Upload folder using huggingface_hub
Browse files- model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +303 -3
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 497774208
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dcf45dc8e7790a173031467425b33f209390ed65dda68e67a91b7ddb3823a0e
|
| 3 |
size 497774208
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 995642298
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cec3c3c1a04720dbe28eb2b587829f6d45d7fe8de1b3bc974856ce359bb820c6
|
| 3 |
size 995642298
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b3711c485a8110fde599bdc760d38ac91164b4fb784ba247f14c643efbbb747
|
| 3 |
size 14244
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e88092444e9bb13863e011a228498e34c469dadf19c7e73fab77666f8575a1bd
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -757,6 +757,306 @@
|
|
| 757 |
"eval_samples_per_second": 12.083,
|
| 758 |
"eval_steps_per_second": 1.511,
|
| 759 |
"step": 5000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 760 |
}
|
| 761 |
],
|
| 762 |
"logging_steps": 100,
|
|
@@ -776,7 +1076,7 @@
|
|
| 776 |
"attributes": {}
|
| 777 |
}
|
| 778 |
},
|
| 779 |
-
"total_flos": 1.
|
| 780 |
"train_batch_size": 8,
|
| 781 |
"trial_name": null,
|
| 782 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7284837131855552,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 7000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 757 |
"eval_samples_per_second": 12.083,
|
| 758 |
"eval_steps_per_second": 1.511,
|
| 759 |
"step": 5000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.5307524196066188,
|
| 763 |
+
"grad_norm": 7.771067142486572,
|
| 764 |
+
"learning_rate": 4.115412633988969e-05,
|
| 765 |
+
"loss": 3.1719,
|
| 766 |
+
"step": 5100
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5307524196066188,
|
| 770 |
+
"eval_loss": 3.8357627391815186,
|
| 771 |
+
"eval_runtime": 3461.6324,
|
| 772 |
+
"eval_samples_per_second": 12.113,
|
| 773 |
+
"eval_steps_per_second": 1.514,
|
| 774 |
+
"step": 5100
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.5411593297949838,
|
| 778 |
+
"grad_norm": 4.688082218170166,
|
| 779 |
+
"learning_rate": 4.098067783675027e-05,
|
| 780 |
+
"loss": 3.2683,
|
| 781 |
+
"step": 5200
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.5411593297949838,
|
| 785 |
+
"eval_loss": 3.847905158996582,
|
| 786 |
+
"eval_runtime": 3464.2184,
|
| 787 |
+
"eval_samples_per_second": 12.104,
|
| 788 |
+
"eval_steps_per_second": 1.513,
|
| 789 |
+
"step": 5200
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.551566239983349,
|
| 793 |
+
"grad_norm": 5.934951305389404,
|
| 794 |
+
"learning_rate": 4.0807229333610856e-05,
|
| 795 |
+
"loss": 3.2195,
|
| 796 |
+
"step": 5300
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.551566239983349,
|
| 800 |
+
"eval_loss": 3.8351073265075684,
|
| 801 |
+
"eval_runtime": 3462.7613,
|
| 802 |
+
"eval_samples_per_second": 12.109,
|
| 803 |
+
"eval_steps_per_second": 1.514,
|
| 804 |
+
"step": 5300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.561973150171714,
|
| 808 |
+
"grad_norm": 8.332444190979004,
|
| 809 |
+
"learning_rate": 4.0633780830471436e-05,
|
| 810 |
+
"loss": 3.2279,
|
| 811 |
+
"step": 5400
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.561973150171714,
|
| 815 |
+
"eval_loss": 3.827998638153076,
|
| 816 |
+
"eval_runtime": 3460.3148,
|
| 817 |
+
"eval_samples_per_second": 12.118,
|
| 818 |
+
"eval_steps_per_second": 1.515,
|
| 819 |
+
"step": 5400
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.572380060360079,
|
| 823 |
+
"grad_norm": 4.6214599609375,
|
| 824 |
+
"learning_rate": 4.046033232733202e-05,
|
| 825 |
+
"loss": 3.2472,
|
| 826 |
+
"step": 5500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.572380060360079,
|
| 830 |
+
"eval_loss": 3.8354616165161133,
|
| 831 |
+
"eval_runtime": 3550.6038,
|
| 832 |
+
"eval_samples_per_second": 11.81,
|
| 833 |
+
"eval_steps_per_second": 1.476,
|
| 834 |
+
"step": 5500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.5827869705484442,
|
| 838 |
+
"grad_norm": 7.2509589195251465,
|
| 839 |
+
"learning_rate": 4.02868838241926e-05,
|
| 840 |
+
"loss": 3.2335,
|
| 841 |
+
"step": 5600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.5827869705484442,
|
| 845 |
+
"eval_loss": 3.829226016998291,
|
| 846 |
+
"eval_runtime": 3930.9878,
|
| 847 |
+
"eval_samples_per_second": 10.667,
|
| 848 |
+
"eval_steps_per_second": 1.334,
|
| 849 |
+
"step": 5600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5931938807368092,
|
| 853 |
+
"grad_norm": 5.364116191864014,
|
| 854 |
+
"learning_rate": 4.0113435321053185e-05,
|
| 855 |
+
"loss": 3.1955,
|
| 856 |
+
"step": 5700
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5931938807368092,
|
| 860 |
+
"eval_loss": 3.8261866569519043,
|
| 861 |
+
"eval_runtime": 3914.6993,
|
| 862 |
+
"eval_samples_per_second": 10.711,
|
| 863 |
+
"eval_steps_per_second": 1.339,
|
| 864 |
+
"step": 5700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.6036007909251743,
|
| 868 |
+
"grad_norm": 7.65122652053833,
|
| 869 |
+
"learning_rate": 3.9939986817913766e-05,
|
| 870 |
+
"loss": 3.1441,
|
| 871 |
+
"step": 5800
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.6036007909251743,
|
| 875 |
+
"eval_loss": 3.842827796936035,
|
| 876 |
+
"eval_runtime": 3638.8184,
|
| 877 |
+
"eval_samples_per_second": 11.524,
|
| 878 |
+
"eval_steps_per_second": 1.441,
|
| 879 |
+
"step": 5800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.6140077011135394,
|
| 883 |
+
"grad_norm": 5.782172203063965,
|
| 884 |
+
"learning_rate": 3.976653831477434e-05,
|
| 885 |
+
"loss": 3.1673,
|
| 886 |
+
"step": 5900
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.6140077011135394,
|
| 890 |
+
"eval_loss": 3.8224923610687256,
|
| 891 |
+
"eval_runtime": 3852.8778,
|
| 892 |
+
"eval_samples_per_second": 10.883,
|
| 893 |
+
"eval_steps_per_second": 1.361,
|
| 894 |
+
"step": 5900
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.6244146113019045,
|
| 898 |
+
"grad_norm": 5.476900100708008,
|
| 899 |
+
"learning_rate": 3.959308981163493e-05,
|
| 900 |
+
"loss": 3.12,
|
| 901 |
+
"step": 6000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.6244146113019045,
|
| 905 |
+
"eval_loss": 3.8422653675079346,
|
| 906 |
+
"eval_runtime": 3460.5604,
|
| 907 |
+
"eval_samples_per_second": 12.117,
|
| 908 |
+
"eval_steps_per_second": 1.515,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.6348215214902695,
|
| 913 |
+
"grad_norm": 4.047117710113525,
|
| 914 |
+
"learning_rate": 3.941964130849551e-05,
|
| 915 |
+
"loss": 3.1838,
|
| 916 |
+
"step": 6100
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.6348215214902695,
|
| 920 |
+
"eval_loss": 3.8332009315490723,
|
| 921 |
+
"eval_runtime": 3464.3723,
|
| 922 |
+
"eval_samples_per_second": 12.104,
|
| 923 |
+
"eval_steps_per_second": 1.513,
|
| 924 |
+
"step": 6100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.6452284316786346,
|
| 928 |
+
"grad_norm": 5.556035995483398,
|
| 929 |
+
"learning_rate": 3.924619280535609e-05,
|
| 930 |
+
"loss": 3.1586,
|
| 931 |
+
"step": 6200
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.6452284316786346,
|
| 935 |
+
"eval_loss": 3.8465607166290283,
|
| 936 |
+
"eval_runtime": 3461.4127,
|
| 937 |
+
"eval_samples_per_second": 12.114,
|
| 938 |
+
"eval_steps_per_second": 1.514,
|
| 939 |
+
"step": 6200
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.6556353418669997,
|
| 943 |
+
"grad_norm": 5.888989448547363,
|
| 944 |
+
"learning_rate": 3.9072744302216676e-05,
|
| 945 |
+
"loss": 3.1398,
|
| 946 |
+
"step": 6300
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.6556353418669997,
|
| 950 |
+
"eval_loss": 3.8334028720855713,
|
| 951 |
+
"eval_runtime": 3490.2152,
|
| 952 |
+
"eval_samples_per_second": 12.014,
|
| 953 |
+
"eval_steps_per_second": 1.502,
|
| 954 |
+
"step": 6300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6660422520553647,
|
| 958 |
+
"grad_norm": 6.468564987182617,
|
| 959 |
+
"learning_rate": 3.889929579907726e-05,
|
| 960 |
+
"loss": 3.193,
|
| 961 |
+
"step": 6400
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6660422520553647,
|
| 965 |
+
"eval_loss": 3.853891372680664,
|
| 966 |
+
"eval_runtime": 3456.8398,
|
| 967 |
+
"eval_samples_per_second": 12.13,
|
| 968 |
+
"eval_steps_per_second": 1.516,
|
| 969 |
+
"step": 6400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.6764491622437299,
|
| 973 |
+
"grad_norm": 5.189215660095215,
|
| 974 |
+
"learning_rate": 3.872584729593784e-05,
|
| 975 |
+
"loss": 3.204,
|
| 976 |
+
"step": 6500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.6764491622437299,
|
| 980 |
+
"eval_loss": 3.8300530910491943,
|
| 981 |
+
"eval_runtime": 3457.5384,
|
| 982 |
+
"eval_samples_per_second": 12.128,
|
| 983 |
+
"eval_steps_per_second": 1.516,
|
| 984 |
+
"step": 6500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.6868560724320949,
|
| 988 |
+
"grad_norm": 5.385128021240234,
|
| 989 |
+
"learning_rate": 3.8552398792798425e-05,
|
| 990 |
+
"loss": 3.0612,
|
| 991 |
+
"step": 6600
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.6868560724320949,
|
| 995 |
+
"eval_loss": 3.8288888931274414,
|
| 996 |
+
"eval_runtime": 3453.9786,
|
| 997 |
+
"eval_samples_per_second": 12.14,
|
| 998 |
+
"eval_steps_per_second": 1.518,
|
| 999 |
+
"step": 6600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.6972629826204599,
|
| 1003 |
+
"grad_norm": 4.424393653869629,
|
| 1004 |
+
"learning_rate": 3.8378950289659006e-05,
|
| 1005 |
+
"loss": 3.1726,
|
| 1006 |
+
"step": 6700
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.6972629826204599,
|
| 1010 |
+
"eval_loss": 3.820075511932373,
|
| 1011 |
+
"eval_runtime": 3454.0788,
|
| 1012 |
+
"eval_samples_per_second": 12.14,
|
| 1013 |
+
"eval_steps_per_second": 1.518,
|
| 1014 |
+
"step": 6700
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.7076698928088251,
|
| 1018 |
+
"grad_norm": 3.365438222885132,
|
| 1019 |
+
"learning_rate": 3.820550178651958e-05,
|
| 1020 |
+
"loss": 3.1238,
|
| 1021 |
+
"step": 6800
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.7076698928088251,
|
| 1025 |
+
"eval_loss": 3.8395471572875977,
|
| 1026 |
+
"eval_runtime": 3808.8554,
|
| 1027 |
+
"eval_samples_per_second": 11.009,
|
| 1028 |
+
"eval_steps_per_second": 1.376,
|
| 1029 |
+
"step": 6800
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.7180768029971901,
|
| 1033 |
+
"grad_norm": 5.473466873168945,
|
| 1034 |
+
"learning_rate": 3.803205328338017e-05,
|
| 1035 |
+
"loss": 3.198,
|
| 1036 |
+
"step": 6900
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 0.7180768029971901,
|
| 1040 |
+
"eval_loss": 3.831690549850464,
|
| 1041 |
+
"eval_runtime": 4160.6602,
|
| 1042 |
+
"eval_samples_per_second": 10.078,
|
| 1043 |
+
"eval_steps_per_second": 1.26,
|
| 1044 |
+
"step": 6900
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.7284837131855552,
|
| 1048 |
+
"grad_norm": 4.136599540710449,
|
| 1049 |
+
"learning_rate": 3.785860478024075e-05,
|
| 1050 |
+
"loss": 3.1025,
|
| 1051 |
+
"step": 7000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.7284837131855552,
|
| 1055 |
+
"eval_loss": 3.831153631210327,
|
| 1056 |
+
"eval_runtime": 3452.4147,
|
| 1057 |
+
"eval_samples_per_second": 12.146,
|
| 1058 |
+
"eval_steps_per_second": 1.518,
|
| 1059 |
+
"step": 7000
|
| 1060 |
}
|
| 1061 |
],
|
| 1062 |
"logging_steps": 100,
|
|
|
|
| 1076 |
"attributes": {}
|
| 1077 |
}
|
| 1078 |
},
|
| 1079 |
+
"total_flos": 1.4632353792e+16,
|
| 1080 |
"train_batch_size": 8,
|
| 1081 |
"trial_name": null,
|
| 1082 |
"trial_params": null
|