4936223328113556,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.5301204819277109,
854
+ "grad_norm": 0.2300383746623993,
855
+ "learning_rate": 4.928004380675941e-05,
856
+ "loss": 0.5209953188896179,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.5345016429353778,
861
+ "grad_norm": 0.1960521787405014,
862
+ "learning_rate": 4.9263088217216544e-05,
863
+ "loss": 0.568078875541687,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.5388828039430449,
868
+ "grad_norm": 0.2199598103761673,
869
+ "learning_rate": 4.9245938270947435e-05,
870
+ "loss": 0.35799849033355713,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.5432639649507119,
875
+ "grad_norm": 0.21960218250751495,
876
+ "learning_rate": 4.922859410532978e-05,
877
+ "loss": 0.5813632011413574,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.547645125958379,
882
+ "grad_norm": 0.2728692591190338,
883
+ "learning_rate": 4.921105585929709e-05,
884
+ "loss": 0.6112500429153442,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.552026286966046,
889
+ "grad_norm": 0.2527194321155548,
890
+ "learning_rate": 4.9193323673337476e-05,
891
+ "loss": 0.3944776952266693,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.556407447973713,
896
+ "grad_norm": 0.23080946505069733,
897
+ "learning_rate": 4.9175397689492614e-05,
898
+ "loss": 0.6670535802841187,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.56078860898138,
903
+ "grad_norm": 0.23454329371452332,
904
+ "learning_rate": 4.915727805135657e-05,
905
+ "loss": 0.5472237467765808,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.5651697699890471,
910
+ "grad_norm": 0.23370526731014252,
911
+ "learning_rate": 4.9138964904074667e-05,
912
+ "loss": 0.4220460057258606,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.5695509309967142,
917
+ "grad_norm": 0.1963150054216385,
918
+ "learning_rate": 4.91204583943423e-05,
919
+ "loss": 0.46484917402267456,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.5739320920043811,
924
+ "grad_norm": 0.2073400616645813,
925
+ "learning_rate": 4.910175867040377e-05,
926
+ "loss": 0.40495163202285767,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.5783132530120482,
931
+ "grad_norm": 0.2338627427816391,
932
+ "learning_rate": 4.90828658820511e-05,
933
+ "loss": 0.507017970085144,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.5826944140197152,
938
+ "grad_norm": 0.21270239353179932,
939
+ "learning_rate": 4.9063780180622845e-05,
940
+ "loss": 0.5104067921638489,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.5870755750273823,
945
+ "grad_norm": 0.21894286572933197,
946
+ "learning_rate": 4.9044501719002844e-05,
947
+ "loss": 0.447110652923584,
948
+ "step": 134
949
  }
950
  ],
951
  "logging_steps": 1,
 
960
  "should_evaluate": false,
961
  "should_log": false,
962
  "should_save": true,
963
+ "should_training_stop": true
964
  },
965
  "attributes": {}
966
  }
967
  },
968
+ "total_flos": 8.637401282723512e+17,
969
  "train_batch_size": 1,
970
  "trial_name": null,
971
  "trial_params": null