Training in progress, step 800, 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 204500912
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d24df5eae34ec3fdf38b1f2b6c298e42ad8fb07bed978b6111473d86177209f2
|
| 3 |
size 204500912
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 104062923
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:789ad1dcf0738c2b103827b3bf6b47de957c032a3a1cd442af3b12552cfbbd4f
|
| 3 |
size 104062923
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d6d8fafcd1ee268414be5acf0366296af5b03d60871978712eac1979cb42d65
|
| 3 |
size 14645
|
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:d72352d4b2a5c34b9f8314745f4274a6ceaf25c7fd28797dfe03562830d2faba
|
| 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":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -908,6 +908,306 @@
|
|
| 908 |
"rewards/margins": 149.69699096679688,
|
| 909 |
"rewards/rejected": -148.73023986816406,
|
| 910 |
"step": 600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 911 |
}
|
| 912 |
],
|
| 913 |
"logging_steps": 10,
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.122299767364573,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 800,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 908 |
"rewards/margins": 149.69699096679688,
|
| 909 |
"rewards/rejected": -148.73023986816406,
|
| 910 |
"step": 600
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 1.6194749086075109,
|
| 914 |
+
"grad_norm": 66.45056915283203,
|
| 915 |
+
"learning_rate": 4.344444444444444e-07,
|
| 916 |
+
"logits/chosen": 7.270118713378906,
|
| 917 |
+
"logits/rejected": 7.5960588455200195,
|
| 918 |
+
"logps/chosen": -503.7193298339844,
|
| 919 |
+
"logps/rejected": -1094.6021728515625,
|
| 920 |
+
"loss": 1.5946972846984864,
|
| 921 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 922 |
+
"rewards/chosen": -3.655120372772217,
|
| 923 |
+
"rewards/margins": 125.68217468261719,
|
| 924 |
+
"rewards/rejected": -129.33731079101562,
|
| 925 |
+
"step": 610
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.646061814556331,
|
| 929 |
+
"grad_norm": 280.2427978515625,
|
| 930 |
+
"learning_rate": 4.2333333333333334e-07,
|
| 931 |
+
"logits/chosen": 7.251768589019775,
|
| 932 |
+
"logits/rejected": 7.520864009857178,
|
| 933 |
+
"logps/chosen": -517.1514892578125,
|
| 934 |
+
"logps/rejected": -1172.587158203125,
|
| 935 |
+
"loss": 2.4477691650390625,
|
| 936 |
+
"rewards/accuracies": 0.925000011920929,
|
| 937 |
+
"rewards/chosen": 0.9268826246261597,
|
| 938 |
+
"rewards/margins": 142.18368530273438,
|
| 939 |
+
"rewards/rejected": -141.2567901611328,
|
| 940 |
+
"step": 620
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.6726487205051512,
|
| 944 |
+
"grad_norm": 1.0393255949020386,
|
| 945 |
+
"learning_rate": 4.122222222222222e-07,
|
| 946 |
+
"logits/chosen": 7.011075019836426,
|
| 947 |
+
"logits/rejected": 7.46621561050415,
|
| 948 |
+
"logps/chosen": -447.34124755859375,
|
| 949 |
+
"logps/rejected": -1143.3458251953125,
|
| 950 |
+
"loss": 1.0738434791564941,
|
| 951 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 952 |
+
"rewards/chosen": 1.6229969263076782,
|
| 953 |
+
"rewards/margins": 142.9796600341797,
|
| 954 |
+
"rewards/rejected": -141.35665893554688,
|
| 955 |
+
"step": 630
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.6992356264539714,
|
| 959 |
+
"grad_norm": 119.75847625732422,
|
| 960 |
+
"learning_rate": 4.0111111111111106e-07,
|
| 961 |
+
"logits/chosen": 6.9999237060546875,
|
| 962 |
+
"logits/rejected": 7.578449249267578,
|
| 963 |
+
"logps/chosen": -469.19012451171875,
|
| 964 |
+
"logps/rejected": -1200.680419921875,
|
| 965 |
+
"loss": 0.9937694549560547,
|
| 966 |
+
"rewards/accuracies": 0.949999988079071,
|
| 967 |
+
"rewards/chosen": 1.2244775295257568,
|
| 968 |
+
"rewards/margins": 147.14993286132812,
|
| 969 |
+
"rewards/rejected": -145.92544555664062,
|
| 970 |
+
"step": 640
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.7258225324027916,
|
| 974 |
+
"grad_norm": 27.75540542602539,
|
| 975 |
+
"learning_rate": 3.8999999999999997e-07,
|
| 976 |
+
"logits/chosen": 6.937554359436035,
|
| 977 |
+
"logits/rejected": 7.374237060546875,
|
| 978 |
+
"logps/chosen": -463.05438232421875,
|
| 979 |
+
"logps/rejected": -1168.0521240234375,
|
| 980 |
+
"loss": 0.39649856090545654,
|
| 981 |
+
"rewards/accuracies": 0.981249988079071,
|
| 982 |
+
"rewards/chosen": 3.329749345779419,
|
| 983 |
+
"rewards/margins": 145.75735473632812,
|
| 984 |
+
"rewards/rejected": -142.4276123046875,
|
| 985 |
+
"step": 650
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.7524094383516118,
|
| 989 |
+
"grad_norm": 31.218721389770508,
|
| 990 |
+
"learning_rate": 3.788888888888889e-07,
|
| 991 |
+
"logits/chosen": 7.072316646575928,
|
| 992 |
+
"logits/rejected": 7.550895690917969,
|
| 993 |
+
"logps/chosen": -483.40234375,
|
| 994 |
+
"logps/rejected": -1184.9073486328125,
|
| 995 |
+
"loss": 0.25033409595489503,
|
| 996 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 997 |
+
"rewards/chosen": -2.29609751701355,
|
| 998 |
+
"rewards/margins": 147.64785766601562,
|
| 999 |
+
"rewards/rejected": -149.94395446777344,
|
| 1000 |
+
"step": 660
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.778996344300432,
|
| 1004 |
+
"grad_norm": 6.414053359549143e-07,
|
| 1005 |
+
"learning_rate": 3.6777777777777774e-07,
|
| 1006 |
+
"logits/chosen": 7.303959846496582,
|
| 1007 |
+
"logits/rejected": 7.623525142669678,
|
| 1008 |
+
"logps/chosen": -508.6453552246094,
|
| 1009 |
+
"logps/rejected": -1223.940673828125,
|
| 1010 |
+
"loss": 0.31205618381500244,
|
| 1011 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1012 |
+
"rewards/chosen": -2.5321922302246094,
|
| 1013 |
+
"rewards/margins": 150.60842895507812,
|
| 1014 |
+
"rewards/rejected": -153.14060974121094,
|
| 1015 |
+
"step": 670
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 1.8055832502492524,
|
| 1019 |
+
"grad_norm": 1.0985974499902462e-12,
|
| 1020 |
+
"learning_rate": 3.5666666666666666e-07,
|
| 1021 |
+
"logits/chosen": 7.344334602355957,
|
| 1022 |
+
"logits/rejected": 7.8254547119140625,
|
| 1023 |
+
"logps/chosen": -532.2833251953125,
|
| 1024 |
+
"logps/rejected": -1228.1844482421875,
|
| 1025 |
+
"loss": 1.0204992294311523,
|
| 1026 |
+
"rewards/accuracies": 0.956250011920929,
|
| 1027 |
+
"rewards/chosen": 3.5795791149139404,
|
| 1028 |
+
"rewards/margins": 159.29647827148438,
|
| 1029 |
+
"rewards/rejected": -155.71688842773438,
|
| 1030 |
+
"step": 680
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.8321701561980723,
|
| 1034 |
+
"grad_norm": 66.1689453125,
|
| 1035 |
+
"learning_rate": 3.4555555555555557e-07,
|
| 1036 |
+
"logits/chosen": 7.0121636390686035,
|
| 1037 |
+
"logits/rejected": 7.367627143859863,
|
| 1038 |
+
"logps/chosen": -453.792236328125,
|
| 1039 |
+
"logps/rejected": -1141.1865234375,
|
| 1040 |
+
"loss": 0.38547022342681886,
|
| 1041 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1042 |
+
"rewards/chosen": -0.3728172183036804,
|
| 1043 |
+
"rewards/margins": 139.9238739013672,
|
| 1044 |
+
"rewards/rejected": -140.29669189453125,
|
| 1045 |
+
"step": 690
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.8587570621468927,
|
| 1049 |
+
"grad_norm": 1.7826409438004044e-20,
|
| 1050 |
+
"learning_rate": 3.3444444444444443e-07,
|
| 1051 |
+
"logits/chosen": 6.914497375488281,
|
| 1052 |
+
"logits/rejected": 7.344313144683838,
|
| 1053 |
+
"logps/chosen": -456.8873596191406,
|
| 1054 |
+
"logps/rejected": -1159.482666015625,
|
| 1055 |
+
"loss": 0.2864746332168579,
|
| 1056 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 1057 |
+
"rewards/chosen": -0.41872739791870117,
|
| 1058 |
+
"rewards/margins": 139.0840606689453,
|
| 1059 |
+
"rewards/rejected": -139.50277709960938,
|
| 1060 |
+
"step": 700
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.8853439680957127,
|
| 1064 |
+
"grad_norm": 0.6577161550521851,
|
| 1065 |
+
"learning_rate": 3.233333333333333e-07,
|
| 1066 |
+
"logits/chosen": 7.24100399017334,
|
| 1067 |
+
"logits/rejected": 7.729971408843994,
|
| 1068 |
+
"logps/chosen": -453.352783203125,
|
| 1069 |
+
"logps/rejected": -1139.1920166015625,
|
| 1070 |
+
"loss": 0.40453357696533204,
|
| 1071 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1072 |
+
"rewards/chosen": -0.5374351739883423,
|
| 1073 |
+
"rewards/margins": 134.17910766601562,
|
| 1074 |
+
"rewards/rejected": -134.71653747558594,
|
| 1075 |
+
"step": 710
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.911930874044533,
|
| 1079 |
+
"grad_norm": 58.065155029296875,
|
| 1080 |
+
"learning_rate": 3.122222222222222e-07,
|
| 1081 |
+
"logits/chosen": 7.2779541015625,
|
| 1082 |
+
"logits/rejected": 7.5862250328063965,
|
| 1083 |
+
"logps/chosen": -502.33489990234375,
|
| 1084 |
+
"logps/rejected": -1174.3145751953125,
|
| 1085 |
+
"loss": 0.2982128143310547,
|
| 1086 |
+
"rewards/accuracies": 0.949999988079071,
|
| 1087 |
+
"rewards/chosen": 0.528018593788147,
|
| 1088 |
+
"rewards/margins": 134.43031311035156,
|
| 1089 |
+
"rewards/rejected": -133.9022979736328,
|
| 1090 |
+
"step": 720
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.9385177799933533,
|
| 1094 |
+
"grad_norm": 3.057793140411377,
|
| 1095 |
+
"learning_rate": 3.011111111111111e-07,
|
| 1096 |
+
"logits/chosen": 7.26782751083374,
|
| 1097 |
+
"logits/rejected": 7.741539001464844,
|
| 1098 |
+
"logps/chosen": -496.8504943847656,
|
| 1099 |
+
"logps/rejected": -1235.9169921875,
|
| 1100 |
+
"loss": 0.8299455642700195,
|
| 1101 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 1102 |
+
"rewards/chosen": -0.2519731819629669,
|
| 1103 |
+
"rewards/margins": 152.95582580566406,
|
| 1104 |
+
"rewards/rejected": -153.20779418945312,
|
| 1105 |
+
"step": 730
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.9651046859421735,
|
| 1109 |
+
"grad_norm": 185.18174743652344,
|
| 1110 |
+
"learning_rate": 2.9e-07,
|
| 1111 |
+
"logits/chosen": 7.201784610748291,
|
| 1112 |
+
"logits/rejected": 7.585198402404785,
|
| 1113 |
+
"logps/chosen": -474.599853515625,
|
| 1114 |
+
"logps/rejected": -1184.7464599609375,
|
| 1115 |
+
"loss": 1.5328912734985352,
|
| 1116 |
+
"rewards/accuracies": 0.9312499761581421,
|
| 1117 |
+
"rewards/chosen": -3.7899742126464844,
|
| 1118 |
+
"rewards/margins": 142.3949432373047,
|
| 1119 |
+
"rewards/rejected": -146.1849365234375,
|
| 1120 |
+
"step": 740
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 1.9916915918909937,
|
| 1124 |
+
"grad_norm": 173.19436645507812,
|
| 1125 |
+
"learning_rate": 2.788888888888889e-07,
|
| 1126 |
+
"logits/chosen": 7.347403526306152,
|
| 1127 |
+
"logits/rejected": 7.8731865882873535,
|
| 1128 |
+
"logps/chosen": -484.5741271972656,
|
| 1129 |
+
"logps/rejected": -1241.9910888671875,
|
| 1130 |
+
"loss": 0.9414227485656739,
|
| 1131 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1132 |
+
"rewards/chosen": -0.26702070236206055,
|
| 1133 |
+
"rewards/margins": 149.2693328857422,
|
| 1134 |
+
"rewards/rejected": -149.53634643554688,
|
| 1135 |
+
"step": 750
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 2.015952143569292,
|
| 1139 |
+
"grad_norm": 277.53521728515625,
|
| 1140 |
+
"learning_rate": 2.6777777777777775e-07,
|
| 1141 |
+
"logits/chosen": 7.591332912445068,
|
| 1142 |
+
"logits/rejected": 7.759430408477783,
|
| 1143 |
+
"logps/chosen": -578.1312866210938,
|
| 1144 |
+
"logps/rejected": -1104.3414306640625,
|
| 1145 |
+
"loss": 0.3479891538619995,
|
| 1146 |
+
"rewards/accuracies": 0.9726027250289917,
|
| 1147 |
+
"rewards/chosen": -3.462564468383789,
|
| 1148 |
+
"rewards/margins": 117.48858642578125,
|
| 1149 |
+
"rewards/rejected": -120.95115661621094,
|
| 1150 |
+
"step": 760
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.0425390495181124,
|
| 1154 |
+
"grad_norm": 81.4224624633789,
|
| 1155 |
+
"learning_rate": 2.5666666666666666e-07,
|
| 1156 |
+
"logits/chosen": 6.820937156677246,
|
| 1157 |
+
"logits/rejected": 7.438076972961426,
|
| 1158 |
+
"logps/chosen": -445.38592529296875,
|
| 1159 |
+
"logps/rejected": -1255.2547607421875,
|
| 1160 |
+
"loss": 0.7632743835449218,
|
| 1161 |
+
"rewards/accuracies": 0.9624999761581421,
|
| 1162 |
+
"rewards/chosen": -1.3056232929229736,
|
| 1163 |
+
"rewards/margins": 162.7215576171875,
|
| 1164 |
+
"rewards/rejected": -164.02719116210938,
|
| 1165 |
+
"step": 770
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 2.0691259554669323,
|
| 1169 |
+
"grad_norm": 80.839111328125,
|
| 1170 |
+
"learning_rate": 2.455555555555555e-07,
|
| 1171 |
+
"logits/chosen": 7.089077949523926,
|
| 1172 |
+
"logits/rejected": 7.562623500823975,
|
| 1173 |
+
"logps/chosen": -479.9771423339844,
|
| 1174 |
+
"logps/rejected": -1212.7470703125,
|
| 1175 |
+
"loss": 0.37755522727966306,
|
| 1176 |
+
"rewards/accuracies": 0.96875,
|
| 1177 |
+
"rewards/chosen": -0.4610620439052582,
|
| 1178 |
+
"rewards/margins": 147.49462890625,
|
| 1179 |
+
"rewards/rejected": -147.9556884765625,
|
| 1180 |
+
"step": 780
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.0957128614157527,
|
| 1184 |
+
"grad_norm": 102.21258544921875,
|
| 1185 |
+
"learning_rate": 2.3444444444444444e-07,
|
| 1186 |
+
"logits/chosen": 7.348860740661621,
|
| 1187 |
+
"logits/rejected": 7.812272548675537,
|
| 1188 |
+
"logps/chosen": -545.9133911132812,
|
| 1189 |
+
"logps/rejected": -1209.3829345703125,
|
| 1190 |
+
"loss": 0.8432134628295899,
|
| 1191 |
+
"rewards/accuracies": 0.9437500238418579,
|
| 1192 |
+
"rewards/chosen": -4.405800819396973,
|
| 1193 |
+
"rewards/margins": 136.8076629638672,
|
| 1194 |
+
"rewards/rejected": -141.2134552001953,
|
| 1195 |
+
"step": 790
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 2.122299767364573,
|
| 1199 |
+
"grad_norm": 7.490438461303711,
|
| 1200 |
+
"learning_rate": 2.2333333333333332e-07,
|
| 1201 |
+
"logits/chosen": 6.928166389465332,
|
| 1202 |
+
"logits/rejected": 7.589695930480957,
|
| 1203 |
+
"logps/chosen": -411.7215881347656,
|
| 1204 |
+
"logps/rejected": -1234.5572509765625,
|
| 1205 |
+
"loss": 0.08373026251792907,
|
| 1206 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 1207 |
+
"rewards/chosen": -1.1743253469467163,
|
| 1208 |
+
"rewards/margins": 159.31773376464844,
|
| 1209 |
+
"rewards/rejected": -160.49208068847656,
|
| 1210 |
+
"step": 800
|
| 1211 |
}
|
| 1212 |
],
|
| 1213 |
"logging_steps": 10,
|