Training in progress, step 271, 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 671149168
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8a905affccc1b6145d62691708c625e614f206127f91356929e645682c277e6
|
| 3 |
size 671149168
|
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 sha256:573d98362e681eb547d72d58d88a55adc74305d83b4d51b48d13bd5c920d68d4
|
| 3 |
+
size 341314644
|
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:b361b8475bb615094c5f4de63cfa4af90651dc59b0efe2e8b589bd0fdbe124f4
|
| 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:e4475ffcc8fe668c2923c634b7b5dbf821babf4359957da4392b7e4aa7d872b2
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.25790610909461975,
|
| 3 |
"best_model_checkpoint": "miner_id_24/checkpoint-200",
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1431,6 +1431,503 @@
|
|
| 1431 |
"eval_samples_per_second": 6.702,
|
| 1432 |
"eval_steps_per_second": 1.676,
|
| 1433 |
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1434 |
}
|
| 1435 |
],
|
| 1436 |
"logging_steps": 1,
|
|
@@ -1454,12 +1951,12 @@
|
|
| 1454 |
"should_evaluate": false,
|
| 1455 |
"should_log": false,
|
| 1456 |
"should_save": true,
|
| 1457 |
-
"should_training_stop":
|
| 1458 |
},
|
| 1459 |
"attributes": {}
|
| 1460 |
}
|
| 1461 |
},
|
| 1462 |
-
"total_flos":
|
| 1463 |
"train_batch_size": 8,
|
| 1464 |
"trial_name": null,
|
| 1465 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.25790610909461975,
|
| 3 |
"best_model_checkpoint": "miner_id_24/checkpoint-200",
|
| 4 |
+
"epoch": 1.0027726432532347,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 271,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1431 |
"eval_samples_per_second": 6.702,
|
| 1432 |
"eval_steps_per_second": 1.676,
|
| 1433 |
"step": 200
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.7430683918669131,
|
| 1437 |
+
"grad_norm": 1.1571354866027832,
|
| 1438 |
+
"learning_rate": 1.7993980802668946e-05,
|
| 1439 |
+
"loss": 1.0016,
|
| 1440 |
+
"step": 201
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 0.7467652495378928,
|
| 1444 |
+
"grad_norm": 1.2866359949111938,
|
| 1445 |
+
"learning_rate": 1.7515702331780753e-05,
|
| 1446 |
+
"loss": 1.1153,
|
| 1447 |
+
"step": 202
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 0.7504621072088724,
|
| 1451 |
+
"grad_norm": 1.2495567798614502,
|
| 1452 |
+
"learning_rate": 1.7042512716335873e-05,
|
| 1453 |
+
"loss": 1.6495,
|
| 1454 |
+
"step": 203
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.7541589648798521,
|
| 1458 |
+
"grad_norm": 1.184030294418335,
|
| 1459 |
+
"learning_rate": 1.6574486084258366e-05,
|
| 1460 |
+
"loss": 0.9913,
|
| 1461 |
+
"step": 204
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 0.7578558225508318,
|
| 1465 |
+
"grad_norm": 1.1039363145828247,
|
| 1466 |
+
"learning_rate": 1.6111695754660667e-05,
|
| 1467 |
+
"loss": 0.9402,
|
| 1468 |
+
"step": 205
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 0.7615526802218114,
|
| 1472 |
+
"grad_norm": 1.1745617389678955,
|
| 1473 |
+
"learning_rate": 1.565421422635782e-05,
|
| 1474 |
+
"loss": 1.0854,
|
| 1475 |
+
"step": 206
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 0.7652495378927912,
|
| 1479 |
+
"grad_norm": 1.162819743156433,
|
| 1480 |
+
"learning_rate": 1.5202113166510057e-05,
|
| 1481 |
+
"loss": 1.1295,
|
| 1482 |
+
"step": 207
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 0.7689463955637708,
|
| 1486 |
+
"grad_norm": 1.0997729301452637,
|
| 1487 |
+
"learning_rate": 1.475546339939568e-05,
|
| 1488 |
+
"loss": 0.9278,
|
| 1489 |
+
"step": 208
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 0.7726432532347505,
|
| 1493 |
+
"grad_norm": 1.1672478914260864,
|
| 1494 |
+
"learning_rate": 1.4314334895316094e-05,
|
| 1495 |
+
"loss": 1.1381,
|
| 1496 |
+
"step": 209
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.7763401109057301,
|
| 1500 |
+
"grad_norm": 1.097519874572754,
|
| 1501 |
+
"learning_rate": 1.3878796759634544e-05,
|
| 1502 |
+
"loss": 0.9361,
|
| 1503 |
+
"step": 210
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 0.7800369685767098,
|
| 1507 |
+
"grad_norm": 1.0834153890609741,
|
| 1508 |
+
"learning_rate": 1.3448917221950263e-05,
|
| 1509 |
+
"loss": 0.8876,
|
| 1510 |
+
"step": 211
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.7837338262476895,
|
| 1514 |
+
"grad_norm": 1.1496264934539795,
|
| 1515 |
+
"learning_rate": 1.3024763625410024e-05,
|
| 1516 |
+
"loss": 0.9791,
|
| 1517 |
+
"step": 212
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 0.7874306839186691,
|
| 1521 |
+
"grad_norm": 1.1406208276748657,
|
| 1522 |
+
"learning_rate": 1.2606402416158392e-05,
|
| 1523 |
+
"loss": 0.964,
|
| 1524 |
+
"step": 213
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 0.7911275415896488,
|
| 1528 |
+
"grad_norm": 1.2033501863479614,
|
| 1529 |
+
"learning_rate": 1.2193899132928537e-05,
|
| 1530 |
+
"loss": 1.0953,
|
| 1531 |
+
"step": 214
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 0.7948243992606284,
|
| 1535 |
+
"grad_norm": 1.1828855276107788,
|
| 1536 |
+
"learning_rate": 1.1787318396775188e-05,
|
| 1537 |
+
"loss": 0.9958,
|
| 1538 |
+
"step": 215
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.7985212569316081,
|
| 1542 |
+
"grad_norm": 1.1799291372299194,
|
| 1543 |
+
"learning_rate": 1.138672390095143e-05,
|
| 1544 |
+
"loss": 1.0501,
|
| 1545 |
+
"step": 216
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 0.8022181146025879,
|
| 1549 |
+
"grad_norm": 1.232038974761963,
|
| 1550 |
+
"learning_rate": 1.0992178400930753e-05,
|
| 1551 |
+
"loss": 1.0139,
|
| 1552 |
+
"step": 217
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.8059149722735675,
|
| 1556 |
+
"grad_norm": 1.1742757558822632,
|
| 1557 |
+
"learning_rate": 1.060374370457599e-05,
|
| 1558 |
+
"loss": 0.8714,
|
| 1559 |
+
"step": 218
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 0.8096118299445472,
|
| 1563 |
+
"grad_norm": 1.2701990604400635,
|
| 1564 |
+
"learning_rate": 1.0221480662456845e-05,
|
| 1565 |
+
"loss": 0.9587,
|
| 1566 |
+
"step": 219
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 0.8133086876155268,
|
| 1570 |
+
"grad_norm": 1.295379400253296,
|
| 1571 |
+
"learning_rate": 9.845449158317215e-06,
|
| 1572 |
+
"loss": 1.1038,
|
| 1573 |
+
"step": 220
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 0.8170055452865065,
|
| 1577 |
+
"grad_norm": 1.2289679050445557,
|
| 1578 |
+
"learning_rate": 9.475708099694124e-06,
|
| 1579 |
+
"loss": 0.9174,
|
| 1580 |
+
"step": 221
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.8207024029574861,
|
| 1584 |
+
"grad_norm": 1.3198155164718628,
|
| 1585 |
+
"learning_rate": 9.112315408689414e-06,
|
| 1586 |
+
"loss": 1.0472,
|
| 1587 |
+
"step": 222
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 0.8243992606284658,
|
| 1591 |
+
"grad_norm": 1.1874481439590454,
|
| 1592 |
+
"learning_rate": 8.755328012896003e-06,
|
| 1593 |
+
"loss": 0.8252,
|
| 1594 |
+
"step": 223
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 0.8280961182994455,
|
| 1598 |
+
"grad_norm": 1.2251365184783936,
|
| 1599 |
+
"learning_rate": 8.404801836479808e-06,
|
| 1600 |
+
"loss": 0.8606,
|
| 1601 |
+
"step": 224
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 0.8317929759704251,
|
| 1605 |
+
"grad_norm": 1.3227380514144897,
|
| 1606 |
+
"learning_rate": 8.060791791418886e-06,
|
| 1607 |
+
"loss": 1.0696,
|
| 1608 |
+
"step": 225
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 0.8354898336414048,
|
| 1612 |
+
"grad_norm": 1.2983530759811401,
|
| 1613 |
+
"learning_rate": 7.723351768901172e-06,
|
| 1614 |
+
"loss": 0.9411,
|
| 1615 |
+
"step": 226
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 0.8391866913123844,
|
| 1619 |
+
"grad_norm": 1.256066083908081,
|
| 1620 |
+
"learning_rate": 7.392534630882092e-06,
|
| 1621 |
+
"loss": 0.8509,
|
| 1622 |
+
"step": 227
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 0.8428835489833642,
|
| 1626 |
+
"grad_norm": 1.2773510217666626,
|
| 1627 |
+
"learning_rate": 7.06839220180342e-06,
|
| 1628 |
+
"loss": 0.8596,
|
| 1629 |
+
"step": 228
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 0.8465804066543438,
|
| 1633 |
+
"grad_norm": 1.3085066080093384,
|
| 1634 |
+
"learning_rate": 6.750975260474718e-06,
|
| 1635 |
+
"loss": 0.885,
|
| 1636 |
+
"step": 229
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 0.8502772643253235,
|
| 1640 |
+
"grad_norm": 1.2915611267089844,
|
| 1641 |
+
"learning_rate": 6.440333532118503e-06,
|
| 1642 |
+
"loss": 0.9502,
|
| 1643 |
+
"step": 230
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 0.8539741219963032,
|
| 1647 |
+
"grad_norm": 1.2121251821517944,
|
| 1648 |
+
"learning_rate": 6.136515680580479e-06,
|
| 1649 |
+
"loss": 0.8245,
|
| 1650 |
+
"step": 231
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 0.8576709796672828,
|
| 1654 |
+
"grad_norm": 1.3207285404205322,
|
| 1655 |
+
"learning_rate": 5.839569300706127e-06,
|
| 1656 |
+
"loss": 0.8947,
|
| 1657 |
+
"step": 232
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 0.8613678373382625,
|
| 1661 |
+
"grad_norm": 1.4474835395812988,
|
| 1662 |
+
"learning_rate": 5.549540910884648e-06,
|
| 1663 |
+
"loss": 1.1095,
|
| 1664 |
+
"step": 233
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.8650646950092421,
|
| 1668 |
+
"grad_norm": 1.3800991773605347,
|
| 1669 |
+
"learning_rate": 5.266475945761562e-06,
|
| 1670 |
+
"loss": 0.8926,
|
| 1671 |
+
"step": 234
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 0.8687615526802218,
|
| 1675 |
+
"grad_norm": 1.2967448234558105,
|
| 1676 |
+
"learning_rate": 4.990418749121178e-06,
|
| 1677 |
+
"loss": 0.8284,
|
| 1678 |
+
"step": 235
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 0.8724584103512015,
|
| 1682 |
+
"grad_norm": 1.3374454975128174,
|
| 1683 |
+
"learning_rate": 4.721412566939804e-06,
|
| 1684 |
+
"loss": 0.9239,
|
| 1685 |
+
"step": 236
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 0.8761552680221811,
|
| 1689 |
+
"grad_norm": 1.404314637184143,
|
| 1690 |
+
"learning_rate": 4.459499540611078e-06,
|
| 1691 |
+
"loss": 0.9797,
|
| 1692 |
+
"step": 237
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 0.8798521256931608,
|
| 1696 |
+
"grad_norm": 1.361304759979248,
|
| 1697 |
+
"learning_rate": 4.2047207003442e-06,
|
| 1698 |
+
"loss": 0.8175,
|
| 1699 |
+
"step": 238
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 0.8835489833641405,
|
| 1703 |
+
"grad_norm": 1.464787244796753,
|
| 1704 |
+
"learning_rate": 3.9571159587363734e-06,
|
| 1705 |
+
"loss": 0.912,
|
| 1706 |
+
"step": 239
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.8872458410351202,
|
| 1710 |
+
"grad_norm": 1.3126565217971802,
|
| 1711 |
+
"learning_rate": 3.7167241045202473e-06,
|
| 1712 |
+
"loss": 0.8109,
|
| 1713 |
+
"step": 240
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 0.8909426987060998,
|
| 1717 |
+
"grad_norm": 1.3885515928268433,
|
| 1718 |
+
"learning_rate": 3.4835827964873945e-06,
|
| 1719 |
+
"loss": 0.8852,
|
| 1720 |
+
"step": 241
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 0.8946395563770795,
|
| 1724 |
+
"grad_norm": 1.48948335647583,
|
| 1725 |
+
"learning_rate": 3.2577285575889017e-06,
|
| 1726 |
+
"loss": 0.9546,
|
| 1727 |
+
"step": 242
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 0.8983364140480592,
|
| 1731 |
+
"grad_norm": 1.467980146408081,
|
| 1732 |
+
"learning_rate": 3.039196769213787e-06,
|
| 1733 |
+
"loss": 0.8802,
|
| 1734 |
+
"step": 243
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 0.9020332717190388,
|
| 1738 |
+
"grad_norm": 1.5345256328582764,
|
| 1739 |
+
"learning_rate": 2.8280216656463408e-06,
|
| 1740 |
+
"loss": 0.835,
|
| 1741 |
+
"step": 244
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 0.9057301293900185,
|
| 1745 |
+
"grad_norm": 1.5734572410583496,
|
| 1746 |
+
"learning_rate": 2.6242363287030613e-06,
|
| 1747 |
+
"loss": 1.0006,
|
| 1748 |
+
"step": 245
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.9094269870609981,
|
| 1752 |
+
"grad_norm": 1.4659297466278076,
|
| 1753 |
+
"learning_rate": 2.4278726825502697e-06,
|
| 1754 |
+
"loss": 0.8125,
|
| 1755 |
+
"step": 246
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 0.9131238447319778,
|
| 1759 |
+
"grad_norm": 1.6777136325836182,
|
| 1760 |
+
"learning_rate": 2.2389614887029566e-06,
|
| 1761 |
+
"loss": 0.9509,
|
| 1762 |
+
"step": 247
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 0.9168207024029574,
|
| 1766 |
+
"grad_norm": 1.7119040489196777,
|
| 1767 |
+
"learning_rate": 2.0575323412058033e-06,
|
| 1768 |
+
"loss": 0.8874,
|
| 1769 |
+
"step": 248
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.9205175600739371,
|
| 1773 |
+
"grad_norm": 1.8174391984939575,
|
| 1774 |
+
"learning_rate": 1.8836136619971466e-06,
|
| 1775 |
+
"loss": 0.9305,
|
| 1776 |
+
"step": 249
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 0.9242144177449169,
|
| 1780 |
+
"grad_norm": 1.8792368173599243,
|
| 1781 |
+
"learning_rate": 1.7172326964564778e-06,
|
| 1782 |
+
"loss": 0.9589,
|
| 1783 |
+
"step": 250
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 0.9279112754158965,
|
| 1787 |
+
"grad_norm": 1.1916697025299072,
|
| 1788 |
+
"learning_rate": 1.5584155091362906e-06,
|
| 1789 |
+
"loss": 1.0869,
|
| 1790 |
+
"step": 251
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 0.9316081330868762,
|
| 1794 |
+
"grad_norm": 1.109428882598877,
|
| 1795 |
+
"learning_rate": 1.4071869796789428e-06,
|
| 1796 |
+
"loss": 0.9649,
|
| 1797 |
+
"step": 252
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 0.9353049907578558,
|
| 1801 |
+
"grad_norm": 1.1600779294967651,
|
| 1802 |
+
"learning_rate": 1.263570798919106e-06,
|
| 1803 |
+
"loss": 1.0584,
|
| 1804 |
+
"step": 253
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 0.9390018484288355,
|
| 1808 |
+
"grad_norm": 1.027513027191162,
|
| 1809 |
+
"learning_rate": 1.1275894651724517e-06,
|
| 1810 |
+
"loss": 0.8654,
|
| 1811 |
+
"step": 254
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 0.9426987060998152,
|
| 1815 |
+
"grad_norm": 1.1321669816970825,
|
| 1816 |
+
"learning_rate": 9.992642807111485e-07,
|
| 1817 |
+
"loss": 0.9776,
|
| 1818 |
+
"step": 255
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 0.9463955637707948,
|
| 1822 |
+
"grad_norm": 1.2419346570968628,
|
| 1823 |
+
"learning_rate": 8.786153484267589e-07,
|
| 1824 |
+
"loss": 1.0648,
|
| 1825 |
+
"step": 256
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 0.9500924214417745,
|
| 1829 |
+
"grad_norm": 1.159534215927124,
|
| 1830 |
+
"learning_rate": 7.656615686809976e-07,
|
| 1831 |
+
"loss": 0.9293,
|
| 1832 |
+
"step": 257
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 0.9537892791127541,
|
| 1836 |
+
"grad_norm": 1.2147092819213867,
|
| 1837 |
+
"learning_rate": 6.604206363448661e-07,
|
| 1838 |
+
"loss": 0.9849,
|
| 1839 |
+
"step": 258
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 0.9574861367837338,
|
| 1843 |
+
"grad_norm": 1.2285771369934082,
|
| 1844 |
+
"learning_rate": 5.629090380266544e-07,
|
| 1845 |
+
"loss": 0.9584,
|
| 1846 |
+
"step": 259
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 0.9611829944547134,
|
| 1850 |
+
"grad_norm": 1.2933838367462158,
|
| 1851 |
+
"learning_rate": 4.7314204948923356e-07,
|
| 1852 |
+
"loss": 0.9436,
|
| 1853 |
+
"step": 260
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 0.9648798521256932,
|
| 1857 |
+
"grad_norm": 1.2568949460983276,
|
| 1858 |
+
"learning_rate": 3.9113373325698754e-07,
|
| 1859 |
+
"loss": 0.9507,
|
| 1860 |
+
"step": 261
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 0.9685767097966729,
|
| 1864 |
+
"grad_norm": 1.251380443572998,
|
| 1865 |
+
"learning_rate": 3.168969364128527e-07,
|
| 1866 |
+
"loss": 0.9798,
|
| 1867 |
+
"step": 262
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 0.9722735674676525,
|
| 1871 |
+
"grad_norm": 1.3915915489196777,
|
| 1872 |
+
"learning_rate": 2.5044328858576106e-07,
|
| 1873 |
+
"loss": 1.047,
|
| 1874 |
+
"step": 263
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 0.9759704251386322,
|
| 1878 |
+
"grad_norm": 1.353722333908081,
|
| 1879 |
+
"learning_rate": 1.917832001287645e-07,
|
| 1880 |
+
"loss": 0.8804,
|
| 1881 |
+
"step": 264
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 0.9796672828096118,
|
| 1885 |
+
"grad_norm": 1.3146427869796753,
|
| 1886 |
+
"learning_rate": 1.4092586048820578e-07,
|
| 1887 |
+
"loss": 0.9416,
|
| 1888 |
+
"step": 265
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 0.9833641404805915,
|
| 1892 |
+
"grad_norm": 1.4524424076080322,
|
| 1893 |
+
"learning_rate": 9.787923676414235e-08,
|
| 1894 |
+
"loss": 1.0211,
|
| 1895 |
+
"step": 266
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 0.9870609981515711,
|
| 1899 |
+
"grad_norm": 1.499550461769104,
|
| 1900 |
+
"learning_rate": 6.265007246223364e-08,
|
| 1901 |
+
"loss": 0.9393,
|
| 1902 |
+
"step": 267
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 0.9907578558225508,
|
| 1906 |
+
"grad_norm": 1.4518331289291382,
|
| 1907 |
+
"learning_rate": 3.524388643736387e-08,
|
| 1908 |
+
"loss": 0.8188,
|
| 1909 |
+
"step": 268
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 0.9944547134935305,
|
| 1913 |
+
"grad_norm": 1.5743730068206787,
|
| 1914 |
+
"learning_rate": 1.566497202904471e-08,
|
| 1915 |
+
"loss": 0.8282,
|
| 1916 |
+
"step": 269
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 0.9981515711645101,
|
| 1920 |
+
"grad_norm": 1.7006173133850098,
|
| 1921 |
+
"learning_rate": 3.91639638886998e-09,
|
| 1922 |
+
"loss": 0.7826,
|
| 1923 |
+
"step": 270
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 1.0027726432532347,
|
| 1927 |
+
"grad_norm": 1.2868958711624146,
|
| 1928 |
+
"learning_rate": 0.0,
|
| 1929 |
+
"loss": 0.9401,
|
| 1930 |
+
"step": 271
|
| 1931 |
}
|
| 1932 |
],
|
| 1933 |
"logging_steps": 1,
|
|
|
|
| 1951 |
"should_evaluate": false,
|
| 1952 |
"should_log": false,
|
| 1953 |
"should_save": true,
|
| 1954 |
+
"should_training_stop": true
|
| 1955 |
},
|
| 1956 |
"attributes": {}
|
| 1957 |
}
|
| 1958 |
},
|
| 1959 |
+
"total_flos": 7.756009208713052e+17,
|
| 1960 |
"train_batch_size": 8,
|
| 1961 |
"trial_name": null,
|
| 1962 |
"trial_params": null
|