Update trainer_state.json
Browse files- trainer_state.json +846 -12
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1340,18 +1340,852 @@
|
|
| 1340 |
"step": 19890
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
-
"epoch":
|
| 1344 |
-
"
|
| 1345 |
-
"
|
| 1346 |
-
"
|
| 1347 |
-
|
| 1348 |
-
|
| 1349 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1350 |
}
|
| 1351 |
],
|
| 1352 |
-
"max_steps":
|
| 1353 |
-
"num_train_epochs":
|
| 1354 |
-
"total_flos":
|
| 1355 |
"trial_name": null,
|
| 1356 |
"trial_params": null
|
| 1357 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 90.0,
|
| 5 |
+
"global_step": 29790,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1340 |
"step": 19890
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
+
"epoch": 60.12,
|
| 1344 |
+
"learning_rate": 8.693541350698569e-09,
|
| 1345 |
+
"loss": 1.5417,
|
| 1346 |
+
"step": 19900
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 60.42,
|
| 1350 |
+
"learning_rate": 6.223106707028002e-08,
|
| 1351 |
+
"loss": 1.3927,
|
| 1352 |
+
"step": 20000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 60.73,
|
| 1356 |
+
"learning_rate": 1.2172156539717049e-07,
|
| 1357 |
+
"loss": 1.3567,
|
| 1358 |
+
"step": 20100
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 61.0,
|
| 1362 |
+
"eval_loss": 3.4576382637023926,
|
| 1363 |
+
"eval_runtime": 0.952,
|
| 1364 |
+
"eval_samples_per_second": 35.715,
|
| 1365 |
+
"eval_steps_per_second": 35.715,
|
| 1366 |
+
"step": 20191
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 61.03,
|
| 1370 |
+
"learning_rate": 1.3426435048201084e-07,
|
| 1371 |
+
"loss": 1.3891,
|
| 1372 |
+
"step": 20200
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 61.33,
|
| 1376 |
+
"learning_rate": 8.877437479246501e-08,
|
| 1377 |
+
"loss": 1.3829,
|
| 1378 |
+
"step": 20300
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 61.63,
|
| 1382 |
+
"learning_rate": 2.5454714388004032e-08,
|
| 1383 |
+
"loss": 1.3482,
|
| 1384 |
+
"step": 20400
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 61.93,
|
| 1388 |
+
"learning_rate": 2.6594266338289214e-10,
|
| 1389 |
+
"loss": 1.3278,
|
| 1390 |
+
"step": 20500
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 62.0,
|
| 1394 |
+
"eval_loss": 3.452810525894165,
|
| 1395 |
+
"eval_runtime": 0.964,
|
| 1396 |
+
"eval_samples_per_second": 35.269,
|
| 1397 |
+
"eval_steps_per_second": 35.269,
|
| 1398 |
+
"step": 20522
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 62.24,
|
| 1402 |
+
"learning_rate": 3.54693620295034e-08,
|
| 1403 |
+
"loss": 1.3495,
|
| 1404 |
+
"step": 20600
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 62.54,
|
| 1408 |
+
"learning_rate": 9.995293681528824e-08,
|
| 1409 |
+
"loss": 1.3372,
|
| 1410 |
+
"step": 20700
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 62.84,
|
| 1414 |
+
"learning_rate": 1.3672745077248097e-07,
|
| 1415 |
+
"loss": 1.3292,
|
| 1416 |
+
"step": 20800
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 63.0,
|
| 1420 |
+
"eval_loss": 3.4468021392822266,
|
| 1421 |
+
"eval_runtime": 0.9878,
|
| 1422 |
+
"eval_samples_per_second": 34.42,
|
| 1423 |
+
"eval_steps_per_second": 34.42,
|
| 1424 |
+
"step": 20853
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 63.14,
|
| 1428 |
+
"learning_rate": 1.1329236813994995e-07,
|
| 1429 |
+
"loss": 1.3209,
|
| 1430 |
+
"step": 20900
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 63.44,
|
| 1434 |
+
"learning_rate": 5.035911808981744e-08,
|
| 1435 |
+
"loss": 1.3127,
|
| 1436 |
+
"step": 21000
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 63.75,
|
| 1440 |
+
"learning_rate": 3.5467727381560775e-09,
|
| 1441 |
+
"loss": 1.3285,
|
| 1442 |
+
"step": 21100
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 64.0,
|
| 1446 |
+
"eval_loss": 3.443075180053711,
|
| 1447 |
+
"eval_runtime": 0.9684,
|
| 1448 |
+
"eval_samples_per_second": 35.111,
|
| 1449 |
+
"eval_steps_per_second": 35.111,
|
| 1450 |
+
"step": 21184
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 64.05,
|
| 1454 |
+
"learning_rate": 1.4227090415615305e-08,
|
| 1455 |
+
"loss": 1.2899,
|
| 1456 |
+
"step": 21200
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 64.35,
|
| 1460 |
+
"learning_rate": 7.296103279743161e-08,
|
| 1461 |
+
"loss": 1.3259,
|
| 1462 |
+
"step": 21300
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 64.65,
|
| 1466 |
+
"learning_rate": 1.2784078685324094e-07,
|
| 1467 |
+
"loss": 1.308,
|
| 1468 |
+
"step": 21400
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 64.95,
|
| 1472 |
+
"learning_rate": 1.303647894806954e-07,
|
| 1473 |
+
"loss": 1.3032,
|
| 1474 |
+
"step": 21500
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 65.0,
|
| 1478 |
+
"eval_loss": 3.437013626098633,
|
| 1479 |
+
"eval_runtime": 0.9607,
|
| 1480 |
+
"eval_samples_per_second": 35.392,
|
| 1481 |
+
"eval_steps_per_second": 35.392,
|
| 1482 |
+
"step": 21515
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 65.26,
|
| 1486 |
+
"learning_rate": 7.830238067025156e-08,
|
| 1487 |
+
"loss": 1.2946,
|
| 1488 |
+
"step": 21600
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 65.56,
|
| 1492 |
+
"learning_rate": 1.766521352829689e-08,
|
| 1493 |
+
"loss": 1.3209,
|
| 1494 |
+
"step": 21700
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 65.86,
|
| 1498 |
+
"learning_rate": 2.043130777021614e-09,
|
| 1499 |
+
"loss": 1.318,
|
| 1500 |
+
"step": 21800
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 66.0,
|
| 1504 |
+
"eval_loss": 3.4345176219940186,
|
| 1505 |
+
"eval_runtime": 0.9583,
|
| 1506 |
+
"eval_samples_per_second": 35.479,
|
| 1507 |
+
"eval_steps_per_second": 35.479,
|
| 1508 |
+
"step": 21846
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 66.16,
|
| 1512 |
+
"learning_rate": 4.5242597940677724e-08,
|
| 1513 |
+
"loss": 1.2637,
|
| 1514 |
+
"step": 21900
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 66.47,
|
| 1518 |
+
"learning_rate": 1.090848418327998e-07,
|
| 1519 |
+
"loss": 1.2813,
|
| 1520 |
+
"step": 22000
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 66.77,
|
| 1524 |
+
"learning_rate": 1.3714744085178092e-07,
|
| 1525 |
+
"loss": 1.3003,
|
| 1526 |
+
"step": 22100
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 67.0,
|
| 1530 |
+
"eval_loss": 3.428901195526123,
|
| 1531 |
+
"eval_runtime": 0.9557,
|
| 1532 |
+
"eval_samples_per_second": 35.575,
|
| 1533 |
+
"eval_steps_per_second": 35.575,
|
| 1534 |
+
"step": 22177
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 67.07,
|
| 1538 |
+
"learning_rate": 1.0462926495336562e-07,
|
| 1539 |
+
"loss": 1.3045,
|
| 1540 |
+
"step": 22200
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 67.37,
|
| 1544 |
+
"learning_rate": 4.026918856301729e-08,
|
| 1545 |
+
"loss": 1.2654,
|
| 1546 |
+
"step": 22300
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 67.67,
|
| 1550 |
+
"learning_rate": 9.472826654548543e-10,
|
| 1551 |
+
"loss": 1.2689,
|
| 1552 |
+
"step": 22400
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 67.98,
|
| 1556 |
+
"learning_rate": 2.141541401022131e-08,
|
| 1557 |
+
"loss": 1.3202,
|
| 1558 |
+
"step": 22500
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 68.0,
|
| 1562 |
+
"eval_loss": 3.427440881729126,
|
| 1563 |
+
"eval_runtime": 0.958,
|
| 1564 |
+
"eval_samples_per_second": 35.49,
|
| 1565 |
+
"eval_steps_per_second": 35.49,
|
| 1566 |
+
"step": 22508
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 68.28,
|
| 1570 |
+
"learning_rate": 8.358428206955252e-08,
|
| 1571 |
+
"loss": 1.3008,
|
| 1572 |
+
"step": 22600
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 68.58,
|
| 1576 |
+
"learning_rate": 1.3251035932481603e-07,
|
| 1577 |
+
"loss": 1.2628,
|
| 1578 |
+
"step": 22700
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 68.88,
|
| 1582 |
+
"learning_rate": 1.2495381600320813e-07,
|
| 1583 |
+
"loss": 1.2643,
|
| 1584 |
+
"step": 22800
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 69.0,
|
| 1588 |
+
"eval_loss": 3.4231972694396973,
|
| 1589 |
+
"eval_runtime": 0.9561,
|
| 1590 |
+
"eval_samples_per_second": 35.563,
|
| 1591 |
+
"eval_steps_per_second": 35.563,
|
| 1592 |
+
"step": 22839
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 69.18,
|
| 1596 |
+
"learning_rate": 6.759296488244086e-08,
|
| 1597 |
+
"loss": 1.2717,
|
| 1598 |
+
"step": 22900
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 69.49,
|
| 1602 |
+
"learning_rate": 1.1122110047934364e-08,
|
| 1603 |
+
"loss": 1.2759,
|
| 1604 |
+
"step": 23000
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 69.79,
|
| 1608 |
+
"learning_rate": 5.44899573655199e-09,
|
| 1609 |
+
"loss": 1.2862,
|
| 1610 |
+
"step": 23100
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 70.0,
|
| 1614 |
+
"eval_loss": 3.4223098754882812,
|
| 1615 |
+
"eval_runtime": 1.0568,
|
| 1616 |
+
"eval_samples_per_second": 32.174,
|
| 1617 |
+
"eval_steps_per_second": 32.174,
|
| 1618 |
+
"step": 23170
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 70.09,
|
| 1622 |
+
"learning_rate": 5.492884226688333e-08,
|
| 1623 |
+
"loss": 1.2962,
|
| 1624 |
+
"step": 23200
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 70.39,
|
| 1628 |
+
"learning_rate": 1.1675016817694061e-07,
|
| 1629 |
+
"loss": 1.2762,
|
| 1630 |
+
"step": 23300
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 70.69,
|
| 1634 |
+
"learning_rate": 1.3601739630719386e-07,
|
| 1635 |
+
"loss": 1.2783,
|
| 1636 |
+
"step": 23400
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 71.0,
|
| 1640 |
+
"learning_rate": 9.570255892993473e-08,
|
| 1641 |
+
"loss": 1.2597,
|
| 1642 |
+
"step": 23500
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 71.0,
|
| 1646 |
+
"eval_loss": 3.418621063232422,
|
| 1647 |
+
"eval_runtime": 1.0452,
|
| 1648 |
+
"eval_samples_per_second": 32.531,
|
| 1649 |
+
"eval_steps_per_second": 32.531,
|
| 1650 |
+
"step": 23501
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 71.3,
|
| 1654 |
+
"learning_rate": 3.143505458484596e-08,
|
| 1655 |
+
"loss": 1.3153,
|
| 1656 |
+
"step": 23600
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 71.6,
|
| 1660 |
+
"learning_rate": 1.3141045706995258e-11,
|
| 1661 |
+
"loss": 1.245,
|
| 1662 |
+
"step": 23700
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 71.9,
|
| 1666 |
+
"learning_rate": 2.920683930724396e-08,
|
| 1667 |
+
"loss": 1.2426,
|
| 1668 |
+
"step": 23800
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 72.0,
|
| 1672 |
+
"eval_loss": 3.4176230430603027,
|
| 1673 |
+
"eval_runtime": 1.0483,
|
| 1674 |
+
"eval_samples_per_second": 32.434,
|
| 1675 |
+
"eval_steps_per_second": 32.434,
|
| 1676 |
+
"step": 23832
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 72.21,
|
| 1680 |
+
"learning_rate": 9.32153777956087e-08,
|
| 1681 |
+
"loss": 1.276,
|
| 1682 |
+
"step": 23900
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 72.51,
|
| 1686 |
+
"learning_rate": 1.3546936727073083e-07,
|
| 1687 |
+
"loss": 1.2707,
|
| 1688 |
+
"step": 24000
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 72.81,
|
| 1692 |
+
"learning_rate": 1.1862562768173958e-07,
|
| 1693 |
+
"loss": 1.2539,
|
| 1694 |
+
"step": 24100
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 73.0,
|
| 1698 |
+
"eval_loss": 3.4151535034179688,
|
| 1699 |
+
"eval_runtime": 1.0259,
|
| 1700 |
+
"eval_samples_per_second": 33.143,
|
| 1701 |
+
"eval_steps_per_second": 33.143,
|
| 1702 |
+
"step": 24163
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 73.11,
|
| 1706 |
+
"learning_rate": 5.7570298965124436e-08,
|
| 1707 |
+
"loss": 1.2827,
|
| 1708 |
+
"step": 24200
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 73.41,
|
| 1712 |
+
"learning_rate": 6.262786177113756e-09,
|
| 1713 |
+
"loss": 1.2277,
|
| 1714 |
+
"step": 24300
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 73.72,
|
| 1718 |
+
"learning_rate": 1.0047581324385604e-08,
|
| 1719 |
+
"loss": 1.2604,
|
| 1720 |
+
"step": 24400
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 74.0,
|
| 1724 |
+
"eval_loss": 3.414668083190918,
|
| 1725 |
+
"eval_runtime": 1.0049,
|
| 1726 |
+
"eval_samples_per_second": 33.834,
|
| 1727 |
+
"eval_steps_per_second": 33.834,
|
| 1728 |
+
"step": 24494
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 74.02,
|
| 1732 |
+
"learning_rate": 6.55797626990514e-08,
|
| 1733 |
+
"loss": 1.2588,
|
| 1734 |
+
"step": 24500
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 74.32,
|
| 1738 |
+
"learning_rate": 1.2378116576678553e-07,
|
| 1739 |
+
"loss": 1.2695,
|
| 1740 |
+
"step": 24600
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 74.62,
|
| 1744 |
+
"learning_rate": 1.3321462445238648e-07,
|
| 1745 |
+
"loss": 1.2576,
|
| 1746 |
+
"step": 24700
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 74.92,
|
| 1750 |
+
"learning_rate": 8.554304797283439e-08,
|
| 1751 |
+
"loss": 1.263,
|
| 1752 |
+
"step": 24800
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 75.0,
|
| 1756 |
+
"eval_loss": 3.412790298461914,
|
| 1757 |
+
"eval_runtime": 1.012,
|
| 1758 |
+
"eval_samples_per_second": 33.597,
|
| 1759 |
+
"eval_steps_per_second": 33.597,
|
| 1760 |
+
"step": 24825
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 75.23,
|
| 1764 |
+
"learning_rate": 2.2897564912636232e-08,
|
| 1765 |
+
"loss": 1.2362,
|
| 1766 |
+
"step": 24900
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 75.53,
|
| 1770 |
+
"learning_rate": 6.429250232458032e-10,
|
| 1771 |
+
"loss": 1.2653,
|
| 1772 |
+
"step": 25000
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 75.83,
|
| 1776 |
+
"learning_rate": 3.844730720753521e-08,
|
| 1777 |
+
"loss": 1.2642,
|
| 1778 |
+
"step": 25100
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 76.0,
|
| 1782 |
+
"eval_loss": 3.412682294845581,
|
| 1783 |
+
"eval_runtime": 1.006,
|
| 1784 |
+
"eval_samples_per_second": 33.797,
|
| 1785 |
+
"eval_steps_per_second": 33.797,
|
| 1786 |
+
"step": 25156
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 76.13,
|
| 1790 |
+
"learning_rate": 1.0289999999999885e-07,
|
| 1791 |
+
"loss": 1.2254,
|
| 1792 |
+
"step": 25200
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 76.44,
|
| 1796 |
+
"learning_rate": 1.370390800033334e-07,
|
| 1797 |
+
"loss": 1.2543,
|
| 1798 |
+
"step": 25300
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 76.74,
|
| 1802 |
+
"learning_rate": 1.106931521097187e-07,
|
| 1803 |
+
"loss": 1.2694,
|
| 1804 |
+
"step": 25400
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 77.0,
|
| 1808 |
+
"eval_loss": 3.4108803272247314,
|
| 1809 |
+
"eval_runtime": 1.004,
|
| 1810 |
+
"eval_samples_per_second": 33.865,
|
| 1811 |
+
"eval_steps_per_second": 33.865,
|
| 1812 |
+
"step": 25487
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 77.04,
|
| 1816 |
+
"learning_rate": 4.7146188818785276e-08,
|
| 1817 |
+
"loss": 1.25,
|
| 1818 |
+
"step": 25500
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 77.34,
|
| 1822 |
+
"learning_rate": 2.559648997196144e-09,
|
| 1823 |
+
"loss": 1.2614,
|
| 1824 |
+
"step": 25600
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 77.64,
|
| 1828 |
+
"learning_rate": 1.6338171238500445e-08,
|
| 1829 |
+
"loss": 1.2549,
|
| 1830 |
+
"step": 25700
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 77.95,
|
| 1834 |
+
"learning_rate": 7.63045897109416e-08,
|
| 1835 |
+
"loss": 1.2251,
|
| 1836 |
+
"step": 25800
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 78.0,
|
| 1840 |
+
"eval_loss": 3.4106197357177734,
|
| 1841 |
+
"eval_runtime": 1.0035,
|
| 1842 |
+
"eval_samples_per_second": 33.882,
|
| 1843 |
+
"eval_steps_per_second": 33.882,
|
| 1844 |
+
"step": 25818
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 78.25,
|
| 1848 |
+
"learning_rate": 1.2946185514557884e-07,
|
| 1849 |
+
"loss": 1.2965,
|
| 1850 |
+
"step": 25900
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 78.55,
|
| 1854 |
+
"learning_rate": 1.288307033641468e-07,
|
| 1855 |
+
"loss": 1.2137,
|
| 1856 |
+
"step": 26000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 78.85,
|
| 1860 |
+
"learning_rate": 7.496893292971942e-08,
|
| 1861 |
+
"loss": 1.2673,
|
| 1862 |
+
"step": 26100
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 79.0,
|
| 1866 |
+
"eval_loss": 3.409691333770752,
|
| 1867 |
+
"eval_runtime": 1.0062,
|
| 1868 |
+
"eval_samples_per_second": 33.792,
|
| 1869 |
+
"eval_steps_per_second": 33.792,
|
| 1870 |
+
"step": 26149
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 79.15,
|
| 1874 |
+
"learning_rate": 1.5478434602829786e-08,
|
| 1875 |
+
"loss": 1.225,
|
| 1876 |
+
"step": 26200
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 79.46,
|
| 1880 |
+
"learning_rate": 2.9356495179890543e-09,
|
| 1881 |
+
"loss": 1.2545,
|
| 1882 |
+
"step": 26300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 79.76,
|
| 1886 |
+
"learning_rate": 4.842562520753367e-08,
|
| 1887 |
+
"loss": 1.233,
|
| 1888 |
+
"step": 26400
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 80.0,
|
| 1892 |
+
"eval_loss": 3.409552574157715,
|
| 1893 |
+
"eval_runtime": 1.0059,
|
| 1894 |
+
"eval_samples_per_second": 33.8,
|
| 1895 |
+
"eval_steps_per_second": 33.8,
|
| 1896 |
+
"step": 26480
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 80.06,
|
| 1900 |
+
"learning_rate": 1.1174528561199565e-07,
|
| 1901 |
+
"loss": 1.2286,
|
| 1902 |
+
"step": 26500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 80.36,
|
| 1906 |
+
"learning_rate": 1.3693405733661706e-07,
|
| 1907 |
+
"loss": 1.2464,
|
| 1908 |
+
"step": 26600
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 80.66,
|
| 1912 |
+
"learning_rate": 1.0173063797049695e-07,
|
| 1913 |
+
"loss": 1.2497,
|
| 1914 |
+
"step": 26700
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 80.97,
|
| 1918 |
+
"learning_rate": 3.7247063184712124e-08,
|
| 1919 |
+
"loss": 1.2408,
|
| 1920 |
+
"step": 26800
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 81.0,
|
| 1924 |
+
"eval_loss": 3.4087181091308594,
|
| 1925 |
+
"eval_runtime": 1.0321,
|
| 1926 |
+
"eval_samples_per_second": 32.942,
|
| 1927 |
+
"eval_steps_per_second": 32.942,
|
| 1928 |
+
"step": 26811
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 81.27,
|
| 1932 |
+
"learning_rate": 4.725492275192098e-10,
|
| 1933 |
+
"loss": 1.2461,
|
| 1934 |
+
"step": 26900
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 81.57,
|
| 1938 |
+
"learning_rate": 2.3400433722385994e-08,
|
| 1939 |
+
"loss": 1.2495,
|
| 1940 |
+
"step": 27000
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 81.87,
|
| 1944 |
+
"learning_rate": 8.619280748313942e-08,
|
| 1945 |
+
"loss": 1.2579,
|
| 1946 |
+
"step": 27100
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 82.0,
|
| 1950 |
+
"eval_loss": 3.408783435821533,
|
| 1951 |
+
"eval_runtime": 1.0924,
|
| 1952 |
+
"eval_samples_per_second": 31.123,
|
| 1953 |
+
"eval_steps_per_second": 31.123,
|
| 1954 |
+
"step": 27142
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 82.18,
|
| 1958 |
+
"learning_rate": 1.3343703098390236e-07,
|
| 1959 |
+
"loss": 1.2153,
|
| 1960 |
+
"step": 27200
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 82.48,
|
| 1964 |
+
"learning_rate": 1.233796611423087e-07,
|
| 1965 |
+
"loss": 1.2398,
|
| 1966 |
+
"step": 27300
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 82.78,
|
| 1970 |
+
"learning_rate": 6.490918819320101e-08,
|
| 1971 |
+
"loss": 1.2346,
|
| 1972 |
+
"step": 27400
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 83.0,
|
| 1976 |
+
"eval_loss": 3.408099412918091,
|
| 1977 |
+
"eval_runtime": 1.0406,
|
| 1978 |
+
"eval_samples_per_second": 32.673,
|
| 1979 |
+
"eval_steps_per_second": 32.673,
|
| 1980 |
+
"step": 27473
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 83.08,
|
| 1984 |
+
"learning_rate": 9.700576468258166e-09,
|
| 1985 |
+
"loss": 1.2543,
|
| 1986 |
+
"step": 27500
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 83.38,
|
| 1990 |
+
"learning_rate": 6.546026505537958e-09,
|
| 1991 |
+
"loss": 1.247,
|
| 1992 |
+
"step": 27600
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 83.69,
|
| 1996 |
+
"learning_rate": 5.823346268095307e-08,
|
| 1997 |
+
"loss": 1.2225,
|
| 1998 |
+
"step": 27700
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 83.99,
|
| 2002 |
+
"learning_rate": 1.190826247529784e-07,
|
| 2003 |
+
"loss": 1.2298,
|
| 2004 |
+
"step": 27800
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 84.0,
|
| 2008 |
+
"eval_loss": 3.4082374572753906,
|
| 2009 |
+
"eval_runtime": 1.0242,
|
| 2010 |
+
"eval_samples_per_second": 33.197,
|
| 2011 |
+
"eval_steps_per_second": 33.197,
|
| 2012 |
+
"step": 27804
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 84.29,
|
| 2016 |
+
"learning_rate": 1.3531631337483422e-07,
|
| 2017 |
+
"loss": 1.2426,
|
| 2018 |
+
"step": 27900
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 84.59,
|
| 2022 |
+
"learning_rate": 9.258753872080096e-08,
|
| 2023 |
+
"loss": 1.2389,
|
| 2024 |
+
"step": 28000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 84.89,
|
| 2028 |
+
"learning_rate": 2.865908591672022e-08,
|
| 2029 |
+
"loss": 1.219,
|
| 2030 |
+
"step": 28100
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 85.0,
|
| 2034 |
+
"eval_loss": 3.407871961593628,
|
| 2035 |
+
"eval_runtime": 0.9538,
|
| 2036 |
+
"eval_samples_per_second": 35.648,
|
| 2037 |
+
"eval_steps_per_second": 35.648,
|
| 2038 |
+
"step": 28135
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 85.2,
|
| 2042 |
+
"learning_rate": 2.956617282787915e-11,
|
| 2043 |
+
"loss": 1.2712,
|
| 2044 |
+
"step": 28200
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 85.5,
|
| 2048 |
+
"learning_rate": 3.2001142050499094e-08,
|
| 2049 |
+
"loss": 1.2184,
|
| 2050 |
+
"step": 28300
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 85.8,
|
| 2054 |
+
"learning_rate": 9.631801263387146e-08,
|
| 2055 |
+
"loss": 1.2515,
|
| 2056 |
+
"step": 28400
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 86.0,
|
| 2060 |
+
"eval_loss": 3.4080073833465576,
|
| 2061 |
+
"eval_runtime": 0.951,
|
| 2062 |
+
"eval_samples_per_second": 35.752,
|
| 2063 |
+
"eval_steps_per_second": 35.752,
|
| 2064 |
+
"step": 28466
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 86.1,
|
| 2068 |
+
"learning_rate": 1.361382913144042e-07,
|
| 2069 |
+
"loss": 1.2196,
|
| 2070 |
+
"step": 28500
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 86.4,
|
| 2074 |
+
"learning_rate": 1.162696600432059e-07,
|
| 2075 |
+
"loss": 1.2085,
|
| 2076 |
+
"step": 28600
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 86.71,
|
| 2080 |
+
"learning_rate": 5.427159389331479e-08,
|
| 2081 |
+
"loss": 1.2316,
|
| 2082 |
+
"step": 28700
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 87.0,
|
| 2086 |
+
"eval_loss": 3.408378839492798,
|
| 2087 |
+
"eval_runtime": 0.9581,
|
| 2088 |
+
"eval_samples_per_second": 35.486,
|
| 2089 |
+
"eval_steps_per_second": 35.486,
|
| 2090 |
+
"step": 28797
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 87.01,
|
| 2094 |
+
"learning_rate": 4.936669277050985e-09,
|
| 2095 |
+
"loss": 1.2578,
|
| 2096 |
+
"step": 28800
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 87.31,
|
| 2100 |
+
"learning_rate": 1.1866046596701257e-08,
|
| 2101 |
+
"loss": 1.24,
|
| 2102 |
+
"step": 28900
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 87.61,
|
| 2106 |
+
"learning_rate": 6.893568909023435e-08,
|
| 2107 |
+
"loss": 1.2342,
|
| 2108 |
+
"step": 29000
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 87.92,
|
| 2112 |
+
"learning_rate": 1.2570865668292598e-07,
|
| 2113 |
+
"loss": 1.2085,
|
| 2114 |
+
"step": 29100
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 88.0,
|
| 2118 |
+
"eval_loss": 3.4085471630096436,
|
| 2119 |
+
"eval_runtime": 1.0334,
|
| 2120 |
+
"eval_samples_per_second": 32.901,
|
| 2121 |
+
"eval_steps_per_second": 32.901,
|
| 2122 |
+
"step": 29128
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 88.22,
|
| 2126 |
+
"learning_rate": 1.32010204287465e-07,
|
| 2127 |
+
"loss": 1.2504,
|
| 2128 |
+
"step": 29200
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 88.52,
|
| 2132 |
+
"learning_rate": 8.227115773311707e-08,
|
| 2133 |
+
"loss": 1.2419,
|
| 2134 |
+
"step": 29300
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 88.82,
|
| 2138 |
+
"learning_rate": 2.0449831823058302e-08,
|
| 2139 |
+
"loss": 1.2334,
|
| 2140 |
+
"step": 29400
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 89.0,
|
| 2144 |
+
"eval_loss": 3.4085144996643066,
|
| 2145 |
+
"eval_runtime": 1.0283,
|
| 2146 |
+
"eval_samples_per_second": 33.063,
|
| 2147 |
+
"eval_steps_per_second": 33.063,
|
| 2148 |
+
"step": 29459
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 89.12,
|
| 2152 |
+
"learning_rate": 1.1826036928064389e-09,
|
| 2153 |
+
"loss": 1.2213,
|
| 2154 |
+
"step": 29500
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 89.43,
|
| 2158 |
+
"learning_rate": 4.088198736612909e-08,
|
| 2159 |
+
"loss": 1.2165,
|
| 2160 |
+
"step": 29600
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 89.73,
|
| 2164 |
+
"learning_rate": 1.0519885794950143e-07,
|
| 2165 |
+
"loss": 1.2263,
|
| 2166 |
+
"step": 29700
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 90.0,
|
| 2170 |
+
"eval_loss": 3.408367156982422,
|
| 2171 |
+
"eval_runtime": 1.0186,
|
| 2172 |
+
"eval_samples_per_second": 33.38,
|
| 2173 |
+
"eval_steps_per_second": 33.38,
|
| 2174 |
+
"step": 29790
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 90.0,
|
| 2178 |
+
"step": 29790,
|
| 2179 |
+
"total_flos": 1.037068075008e+16,
|
| 2180 |
+
"train_loss": 0.13739024364296004,
|
| 2181 |
+
"train_runtime": 461.3754,
|
| 2182 |
+
"train_samples_per_second": 64.568,
|
| 2183 |
+
"train_steps_per_second": 64.568
|
| 2184 |
}
|
| 2185 |
],
|
| 2186 |
+
"max_steps": 29790,
|
| 2187 |
+
"num_train_epochs": 90,
|
| 2188 |
+
"total_flos": 1.037068075008e+16,
|
| 2189 |
"trial_name": null,
|
| 2190 |
"trial_params": null
|
| 2191 |
}
|