training state at step 100
Browse files- trainer_state.json +78 -3
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 1.
|
| 5 |
"eval_steps": 1,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1387,6 +1387,81 @@
|
|
| 1387 |
"eval_samples_per_second": 1.098,
|
| 1388 |
"eval_steps_per_second": 0.549,
|
| 1389 |
"step": 94
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1390 |
}
|
| 1391 |
],
|
| 1392 |
"logging_steps": 1,
|
|
@@ -1406,7 +1481,7 @@
|
|
| 1406 |
"attributes": {}
|
| 1407 |
}
|
| 1408 |
},
|
| 1409 |
-
"total_flos": 1.
|
| 1410 |
"train_batch_size": 2,
|
| 1411 |
"trial_name": null,
|
| 1412 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.3793103448275863,
|
| 5 |
"eval_steps": 1,
|
| 6 |
+
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1387 |
"eval_samples_per_second": 1.098,
|
| 1388 |
"eval_steps_per_second": 0.549,
|
| 1389 |
"step": 94
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.3103448275862069,
|
| 1393 |
+
"grad_norm": 2.077143669128418,
|
| 1394 |
+
"learning_rate": 4.894543310469968e-05,
|
| 1395 |
+
"loss": 1.2378,
|
| 1396 |
+
"step": 95
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.3103448275862069,
|
| 1400 |
+
"eval_loss": 0.6547893285751343,
|
| 1401 |
+
"eval_runtime": 18.7488,
|
| 1402 |
+
"eval_samples_per_second": 1.067,
|
| 1403 |
+
"eval_steps_per_second": 0.533,
|
| 1404 |
+
"step": 95
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 1.3241379310344827,
|
| 1408 |
+
"grad_norm": 1.9517972469329834,
|
| 1409 |
+
"learning_rate": 4.882595527372152e-05,
|
| 1410 |
+
"loss": 1.0997,
|
| 1411 |
+
"step": 96
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.3241379310344827,
|
| 1415 |
+
"eval_loss": 0.6498640775680542,
|
| 1416 |
+
"eval_runtime": 18.4304,
|
| 1417 |
+
"eval_samples_per_second": 1.085,
|
| 1418 |
+
"eval_steps_per_second": 0.543,
|
| 1419 |
+
"step": 96
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.3379310344827586,
|
| 1423 |
+
"grad_norm": 2.0447959899902344,
|
| 1424 |
+
"learning_rate": 4.870022949890676e-05,
|
| 1425 |
+
"loss": 0.9613,
|
| 1426 |
+
"step": 97
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.3379310344827586,
|
| 1430 |
+
"eval_loss": 0.6370054483413696,
|
| 1431 |
+
"eval_runtime": 18.252,
|
| 1432 |
+
"eval_samples_per_second": 1.096,
|
| 1433 |
+
"eval_steps_per_second": 0.548,
|
| 1434 |
+
"step": 97
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 1.3517241379310345,
|
| 1438 |
+
"grad_norm": 2.078657865524292,
|
| 1439 |
+
"learning_rate": 4.856828874966086e-05,
|
| 1440 |
+
"loss": 1.1216,
|
| 1441 |
+
"step": 98
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.3517241379310345,
|
| 1445 |
+
"eval_loss": 0.6291982531547546,
|
| 1446 |
+
"eval_runtime": 18.2386,
|
| 1447 |
+
"eval_samples_per_second": 1.097,
|
| 1448 |
+
"eval_steps_per_second": 0.548,
|
| 1449 |
+
"step": 98
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 1.3655172413793104,
|
| 1453 |
+
"grad_norm": 2.0556623935699463,
|
| 1454 |
+
"learning_rate": 4.8430167625158595e-05,
|
| 1455 |
+
"loss": 1.0718,
|
| 1456 |
+
"step": 99
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.3655172413793104,
|
| 1460 |
+
"eval_loss": 0.6218433380126953,
|
| 1461 |
+
"eval_runtime": 18.1671,
|
| 1462 |
+
"eval_samples_per_second": 1.101,
|
| 1463 |
+
"eval_steps_per_second": 0.55,
|
| 1464 |
+
"step": 99
|
| 1465 |
}
|
| 1466 |
],
|
| 1467 |
"logging_steps": 1,
|
|
|
|
| 1481 |
"attributes": {}
|
| 1482 |
}
|
| 1483 |
},
|
| 1484 |
+
"total_flos": 1.2709753377329971e+17,
|
| 1485 |
"train_batch_size": 2,
|
| 1486 |
"trial_name": null,
|
| 1487 |
"trial_params": null
|