Training in progress, step 145, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3537299144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13280e061892483b17784591c0537b2c9172267f68cfb64f88a6617afe581e9b
|
| 3 |
size 3537299144
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1798933287
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fcb1b8b6d38f585a891b6686134756afe1d409bd66adb0e6152848ac80c914e
|
| 3 |
size 1798933287
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:796ce27fe4783162e069352eb3f4da0f6937b8a24163640ebf02a34d44d692a2
|
| 3 |
size 1465
|
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
|
| 6 |
"eval_steps": 50,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -724,6 +724,321 @@
|
|
| 724 |
"eval_samples_per_second": 2.399,
|
| 725 |
"eval_steps_per_second": 0.602,
|
| 726 |
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 727 |
}
|
| 728 |
],
|
| 729 |
"logging_steps": 1,
|
|
@@ -738,12 +1053,12 @@
|
|
| 738 |
"should_evaluate": false,
|
| 739 |
"should_log": false,
|
| 740 |
"should_save": true,
|
| 741 |
-
"should_training_stop":
|
| 742 |
},
|
| 743 |
"attributes": {}
|
| 744 |
}
|
| 745 |
},
|
| 746 |
-
"total_flos": 2.
|
| 747 |
"train_batch_size": 8,
|
| 748 |
"trial_name": null,
|
| 749 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
"eval_steps": 50,
|
| 7 |
+
"global_step": 145,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 724 |
"eval_samples_per_second": 2.399,
|
| 725 |
"eval_steps_per_second": 0.602,
|
| 726 |
"step": 100
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.697754749568221,
|
| 730 |
+
"grad_norm": 0.25159376859664917,
|
| 731 |
+
"learning_rate": 2.339839617423318e-06,
|
| 732 |
+
"loss": 1.7943,
|
| 733 |
+
"step": 101
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.7046632124352331,
|
| 737 |
+
"grad_norm": 0.2905914783477783,
|
| 738 |
+
"learning_rate": 2.245515092739488e-06,
|
| 739 |
+
"loss": 2.2342,
|
| 740 |
+
"step": 102
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.7115716753022453,
|
| 744 |
+
"grad_norm": 0.27230361104011536,
|
| 745 |
+
"learning_rate": 2.1525775341577404e-06,
|
| 746 |
+
"loss": 1.9488,
|
| 747 |
+
"step": 103
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.7184801381692574,
|
| 751 |
+
"grad_norm": 0.2634965181350708,
|
| 752 |
+
"learning_rate": 2.061073738537635e-06,
|
| 753 |
+
"loss": 2.0043,
|
| 754 |
+
"step": 104
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.7253886010362695,
|
| 758 |
+
"grad_norm": 0.2540317475795746,
|
| 759 |
+
"learning_rate": 1.971049780795901e-06,
|
| 760 |
+
"loss": 1.8396,
|
| 761 |
+
"step": 105
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.7322970639032815,
|
| 765 |
+
"grad_norm": 0.29721683263778687,
|
| 766 |
+
"learning_rate": 1.8825509907063328e-06,
|
| 767 |
+
"loss": 2.0895,
|
| 768 |
+
"step": 106
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.7392055267702936,
|
| 772 |
+
"grad_norm": 0.26594671607017517,
|
| 773 |
+
"learning_rate": 1.7956219300748796e-06,
|
| 774 |
+
"loss": 1.9051,
|
| 775 |
+
"step": 107
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.7461139896373057,
|
| 779 |
+
"grad_norm": 0.25168678164482117,
|
| 780 |
+
"learning_rate": 1.7103063703014372e-06,
|
| 781 |
+
"loss": 1.8642,
|
| 782 |
+
"step": 108
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.7530224525043178,
|
| 786 |
+
"grad_norm": 0.2530485987663269,
|
| 787 |
+
"learning_rate": 1.6266472703396286e-06,
|
| 788 |
+
"loss": 1.8058,
|
| 789 |
+
"step": 109
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.7599309153713298,
|
| 793 |
+
"grad_norm": 0.2530233561992645,
|
| 794 |
+
"learning_rate": 1.544686755065677e-06,
|
| 795 |
+
"loss": 1.8528,
|
| 796 |
+
"step": 110
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.7668393782383419,
|
| 800 |
+
"grad_norm": 0.22255302965641022,
|
| 801 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 802 |
+
"loss": 1.6746,
|
| 803 |
+
"step": 111
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.7737478411053541,
|
| 807 |
+
"grad_norm": 0.2742765247821808,
|
| 808 |
+
"learning_rate": 1.3860256808630429e-06,
|
| 809 |
+
"loss": 2.0871,
|
| 810 |
+
"step": 112
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.7806563039723662,
|
| 814 |
+
"grad_norm": 0.2873839735984802,
|
| 815 |
+
"learning_rate": 1.3094050125632973e-06,
|
| 816 |
+
"loss": 2.0781,
|
| 817 |
+
"step": 113
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.7875647668393783,
|
| 821 |
+
"grad_norm": 0.2940923869609833,
|
| 822 |
+
"learning_rate": 1.234642669981946e-06,
|
| 823 |
+
"loss": 1.992,
|
| 824 |
+
"step": 114
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.7944732297063903,
|
| 828 |
+
"grad_norm": 0.2260187417268753,
|
| 829 |
+
"learning_rate": 1.1617762982099446e-06,
|
| 830 |
+
"loss": 1.791,
|
| 831 |
+
"step": 115
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.8013816925734024,
|
| 835 |
+
"grad_norm": 0.21561641991138458,
|
| 836 |
+
"learning_rate": 1.0908425876598512e-06,
|
| 837 |
+
"loss": 1.6785,
|
| 838 |
+
"step": 116
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.8082901554404145,
|
| 842 |
+
"grad_norm": 0.2794196605682373,
|
| 843 |
+
"learning_rate": 1.0218772555910955e-06,
|
| 844 |
+
"loss": 2.0029,
|
| 845 |
+
"step": 117
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.8151986183074266,
|
| 849 |
+
"grad_norm": 0.2553684711456299,
|
| 850 |
+
"learning_rate": 9.549150281252633e-07,
|
| 851 |
+
"loss": 1.906,
|
| 852 |
+
"step": 118
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.8221070811744386,
|
| 856 |
+
"grad_norm": 0.2819472849369049,
|
| 857 |
+
"learning_rate": 8.899896227604509e-07,
|
| 858 |
+
"loss": 2.0208,
|
| 859 |
+
"step": 119
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.8290155440414507,
|
| 863 |
+
"grad_norm": 0.2706139385700226,
|
| 864 |
+
"learning_rate": 8.271337313934869e-07,
|
| 865 |
+
"loss": 2.0251,
|
| 866 |
+
"step": 120
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.8359240069084629,
|
| 870 |
+
"grad_norm": 0.2831857204437256,
|
| 871 |
+
"learning_rate": 7.663790038585794e-07,
|
| 872 |
+
"loss": 2.0147,
|
| 873 |
+
"step": 121
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.842832469775475,
|
| 877 |
+
"grad_norm": 0.23544837534427643,
|
| 878 |
+
"learning_rate": 7.077560319906696e-07,
|
| 879 |
+
"loss": 1.7159,
|
| 880 |
+
"step": 122
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.8497409326424871,
|
| 884 |
+
"grad_norm": 0.23021404445171356,
|
| 885 |
+
"learning_rate": 6.512943342215234e-07,
|
| 886 |
+
"loss": 1.7992,
|
| 887 |
+
"step": 123
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.8566493955094991,
|
| 891 |
+
"grad_norm": 0.29619669914245605,
|
| 892 |
+
"learning_rate": 5.9702234071631e-07,
|
| 893 |
+
"loss": 2.1118,
|
| 894 |
+
"step": 124
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.8635578583765112,
|
| 898 |
+
"grad_norm": 0.2445152997970581,
|
| 899 |
+
"learning_rate": 5.449673790581611e-07,
|
| 900 |
+
"loss": 1.8761,
|
| 901 |
+
"step": 125
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.8704663212435233,
|
| 905 |
+
"grad_norm": 0.25963884592056274,
|
| 906 |
+
"learning_rate": 4.951556604879049e-07,
|
| 907 |
+
"loss": 2.0455,
|
| 908 |
+
"step": 126
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.8773747841105354,
|
| 912 |
+
"grad_norm": 0.24869026243686676,
|
| 913 |
+
"learning_rate": 4.4761226670592074e-07,
|
| 914 |
+
"loss": 1.8509,
|
| 915 |
+
"step": 127
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.8842832469775475,
|
| 919 |
+
"grad_norm": 0.24959927797317505,
|
| 920 |
+
"learning_rate": 4.0236113724274716e-07,
|
| 921 |
+
"loss": 1.8741,
|
| 922 |
+
"step": 128
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.8911917098445595,
|
| 926 |
+
"grad_norm": 0.26798906922340393,
|
| 927 |
+
"learning_rate": 3.5942505740480583e-07,
|
| 928 |
+
"loss": 1.8425,
|
| 929 |
+
"step": 129
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.8981001727115717,
|
| 933 |
+
"grad_norm": 0.2523483335971832,
|
| 934 |
+
"learning_rate": 3.18825646801314e-07,
|
| 935 |
+
"loss": 1.9115,
|
| 936 |
+
"step": 130
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.9050086355785838,
|
| 940 |
+
"grad_norm": 0.267639696598053,
|
| 941 |
+
"learning_rate": 2.8058334845816214e-07,
|
| 942 |
+
"loss": 2.0045,
|
| 943 |
+
"step": 131
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.9119170984455959,
|
| 947 |
+
"grad_norm": 0.27384859323501587,
|
| 948 |
+
"learning_rate": 2.447174185242324e-07,
|
| 949 |
+
"loss": 2.0145,
|
| 950 |
+
"step": 132
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.918825561312608,
|
| 954 |
+
"grad_norm": 0.2487836331129074,
|
| 955 |
+
"learning_rate": 2.1124591657534776e-07,
|
| 956 |
+
"loss": 1.7049,
|
| 957 |
+
"step": 133
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 0.92573402417962,
|
| 961 |
+
"grad_norm": 0.2451385110616684,
|
| 962 |
+
"learning_rate": 1.801856965207338e-07,
|
| 963 |
+
"loss": 1.7526,
|
| 964 |
+
"step": 134
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.9326424870466321,
|
| 968 |
+
"grad_norm": 0.2616844177246094,
|
| 969 |
+
"learning_rate": 1.5155239811656562e-07,
|
| 970 |
+
"loss": 1.868,
|
| 971 |
+
"step": 135
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.9395509499136442,
|
| 975 |
+
"grad_norm": 0.23664423823356628,
|
| 976 |
+
"learning_rate": 1.253604390908819e-07,
|
| 977 |
+
"loss": 1.6551,
|
| 978 |
+
"step": 136
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.9464594127806563,
|
| 982 |
+
"grad_norm": 0.2638615667819977,
|
| 983 |
+
"learning_rate": 1.0162300788382263e-07,
|
| 984 |
+
"loss": 2.0231,
|
| 985 |
+
"step": 137
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.9533678756476683,
|
| 989 |
+
"grad_norm": 0.24653677642345428,
|
| 990 |
+
"learning_rate": 8.035205700685167e-08,
|
| 991 |
+
"loss": 1.8138,
|
| 992 |
+
"step": 138
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.9602763385146805,
|
| 996 |
+
"grad_norm": 0.3076398968696594,
|
| 997 |
+
"learning_rate": 6.15582970243117e-08,
|
| 998 |
+
"loss": 2.2117,
|
| 999 |
+
"step": 139
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.9671848013816926,
|
| 1003 |
+
"grad_norm": 0.3007378876209259,
|
| 1004 |
+
"learning_rate": 4.52511911603265e-08,
|
| 1005 |
+
"loss": 2.063,
|
| 1006 |
+
"step": 140
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.9740932642487047,
|
| 1010 |
+
"grad_norm": 0.295623242855072,
|
| 1011 |
+
"learning_rate": 3.143895053378698e-08,
|
| 1012 |
+
"loss": 2.2005,
|
| 1013 |
+
"step": 141
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.9810017271157168,
|
| 1017 |
+
"grad_norm": 0.2502221465110779,
|
| 1018 |
+
"learning_rate": 2.012853002380466e-08,
|
| 1019 |
+
"loss": 1.9277,
|
| 1020 |
+
"step": 142
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.9879101899827288,
|
| 1024 |
+
"grad_norm": 0.26931339502334595,
|
| 1025 |
+
"learning_rate": 1.132562476771959e-08,
|
| 1026 |
+
"loss": 1.9401,
|
| 1027 |
+
"step": 143
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 0.9948186528497409,
|
| 1031 |
+
"grad_norm": 0.2606675922870636,
|
| 1032 |
+
"learning_rate": 5.034667293427053e-09,
|
| 1033 |
+
"loss": 1.9991,
|
| 1034 |
+
"step": 144
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.0,
|
| 1038 |
+
"grad_norm": 0.38600924611091614,
|
| 1039 |
+
"learning_rate": 1.2588252874673469e-09,
|
| 1040 |
+
"loss": 1.5566,
|
| 1041 |
+
"step": 145
|
| 1042 |
}
|
| 1043 |
],
|
| 1044 |
"logging_steps": 1,
|
|
|
|
| 1053 |
"should_evaluate": false,
|
| 1054 |
"should_log": false,
|
| 1055 |
"should_save": true,
|
| 1056 |
+
"should_training_stop": true
|
| 1057 |
},
|
| 1058 |
"attributes": {}
|
| 1059 |
}
|
| 1060 |
},
|
| 1061 |
+
"total_flos": 2.9813229542059776e+18,
|
| 1062 |
"train_batch_size": 8,
|
| 1063 |
"trial_name": null,
|
| 1064 |
"trial_params": null
|