Training in progress, step 150, 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 80013120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d578bb35c317e7e525db203822ae41cce5d685a3cde394257758d4000ede4c6
|
| 3 |
size 80013120
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 41119636
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d5812584bde22bf01fc0c56c6dbe8f3c19965476007b408bc249ebf8cf4edbd
|
| 3 |
size 41119636
|
last-checkpoint/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:eafb18d98f4c54a0a319ed9fd7490c4afd4dd5b2d57902a27826238fb340a960
|
| 3 |
size 14244
|
last-checkpoint/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 sha256:2231a499fc6249ec0c6a54e630ff27aff6a281425b76ecea459adede6c9680b9
|
| 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": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -707,6 +707,356 @@
|
|
| 707 |
"learning_rate": 9.999991128530895e-05,
|
| 708 |
"loss": 0.5321,
|
| 709 |
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 710 |
}
|
| 711 |
],
|
| 712 |
"logging_steps": 1,
|
|
@@ -726,7 +1076,7 @@
|
|
| 726 |
"attributes": {}
|
| 727 |
}
|
| 728 |
},
|
| 729 |
-
"total_flos":
|
| 730 |
"train_batch_size": 4,
|
| 731 |
"trial_name": null,
|
| 732 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9463722397476341,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 150,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 707 |
"learning_rate": 9.999991128530895e-05,
|
| 708 |
"loss": 0.5321,
|
| 709 |
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.637223974763407,
|
| 713 |
+
"grad_norm": 0.3380476236343384,
|
| 714 |
+
"learning_rate": 9.999990940780191e-05,
|
| 715 |
+
"loss": 0.6164,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.6435331230283912,
|
| 720 |
+
"grad_norm": 0.3839210867881775,
|
| 721 |
+
"learning_rate": 9.999990751063516e-05,
|
| 722 |
+
"loss": 0.6901,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6498422712933754,
|
| 727 |
+
"grad_norm": 0.36936458945274353,
|
| 728 |
+
"learning_rate": 9.999990559380867e-05,
|
| 729 |
+
"loss": 0.7658,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.6561514195583596,
|
| 734 |
+
"grad_norm": 0.37085050344467163,
|
| 735 |
+
"learning_rate": 9.999990365732244e-05,
|
| 736 |
+
"loss": 0.7844,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.6624605678233438,
|
| 741 |
+
"grad_norm": 0.36542758345603943,
|
| 742 |
+
"learning_rate": 9.999990170117648e-05,
|
| 743 |
+
"loss": 0.7223,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.668769716088328,
|
| 748 |
+
"grad_norm": 0.38479429483413696,
|
| 749 |
+
"learning_rate": 9.999989972537079e-05,
|
| 750 |
+
"loss": 0.7429,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.6750788643533123,
|
| 755 |
+
"grad_norm": 0.4295928180217743,
|
| 756 |
+
"learning_rate": 9.999989772990536e-05,
|
| 757 |
+
"loss": 0.6426,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.6813880126182965,
|
| 762 |
+
"grad_norm": 0.4287479817867279,
|
| 763 |
+
"learning_rate": 9.999989571478021e-05,
|
| 764 |
+
"loss": 0.7149,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.6876971608832808,
|
| 769 |
+
"grad_norm": 0.4698425531387329,
|
| 770 |
+
"learning_rate": 9.999989367999532e-05,
|
| 771 |
+
"loss": 0.707,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.694006309148265,
|
| 776 |
+
"grad_norm": 0.48254162073135376,
|
| 777 |
+
"learning_rate": 9.999989162555071e-05,
|
| 778 |
+
"loss": 0.7056,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.7003154574132492,
|
| 783 |
+
"grad_norm": 0.5717018842697144,
|
| 784 |
+
"learning_rate": 9.999988955144637e-05,
|
| 785 |
+
"loss": 0.6857,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.7066246056782335,
|
| 790 |
+
"grad_norm": 0.60892254114151,
|
| 791 |
+
"learning_rate": 9.99998874576823e-05,
|
| 792 |
+
"loss": 0.6499,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7129337539432177,
|
| 797 |
+
"grad_norm": 0.7676037549972534,
|
| 798 |
+
"learning_rate": 9.99998853442585e-05,
|
| 799 |
+
"loss": 0.7463,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.7192429022082019,
|
| 804 |
+
"grad_norm": 0.9815330505371094,
|
| 805 |
+
"learning_rate": 9.999988321117497e-05,
|
| 806 |
+
"loss": 0.8505,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.7255520504731862,
|
| 811 |
+
"grad_norm": 1.2696208953857422,
|
| 812 |
+
"learning_rate": 9.999988105843173e-05,
|
| 813 |
+
"loss": 0.7409,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.7318611987381703,
|
| 818 |
+
"grad_norm": 1.1130510568618774,
|
| 819 |
+
"learning_rate": 9.999987888602875e-05,
|
| 820 |
+
"loss": 0.3319,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.7381703470031545,
|
| 825 |
+
"grad_norm": 0.7321246862411499,
|
| 826 |
+
"learning_rate": 9.999987669396606e-05,
|
| 827 |
+
"loss": 0.38,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.7444794952681388,
|
| 832 |
+
"grad_norm": 0.27002546191215515,
|
| 833 |
+
"learning_rate": 9.999987448224363e-05,
|
| 834 |
+
"loss": 0.8287,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.750788643533123,
|
| 839 |
+
"grad_norm": 0.24602839350700378,
|
| 840 |
+
"learning_rate": 9.99998722508615e-05,
|
| 841 |
+
"loss": 0.9131,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.7570977917981072,
|
| 846 |
+
"grad_norm": 0.25864341855049133,
|
| 847 |
+
"learning_rate": 9.999986999981963e-05,
|
| 848 |
+
"loss": 0.9195,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.7634069400630915,
|
| 853 |
+
"grad_norm": 0.2722395956516266,
|
| 854 |
+
"learning_rate": 9.999986772911804e-05,
|
| 855 |
+
"loss": 0.9736,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.7697160883280757,
|
| 860 |
+
"grad_norm": 0.27622660994529724,
|
| 861 |
+
"learning_rate": 9.999986543875674e-05,
|
| 862 |
+
"loss": 0.867,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.7760252365930599,
|
| 867 |
+
"grad_norm": 0.3594439625740051,
|
| 868 |
+
"learning_rate": 9.999986312873572e-05,
|
| 869 |
+
"loss": 0.9813,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.7823343848580442,
|
| 874 |
+
"grad_norm": 0.3490639925003052,
|
| 875 |
+
"learning_rate": 9.999986079905499e-05,
|
| 876 |
+
"loss": 0.8835,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.7886435331230284,
|
| 881 |
+
"grad_norm": 0.34569522738456726,
|
| 882 |
+
"learning_rate": 9.999985844971453e-05,
|
| 883 |
+
"loss": 0.8828,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.7949526813880127,
|
| 888 |
+
"grad_norm": 0.3032309412956238,
|
| 889 |
+
"learning_rate": 9.999985608071438e-05,
|
| 890 |
+
"loss": 0.7422,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.8012618296529969,
|
| 895 |
+
"grad_norm": 0.2762659192085266,
|
| 896 |
+
"learning_rate": 9.99998536920545e-05,
|
| 897 |
+
"loss": 0.7403,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.807570977917981,
|
| 902 |
+
"grad_norm": 0.29897361993789673,
|
| 903 |
+
"learning_rate": 9.999985128373489e-05,
|
| 904 |
+
"loss": 0.852,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.8138801261829653,
|
| 909 |
+
"grad_norm": 0.2828134000301361,
|
| 910 |
+
"learning_rate": 9.999984885575557e-05,
|
| 911 |
+
"loss": 0.7355,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.8201892744479495,
|
| 916 |
+
"grad_norm": 0.28995025157928467,
|
| 917 |
+
"learning_rate": 9.999984640811656e-05,
|
| 918 |
+
"loss": 0.8051,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.8264984227129337,
|
| 923 |
+
"grad_norm": 0.2877126932144165,
|
| 924 |
+
"learning_rate": 9.999984394081783e-05,
|
| 925 |
+
"loss": 0.684,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.832807570977918,
|
| 930 |
+
"grad_norm": 0.32978740334510803,
|
| 931 |
+
"learning_rate": 9.999984145385939e-05,
|
| 932 |
+
"loss": 0.7361,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.8391167192429022,
|
| 937 |
+
"grad_norm": 0.2849004566669464,
|
| 938 |
+
"learning_rate": 9.999983894724123e-05,
|
| 939 |
+
"loss": 0.8116,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.8454258675078864,
|
| 944 |
+
"grad_norm": 0.2859273850917816,
|
| 945 |
+
"learning_rate": 9.999983642096338e-05,
|
| 946 |
+
"loss": 0.5324,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.8517350157728707,
|
| 951 |
+
"grad_norm": 0.3232719302177429,
|
| 952 |
+
"learning_rate": 9.999983387502581e-05,
|
| 953 |
+
"loss": 0.7514,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.8580441640378549,
|
| 958 |
+
"grad_norm": 0.3699396252632141,
|
| 959 |
+
"learning_rate": 9.999983130942854e-05,
|
| 960 |
+
"loss": 0.9407,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.8643533123028391,
|
| 965 |
+
"grad_norm": 0.3257301449775696,
|
| 966 |
+
"learning_rate": 9.999982872417156e-05,
|
| 967 |
+
"loss": 0.6552,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.8706624605678234,
|
| 972 |
+
"grad_norm": 0.34637680649757385,
|
| 973 |
+
"learning_rate": 9.999982611925488e-05,
|
| 974 |
+
"loss": 0.7539,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.8769716088328076,
|
| 979 |
+
"grad_norm": 0.38663750886917114,
|
| 980 |
+
"learning_rate": 9.99998234946785e-05,
|
| 981 |
+
"loss": 0.7163,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.8832807570977917,
|
| 986 |
+
"grad_norm": 0.33887526392936707,
|
| 987 |
+
"learning_rate": 9.999982085044242e-05,
|
| 988 |
+
"loss": 0.6362,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.889589905362776,
|
| 993 |
+
"grad_norm": 0.3487693667411804,
|
| 994 |
+
"learning_rate": 9.999981818654662e-05,
|
| 995 |
+
"loss": 0.7196,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.8958990536277602,
|
| 1000 |
+
"grad_norm": 0.3876301348209381,
|
| 1001 |
+
"learning_rate": 9.999981550299115e-05,
|
| 1002 |
+
"loss": 0.8779,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.9022082018927445,
|
| 1007 |
+
"grad_norm": 0.39342001080513,
|
| 1008 |
+
"learning_rate": 9.999981279977596e-05,
|
| 1009 |
+
"loss": 0.6961,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.9085173501577287,
|
| 1014 |
+
"grad_norm": 0.37053290009498596,
|
| 1015 |
+
"learning_rate": 9.999981007690108e-05,
|
| 1016 |
+
"loss": 0.5283,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.9148264984227129,
|
| 1021 |
+
"grad_norm": 0.39710113406181335,
|
| 1022 |
+
"learning_rate": 9.99998073343665e-05,
|
| 1023 |
+
"loss": 0.6993,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.9211356466876972,
|
| 1028 |
+
"grad_norm": 0.3932151794433594,
|
| 1029 |
+
"learning_rate": 9.999980457217224e-05,
|
| 1030 |
+
"loss": 0.5497,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.9274447949526814,
|
| 1035 |
+
"grad_norm": 0.41066697239875793,
|
| 1036 |
+
"learning_rate": 9.999980179031826e-05,
|
| 1037 |
+
"loss": 0.5879,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.9337539432176656,
|
| 1042 |
+
"grad_norm": 0.5037790536880493,
|
| 1043 |
+
"learning_rate": 9.99997989888046e-05,
|
| 1044 |
+
"loss": 0.8422,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.9400630914826499,
|
| 1049 |
+
"grad_norm": 0.5321910381317139,
|
| 1050 |
+
"learning_rate": 9.999979616763126e-05,
|
| 1051 |
+
"loss": 0.7902,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.9463722397476341,
|
| 1056 |
+
"grad_norm": 0.5393354892730713,
|
| 1057 |
+
"learning_rate": 9.99997933267982e-05,
|
| 1058 |
+
"loss": 0.6235,
|
| 1059 |
+
"step": 150
|
| 1060 |
}
|
| 1061 |
],
|
| 1062 |
"logging_steps": 1,
|
|
|
|
| 1076 |
"attributes": {}
|
| 1077 |
}
|
| 1078 |
},
|
| 1079 |
+
"total_flos": 1.1026632368612966e+17,
|
| 1080 |
"train_batch_size": 4,
|
| 1081 |
"trial_name": null,
|
| 1082 |
"trial_params": null
|