Training in progress, step 2000, 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 2718107304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd14d50acc2f98517bbef06b030c6c35a3358709f1d81f28d0924632e3a3327d
|
| 3 |
size 2718107304
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 145486330
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:732641cda8fbb415a9820a5a3bca30e63a51107e27c0f0f06200240f25dc5c64
|
| 3 |
size 145486330
|
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 sha256:6cc3defa9ba1ea15f769731d66d430abc7783c906f059b00b857dbeb10740c1d
|
| 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 sha256:181640f7107f64b366f0a1e14f0d45b5ddc4bab25cf83c8ffe834db0b38e9f04
|
| 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": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1065,6 +1065,364 @@
|
|
| 1065 |
"learning_rate": 9.042494563884404e-05,
|
| 1066 |
"loss": 2.8108,
|
| 1067 |
"step": 1500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1068 |
}
|
| 1069 |
],
|
| 1070 |
"logging_steps": 10,
|
|
@@ -1084,7 +1442,7 @@
|
|
| 1084 |
"attributes": {}
|
| 1085 |
}
|
| 1086 |
},
|
| 1087 |
-
"total_flos":
|
| 1088 |
"train_batch_size": 4,
|
| 1089 |
"trial_name": null,
|
| 1090 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.3640997633351538,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1065 |
"learning_rate": 9.042494563884404e-05,
|
| 1066 |
"loss": 2.8108,
|
| 1067 |
"step": 1500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.2748953213180411,
|
| 1071 |
+
"grad_norm": 7.8658061027526855,
|
| 1072 |
+
"learning_rate": 9.023900547419646e-05,
|
| 1073 |
+
"loss": 2.7663,
|
| 1074 |
+
"step": 1510
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.2767158201347169,
|
| 1078 |
+
"grad_norm": 15.445107460021973,
|
| 1079 |
+
"learning_rate": 9.005147228518174e-05,
|
| 1080 |
+
"loss": 2.7878,
|
| 1081 |
+
"step": 1520
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.2785363189513927,
|
| 1085 |
+
"grad_norm": 12.650901794433594,
|
| 1086 |
+
"learning_rate": 8.986235349606238e-05,
|
| 1087 |
+
"loss": 2.8219,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.2803568177680685,
|
| 1092 |
+
"grad_norm": 10.226774215698242,
|
| 1093 |
+
"learning_rate": 8.967165659387331e-05,
|
| 1094 |
+
"loss": 2.742,
|
| 1095 |
+
"step": 1540
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.2821773165847442,
|
| 1099 |
+
"grad_norm": 138.37210083007812,
|
| 1100 |
+
"learning_rate": 8.947938912812548e-05,
|
| 1101 |
+
"loss": 2.9524,
|
| 1102 |
+
"step": 1550
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.28399781540142,
|
| 1106 |
+
"grad_norm": 16.11450958251953,
|
| 1107 |
+
"learning_rate": 8.928555871050693e-05,
|
| 1108 |
+
"loss": 2.7966,
|
| 1109 |
+
"step": 1560
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.2858183142180958,
|
| 1113 |
+
"grad_norm": 10.201882362365723,
|
| 1114 |
+
"learning_rate": 8.909017301458156e-05,
|
| 1115 |
+
"loss": 2.8389,
|
| 1116 |
+
"step": 1570
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.2876388130347715,
|
| 1120 |
+
"grad_norm": 10.96867847442627,
|
| 1121 |
+
"learning_rate": 8.889323977548521e-05,
|
| 1122 |
+
"loss": 2.7495,
|
| 1123 |
+
"step": 1580
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.2894593118514473,
|
| 1127 |
+
"grad_norm": 10.814942359924316,
|
| 1128 |
+
"learning_rate": 8.869476678961954e-05,
|
| 1129 |
+
"loss": 2.7676,
|
| 1130 |
+
"step": 1590
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.2912798106681231,
|
| 1134 |
+
"grad_norm": 6.535337448120117,
|
| 1135 |
+
"learning_rate": 8.849476191434334e-05,
|
| 1136 |
+
"loss": 2.7589,
|
| 1137 |
+
"step": 1600
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.2931003094847988,
|
| 1141 |
+
"grad_norm": 7.036696434020996,
|
| 1142 |
+
"learning_rate": 8.829323306766142e-05,
|
| 1143 |
+
"loss": 2.7921,
|
| 1144 |
+
"step": 1610
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.2949208083014746,
|
| 1148 |
+
"grad_norm": 5.92086124420166,
|
| 1149 |
+
"learning_rate": 8.809018822791121e-05,
|
| 1150 |
+
"loss": 2.7267,
|
| 1151 |
+
"step": 1620
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.2967413071181504,
|
| 1155 |
+
"grad_norm": 6.186739921569824,
|
| 1156 |
+
"learning_rate": 8.788563543344688e-05,
|
| 1157 |
+
"loss": 2.795,
|
| 1158 |
+
"step": 1630
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.2985618059348261,
|
| 1162 |
+
"grad_norm": 8.154546737670898,
|
| 1163 |
+
"learning_rate": 8.767958278232112e-05,
|
| 1164 |
+
"loss": 2.7627,
|
| 1165 |
+
"step": 1640
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.3003823047515019,
|
| 1169 |
+
"grad_norm": 7.674529075622559,
|
| 1170 |
+
"learning_rate": 8.74720384319645e-05,
|
| 1171 |
+
"loss": 2.7996,
|
| 1172 |
+
"step": 1650
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.3022028035681777,
|
| 1176 |
+
"grad_norm": 6.348474025726318,
|
| 1177 |
+
"learning_rate": 8.726301059886259e-05,
|
| 1178 |
+
"loss": 2.7704,
|
| 1179 |
+
"step": 1660
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.3040233023848534,
|
| 1183 |
+
"grad_norm": 10.496267318725586,
|
| 1184 |
+
"learning_rate": 8.705250755823064e-05,
|
| 1185 |
+
"loss": 2.7591,
|
| 1186 |
+
"step": 1670
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.3058438012015292,
|
| 1190 |
+
"grad_norm": 102.05543518066406,
|
| 1191 |
+
"learning_rate": 8.684053764368598e-05,
|
| 1192 |
+
"loss": 2.8027,
|
| 1193 |
+
"step": 1680
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.307664300018205,
|
| 1197 |
+
"grad_norm": 8.403404235839844,
|
| 1198 |
+
"learning_rate": 8.662710924691805e-05,
|
| 1199 |
+
"loss": 2.8801,
|
| 1200 |
+
"step": 1690
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.3094847988348808,
|
| 1204 |
+
"grad_norm": 7.355569839477539,
|
| 1205 |
+
"learning_rate": 8.64122308173563e-05,
|
| 1206 |
+
"loss": 2.8346,
|
| 1207 |
+
"step": 1700
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.3113052976515565,
|
| 1211 |
+
"grad_norm": 12.551121711730957,
|
| 1212 |
+
"learning_rate": 8.61959108618356e-05,
|
| 1213 |
+
"loss": 2.8381,
|
| 1214 |
+
"step": 1710
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.3131257964682323,
|
| 1218 |
+
"grad_norm": 116.6989517211914,
|
| 1219 |
+
"learning_rate": 8.597815794425943e-05,
|
| 1220 |
+
"loss": 2.814,
|
| 1221 |
+
"step": 1720
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.3149462952849081,
|
| 1225 |
+
"grad_norm": 21.63788604736328,
|
| 1226 |
+
"learning_rate": 8.575898068526093e-05,
|
| 1227 |
+
"loss": 2.8389,
|
| 1228 |
+
"step": 1730
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.3167667941015838,
|
| 1232 |
+
"grad_norm": 143.42408752441406,
|
| 1233 |
+
"learning_rate": 8.553838776186158e-05,
|
| 1234 |
+
"loss": 2.8534,
|
| 1235 |
+
"step": 1740
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.3185872929182596,
|
| 1239 |
+
"grad_norm": 9.04028034210205,
|
| 1240 |
+
"learning_rate": 8.531638790712765e-05,
|
| 1241 |
+
"loss": 2.8186,
|
| 1242 |
+
"step": 1750
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.3204077917349354,
|
| 1246 |
+
"grad_norm": 11.659414291381836,
|
| 1247 |
+
"learning_rate": 8.509298990982453e-05,
|
| 1248 |
+
"loss": 2.8078,
|
| 1249 |
+
"step": 1760
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.3222282905516111,
|
| 1253 |
+
"grad_norm": 7.934113502502441,
|
| 1254 |
+
"learning_rate": 8.486820261406873e-05,
|
| 1255 |
+
"loss": 2.792,
|
| 1256 |
+
"step": 1770
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.3240487893682869,
|
| 1260 |
+
"grad_norm": 12.919567108154297,
|
| 1261 |
+
"learning_rate": 8.464203491897779e-05,
|
| 1262 |
+
"loss": 2.8111,
|
| 1263 |
+
"step": 1780
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.3258692881849627,
|
| 1267 |
+
"grad_norm": 13.67540454864502,
|
| 1268 |
+
"learning_rate": 8.441449577831801e-05,
|
| 1269 |
+
"loss": 2.8085,
|
| 1270 |
+
"step": 1790
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.3276897870016384,
|
| 1274 |
+
"grad_norm": 7.7655110359191895,
|
| 1275 |
+
"learning_rate": 8.418559420014984e-05,
|
| 1276 |
+
"loss": 2.7689,
|
| 1277 |
+
"step": 1800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.3295102858183142,
|
| 1281 |
+
"grad_norm": 8.168259620666504,
|
| 1282 |
+
"learning_rate": 8.395533924647141e-05,
|
| 1283 |
+
"loss": 2.7534,
|
| 1284 |
+
"step": 1810
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.33133078463499,
|
| 1288 |
+
"grad_norm": 14.387748718261719,
|
| 1289 |
+
"learning_rate": 8.372374003285968e-05,
|
| 1290 |
+
"loss": 2.8353,
|
| 1291 |
+
"step": 1820
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.3331512834516658,
|
| 1295 |
+
"grad_norm": 9.209723472595215,
|
| 1296 |
+
"learning_rate": 8.349080572810965e-05,
|
| 1297 |
+
"loss": 2.7837,
|
| 1298 |
+
"step": 1830
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.3349717822683415,
|
| 1302 |
+
"grad_norm": 9.160303115844727,
|
| 1303 |
+
"learning_rate": 8.325654555387123e-05,
|
| 1304 |
+
"loss": 2.8186,
|
| 1305 |
+
"step": 1840
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.3367922810850173,
|
| 1309 |
+
"grad_norm": 20.171415328979492,
|
| 1310 |
+
"learning_rate": 8.302096878428438e-05,
|
| 1311 |
+
"loss": 2.8011,
|
| 1312 |
+
"step": 1850
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.3386127799016931,
|
| 1316 |
+
"grad_norm": 29.545217514038086,
|
| 1317 |
+
"learning_rate": 8.278408474561169e-05,
|
| 1318 |
+
"loss": 2.7971,
|
| 1319 |
+
"step": 1860
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.3404332787183688,
|
| 1323 |
+
"grad_norm": 19.314136505126953,
|
| 1324 |
+
"learning_rate": 8.254590281586942e-05,
|
| 1325 |
+
"loss": 2.7983,
|
| 1326 |
+
"step": 1870
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.3422537775350446,
|
| 1330 |
+
"grad_norm": 8.010175704956055,
|
| 1331 |
+
"learning_rate": 8.230643242445605e-05,
|
| 1332 |
+
"loss": 2.7921,
|
| 1333 |
+
"step": 1880
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.3440742763517204,
|
| 1337 |
+
"grad_norm": 24.649381637573242,
|
| 1338 |
+
"learning_rate": 8.206568305177907e-05,
|
| 1339 |
+
"loss": 2.7962,
|
| 1340 |
+
"step": 1890
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.3458947751683961,
|
| 1344 |
+
"grad_norm": 8.272650718688965,
|
| 1345 |
+
"learning_rate": 8.182366422887964e-05,
|
| 1346 |
+
"loss": 2.7439,
|
| 1347 |
+
"step": 1900
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.3477152739850719,
|
| 1351 |
+
"grad_norm": 7.553550720214844,
|
| 1352 |
+
"learning_rate": 8.158038553705524e-05,
|
| 1353 |
+
"loss": 2.7845,
|
| 1354 |
+
"step": 1910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.3495357728017477,
|
| 1358 |
+
"grad_norm": 8.573986053466797,
|
| 1359 |
+
"learning_rate": 8.13358566074804e-05,
|
| 1360 |
+
"loss": 2.7003,
|
| 1361 |
+
"step": 1920
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.3513562716184234,
|
| 1365 |
+
"grad_norm": 10.316489219665527,
|
| 1366 |
+
"learning_rate": 8.109008712082538e-05,
|
| 1367 |
+
"loss": 2.7627,
|
| 1368 |
+
"step": 1930
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.3531767704350992,
|
| 1372 |
+
"grad_norm": 8.462483406066895,
|
| 1373 |
+
"learning_rate": 8.084308680687287e-05,
|
| 1374 |
+
"loss": 2.7281,
|
| 1375 |
+
"step": 1940
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.354997269251775,
|
| 1379 |
+
"grad_norm": 20.140274047851562,
|
| 1380 |
+
"learning_rate": 8.059486544413298e-05,
|
| 1381 |
+
"loss": 2.6906,
|
| 1382 |
+
"step": 1950
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.3568177680684508,
|
| 1386 |
+
"grad_norm": 7.473912239074707,
|
| 1387 |
+
"learning_rate": 8.034543285945584e-05,
|
| 1388 |
+
"loss": 2.8117,
|
| 1389 |
+
"step": 1960
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.3586382668851265,
|
| 1393 |
+
"grad_norm": 38.26898193359375,
|
| 1394 |
+
"learning_rate": 8.009479892764284e-05,
|
| 1395 |
+
"loss": 2.7456,
|
| 1396 |
+
"step": 1970
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.3604587657018023,
|
| 1400 |
+
"grad_norm": 48.63120651245117,
|
| 1401 |
+
"learning_rate": 7.984297357105552e-05,
|
| 1402 |
+
"loss": 2.7224,
|
| 1403 |
+
"step": 1980
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.3622792645184781,
|
| 1407 |
+
"grad_norm": 10.31283187866211,
|
| 1408 |
+
"learning_rate": 7.95899667592228e-05,
|
| 1409 |
+
"loss": 2.7108,
|
| 1410 |
+
"step": 1990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.3640997633351538,
|
| 1414 |
+
"grad_norm": 6.465616703033447,
|
| 1415 |
+
"learning_rate": 7.933578850844636e-05,
|
| 1416 |
+
"loss": 2.6901,
|
| 1417 |
+
"step": 2000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.3640997633351538,
|
| 1421 |
+
"eval_loss": 2.735260248184204,
|
| 1422 |
+
"eval_runtime": 1011.986,
|
| 1423 |
+
"eval_samples_per_second": 9.65,
|
| 1424 |
+
"eval_steps_per_second": 1.207,
|
| 1425 |
+
"step": 2000
|
| 1426 |
}
|
| 1427 |
],
|
| 1428 |
"logging_steps": 10,
|
|
|
|
| 1442 |
"attributes": {}
|
| 1443 |
}
|
| 1444 |
},
|
| 1445 |
+
"total_flos": 4.67628931940352e+17,
|
| 1446 |
"train_batch_size": 4,
|
| 1447 |
"trial_name": null,
|
| 1448 |
"trial_params": null
|