Training in progress, step 4000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151061672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:948197f12c590c0dda47ebc1917b48df09d620ce9166a3a391d7061f05435036
|
| 3 |
size 151061672
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 297616186
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f16ff8a7474b79d86fe40d0ca22c7f7ed33fcee156c67270373e05fb5778508
|
| 3 |
size 297616186
|
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 sha256:1b63186d5b05381af62b3d7173fde5bc46daf7e9a7d51b3c424f534f7bee96e2
|
| 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 sha256:5638f21d0f0354e92b9a9f8dfa6255986109c390d6b1c6955ecb1e5acd3b9eab
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -874,6 +874,295 @@
|
|
| 874 |
"eval_steps_per_second": 0.223,
|
| 875 |
"eval_wer": 64.5190036982578,
|
| 876 |
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 877 |
}
|
| 878 |
],
|
| 879 |
"logging_steps": 25,
|
|
@@ -893,7 +1182,7 @@
|
|
| 893 |
"attributes": {}
|
| 894 |
}
|
| 895 |
},
|
| 896 |
-
"total_flos":
|
| 897 |
"train_batch_size": 32,
|
| 898 |
"trial_name": null,
|
| 899 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 63.24466426057123,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-4000",
|
| 4 |
+
"epoch": 20.02,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 4000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 874 |
"eval_steps_per_second": 0.223,
|
| 875 |
"eval_wer": 64.5190036982578,
|
| 876 |
"step": 3000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 15.02,
|
| 880 |
+
"grad_norm": 5.850539207458496,
|
| 881 |
+
"learning_rate": 4.395555555555556e-06,
|
| 882 |
+
"loss": 0.0914,
|
| 883 |
+
"step": 3025
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 15.025,
|
| 887 |
+
"grad_norm": 4.085538387298584,
|
| 888 |
+
"learning_rate": 4.34e-06,
|
| 889 |
+
"loss": 0.1051,
|
| 890 |
+
"step": 3050
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 15.03,
|
| 894 |
+
"grad_norm": 3.830420732498169,
|
| 895 |
+
"learning_rate": 4.284444444444445e-06,
|
| 896 |
+
"loss": 0.0627,
|
| 897 |
+
"step": 3075
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 15.035,
|
| 901 |
+
"grad_norm": 4.171704292297363,
|
| 902 |
+
"learning_rate": 4.228888888888889e-06,
|
| 903 |
+
"loss": 0.072,
|
| 904 |
+
"step": 3100
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 16.001,
|
| 908 |
+
"grad_norm": 6.426929950714111,
|
| 909 |
+
"learning_rate": 4.173333333333334e-06,
|
| 910 |
+
"loss": 0.0768,
|
| 911 |
+
"step": 3125
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 16.006,
|
| 915 |
+
"grad_norm": 2.9023313522338867,
|
| 916 |
+
"learning_rate": 4.117777777777779e-06,
|
| 917 |
+
"loss": 0.093,
|
| 918 |
+
"step": 3150
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 16.011,
|
| 922 |
+
"grad_norm": 3.1878390312194824,
|
| 923 |
+
"learning_rate": 4.062222222222223e-06,
|
| 924 |
+
"loss": 0.0799,
|
| 925 |
+
"step": 3175
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 16.016,
|
| 929 |
+
"grad_norm": 3.0205600261688232,
|
| 930 |
+
"learning_rate": 4.006666666666667e-06,
|
| 931 |
+
"loss": 0.0809,
|
| 932 |
+
"step": 3200
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 16.021,
|
| 936 |
+
"grad_norm": 4.082855701446533,
|
| 937 |
+
"learning_rate": 3.951111111111112e-06,
|
| 938 |
+
"loss": 0.0905,
|
| 939 |
+
"step": 3225
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 16.026,
|
| 943 |
+
"grad_norm": 3.3701670169830322,
|
| 944 |
+
"learning_rate": 3.895555555555556e-06,
|
| 945 |
+
"loss": 0.0874,
|
| 946 |
+
"step": 3250
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 16.031,
|
| 950 |
+
"grad_norm": 4.763386249542236,
|
| 951 |
+
"learning_rate": 3.8400000000000005e-06,
|
| 952 |
+
"loss": 0.0602,
|
| 953 |
+
"step": 3275
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 16.036,
|
| 957 |
+
"grad_norm": 3.008446216583252,
|
| 958 |
+
"learning_rate": 3.784444444444445e-06,
|
| 959 |
+
"loss": 0.0632,
|
| 960 |
+
"step": 3300
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 17.002,
|
| 964 |
+
"grad_norm": 4.296396732330322,
|
| 965 |
+
"learning_rate": 3.728888888888889e-06,
|
| 966 |
+
"loss": 0.0774,
|
| 967 |
+
"step": 3325
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 17.007,
|
| 971 |
+
"grad_norm": 2.4229137897491455,
|
| 972 |
+
"learning_rate": 3.673333333333334e-06,
|
| 973 |
+
"loss": 0.0821,
|
| 974 |
+
"step": 3350
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 17.012,
|
| 978 |
+
"grad_norm": 3.1641666889190674,
|
| 979 |
+
"learning_rate": 3.617777777777778e-06,
|
| 980 |
+
"loss": 0.0744,
|
| 981 |
+
"step": 3375
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 17.017,
|
| 985 |
+
"grad_norm": 3.2084319591522217,
|
| 986 |
+
"learning_rate": 3.5622222222222224e-06,
|
| 987 |
+
"loss": 0.0749,
|
| 988 |
+
"step": 3400
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 17.022,
|
| 992 |
+
"grad_norm": 3.579460620880127,
|
| 993 |
+
"learning_rate": 3.5066666666666673e-06,
|
| 994 |
+
"loss": 0.089,
|
| 995 |
+
"step": 3425
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 17.027,
|
| 999 |
+
"grad_norm": 4.045797824859619,
|
| 1000 |
+
"learning_rate": 3.4511111111111113e-06,
|
| 1001 |
+
"loss": 0.073,
|
| 1002 |
+
"step": 3450
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 17.032,
|
| 1006 |
+
"grad_norm": 4.456791400909424,
|
| 1007 |
+
"learning_rate": 3.3955555555555558e-06,
|
| 1008 |
+
"loss": 0.0558,
|
| 1009 |
+
"step": 3475
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 17.037,
|
| 1013 |
+
"grad_norm": 4.2783203125,
|
| 1014 |
+
"learning_rate": 3.3400000000000006e-06,
|
| 1015 |
+
"loss": 0.0571,
|
| 1016 |
+
"step": 3500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 18.003,
|
| 1020 |
+
"grad_norm": 2.9448039531707764,
|
| 1021 |
+
"learning_rate": 3.2844444444444447e-06,
|
| 1022 |
+
"loss": 0.0781,
|
| 1023 |
+
"step": 3525
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 18.008,
|
| 1027 |
+
"grad_norm": 2.74501371383667,
|
| 1028 |
+
"learning_rate": 3.228888888888889e-06,
|
| 1029 |
+
"loss": 0.0748,
|
| 1030 |
+
"step": 3550
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 18.013,
|
| 1034 |
+
"grad_norm": 2.291656494140625,
|
| 1035 |
+
"learning_rate": 3.173333333333334e-06,
|
| 1036 |
+
"loss": 0.0692,
|
| 1037 |
+
"step": 3575
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 18.018,
|
| 1041 |
+
"grad_norm": 2.8000614643096924,
|
| 1042 |
+
"learning_rate": 3.117777777777778e-06,
|
| 1043 |
+
"loss": 0.0696,
|
| 1044 |
+
"step": 3600
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 18.023,
|
| 1048 |
+
"grad_norm": 3.9665687084198,
|
| 1049 |
+
"learning_rate": 3.0622222222222225e-06,
|
| 1050 |
+
"loss": 0.0883,
|
| 1051 |
+
"step": 3625
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 18.028,
|
| 1055 |
+
"grad_norm": 3.8955137729644775,
|
| 1056 |
+
"learning_rate": 3.0066666666666674e-06,
|
| 1057 |
+
"loss": 0.063,
|
| 1058 |
+
"step": 3650
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 18.033,
|
| 1062 |
+
"grad_norm": 3.781052827835083,
|
| 1063 |
+
"learning_rate": 2.9511111111111114e-06,
|
| 1064 |
+
"loss": 0.0479,
|
| 1065 |
+
"step": 3675
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 18.038,
|
| 1069 |
+
"grad_norm": 3.0369510650634766,
|
| 1070 |
+
"learning_rate": 2.895555555555556e-06,
|
| 1071 |
+
"loss": 0.0542,
|
| 1072 |
+
"step": 3700
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 19.004,
|
| 1076 |
+
"grad_norm": 3.2044124603271484,
|
| 1077 |
+
"learning_rate": 2.84e-06,
|
| 1078 |
+
"loss": 0.0774,
|
| 1079 |
+
"step": 3725
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 19.009,
|
| 1083 |
+
"grad_norm": 3.152061939239502,
|
| 1084 |
+
"learning_rate": 2.784444444444445e-06,
|
| 1085 |
+
"loss": 0.0698,
|
| 1086 |
+
"step": 3750
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 19.014,
|
| 1090 |
+
"grad_norm": 2.7949397563934326,
|
| 1091 |
+
"learning_rate": 2.7288888888888893e-06,
|
| 1092 |
+
"loss": 0.0661,
|
| 1093 |
+
"step": 3775
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 19.019,
|
| 1097 |
+
"grad_norm": 2.393399477005005,
|
| 1098 |
+
"learning_rate": 2.6733333333333333e-06,
|
| 1099 |
+
"loss": 0.0644,
|
| 1100 |
+
"step": 3800
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 19.024,
|
| 1104 |
+
"grad_norm": 3.772813558578491,
|
| 1105 |
+
"learning_rate": 2.617777777777778e-06,
|
| 1106 |
+
"loss": 0.0869,
|
| 1107 |
+
"step": 3825
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 19.029,
|
| 1111 |
+
"grad_norm": 3.960970401763916,
|
| 1112 |
+
"learning_rate": 2.5622222222222226e-06,
|
| 1113 |
+
"loss": 0.0521,
|
| 1114 |
+
"step": 3850
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 19.034,
|
| 1118 |
+
"grad_norm": 4.602725505828857,
|
| 1119 |
+
"learning_rate": 2.5066666666666667e-06,
|
| 1120 |
+
"loss": 0.0473,
|
| 1121 |
+
"step": 3875
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 19.039,
|
| 1125 |
+
"grad_norm": 0.9137289524078369,
|
| 1126 |
+
"learning_rate": 2.451111111111111e-06,
|
| 1127 |
+
"loss": 0.0497,
|
| 1128 |
+
"step": 3900
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 20.005,
|
| 1132 |
+
"grad_norm": 3.5569727420806885,
|
| 1133 |
+
"learning_rate": 2.3955555555555556e-06,
|
| 1134 |
+
"loss": 0.0774,
|
| 1135 |
+
"step": 3925
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 20.01,
|
| 1139 |
+
"grad_norm": 2.2422502040863037,
|
| 1140 |
+
"learning_rate": 2.3400000000000005e-06,
|
| 1141 |
+
"loss": 0.0629,
|
| 1142 |
+
"step": 3950
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 20.015,
|
| 1146 |
+
"grad_norm": 2.4179162979125977,
|
| 1147 |
+
"learning_rate": 2.2844444444444445e-06,
|
| 1148 |
+
"loss": 0.0637,
|
| 1149 |
+
"step": 3975
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 20.02,
|
| 1153 |
+
"grad_norm": 3.352149724960327,
|
| 1154 |
+
"learning_rate": 2.228888888888889e-06,
|
| 1155 |
+
"loss": 0.0655,
|
| 1156 |
+
"step": 4000
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 20.02,
|
| 1160 |
+
"eval_loss": 0.20272822678089142,
|
| 1161 |
+
"eval_runtime": 353.3869,
|
| 1162 |
+
"eval_samples_per_second": 1.803,
|
| 1163 |
+
"eval_steps_per_second": 0.226,
|
| 1164 |
+
"eval_wer": 63.24466426057123,
|
| 1165 |
+
"step": 4000
|
| 1166 |
}
|
| 1167 |
],
|
| 1168 |
"logging_steps": 25,
|
|
|
|
| 1182 |
"attributes": {}
|
| 1183 |
}
|
| 1184 |
},
|
| 1185 |
+
"total_flos": 3.1389078528e+18,
|
| 1186 |
"train_batch_size": 32,
|
| 1187 |
"trial_name": null,
|
| 1188 |
"trial_params": null
|