Training in progress, step 249, 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 216151256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12fef18f99c9bf3ec9eedf986ccbe12d2b84ec11b66e3b9788ae7ec43065b3d7
|
| 3 |
size 216151256
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 432640054
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:541a7b0bf8fbbc06854b3c570354a27505018117b0f0d67f11955711b3bef1b4
|
| 3 |
size 432640054
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:341f7be18cc89c2ad2dec55ac567729ae8e5db65bdf39a5bc196fbd79e2cbf16
|
| 3 |
size 14512
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d870e1e8472cc5e0d2cb8fe273473f96e2cec9ffdf9fce6a51cdd5b21cd3bae6
|
| 3 |
size 14512
|
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 sha256:3f919157faf64362df2e66ee2a7671eb7f6cf8287caadba981f351111997c856
|
| 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": 83,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1193,6 +1193,595 @@
|
|
| 1193 |
"eval_samples_per_second": 6.002,
|
| 1194 |
"eval_steps_per_second": 1.503,
|
| 1195 |
"step": 166
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1196 |
}
|
| 1197 |
],
|
| 1198 |
"logging_steps": 1,
|
|
@@ -1212,7 +1801,7 @@
|
|
| 1212 |
"attributes": {}
|
| 1213 |
}
|
| 1214 |
},
|
| 1215 |
-
"total_flos":
|
| 1216 |
"train_batch_size": 2,
|
| 1217 |
"trial_name": null,
|
| 1218 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7502824858757062,
|
| 5 |
"eval_steps": 83,
|
| 6 |
+
"global_step": 249,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1193 |
"eval_samples_per_second": 6.002,
|
| 1194 |
"eval_steps_per_second": 1.503,
|
| 1195 |
"step": 166
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.5032015065913371,
|
| 1199 |
+
"grad_norm": 1.2678625583648682,
|
| 1200 |
+
"learning_rate": 6.298594285815584e-05,
|
| 1201 |
+
"loss": 0.2982,
|
| 1202 |
+
"step": 167
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.5062146892655367,
|
| 1206 |
+
"grad_norm": 1.2661633491516113,
|
| 1207 |
+
"learning_rate": 6.244532285066382e-05,
|
| 1208 |
+
"loss": 0.3381,
|
| 1209 |
+
"step": 168
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.5092278719397364,
|
| 1213 |
+
"grad_norm": 1.160649061203003,
|
| 1214 |
+
"learning_rate": 6.190314727607196e-05,
|
| 1215 |
+
"loss": 0.3428,
|
| 1216 |
+
"step": 169
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.512241054613936,
|
| 1220 |
+
"grad_norm": 2.1445140838623047,
|
| 1221 |
+
"learning_rate": 6.13594839020466e-05,
|
| 1222 |
+
"loss": 0.6844,
|
| 1223 |
+
"step": 170
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 0.5152542372881356,
|
| 1227 |
+
"grad_norm": 1.9892338514328003,
|
| 1228 |
+
"learning_rate": 6.0814400682217234e-05,
|
| 1229 |
+
"loss": 0.559,
|
| 1230 |
+
"step": 171
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.5182674199623352,
|
| 1234 |
+
"grad_norm": 3.0277585983276367,
|
| 1235 |
+
"learning_rate": 6.026796574768288e-05,
|
| 1236 |
+
"loss": 0.6495,
|
| 1237 |
+
"step": 172
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 0.5212806026365349,
|
| 1241 |
+
"grad_norm": 3.0984301567077637,
|
| 1242 |
+
"learning_rate": 5.972024739849622e-05,
|
| 1243 |
+
"loss": 0.4114,
|
| 1244 |
+
"step": 173
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.5242937853107345,
|
| 1248 |
+
"grad_norm": 3.296858549118042,
|
| 1249 |
+
"learning_rate": 5.917131409512663e-05,
|
| 1250 |
+
"loss": 0.5272,
|
| 1251 |
+
"step": 174
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 0.527306967984934,
|
| 1255 |
+
"grad_norm": 4.093991756439209,
|
| 1256 |
+
"learning_rate": 5.862123444990318e-05,
|
| 1257 |
+
"loss": 0.5134,
|
| 1258 |
+
"step": 175
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.5303201506591337,
|
| 1262 |
+
"grad_norm": 1.48560631275177,
|
| 1263 |
+
"learning_rate": 5.807007721843861e-05,
|
| 1264 |
+
"loss": 0.5482,
|
| 1265 |
+
"step": 176
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 0.5333333333333333,
|
| 1269 |
+
"grad_norm": 1.6580932140350342,
|
| 1270 |
+
"learning_rate": 5.751791129103544e-05,
|
| 1271 |
+
"loss": 0.5894,
|
| 1272 |
+
"step": 177
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 0.5363465160075329,
|
| 1276 |
+
"grad_norm": 1.1557704210281372,
|
| 1277 |
+
"learning_rate": 5.696480568407523e-05,
|
| 1278 |
+
"loss": 0.4388,
|
| 1279 |
+
"step": 178
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 0.5393596986817326,
|
| 1283 |
+
"grad_norm": 1.1503976583480835,
|
| 1284 |
+
"learning_rate": 5.6410829531392006e-05,
|
| 1285 |
+
"loss": 0.3841,
|
| 1286 |
+
"step": 179
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.5423728813559322,
|
| 1290 |
+
"grad_norm": 1.0935317277908325,
|
| 1291 |
+
"learning_rate": 5.585605207563124e-05,
|
| 1292 |
+
"loss": 0.38,
|
| 1293 |
+
"step": 180
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 0.5453860640301318,
|
| 1297 |
+
"grad_norm": 1.0500688552856445,
|
| 1298 |
+
"learning_rate": 5.5300542659594854e-05,
|
| 1299 |
+
"loss": 0.37,
|
| 1300 |
+
"step": 181
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.5483992467043315,
|
| 1304 |
+
"grad_norm": 0.910677433013916,
|
| 1305 |
+
"learning_rate": 5.47443707175741e-05,
|
| 1306 |
+
"loss": 0.335,
|
| 1307 |
+
"step": 182
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 0.5514124293785311,
|
| 1311 |
+
"grad_norm": 1.0660496950149536,
|
| 1312 |
+
"learning_rate": 5.418760576667071e-05,
|
| 1313 |
+
"loss": 0.3516,
|
| 1314 |
+
"step": 183
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.5544256120527307,
|
| 1318 |
+
"grad_norm": 0.8869411945343018,
|
| 1319 |
+
"learning_rate": 5.3630317398107864e-05,
|
| 1320 |
+
"loss": 0.3249,
|
| 1321 |
+
"step": 184
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.5574387947269304,
|
| 1325 |
+
"grad_norm": 1.6419062614440918,
|
| 1326 |
+
"learning_rate": 5.3072575268531835e-05,
|
| 1327 |
+
"loss": 0.3531,
|
| 1328 |
+
"step": 185
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.56045197740113,
|
| 1332 |
+
"grad_norm": 0.8144354820251465,
|
| 1333 |
+
"learning_rate": 5.2514449091305375e-05,
|
| 1334 |
+
"loss": 0.3002,
|
| 1335 |
+
"step": 186
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 0.5634651600753295,
|
| 1339 |
+
"grad_norm": 1.1171809434890747,
|
| 1340 |
+
"learning_rate": 5.195600862779421e-05,
|
| 1341 |
+
"loss": 0.3776,
|
| 1342 |
+
"step": 187
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 0.5664783427495292,
|
| 1346 |
+
"grad_norm": 1.14911687374115,
|
| 1347 |
+
"learning_rate": 5.139732367864736e-05,
|
| 1348 |
+
"loss": 0.3215,
|
| 1349 |
+
"step": 188
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 0.5694915254237288,
|
| 1353 |
+
"grad_norm": 0.928807258605957,
|
| 1354 |
+
"learning_rate": 5.083846407507263e-05,
|
| 1355 |
+
"loss": 0.2747,
|
| 1356 |
+
"step": 189
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 0.5725047080979284,
|
| 1360 |
+
"grad_norm": 1.024272084236145,
|
| 1361 |
+
"learning_rate": 5.0279499670108245e-05,
|
| 1362 |
+
"loss": 0.3024,
|
| 1363 |
+
"step": 190
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 0.5755178907721281,
|
| 1367 |
+
"grad_norm": 1.2534008026123047,
|
| 1368 |
+
"learning_rate": 4.972050032989175e-05,
|
| 1369 |
+
"loss": 0.3704,
|
| 1370 |
+
"step": 191
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.5785310734463277,
|
| 1374 |
+
"grad_norm": 0.9777889847755432,
|
| 1375 |
+
"learning_rate": 4.9161535924927374e-05,
|
| 1376 |
+
"loss": 0.271,
|
| 1377 |
+
"step": 192
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 0.5815442561205273,
|
| 1381 |
+
"grad_norm": 1.0151127576828003,
|
| 1382 |
+
"learning_rate": 4.860267632135265e-05,
|
| 1383 |
+
"loss": 0.3179,
|
| 1384 |
+
"step": 193
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 0.5845574387947269,
|
| 1388 |
+
"grad_norm": 0.9281553030014038,
|
| 1389 |
+
"learning_rate": 4.80439913722058e-05,
|
| 1390 |
+
"loss": 0.2508,
|
| 1391 |
+
"step": 194
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 0.5875706214689266,
|
| 1395 |
+
"grad_norm": 1.6091618537902832,
|
| 1396 |
+
"learning_rate": 4.748555090869464e-05,
|
| 1397 |
+
"loss": 0.3701,
|
| 1398 |
+
"step": 195
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.5905838041431262,
|
| 1402 |
+
"grad_norm": 2.309112548828125,
|
| 1403 |
+
"learning_rate": 4.692742473146818e-05,
|
| 1404 |
+
"loss": 0.5701,
|
| 1405 |
+
"step": 196
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 0.5935969868173258,
|
| 1409 |
+
"grad_norm": 1.9035515785217285,
|
| 1410 |
+
"learning_rate": 4.636968260189214e-05,
|
| 1411 |
+
"loss": 0.5654,
|
| 1412 |
+
"step": 197
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 0.5966101694915255,
|
| 1416 |
+
"grad_norm": 2.222712516784668,
|
| 1417 |
+
"learning_rate": 4.5812394233329305e-05,
|
| 1418 |
+
"loss": 0.51,
|
| 1419 |
+
"step": 198
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 0.599623352165725,
|
| 1423 |
+
"grad_norm": 3.1770148277282715,
|
| 1424 |
+
"learning_rate": 4.525562928242592e-05,
|
| 1425 |
+
"loss": 0.6322,
|
| 1426 |
+
"step": 199
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 0.6026365348399246,
|
| 1430 |
+
"grad_norm": 3.6650121212005615,
|
| 1431 |
+
"learning_rate": 4.4699457340405164e-05,
|
| 1432 |
+
"loss": 0.4471,
|
| 1433 |
+
"step": 200
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.6056497175141243,
|
| 1437 |
+
"grad_norm": 1.350001573562622,
|
| 1438 |
+
"learning_rate": 4.414394792436877e-05,
|
| 1439 |
+
"loss": 0.4445,
|
| 1440 |
+
"step": 201
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 0.6086629001883239,
|
| 1444 |
+
"grad_norm": 1.3186067342758179,
|
| 1445 |
+
"learning_rate": 4.3589170468607985e-05,
|
| 1446 |
+
"loss": 0.4268,
|
| 1447 |
+
"step": 202
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 0.6116760828625235,
|
| 1451 |
+
"grad_norm": 1.0820094347000122,
|
| 1452 |
+
"learning_rate": 4.3035194315924785e-05,
|
| 1453 |
+
"loss": 0.349,
|
| 1454 |
+
"step": 203
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.6146892655367232,
|
| 1458 |
+
"grad_norm": 1.016855001449585,
|
| 1459 |
+
"learning_rate": 4.248208870896456e-05,
|
| 1460 |
+
"loss": 0.3657,
|
| 1461 |
+
"step": 204
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 0.6177024482109228,
|
| 1465 |
+
"grad_norm": 0.7627567648887634,
|
| 1466 |
+
"learning_rate": 4.192992278156141e-05,
|
| 1467 |
+
"loss": 0.282,
|
| 1468 |
+
"step": 205
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 0.6207156308851224,
|
| 1472 |
+
"grad_norm": 0.8734930753707886,
|
| 1473 |
+
"learning_rate": 4.1378765550096835e-05,
|
| 1474 |
+
"loss": 0.3205,
|
| 1475 |
+
"step": 206
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 0.6237288135593221,
|
| 1479 |
+
"grad_norm": 0.9233391880989075,
|
| 1480 |
+
"learning_rate": 4.082868590487339e-05,
|
| 1481 |
+
"loss": 0.2682,
|
| 1482 |
+
"step": 207
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 0.6267419962335217,
|
| 1486 |
+
"grad_norm": 0.7341740131378174,
|
| 1487 |
+
"learning_rate": 4.027975260150381e-05,
|
| 1488 |
+
"loss": 0.2488,
|
| 1489 |
+
"step": 208
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 0.6297551789077213,
|
| 1493 |
+
"grad_norm": 0.8570201992988586,
|
| 1494 |
+
"learning_rate": 3.973203425231715e-05,
|
| 1495 |
+
"loss": 0.2644,
|
| 1496 |
+
"step": 209
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.632768361581921,
|
| 1500 |
+
"grad_norm": 0.8284196853637695,
|
| 1501 |
+
"learning_rate": 3.918559931778277e-05,
|
| 1502 |
+
"loss": 0.3093,
|
| 1503 |
+
"step": 210
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 0.6357815442561205,
|
| 1507 |
+
"grad_norm": 0.7870326638221741,
|
| 1508 |
+
"learning_rate": 3.8640516097953405e-05,
|
| 1509 |
+
"loss": 0.2577,
|
| 1510 |
+
"step": 211
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.6387947269303201,
|
| 1514 |
+
"grad_norm": 0.9884381294250488,
|
| 1515 |
+
"learning_rate": 3.809685272392804e-05,
|
| 1516 |
+
"loss": 0.3252,
|
| 1517 |
+
"step": 212
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 0.6418079096045197,
|
| 1521 |
+
"grad_norm": 1.1668404340744019,
|
| 1522 |
+
"learning_rate": 3.755467714933619e-05,
|
| 1523 |
+
"loss": 0.3414,
|
| 1524 |
+
"step": 213
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 0.6448210922787194,
|
| 1528 |
+
"grad_norm": 1.1960536241531372,
|
| 1529 |
+
"learning_rate": 3.701405714184416e-05,
|
| 1530 |
+
"loss": 0.3029,
|
| 1531 |
+
"step": 214
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 0.647834274952919,
|
| 1535 |
+
"grad_norm": 0.7987526059150696,
|
| 1536 |
+
"learning_rate": 3.647506027468467e-05,
|
| 1537 |
+
"loss": 0.2501,
|
| 1538 |
+
"step": 215
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.6508474576271186,
|
| 1542 |
+
"grad_norm": 1.0721232891082764,
|
| 1543 |
+
"learning_rate": 3.59377539182107e-05,
|
| 1544 |
+
"loss": 0.3132,
|
| 1545 |
+
"step": 216
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 0.6538606403013183,
|
| 1549 |
+
"grad_norm": 0.9739212393760681,
|
| 1550 |
+
"learning_rate": 3.5402205231474736e-05,
|
| 1551 |
+
"loss": 0.2644,
|
| 1552 |
+
"step": 217
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.6568738229755179,
|
| 1556 |
+
"grad_norm": 1.1942991018295288,
|
| 1557 |
+
"learning_rate": 3.486848115383445e-05,
|
| 1558 |
+
"loss": 0.3206,
|
| 1559 |
+
"step": 218
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 0.6598870056497175,
|
| 1563 |
+
"grad_norm": 1.3838952779769897,
|
| 1564 |
+
"learning_rate": 3.4336648396585776e-05,
|
| 1565 |
+
"loss": 0.3569,
|
| 1566 |
+
"step": 219
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 0.6629001883239172,
|
| 1570 |
+
"grad_norm": 2.448126792907715,
|
| 1571 |
+
"learning_rate": 3.380677343462447e-05,
|
| 1572 |
+
"loss": 0.5818,
|
| 1573 |
+
"step": 220
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 0.6659133709981168,
|
| 1577 |
+
"grad_norm": 1.8376816511154175,
|
| 1578 |
+
"learning_rate": 3.327892249813745e-05,
|
| 1579 |
+
"loss": 0.4343,
|
| 1580 |
+
"step": 221
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.6689265536723163,
|
| 1584 |
+
"grad_norm": 2.102494478225708,
|
| 1585 |
+
"learning_rate": 3.275316156432434e-05,
|
| 1586 |
+
"loss": 0.4626,
|
| 1587 |
+
"step": 222
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 0.671939736346516,
|
| 1591 |
+
"grad_norm": 2.167078733444214,
|
| 1592 |
+
"learning_rate": 3.2229556349150945e-05,
|
| 1593 |
+
"loss": 0.4407,
|
| 1594 |
+
"step": 223
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 0.6749529190207156,
|
| 1598 |
+
"grad_norm": 2.149308204650879,
|
| 1599 |
+
"learning_rate": 3.170817229913526e-05,
|
| 1600 |
+
"loss": 0.3198,
|
| 1601 |
+
"step": 224
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 0.6779661016949152,
|
| 1605 |
+
"grad_norm": 4.306909084320068,
|
| 1606 |
+
"learning_rate": 3.118907458316722e-05,
|
| 1607 |
+
"loss": 0.5187,
|
| 1608 |
+
"step": 225
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 0.6809792843691149,
|
| 1612 |
+
"grad_norm": 1.0858993530273438,
|
| 1613 |
+
"learning_rate": 3.067232808436299e-05,
|
| 1614 |
+
"loss": 0.3973,
|
| 1615 |
+
"step": 226
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 0.6839924670433145,
|
| 1619 |
+
"grad_norm": 0.9466880559921265,
|
| 1620 |
+
"learning_rate": 3.0157997391955172e-05,
|
| 1621 |
+
"loss": 0.2911,
|
| 1622 |
+
"step": 227
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 0.6870056497175141,
|
| 1626 |
+
"grad_norm": 1.053808569908142,
|
| 1627 |
+
"learning_rate": 2.964614679321966e-05,
|
| 1628 |
+
"loss": 0.3425,
|
| 1629 |
+
"step": 228
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 0.6900188323917138,
|
| 1633 |
+
"grad_norm": 1.020087718963623,
|
| 1634 |
+
"learning_rate": 2.913684026544021e-05,
|
| 1635 |
+
"loss": 0.3171,
|
| 1636 |
+
"step": 229
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 0.6930320150659134,
|
| 1640 |
+
"grad_norm": 0.9112816452980042,
|
| 1641 |
+
"learning_rate": 2.8630141467911775e-05,
|
| 1642 |
+
"loss": 0.289,
|
| 1643 |
+
"step": 230
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 0.696045197740113,
|
| 1647 |
+
"grad_norm": 0.9472637176513672,
|
| 1648 |
+
"learning_rate": 2.812611373398365e-05,
|
| 1649 |
+
"loss": 0.2909,
|
| 1650 |
+
"step": 231
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 0.6990583804143126,
|
| 1654 |
+
"grad_norm": 0.8144400715827942,
|
| 1655 |
+
"learning_rate": 2.762482006314324e-05,
|
| 1656 |
+
"loss": 0.2527,
|
| 1657 |
+
"step": 232
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 0.7020715630885123,
|
| 1661 |
+
"grad_norm": 0.8899109363555908,
|
| 1662 |
+
"learning_rate": 2.712632311314165e-05,
|
| 1663 |
+
"loss": 0.2814,
|
| 1664 |
+
"step": 233
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.7050847457627119,
|
| 1668 |
+
"grad_norm": 0.8338634967803955,
|
| 1669 |
+
"learning_rate": 2.6630685192161992e-05,
|
| 1670 |
+
"loss": 0.2684,
|
| 1671 |
+
"step": 234
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 0.7080979284369114,
|
| 1675 |
+
"grad_norm": 1.1928447484970093,
|
| 1676 |
+
"learning_rate": 2.6137968251031287e-05,
|
| 1677 |
+
"loss": 0.327,
|
| 1678 |
+
"step": 235
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 0.7111111111111111,
|
| 1682 |
+
"grad_norm": 0.9519332647323608,
|
| 1683 |
+
"learning_rate": 2.5648233875477157e-05,
|
| 1684 |
+
"loss": 0.2797,
|
| 1685 |
+
"step": 236
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 0.7141242937853107,
|
| 1689 |
+
"grad_norm": 1.1364781856536865,
|
| 1690 |
+
"learning_rate": 2.5161543278430054e-05,
|
| 1691 |
+
"loss": 0.3121,
|
| 1692 |
+
"step": 237
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 0.7171374764595103,
|
| 1696 |
+
"grad_norm": 1.0575398206710815,
|
| 1697 |
+
"learning_rate": 2.4677957292372167e-05,
|
| 1698 |
+
"loss": 0.2866,
|
| 1699 |
+
"step": 238
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 0.72015065913371,
|
| 1703 |
+
"grad_norm": 0.7296847105026245,
|
| 1704 |
+
"learning_rate": 2.419753636173379e-05,
|
| 1705 |
+
"loss": 0.2432,
|
| 1706 |
+
"step": 239
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.7231638418079096,
|
| 1710 |
+
"grad_norm": 0.7573246955871582,
|
| 1711 |
+
"learning_rate": 2.3720340535338348e-05,
|
| 1712 |
+
"loss": 0.2545,
|
| 1713 |
+
"step": 240
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 0.7261770244821092,
|
| 1717 |
+
"grad_norm": 0.8128641247749329,
|
| 1718 |
+
"learning_rate": 2.3246429458896634e-05,
|
| 1719 |
+
"loss": 0.2548,
|
| 1720 |
+
"step": 241
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 0.7291902071563089,
|
| 1724 |
+
"grad_norm": 0.8031213283538818,
|
| 1725 |
+
"learning_rate": 2.2775862367551644e-05,
|
| 1726 |
+
"loss": 0.2509,
|
| 1727 |
+
"step": 242
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 0.7322033898305085,
|
| 1731 |
+
"grad_norm": 0.9279189109802246,
|
| 1732 |
+
"learning_rate": 2.2308698078474645e-05,
|
| 1733 |
+
"loss": 0.264,
|
| 1734 |
+
"step": 243
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 0.7352165725047081,
|
| 1738 |
+
"grad_norm": 1.0536640882492065,
|
| 1739 |
+
"learning_rate": 2.1844994983513467e-05,
|
| 1740 |
+
"loss": 0.3295,
|
| 1741 |
+
"step": 244
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 0.7382297551789078,
|
| 1745 |
+
"grad_norm": 2.177769422531128,
|
| 1746 |
+
"learning_rate": 2.1384811041894055e-05,
|
| 1747 |
+
"loss": 0.4975,
|
| 1748 |
+
"step": 245
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.7412429378531074,
|
| 1752 |
+
"grad_norm": 1.5739905834197998,
|
| 1753 |
+
"learning_rate": 2.0928203772975917e-05,
|
| 1754 |
+
"loss": 0.4395,
|
| 1755 |
+
"step": 246
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 0.7442561205273069,
|
| 1759 |
+
"grad_norm": 1.7161879539489746,
|
| 1760 |
+
"learning_rate": 2.0475230249062725e-05,
|
| 1761 |
+
"loss": 0.3947,
|
| 1762 |
+
"step": 247
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 0.7472693032015066,
|
| 1766 |
+
"grad_norm": 3.185561180114746,
|
| 1767 |
+
"learning_rate": 2.0025947088268717e-05,
|
| 1768 |
+
"loss": 0.6166,
|
| 1769 |
+
"step": 248
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.7502824858757062,
|
| 1773 |
+
"grad_norm": 2.201467275619507,
|
| 1774 |
+
"learning_rate": 1.958041044744186e-05,
|
| 1775 |
+
"loss": 0.3779,
|
| 1776 |
+
"step": 249
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 0.7502824858757062,
|
| 1780 |
+
"eval_loss": NaN,
|
| 1781 |
+
"eval_runtime": 92.9008,
|
| 1782 |
+
"eval_samples_per_second": 6.017,
|
| 1783 |
+
"eval_steps_per_second": 1.507,
|
| 1784 |
+
"step": 249
|
| 1785 |
}
|
| 1786 |
],
|
| 1787 |
"logging_steps": 1,
|
|
|
|
| 1801 |
"attributes": {}
|
| 1802 |
}
|
| 1803 |
},
|
| 1804 |
+
"total_flos": 8.20318075147518e+17,
|
| 1805 |
"train_batch_size": 2,
|
| 1806 |
"trial_name": null,
|
| 1807 |
"trial_params": null
|