Training in progress, step 279, 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 80013120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:710f323d2d148d19a77874e7041b6682ca6fdf00fb8ff65d993905368b01d329
|
| 3 |
size 80013120
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d28277b8c5696e9609288685953a64988fc3bd9fe09e1d48c95ba342438e1db6
|
| 3 |
+
size 41120084
|
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:d290d0f18d2c63d334eda98204765110cec7c5f5c7d088e8f0e88675b235ebea
|
| 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:bed55d74992475f85034de9808b502db25c265c1fbe10ac1ead4e6ef3743a36b
|
| 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": 93,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1325,6 +1325,665 @@
|
|
| 1325 |
"eval_samples_per_second": 14.676,
|
| 1326 |
"eval_steps_per_second": 7.385,
|
| 1327 |
"step": 186
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1328 |
}
|
| 1329 |
],
|
| 1330 |
"logging_steps": 1,
|
|
@@ -1344,7 +2003,7 @@
|
|
| 1344 |
"attributes": {}
|
| 1345 |
}
|
| 1346 |
},
|
| 1347 |
-
"total_flos":
|
| 1348 |
"train_batch_size": 2,
|
| 1349 |
"trial_name": null,
|
| 1350 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7505043712172159,
|
| 5 |
"eval_steps": 93,
|
| 6 |
+
"global_step": 279,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1325 |
"eval_samples_per_second": 14.676,
|
| 1326 |
"eval_steps_per_second": 7.385,
|
| 1327 |
"step": 186
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.5030262273032953,
|
| 1331 |
+
"grad_norm": 4.27438497543335,
|
| 1332 |
+
"learning_rate": 0.0001034706759475182,
|
| 1333 |
+
"loss": 0.7565,
|
| 1334 |
+
"step": 187
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.5057162071284466,
|
| 1338 |
+
"grad_norm": 6.716769218444824,
|
| 1339 |
+
"learning_rate": 0.0001026032357053512,
|
| 1340 |
+
"loss": 0.8371,
|
| 1341 |
+
"step": 188
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.5084061869535978,
|
| 1345 |
+
"grad_norm": 2.8589985370635986,
|
| 1346 |
+
"learning_rate": 0.0001017355994012102,
|
| 1347 |
+
"loss": 0.2835,
|
| 1348 |
+
"step": 189
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.5110961667787491,
|
| 1352 |
+
"grad_norm": 2.675102949142456,
|
| 1353 |
+
"learning_rate": 0.00010086783238088244,
|
| 1354 |
+
"loss": 0.2332,
|
| 1355 |
+
"step": 190
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.5137861466039004,
|
| 1359 |
+
"grad_norm": 5.922438144683838,
|
| 1360 |
+
"learning_rate": 0.0001,
|
| 1361 |
+
"loss": 0.3974,
|
| 1362 |
+
"step": 191
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.5164761264290518,
|
| 1366 |
+
"grad_norm": 9.693557739257812,
|
| 1367 |
+
"learning_rate": 9.913216761911755e-05,
|
| 1368 |
+
"loss": 0.8956,
|
| 1369 |
+
"step": 192
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.5191661062542031,
|
| 1373 |
+
"grad_norm": 5.175564765930176,
|
| 1374 |
+
"learning_rate": 9.826440059878982e-05,
|
| 1375 |
+
"loss": 0.4924,
|
| 1376 |
+
"step": 193
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.5218560860793544,
|
| 1380 |
+
"grad_norm": 4.789131164550781,
|
| 1381 |
+
"learning_rate": 9.739676429464881e-05,
|
| 1382 |
+
"loss": 0.6214,
|
| 1383 |
+
"step": 194
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 0.5245460659045057,
|
| 1387 |
+
"grad_norm": 4.584465026855469,
|
| 1388 |
+
"learning_rate": 9.652932405248181e-05,
|
| 1389 |
+
"loss": 0.6807,
|
| 1390 |
+
"step": 195
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.527236045729657,
|
| 1394 |
+
"grad_norm": 8.758447647094727,
|
| 1395 |
+
"learning_rate": 9.566214520330966e-05,
|
| 1396 |
+
"loss": 1.2434,
|
| 1397 |
+
"step": 196
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 0.5299260255548084,
|
| 1401 |
+
"grad_norm": 8.869791984558105,
|
| 1402 |
+
"learning_rate": 9.479529305846652e-05,
|
| 1403 |
+
"loss": 0.903,
|
| 1404 |
+
"step": 197
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.5326160053799597,
|
| 1408 |
+
"grad_norm": 3.716257333755493,
|
| 1409 |
+
"learning_rate": 9.392883290468083e-05,
|
| 1410 |
+
"loss": 0.3255,
|
| 1411 |
+
"step": 198
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.535305985205111,
|
| 1415 |
+
"grad_norm": 5.400476455688477,
|
| 1416 |
+
"learning_rate": 9.306282999915839e-05,
|
| 1417 |
+
"loss": 0.4036,
|
| 1418 |
+
"step": 199
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.5379959650302623,
|
| 1422 |
+
"grad_norm": 3.5948078632354736,
|
| 1423 |
+
"learning_rate": 9.219734956466752e-05,
|
| 1424 |
+
"loss": 0.1489,
|
| 1425 |
+
"step": 200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.5406859448554135,
|
| 1429 |
+
"grad_norm": 3.1145272254943848,
|
| 1430 |
+
"learning_rate": 9.133245678462663e-05,
|
| 1431 |
+
"loss": 1.6383,
|
| 1432 |
+
"step": 201
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.543375924680565,
|
| 1436 |
+
"grad_norm": 3.916707754135132,
|
| 1437 |
+
"learning_rate": 9.046821679819527e-05,
|
| 1438 |
+
"loss": 1.7301,
|
| 1439 |
+
"step": 202
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.5460659045057162,
|
| 1443 |
+
"grad_norm": 3.5734024047851562,
|
| 1444 |
+
"learning_rate": 8.960469469536786e-05,
|
| 1445 |
+
"loss": 1.1493,
|
| 1446 |
+
"step": 203
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.5487558843308675,
|
| 1450 |
+
"grad_norm": 3.514202356338501,
|
| 1451 |
+
"learning_rate": 8.874195551207174e-05,
|
| 1452 |
+
"loss": 1.2045,
|
| 1453 |
+
"step": 204
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.5514458641560188,
|
| 1457 |
+
"grad_norm": 3.9733996391296387,
|
| 1458 |
+
"learning_rate": 8.788006422526881e-05,
|
| 1459 |
+
"loss": 1.8557,
|
| 1460 |
+
"step": 205
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.5541358439811701,
|
| 1464 |
+
"grad_norm": 4.543676853179932,
|
| 1465 |
+
"learning_rate": 8.701908574806197e-05,
|
| 1466 |
+
"loss": 1.3852,
|
| 1467 |
+
"step": 206
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.5568258238063215,
|
| 1471 |
+
"grad_norm": 3.916227102279663,
|
| 1472 |
+
"learning_rate": 8.615908492480598e-05,
|
| 1473 |
+
"loss": 1.1968,
|
| 1474 |
+
"step": 207
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.5595158036314728,
|
| 1478 |
+
"grad_norm": 4.097233295440674,
|
| 1479 |
+
"learning_rate": 8.530012652622397e-05,
|
| 1480 |
+
"loss": 1.2521,
|
| 1481 |
+
"step": 208
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.5622057834566241,
|
| 1485 |
+
"grad_norm": 3.402857780456543,
|
| 1486 |
+
"learning_rate": 8.444227524452918e-05,
|
| 1487 |
+
"loss": 0.8308,
|
| 1488 |
+
"step": 209
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.5648957632817754,
|
| 1492 |
+
"grad_norm": 4.228405952453613,
|
| 1493 |
+
"learning_rate": 8.358559568855249e-05,
|
| 1494 |
+
"loss": 1.1485,
|
| 1495 |
+
"step": 210
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.5675857431069267,
|
| 1499 |
+
"grad_norm": 3.699495553970337,
|
| 1500 |
+
"learning_rate": 8.273015237887673e-05,
|
| 1501 |
+
"loss": 0.8204,
|
| 1502 |
+
"step": 211
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.570275722932078,
|
| 1506 |
+
"grad_norm": 3.4216742515563965,
|
| 1507 |
+
"learning_rate": 8.187600974297714e-05,
|
| 1508 |
+
"loss": 0.7515,
|
| 1509 |
+
"step": 212
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.5729657027572294,
|
| 1513 |
+
"grad_norm": 3.6286754608154297,
|
| 1514 |
+
"learning_rate": 8.102323211036904e-05,
|
| 1515 |
+
"loss": 0.6262,
|
| 1516 |
+
"step": 213
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.5756556825823806,
|
| 1520 |
+
"grad_norm": 4.869045734405518,
|
| 1521 |
+
"learning_rate": 8.017188370776292e-05,
|
| 1522 |
+
"loss": 0.9851,
|
| 1523 |
+
"step": 214
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.5783456624075319,
|
| 1527 |
+
"grad_norm": 5.106837749481201,
|
| 1528 |
+
"learning_rate": 7.932202865422726e-05,
|
| 1529 |
+
"loss": 1.122,
|
| 1530 |
+
"step": 215
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.5810356422326832,
|
| 1534 |
+
"grad_norm": 4.490080833435059,
|
| 1535 |
+
"learning_rate": 7.847373095635937e-05,
|
| 1536 |
+
"loss": 0.8354,
|
| 1537 |
+
"step": 216
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.5837256220578345,
|
| 1541 |
+
"grad_norm": 3.4671199321746826,
|
| 1542 |
+
"learning_rate": 7.762705450346462e-05,
|
| 1543 |
+
"loss": 0.8034,
|
| 1544 |
+
"step": 217
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.5864156018829859,
|
| 1548 |
+
"grad_norm": 5.198472499847412,
|
| 1549 |
+
"learning_rate": 7.678206306274495e-05,
|
| 1550 |
+
"loss": 0.9827,
|
| 1551 |
+
"step": 218
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 0.5891055817081372,
|
| 1555 |
+
"grad_norm": 3.910879135131836,
|
| 1556 |
+
"learning_rate": 7.59388202744959e-05,
|
| 1557 |
+
"loss": 0.5839,
|
| 1558 |
+
"step": 219
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.5917955615332885,
|
| 1562 |
+
"grad_norm": 3.6499719619750977,
|
| 1563 |
+
"learning_rate": 7.509738964731389e-05,
|
| 1564 |
+
"loss": 0.4724,
|
| 1565 |
+
"step": 220
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 0.5944855413584398,
|
| 1569 |
+
"grad_norm": 4.461284160614014,
|
| 1570 |
+
"learning_rate": 7.425783455331281e-05,
|
| 1571 |
+
"loss": 0.6628,
|
| 1572 |
+
"step": 221
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.5971755211835911,
|
| 1576 |
+
"grad_norm": 3.570058584213257,
|
| 1577 |
+
"learning_rate": 7.342021822335143e-05,
|
| 1578 |
+
"loss": 0.455,
|
| 1579 |
+
"step": 222
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 0.5998655010087425,
|
| 1583 |
+
"grad_norm": 4.93247652053833,
|
| 1584 |
+
"learning_rate": 7.258460374227085e-05,
|
| 1585 |
+
"loss": 1.1265,
|
| 1586 |
+
"step": 223
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.6025554808338938,
|
| 1590 |
+
"grad_norm": 3.4849631786346436,
|
| 1591 |
+
"learning_rate": 7.175105404414362e-05,
|
| 1592 |
+
"loss": 0.3686,
|
| 1593 |
+
"step": 224
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.605245460659045,
|
| 1597 |
+
"grad_norm": 2.503511428833008,
|
| 1598 |
+
"learning_rate": 7.091963190753376e-05,
|
| 1599 |
+
"loss": 0.3223,
|
| 1600 |
+
"step": 225
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.6079354404841963,
|
| 1604 |
+
"grad_norm": 4.293323040008545,
|
| 1605 |
+
"learning_rate": 7.009039995076844e-05,
|
| 1606 |
+
"loss": 0.5115,
|
| 1607 |
+
"step": 226
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.6106254203093476,
|
| 1611 |
+
"grad_norm": 4.681167125701904,
|
| 1612 |
+
"learning_rate": 6.926342062722223e-05,
|
| 1613 |
+
"loss": 0.8861,
|
| 1614 |
+
"step": 227
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.613315400134499,
|
| 1618 |
+
"grad_norm": 4.363972187042236,
|
| 1619 |
+
"learning_rate": 6.843875622061304e-05,
|
| 1620 |
+
"loss": 0.714,
|
| 1621 |
+
"step": 228
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.6160053799596503,
|
| 1625 |
+
"grad_norm": 4.711676597595215,
|
| 1626 |
+
"learning_rate": 6.761646884031164e-05,
|
| 1627 |
+
"loss": 0.9218,
|
| 1628 |
+
"step": 229
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.6186953597848016,
|
| 1632 |
+
"grad_norm": 5.378860950469971,
|
| 1633 |
+
"learning_rate": 6.679662041666362e-05,
|
| 1634 |
+
"loss": 0.8466,
|
| 1635 |
+
"step": 230
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.6213853396099529,
|
| 1639 |
+
"grad_norm": 4.945761203765869,
|
| 1640 |
+
"learning_rate": 6.597927269632526e-05,
|
| 1641 |
+
"loss": 0.8303,
|
| 1642 |
+
"step": 231
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.6240753194351042,
|
| 1646 |
+
"grad_norm": 5.643571853637695,
|
| 1647 |
+
"learning_rate": 6.516448723761315e-05,
|
| 1648 |
+
"loss": 0.9418,
|
| 1649 |
+
"step": 232
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.6267652992602556,
|
| 1653 |
+
"grad_norm": 3.7865352630615234,
|
| 1654 |
+
"learning_rate": 6.435232540586763e-05,
|
| 1655 |
+
"loss": 0.447,
|
| 1656 |
+
"step": 233
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 0.6294552790854069,
|
| 1660 |
+
"grad_norm": 4.775557994842529,
|
| 1661 |
+
"learning_rate": 6.354284836883156e-05,
|
| 1662 |
+
"loss": 0.7068,
|
| 1663 |
+
"step": 234
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 0.6321452589105582,
|
| 1667 |
+
"grad_norm": 4.313781261444092,
|
| 1668 |
+
"learning_rate": 6.273611709204304e-05,
|
| 1669 |
+
"loss": 0.4621,
|
| 1670 |
+
"step": 235
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.6348352387357095,
|
| 1674 |
+
"grad_norm": 3.1749980449676514,
|
| 1675 |
+
"learning_rate": 6.193219233424414e-05,
|
| 1676 |
+
"loss": 0.2996,
|
| 1677 |
+
"step": 236
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 0.6375252185608608,
|
| 1681 |
+
"grad_norm": 5.823219299316406,
|
| 1682 |
+
"learning_rate": 6.11311346428046e-05,
|
| 1683 |
+
"loss": 0.7987,
|
| 1684 |
+
"step": 237
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.6402151983860122,
|
| 1688 |
+
"grad_norm": 4.825180530548096,
|
| 1689 |
+
"learning_rate": 6.033300434916203e-05,
|
| 1690 |
+
"loss": 0.9309,
|
| 1691 |
+
"step": 238
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.6429051782111634,
|
| 1695 |
+
"grad_norm": 7.609663486480713,
|
| 1696 |
+
"learning_rate": 5.9537861564277654e-05,
|
| 1697 |
+
"loss": 0.6973,
|
| 1698 |
+
"step": 239
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.6455951580363147,
|
| 1702 |
+
"grad_norm": 3.5007476806640625,
|
| 1703 |
+
"learning_rate": 5.8745766174109495e-05,
|
| 1704 |
+
"loss": 0.5831,
|
| 1705 |
+
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.648285137861466,
|
| 1709 |
+
"grad_norm": 4.732518196105957,
|
| 1710 |
+
"learning_rate": 5.795677783510187e-05,
|
| 1711 |
+
"loss": 0.447,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.6509751176866173,
|
| 1716 |
+
"grad_norm": 4.874922752380371,
|
| 1717 |
+
"learning_rate": 5.7170955969692265e-05,
|
| 1718 |
+
"loss": 0.429,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.6536650975117687,
|
| 1723 |
+
"grad_norm": 2.738624095916748,
|
| 1724 |
+
"learning_rate": 5.638835976183627e-05,
|
| 1725 |
+
"loss": 0.316,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.65635507733692,
|
| 1730 |
+
"grad_norm": 4.868707180023193,
|
| 1731 |
+
"learning_rate": 5.5609048152549794e-05,
|
| 1732 |
+
"loss": 0.528,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.6590450571620713,
|
| 1737 |
+
"grad_norm": 2.545112371444702,
|
| 1738 |
+
"learning_rate": 5.483307983547026e-05,
|
| 1739 |
+
"loss": 0.2108,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.6617350369872226,
|
| 1744 |
+
"grad_norm": 5.225026607513428,
|
| 1745 |
+
"learning_rate": 5.406051325243586e-05,
|
| 1746 |
+
"loss": 0.6883,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.6644250168123739,
|
| 1751 |
+
"grad_norm": 6.781031608581543,
|
| 1752 |
+
"learning_rate": 5.329140658908423e-05,
|
| 1753 |
+
"loss": 0.7684,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.6671149966375253,
|
| 1758 |
+
"grad_norm": 4.9204630851745605,
|
| 1759 |
+
"learning_rate": 5.2525817770470084e-05,
|
| 1760 |
+
"loss": 0.529,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.6698049764626766,
|
| 1765 |
+
"grad_norm": 4.677596092224121,
|
| 1766 |
+
"learning_rate": 5.1763804456702545e-05,
|
| 1767 |
+
"loss": 0.1916,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 0.6724949562878278,
|
| 1772 |
+
"grad_norm": 5.289106369018555,
|
| 1773 |
+
"learning_rate": 5.1005424038602724e-05,
|
| 1774 |
+
"loss": 0.5018,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.6751849361129791,
|
| 1779 |
+
"grad_norm": 3.4332501888275146,
|
| 1780 |
+
"learning_rate": 5.025073363338111e-05,
|
| 1781 |
+
"loss": 1.4558,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 0.6778749159381304,
|
| 1786 |
+
"grad_norm": 3.6167006492614746,
|
| 1787 |
+
"learning_rate": 4.949979008033596e-05,
|
| 1788 |
+
"loss": 1.3181,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.6805648957632818,
|
| 1793 |
+
"grad_norm": 4.852591514587402,
|
| 1794 |
+
"learning_rate": 4.8752649936572304e-05,
|
| 1795 |
+
"loss": 1.5339,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.6832548755884331,
|
| 1800 |
+
"grad_norm": 4.116457462310791,
|
| 1801 |
+
"learning_rate": 4.800936947274255e-05,
|
| 1802 |
+
"loss": 1.8746,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.6859448554135844,
|
| 1807 |
+
"grad_norm": 3.9800620079040527,
|
| 1808 |
+
"learning_rate": 4.7270004668808397e-05,
|
| 1809 |
+
"loss": 1.8474,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.6886348352387357,
|
| 1814 |
+
"grad_norm": 3.9870662689208984,
|
| 1815 |
+
"learning_rate": 4.65346112098246e-05,
|
| 1816 |
+
"loss": 1.148,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.691324815063887,
|
| 1821 |
+
"grad_norm": 4.0448503494262695,
|
| 1822 |
+
"learning_rate": 4.5803244481745275e-05,
|
| 1823 |
+
"loss": 1.0557,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.6940147948890383,
|
| 1828 |
+
"grad_norm": 4.282130241394043,
|
| 1829 |
+
"learning_rate": 4.5075959567252335e-05,
|
| 1830 |
+
"loss": 1.2731,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.6967047747141897,
|
| 1835 |
+
"grad_norm": 4.488638401031494,
|
| 1836 |
+
"learning_rate": 4.435281124160715e-05,
|
| 1837 |
+
"loss": 1.3722,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.699394754539341,
|
| 1842 |
+
"grad_norm": 3.9812207221984863,
|
| 1843 |
+
"learning_rate": 4.363385396852491e-05,
|
| 1844 |
+
"loss": 1.3536,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.7020847343644923,
|
| 1849 |
+
"grad_norm": 5.047592639923096,
|
| 1850 |
+
"learning_rate": 4.291914189607297e-05,
|
| 1851 |
+
"loss": 0.7947,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.7047747141896435,
|
| 1856 |
+
"grad_norm": 4.314056396484375,
|
| 1857 |
+
"learning_rate": 4.220872885259247e-05,
|
| 1858 |
+
"loss": 0.6368,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.7074646940147948,
|
| 1863 |
+
"grad_norm": 4.209965705871582,
|
| 1864 |
+
"learning_rate": 4.1502668342644455e-05,
|
| 1865 |
+
"loss": 0.7786,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.7101546738399462,
|
| 1870 |
+
"grad_norm": 5.0463151931762695,
|
| 1871 |
+
"learning_rate": 4.080101354298016e-05,
|
| 1872 |
+
"loss": 0.5957,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.7128446536650975,
|
| 1877 |
+
"grad_norm": 5.0057373046875,
|
| 1878 |
+
"learning_rate": 4.0103817298535794e-05,
|
| 1879 |
+
"loss": 1.1632,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.7155346334902488,
|
| 1884 |
+
"grad_norm": 4.640236854553223,
|
| 1885 |
+
"learning_rate": 3.9411132118452896e-05,
|
| 1886 |
+
"loss": 1.523,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 0.7182246133154001,
|
| 1891 |
+
"grad_norm": 3.843372344970703,
|
| 1892 |
+
"learning_rate": 3.872301017212337e-05,
|
| 1893 |
+
"loss": 1.0536,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.7209145931405514,
|
| 1898 |
+
"grad_norm": 4.180513858795166,
|
| 1899 |
+
"learning_rate": 3.8039503285260506e-05,
|
| 1900 |
+
"loss": 0.7683,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.7236045729657028,
|
| 1905 |
+
"grad_norm": 7.624268054962158,
|
| 1906 |
+
"learning_rate": 3.73606629359955e-05,
|
| 1907 |
+
"loss": 0.817,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.7262945527908541,
|
| 1912 |
+
"grad_norm": 5.006471633911133,
|
| 1913 |
+
"learning_rate": 3.6686540251000756e-05,
|
| 1914 |
+
"loss": 0.899,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.7289845326160054,
|
| 1919 |
+
"grad_norm": 4.611303806304932,
|
| 1920 |
+
"learning_rate": 3.6017186001639036e-05,
|
| 1921 |
+
"loss": 0.3317,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.7316745124411567,
|
| 1926 |
+
"grad_norm": 4.701048374176025,
|
| 1927 |
+
"learning_rate": 3.535265060013965e-05,
|
| 1928 |
+
"loss": 0.9114,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.734364492266308,
|
| 1933 |
+
"grad_norm": 5.010507583618164,
|
| 1934 |
+
"learning_rate": 3.4692984095801796e-05,
|
| 1935 |
+
"loss": 0.6747,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.7370544720914594,
|
| 1940 |
+
"grad_norm": 4.114429473876953,
|
| 1941 |
+
"learning_rate": 3.4038236171224946e-05,
|
| 1942 |
+
"loss": 0.5408,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.7397444519166106,
|
| 1947 |
+
"grad_norm": 5.027583122253418,
|
| 1948 |
+
"learning_rate": 3.3388456138567225e-05,
|
| 1949 |
+
"loss": 0.4589,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.7424344317417619,
|
| 1954 |
+
"grad_norm": 5.864731788635254,
|
| 1955 |
+
"learning_rate": 3.274369293583121e-05,
|
| 1956 |
+
"loss": 0.9165,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 0.7451244115669132,
|
| 1961 |
+
"grad_norm": 4.747946739196777,
|
| 1962 |
+
"learning_rate": 3.210399512317849e-05,
|
| 1963 |
+
"loss": 0.5087,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.7478143913920645,
|
| 1968 |
+
"grad_norm": 3.576078414916992,
|
| 1969 |
+
"learning_rate": 3.146941087927203e-05,
|
| 1970 |
+
"loss": 0.5728,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 0.7505043712172159,
|
| 1975 |
+
"grad_norm": 2.740264892578125,
|
| 1976 |
+
"learning_rate": 3.0839987997647935e-05,
|
| 1977 |
+
"loss": 0.2307,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.7505043712172159,
|
| 1982 |
+
"eval_loss": 0.7389675974845886,
|
| 1983 |
+
"eval_runtime": 10.7313,
|
| 1984 |
+
"eval_samples_per_second": 14.63,
|
| 1985 |
+
"eval_steps_per_second": 7.362,
|
| 1986 |
+
"step": 279
|
| 1987 |
}
|
| 1988 |
],
|
| 1989 |
"logging_steps": 1,
|
|
|
|
| 2003 |
"attributes": {}
|
| 2004 |
}
|
| 2005 |
},
|
| 2006 |
+
"total_flos": 9.088420770584986e+16,
|
| 2007 |
"train_batch_size": 2,
|
| 2008 |
"trial_name": null,
|
| 2009 |
"trial_params": null
|