Training in progress, step 2500, 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 1235573136
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddf2e82fdc216f548a3cf93873f86d88d21716a46eb1b097e00ae1e6e6db9409
|
| 3 |
size 1235573136
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2471218763
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e32e38acf911ca92a84a8abde45fc1c319e1ef3f4b496f623e0732f90c10f24
|
| 3 |
size 2471218763
|
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:ea11996454b5587fcf33ae0ab5cf14b2031bf5f53f8c2ed5a48e87de31e29c84
|
| 3 |
size 14645
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5226e9e9fa1b9d11f56c6419ad51f6b8f1b208ee0ba1896a448175d1c43d275c
|
| 3 |
size 1383
|
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:db851a7a188e8825ebf9f1146ef7eaaf0e8f60e7eaa5a6238d3f38425f758d1c
|
| 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,
|
|
@@ -1128,6 +1128,636 @@
|
|
| 1128 |
"learning_rate": 0.00029098797595190375,
|
| 1129 |
"loss": 3103321489408.0,
|
| 1130 |
"step": 1600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1131 |
}
|
| 1132 |
],
|
| 1133 |
"logging_steps": 10,
|
|
@@ -1147,7 +1777,7 @@
|
|
| 1147 |
"attributes": {}
|
| 1148 |
}
|
| 1149 |
},
|
| 1150 |
-
"total_flos": 2.
|
| 1151 |
"train_batch_size": 2,
|
| 1152 |
"trial_name": null,
|
| 1153 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 9.0012,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 2500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1128 |
"learning_rate": 0.00029098797595190375,
|
| 1129 |
"loss": 3103321489408.0,
|
| 1130 |
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.0002,
|
| 1134 |
+
"grad_norm": 0.0,
|
| 1135 |
+
"learning_rate": 0.0002909278557114228,
|
| 1136 |
+
"loss": 95.5510498046875,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.0004,
|
| 1141 |
+
"grad_norm": 0.0,
|
| 1142 |
+
"learning_rate": 0.00029086773547094184,
|
| 1143 |
+
"loss": 98.69436645507812,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.0006,
|
| 1148 |
+
"grad_norm": 0.0,
|
| 1149 |
+
"learning_rate": 0.0002908076152304609,
|
| 1150 |
+
"loss": 99.0325439453125,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.0008,
|
| 1155 |
+
"grad_norm": 0.0,
|
| 1156 |
+
"learning_rate": 0.00029074749498997994,
|
| 1157 |
+
"loss": 97.31904296875,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.001,
|
| 1162 |
+
"grad_norm": 0.0,
|
| 1163 |
+
"learning_rate": 0.000290687374749499,
|
| 1164 |
+
"loss": 98.40114135742188,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.0012,
|
| 1169 |
+
"grad_norm": 0.0,
|
| 1170 |
+
"learning_rate": 0.00029062725450901803,
|
| 1171 |
+
"loss": 99.31021728515626,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.0014,
|
| 1176 |
+
"grad_norm": 0.0,
|
| 1177 |
+
"learning_rate": 0.00029056713426853703,
|
| 1178 |
+
"loss": 99.4878662109375,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.0016,
|
| 1183 |
+
"grad_norm": 0.0,
|
| 1184 |
+
"learning_rate": 0.0002905070140280561,
|
| 1185 |
+
"loss": 99.3884521484375,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.0018,
|
| 1190 |
+
"grad_norm": 0.0,
|
| 1191 |
+
"learning_rate": 0.0002904468937875751,
|
| 1192 |
+
"loss": 99.46731567382812,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.002,
|
| 1197 |
+
"grad_norm": 0.0,
|
| 1198 |
+
"learning_rate": 0.00029038677354709417,
|
| 1199 |
+
"loss": 99.4464599609375,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 2.0001,
|
| 1204 |
+
"grad_norm": 0.0,
|
| 1205 |
+
"learning_rate": 0.00029032665330661317,
|
| 1206 |
+
"loss": 104.66427001953124,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 2.0003,
|
| 1211 |
+
"grad_norm": 0.0,
|
| 1212 |
+
"learning_rate": 0.0002902665330661322,
|
| 1213 |
+
"loss": 98.24152221679688,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 2.0005,
|
| 1218 |
+
"grad_norm": 0.0,
|
| 1219 |
+
"learning_rate": 0.0002902064128256513,
|
| 1220 |
+
"loss": 99.4651123046875,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 2.0007,
|
| 1225 |
+
"grad_norm": 0.0,
|
| 1226 |
+
"learning_rate": 0.0002901462925851703,
|
| 1227 |
+
"loss": 99.3810546875,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 2.0009,
|
| 1232 |
+
"grad_norm": 0.0,
|
| 1233 |
+
"learning_rate": 0.00029008617234468936,
|
| 1234 |
+
"loss": 96.20846557617188,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 2.0011,
|
| 1239 |
+
"grad_norm": 0.0,
|
| 1240 |
+
"learning_rate": 0.0002900260521042084,
|
| 1241 |
+
"loss": 99.44166259765625,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 2.0013,
|
| 1246 |
+
"grad_norm": 0.0,
|
| 1247 |
+
"learning_rate": 0.00028996593186372745,
|
| 1248 |
+
"loss": 99.43170776367188,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 2.0015,
|
| 1253 |
+
"grad_norm": 0.0,
|
| 1254 |
+
"learning_rate": 0.00028990581162324644,
|
| 1255 |
+
"loss": 99.37213134765625,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.0017,
|
| 1260 |
+
"grad_norm": 0.0,
|
| 1261 |
+
"learning_rate": 0.0002898456913827655,
|
| 1262 |
+
"loss": 99.51829223632812,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.0019,
|
| 1267 |
+
"grad_norm": 0.0,
|
| 1268 |
+
"learning_rate": 0.00028978557114228454,
|
| 1269 |
+
"loss": 99.27789306640625,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.0021,
|
| 1274 |
+
"grad_norm": 0.0,
|
| 1275 |
+
"learning_rate": 0.0002897254509018036,
|
| 1276 |
+
"loss": 99.44300537109375,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 3.0002,
|
| 1281 |
+
"grad_norm": 0.0,
|
| 1282 |
+
"learning_rate": 0.00028966533066132264,
|
| 1283 |
+
"loss": 103.58494873046875,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 3.0004,
|
| 1288 |
+
"grad_norm": 0.0,
|
| 1289 |
+
"learning_rate": 0.0002896052104208417,
|
| 1290 |
+
"loss": 99.35955810546875,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 3.0006,
|
| 1295 |
+
"grad_norm": 0.0,
|
| 1296 |
+
"learning_rate": 0.00028954509018036073,
|
| 1297 |
+
"loss": 99.25761108398437,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 3.0008,
|
| 1302 |
+
"grad_norm": 0.0,
|
| 1303 |
+
"learning_rate": 0.0002894849699398797,
|
| 1304 |
+
"loss": 97.34839477539063,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 3.001,
|
| 1309 |
+
"grad_norm": 0.0,
|
| 1310 |
+
"learning_rate": 0.00028942484969939877,
|
| 1311 |
+
"loss": 98.359326171875,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 3.0012,
|
| 1316 |
+
"grad_norm": 0.0,
|
| 1317 |
+
"learning_rate": 0.0002893647294589178,
|
| 1318 |
+
"loss": 99.25703735351563,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 3.0014,
|
| 1323 |
+
"grad_norm": 0.0,
|
| 1324 |
+
"learning_rate": 0.00028930460921843687,
|
| 1325 |
+
"loss": 99.42723388671875,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 3.0016,
|
| 1330 |
+
"grad_norm": 0.0,
|
| 1331 |
+
"learning_rate": 0.00028924448897795586,
|
| 1332 |
+
"loss": 99.32404174804688,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 3.0018,
|
| 1337 |
+
"grad_norm": 0.0,
|
| 1338 |
+
"learning_rate": 0.0002891843687374749,
|
| 1339 |
+
"loss": 99.3990966796875,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 3.002,
|
| 1344 |
+
"grad_norm": 0.0,
|
| 1345 |
+
"learning_rate": 0.00028912424849699396,
|
| 1346 |
+
"loss": 99.39078369140626,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 4.0001,
|
| 1351 |
+
"grad_norm": 0.0,
|
| 1352 |
+
"learning_rate": 0.000289064128256513,
|
| 1353 |
+
"loss": 104.5905517578125,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 4.0003,
|
| 1358 |
+
"grad_norm": 0.0,
|
| 1359 |
+
"learning_rate": 0.00028900400801603205,
|
| 1360 |
+
"loss": 98.17286987304688,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 4.0005,
|
| 1365 |
+
"grad_norm": 0.0,
|
| 1366 |
+
"learning_rate": 0.0002889438877755511,
|
| 1367 |
+
"loss": 99.40567016601562,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 4.0007,
|
| 1372 |
+
"grad_norm": 0.0,
|
| 1373 |
+
"learning_rate": 0.00028888376753507015,
|
| 1374 |
+
"loss": 99.32139892578125,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 4.0009,
|
| 1379 |
+
"grad_norm": 0.0,
|
| 1380 |
+
"learning_rate": 0.00028882364729458914,
|
| 1381 |
+
"loss": 96.15136108398437,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 4.0011,
|
| 1386 |
+
"grad_norm": 0.0,
|
| 1387 |
+
"learning_rate": 0.0002887635270541082,
|
| 1388 |
+
"loss": 99.3803466796875,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 4.0013,
|
| 1393 |
+
"grad_norm": 0.0,
|
| 1394 |
+
"learning_rate": 0.00028870340681362724,
|
| 1395 |
+
"loss": 99.37009887695312,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 4.0015,
|
| 1400 |
+
"grad_norm": 0.0,
|
| 1401 |
+
"learning_rate": 0.0002886432865731463,
|
| 1402 |
+
"loss": 99.310498046875,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 4.0017,
|
| 1407 |
+
"grad_norm": 0.0,
|
| 1408 |
+
"learning_rate": 0.0002885831663326653,
|
| 1409 |
+
"loss": 99.45986938476562,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 4.0019,
|
| 1414 |
+
"grad_norm": 0.0,
|
| 1415 |
+
"learning_rate": 0.0002885230460921843,
|
| 1416 |
+
"loss": 99.21234741210938,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 4.0021,
|
| 1421 |
+
"grad_norm": 0.0,
|
| 1422 |
+
"learning_rate": 0.00028846292585170337,
|
| 1423 |
+
"loss": 99.381591796875,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 5.0002,
|
| 1428 |
+
"grad_norm": 0.0,
|
| 1429 |
+
"learning_rate": 0.0002884028056112224,
|
| 1430 |
+
"loss": 103.5211669921875,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 5.0004,
|
| 1435 |
+
"grad_norm": 0.0,
|
| 1436 |
+
"learning_rate": 0.00028834268537074147,
|
| 1437 |
+
"loss": 99.30160522460938,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 5.0006,
|
| 1442 |
+
"grad_norm": 0.0,
|
| 1443 |
+
"learning_rate": 0.0002882825651302605,
|
| 1444 |
+
"loss": 99.195703125,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 5.0008,
|
| 1449 |
+
"grad_norm": 0.0,
|
| 1450 |
+
"learning_rate": 0.0002882224448897795,
|
| 1451 |
+
"loss": 97.29168090820312,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 5.001,
|
| 1456 |
+
"grad_norm": 0.0,
|
| 1457 |
+
"learning_rate": 0.00028816232464929856,
|
| 1458 |
+
"loss": 98.30964965820313,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 5.0012,
|
| 1463 |
+
"grad_norm": 0.0,
|
| 1464 |
+
"learning_rate": 0.0002881022044088176,
|
| 1465 |
+
"loss": 99.19823608398437,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 5.0014,
|
| 1470 |
+
"grad_norm": 0.0,
|
| 1471 |
+
"learning_rate": 0.00028804208416833665,
|
| 1472 |
+
"loss": 99.3675048828125,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 5.0016,
|
| 1477 |
+
"grad_norm": 0.0,
|
| 1478 |
+
"learning_rate": 0.0002879819639278557,
|
| 1479 |
+
"loss": 99.27457275390626,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 5.0018,
|
| 1484 |
+
"grad_norm": 0.0,
|
| 1485 |
+
"learning_rate": 0.0002879218436873747,
|
| 1486 |
+
"loss": 99.3427490234375,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 5.002,
|
| 1491 |
+
"grad_norm": 0.0,
|
| 1492 |
+
"learning_rate": 0.0002878617234468938,
|
| 1493 |
+
"loss": 99.33101806640624,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 6.0001,
|
| 1498 |
+
"grad_norm": 0.0,
|
| 1499 |
+
"learning_rate": 0.0002878016032064128,
|
| 1500 |
+
"loss": 104.52410888671875,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 6.0003,
|
| 1505 |
+
"grad_norm": 0.0,
|
| 1506 |
+
"learning_rate": 0.00028774148296593184,
|
| 1507 |
+
"loss": 98.11661376953126,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 6.0005,
|
| 1512 |
+
"grad_norm": 0.0,
|
| 1513 |
+
"learning_rate": 0.0002876813627254509,
|
| 1514 |
+
"loss": 99.35023193359375,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 6.0007,
|
| 1519 |
+
"grad_norm": 0.0,
|
| 1520 |
+
"learning_rate": 0.00028762124248496993,
|
| 1521 |
+
"loss": 99.27654418945312,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 6.0009,
|
| 1526 |
+
"grad_norm": 0.0,
|
| 1527 |
+
"learning_rate": 0.0002875611222444889,
|
| 1528 |
+
"loss": 96.09561767578126,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 6.0011,
|
| 1533 |
+
"grad_norm": 0.0,
|
| 1534 |
+
"learning_rate": 0.000287501002004008,
|
| 1535 |
+
"loss": 99.32379760742188,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 6.0013,
|
| 1540 |
+
"grad_norm": 0.0,
|
| 1541 |
+
"learning_rate": 0.000287440881763527,
|
| 1542 |
+
"loss": 99.31390991210938,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 6.0015,
|
| 1547 |
+
"grad_norm": 0.0,
|
| 1548 |
+
"learning_rate": 0.00028738076152304607,
|
| 1549 |
+
"loss": 99.25233764648438,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 6.0017,
|
| 1554 |
+
"grad_norm": 0.0,
|
| 1555 |
+
"learning_rate": 0.0002873206412825651,
|
| 1556 |
+
"loss": 99.400146484375,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 6.0019,
|
| 1561 |
+
"grad_norm": 0.0,
|
| 1562 |
+
"learning_rate": 0.00028726052104208416,
|
| 1563 |
+
"loss": 99.16478271484375,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 6.0021,
|
| 1568 |
+
"grad_norm": 0.0,
|
| 1569 |
+
"learning_rate": 0.0002872004008016032,
|
| 1570 |
+
"loss": 99.3350830078125,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 7.0002,
|
| 1575 |
+
"grad_norm": 0.0,
|
| 1576 |
+
"learning_rate": 0.0002871402805611222,
|
| 1577 |
+
"loss": 103.46038818359375,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 7.0004,
|
| 1582 |
+
"grad_norm": 0.0,
|
| 1583 |
+
"learning_rate": 0.00028708016032064125,
|
| 1584 |
+
"loss": 99.24920043945312,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 7.0006,
|
| 1589 |
+
"grad_norm": 0.0,
|
| 1590 |
+
"learning_rate": 0.0002870200400801603,
|
| 1591 |
+
"loss": 99.14209594726563,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 7.0008,
|
| 1596 |
+
"grad_norm": 0.0,
|
| 1597 |
+
"learning_rate": 0.00028695991983967935,
|
| 1598 |
+
"loss": 97.23970336914063,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 7.001,
|
| 1603 |
+
"grad_norm": 0.0,
|
| 1604 |
+
"learning_rate": 0.00028689979959919834,
|
| 1605 |
+
"loss": 98.25855712890625,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 7.0012,
|
| 1610 |
+
"grad_norm": 0.0,
|
| 1611 |
+
"learning_rate": 0.0002868396793587174,
|
| 1612 |
+
"loss": 99.14596557617188,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 7.0014,
|
| 1617 |
+
"grad_norm": 0.0,
|
| 1618 |
+
"learning_rate": 0.00028677955911823644,
|
| 1619 |
+
"loss": 99.317333984375,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 7.0016,
|
| 1624 |
+
"grad_norm": 0.0,
|
| 1625 |
+
"learning_rate": 0.0002867194388777555,
|
| 1626 |
+
"loss": 99.22235107421875,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 7.0018,
|
| 1631 |
+
"grad_norm": 0.0,
|
| 1632 |
+
"learning_rate": 0.00028665931863727453,
|
| 1633 |
+
"loss": 99.2902587890625,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 7.002,
|
| 1638 |
+
"grad_norm": 0.0,
|
| 1639 |
+
"learning_rate": 0.0002865991983967936,
|
| 1640 |
+
"loss": 99.27511596679688,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 8.0001,
|
| 1645 |
+
"grad_norm": 0.0,
|
| 1646 |
+
"learning_rate": 0.00028653907815631263,
|
| 1647 |
+
"loss": 104.4739990234375,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 8.0003,
|
| 1652 |
+
"grad_norm": 0.0,
|
| 1653 |
+
"learning_rate": 0.0002864789579158316,
|
| 1654 |
+
"loss": 98.07261352539062,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 8.0005,
|
| 1659 |
+
"grad_norm": 0.0,
|
| 1660 |
+
"learning_rate": 0.00028641883767535067,
|
| 1661 |
+
"loss": 99.2950439453125,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 8.0007,
|
| 1666 |
+
"grad_norm": 0.0,
|
| 1667 |
+
"learning_rate": 0.0002863587174348697,
|
| 1668 |
+
"loss": 99.22308959960938,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 8.0009,
|
| 1673 |
+
"grad_norm": 0.0,
|
| 1674 |
+
"learning_rate": 0.00028629859719438876,
|
| 1675 |
+
"loss": 96.060498046875,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 8.0011,
|
| 1680 |
+
"grad_norm": 0.0,
|
| 1681 |
+
"learning_rate": 0.00028623847695390776,
|
| 1682 |
+
"loss": 99.27498168945313,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 8.0013,
|
| 1687 |
+
"grad_norm": 0.0,
|
| 1688 |
+
"learning_rate": 0.0002861783567134268,
|
| 1689 |
+
"loss": 99.27508544921875,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 8.0015,
|
| 1694 |
+
"grad_norm": 0.0,
|
| 1695 |
+
"learning_rate": 0.0002861182364729459,
|
| 1696 |
+
"loss": 99.20576782226563,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 8.0017,
|
| 1701 |
+
"grad_norm": 0.0,
|
| 1702 |
+
"learning_rate": 0.0002860581162324649,
|
| 1703 |
+
"loss": 99.35106811523437,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 8.0019,
|
| 1708 |
+
"grad_norm": 0.0,
|
| 1709 |
+
"learning_rate": 0.00028599799599198395,
|
| 1710 |
+
"loss": 99.11838989257812,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 8.0021,
|
| 1715 |
+
"grad_norm": 0.0,
|
| 1716 |
+
"learning_rate": 0.000285937875751503,
|
| 1717 |
+
"loss": 99.290185546875,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 9.0002,
|
| 1722 |
+
"grad_norm": 0.0,
|
| 1723 |
+
"learning_rate": 0.00028587775551102204,
|
| 1724 |
+
"loss": 103.40933837890626,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 9.0004,
|
| 1729 |
+
"grad_norm": 0.0,
|
| 1730 |
+
"learning_rate": 0.00028581763527054104,
|
| 1731 |
+
"loss": 99.20323486328125,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 9.0006,
|
| 1736 |
+
"grad_norm": 0.0,
|
| 1737 |
+
"learning_rate": 0.0002857575150300601,
|
| 1738 |
+
"loss": 99.10476684570312,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 9.0008,
|
| 1743 |
+
"grad_norm": 0.0,
|
| 1744 |
+
"learning_rate": 0.00028569739478957913,
|
| 1745 |
+
"loss": 97.20264282226563,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 9.001,
|
| 1750 |
+
"grad_norm": 0.0,
|
| 1751 |
+
"learning_rate": 0.0002856372745490982,
|
| 1752 |
+
"loss": 98.21160278320312,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 9.0012,
|
| 1757 |
+
"grad_norm": 0.0,
|
| 1758 |
+
"learning_rate": 0.0002855771543086172,
|
| 1759 |
+
"loss": 99.10530395507813,
|
| 1760 |
+
"step": 2500
|
| 1761 |
}
|
| 1762 |
],
|
| 1763 |
"logging_steps": 10,
|
|
|
|
| 1777 |
"attributes": {}
|
| 1778 |
}
|
| 1779 |
},
|
| 1780 |
+
"total_flos": 2.722244994618624e+16,
|
| 1781 |
"train_batch_size": 2,
|
| 1782 |
"trial_name": null,
|
| 1783 |
"trial_params": null
|