Uploaded checkpoint-3000
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +719 -3
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 119975656
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62872446dd68a11022c82656dd80183dd34c09e92c29d79491b5aa450e74f22f
|
3 |
size 119975656
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 240145026
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a75c8fefe10021904a2546c5f68c5efa8f9bcf35bb732e559d62b4c792c9dbbb
|
3 |
size 240145026
|
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:f7eeee07b40fef8c7bdf027c427b1fc8d6a45d979762d8d637d73e82015e5add
|
3 |
size 14244
|
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 sha256:770db92ac44ccb712216aece2abb8a41e68fd6d952c7ae7884e9032fb3cc3f81
|
3 |
size 1064
|
trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": 0.018313532695174217,
|
3 |
"best_model_checkpoint": "runs/deepseek_lora_20240423-133229/checkpoint-2000",
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -1439,6 +1439,722 @@
|
|
1439 |
"eval_samples_per_second": 16.107,
|
1440 |
"eval_steps_per_second": 16.107,
|
1441 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1442 |
}
|
1443 |
],
|
1444 |
"logging_steps": 10,
|
@@ -1446,7 +2162,7 @@
|
|
1446 |
"num_input_tokens_seen": 0,
|
1447 |
"num_train_epochs": 2,
|
1448 |
"save_steps": 1000,
|
1449 |
-
"total_flos":
|
1450 |
"train_batch_size": 1,
|
1451 |
"trial_name": null,
|
1452 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": 0.018313532695174217,
|
3 |
"best_model_checkpoint": "runs/deepseek_lora_20240423-133229/checkpoint-2000",
|
4 |
+
"epoch": 0.9399232396020991,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 3000,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
1439 |
"eval_samples_per_second": 16.107,
|
1440 |
"eval_steps_per_second": 16.107,
|
1441 |
"step": 2000
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 0.63,
|
1445 |
+
"grad_norm": 0.0025979657657444477,
|
1446 |
+
"learning_rate": 1.3288888888888889e-05,
|
1447 |
+
"loss": 0.0485,
|
1448 |
+
"step": 2010
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 0.63,
|
1452 |
+
"grad_norm": 0.00765944505110383,
|
1453 |
+
"learning_rate": 1.3244444444444447e-05,
|
1454 |
+
"loss": 0.0292,
|
1455 |
+
"step": 2020
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 0.64,
|
1459 |
+
"grad_norm": 0.9524846076965332,
|
1460 |
+
"learning_rate": 1.3200000000000002e-05,
|
1461 |
+
"loss": 0.0545,
|
1462 |
+
"step": 2030
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 0.64,
|
1466 |
+
"grad_norm": 0.0015762520488351583,
|
1467 |
+
"learning_rate": 1.3155555555555558e-05,
|
1468 |
+
"loss": 0.0003,
|
1469 |
+
"step": 2040
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 0.64,
|
1473 |
+
"grad_norm": 0.12439166754484177,
|
1474 |
+
"learning_rate": 1.3111111111111113e-05,
|
1475 |
+
"loss": 0.0346,
|
1476 |
+
"step": 2050
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 0.65,
|
1480 |
+
"grad_norm": 0.002841875422745943,
|
1481 |
+
"learning_rate": 1.3066666666666668e-05,
|
1482 |
+
"loss": 0.0785,
|
1483 |
+
"step": 2060
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 0.65,
|
1487 |
+
"grad_norm": 1.3388820886611938,
|
1488 |
+
"learning_rate": 1.3022222222222223e-05,
|
1489 |
+
"loss": 0.0748,
|
1490 |
+
"step": 2070
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 0.65,
|
1494 |
+
"grad_norm": 0.003183074528351426,
|
1495 |
+
"learning_rate": 1.2977777777777779e-05,
|
1496 |
+
"loss": 0.0653,
|
1497 |
+
"step": 2080
|
1498 |
+
},
|
1499 |
+
{
|
1500 |
+
"epoch": 0.65,
|
1501 |
+
"grad_norm": 1.0637716054916382,
|
1502 |
+
"learning_rate": 1.2933333333333334e-05,
|
1503 |
+
"loss": 0.0522,
|
1504 |
+
"step": 2090
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 0.66,
|
1508 |
+
"grad_norm": 1.1531606912612915,
|
1509 |
+
"learning_rate": 1.288888888888889e-05,
|
1510 |
+
"loss": 0.0968,
|
1511 |
+
"step": 2100
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 0.66,
|
1515 |
+
"grad_norm": 0.0031077053863555193,
|
1516 |
+
"learning_rate": 1.2844444444444446e-05,
|
1517 |
+
"loss": 0.0419,
|
1518 |
+
"step": 2110
|
1519 |
+
},
|
1520 |
+
{
|
1521 |
+
"epoch": 0.66,
|
1522 |
+
"grad_norm": 0.011700381524860859,
|
1523 |
+
"learning_rate": 1.2800000000000001e-05,
|
1524 |
+
"loss": 0.0525,
|
1525 |
+
"step": 2120
|
1526 |
+
},
|
1527 |
+
{
|
1528 |
+
"epoch": 0.67,
|
1529 |
+
"grad_norm": 0.0015547301154583693,
|
1530 |
+
"learning_rate": 1.2755555555555556e-05,
|
1531 |
+
"loss": 0.0473,
|
1532 |
+
"step": 2130
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 0.67,
|
1536 |
+
"grad_norm": 0.0020142821595072746,
|
1537 |
+
"learning_rate": 1.2711111111111112e-05,
|
1538 |
+
"loss": 0.044,
|
1539 |
+
"step": 2140
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 0.67,
|
1543 |
+
"grad_norm": 0.005582904908806086,
|
1544 |
+
"learning_rate": 1.2666666666666667e-05,
|
1545 |
+
"loss": 0.0789,
|
1546 |
+
"step": 2150
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 0.68,
|
1550 |
+
"grad_norm": 0.0064437431283295155,
|
1551 |
+
"learning_rate": 1.2622222222222222e-05,
|
1552 |
+
"loss": 0.099,
|
1553 |
+
"step": 2160
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 0.68,
|
1557 |
+
"grad_norm": 0.0012363777495920658,
|
1558 |
+
"learning_rate": 1.257777777777778e-05,
|
1559 |
+
"loss": 0.0411,
|
1560 |
+
"step": 2170
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 0.68,
|
1564 |
+
"grad_norm": 0.14603936672210693,
|
1565 |
+
"learning_rate": 1.2533333333333336e-05,
|
1566 |
+
"loss": 0.006,
|
1567 |
+
"step": 2180
|
1568 |
+
},
|
1569 |
+
{
|
1570 |
+
"epoch": 0.69,
|
1571 |
+
"grad_norm": 0.001298425835557282,
|
1572 |
+
"learning_rate": 1.2488888888888891e-05,
|
1573 |
+
"loss": 0.1263,
|
1574 |
+
"step": 2190
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 0.69,
|
1578 |
+
"grad_norm": 0.001547910855151713,
|
1579 |
+
"learning_rate": 1.2444444444444446e-05,
|
1580 |
+
"loss": 0.0342,
|
1581 |
+
"step": 2200
|
1582 |
+
},
|
1583 |
+
{
|
1584 |
+
"epoch": 0.69,
|
1585 |
+
"grad_norm": 0.0013805734924972057,
|
1586 |
+
"learning_rate": 1.2400000000000002e-05,
|
1587 |
+
"loss": 0.0824,
|
1588 |
+
"step": 2210
|
1589 |
+
},
|
1590 |
+
{
|
1591 |
+
"epoch": 0.7,
|
1592 |
+
"grad_norm": 1.123624563217163,
|
1593 |
+
"learning_rate": 1.2355555555555557e-05,
|
1594 |
+
"loss": 0.0202,
|
1595 |
+
"step": 2220
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 0.7,
|
1599 |
+
"grad_norm": 0.0014894693158566952,
|
1600 |
+
"learning_rate": 1.2311111111111112e-05,
|
1601 |
+
"loss": 0.0313,
|
1602 |
+
"step": 2230
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 0.7,
|
1606 |
+
"grad_norm": 2.194322109222412,
|
1607 |
+
"learning_rate": 1.2266666666666667e-05,
|
1608 |
+
"loss": 0.123,
|
1609 |
+
"step": 2240
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 0.7,
|
1613 |
+
"grad_norm": 2.4329869747161865,
|
1614 |
+
"learning_rate": 1.2222222222222224e-05,
|
1615 |
+
"loss": 0.0619,
|
1616 |
+
"step": 2250
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 0.71,
|
1620 |
+
"grad_norm": 0.0048148720525205135,
|
1621 |
+
"learning_rate": 1.217777777777778e-05,
|
1622 |
+
"loss": 0.0105,
|
1623 |
+
"step": 2260
|
1624 |
+
},
|
1625 |
+
{
|
1626 |
+
"epoch": 0.71,
|
1627 |
+
"grad_norm": 0.8393607139587402,
|
1628 |
+
"learning_rate": 1.2133333333333335e-05,
|
1629 |
+
"loss": 0.0763,
|
1630 |
+
"step": 2270
|
1631 |
+
},
|
1632 |
+
{
|
1633 |
+
"epoch": 0.71,
|
1634 |
+
"grad_norm": 0.0013370034284889698,
|
1635 |
+
"learning_rate": 1.208888888888889e-05,
|
1636 |
+
"loss": 0.0263,
|
1637 |
+
"step": 2280
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 0.72,
|
1641 |
+
"grad_norm": 0.0011019165394827724,
|
1642 |
+
"learning_rate": 1.2044444444444445e-05,
|
1643 |
+
"loss": 0.0451,
|
1644 |
+
"step": 2290
|
1645 |
+
},
|
1646 |
+
{
|
1647 |
+
"epoch": 0.72,
|
1648 |
+
"grad_norm": 0.0017341958591714501,
|
1649 |
+
"learning_rate": 1.2e-05,
|
1650 |
+
"loss": 0.0612,
|
1651 |
+
"step": 2300
|
1652 |
+
},
|
1653 |
+
{
|
1654 |
+
"epoch": 0.72,
|
1655 |
+
"grad_norm": 0.0012070784578099847,
|
1656 |
+
"learning_rate": 1.1955555555555556e-05,
|
1657 |
+
"loss": 0.0383,
|
1658 |
+
"step": 2310
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 0.73,
|
1662 |
+
"grad_norm": 0.004143261816352606,
|
1663 |
+
"learning_rate": 1.191111111111111e-05,
|
1664 |
+
"loss": 0.0523,
|
1665 |
+
"step": 2320
|
1666 |
+
},
|
1667 |
+
{
|
1668 |
+
"epoch": 0.73,
|
1669 |
+
"grad_norm": 0.0022921899799257517,
|
1670 |
+
"learning_rate": 1.186666666666667e-05,
|
1671 |
+
"loss": 0.054,
|
1672 |
+
"step": 2330
|
1673 |
+
},
|
1674 |
+
{
|
1675 |
+
"epoch": 0.73,
|
1676 |
+
"grad_norm": 1.052677869796753,
|
1677 |
+
"learning_rate": 1.1822222222222225e-05,
|
1678 |
+
"loss": 0.0329,
|
1679 |
+
"step": 2340
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 0.74,
|
1683 |
+
"grad_norm": 0.5296663641929626,
|
1684 |
+
"learning_rate": 1.177777777777778e-05,
|
1685 |
+
"loss": 0.075,
|
1686 |
+
"step": 2350
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 0.74,
|
1690 |
+
"grad_norm": 0.001526731881313026,
|
1691 |
+
"learning_rate": 1.1733333333333335e-05,
|
1692 |
+
"loss": 0.0386,
|
1693 |
+
"step": 2360
|
1694 |
+
},
|
1695 |
+
{
|
1696 |
+
"epoch": 0.74,
|
1697 |
+
"grad_norm": 4.293755531311035,
|
1698 |
+
"learning_rate": 1.168888888888889e-05,
|
1699 |
+
"loss": 0.0576,
|
1700 |
+
"step": 2370
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 0.75,
|
1704 |
+
"grad_norm": 0.002120732795447111,
|
1705 |
+
"learning_rate": 1.1644444444444446e-05,
|
1706 |
+
"loss": 0.0398,
|
1707 |
+
"step": 2380
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 0.75,
|
1711 |
+
"grad_norm": 0.001393395708873868,
|
1712 |
+
"learning_rate": 1.16e-05,
|
1713 |
+
"loss": 0.0335,
|
1714 |
+
"step": 2390
|
1715 |
+
},
|
1716 |
+
{
|
1717 |
+
"epoch": 0.75,
|
1718 |
+
"grad_norm": 0.001357159111648798,
|
1719 |
+
"learning_rate": 1.1555555555555556e-05,
|
1720 |
+
"loss": 0.0186,
|
1721 |
+
"step": 2400
|
1722 |
+
},
|
1723 |
+
{
|
1724 |
+
"epoch": 0.76,
|
1725 |
+
"grad_norm": 2.845559597015381,
|
1726 |
+
"learning_rate": 1.1511111111111113e-05,
|
1727 |
+
"loss": 0.0231,
|
1728 |
+
"step": 2410
|
1729 |
+
},
|
1730 |
+
{
|
1731 |
+
"epoch": 0.76,
|
1732 |
+
"grad_norm": 0.001933318912051618,
|
1733 |
+
"learning_rate": 1.1466666666666668e-05,
|
1734 |
+
"loss": 0.0572,
|
1735 |
+
"step": 2420
|
1736 |
+
},
|
1737 |
+
{
|
1738 |
+
"epoch": 0.76,
|
1739 |
+
"grad_norm": 0.08605944365262985,
|
1740 |
+
"learning_rate": 1.1422222222222223e-05,
|
1741 |
+
"loss": 0.0361,
|
1742 |
+
"step": 2430
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 0.76,
|
1746 |
+
"grad_norm": 2.6697850227355957,
|
1747 |
+
"learning_rate": 1.1377777777777779e-05,
|
1748 |
+
"loss": 0.1098,
|
1749 |
+
"step": 2440
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 0.77,
|
1753 |
+
"grad_norm": 1.6600360870361328,
|
1754 |
+
"learning_rate": 1.1333333333333334e-05,
|
1755 |
+
"loss": 0.0599,
|
1756 |
+
"step": 2450
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 0.77,
|
1760 |
+
"grad_norm": 2.8962271213531494,
|
1761 |
+
"learning_rate": 1.1288888888888889e-05,
|
1762 |
+
"loss": 0.0333,
|
1763 |
+
"step": 2460
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 0.77,
|
1767 |
+
"grad_norm": 0.3499470055103302,
|
1768 |
+
"learning_rate": 1.1244444444444444e-05,
|
1769 |
+
"loss": 0.0276,
|
1770 |
+
"step": 2470
|
1771 |
+
},
|
1772 |
+
{
|
1773 |
+
"epoch": 0.78,
|
1774 |
+
"grad_norm": 0.0011357500916346908,
|
1775 |
+
"learning_rate": 1.1200000000000001e-05,
|
1776 |
+
"loss": 0.0662,
|
1777 |
+
"step": 2480
|
1778 |
+
},
|
1779 |
+
{
|
1780 |
+
"epoch": 0.78,
|
1781 |
+
"grad_norm": 0.9571266174316406,
|
1782 |
+
"learning_rate": 1.1155555555555556e-05,
|
1783 |
+
"loss": 0.0403,
|
1784 |
+
"step": 2490
|
1785 |
+
},
|
1786 |
+
{
|
1787 |
+
"epoch": 0.78,
|
1788 |
+
"grad_norm": 0.0018001727294176817,
|
1789 |
+
"learning_rate": 1.1111111111111113e-05,
|
1790 |
+
"loss": 0.0549,
|
1791 |
+
"step": 2500
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 0.78,
|
1795 |
+
"eval_loss": 0.019961679354310036,
|
1796 |
+
"eval_runtime": 62.0505,
|
1797 |
+
"eval_samples_per_second": 16.116,
|
1798 |
+
"eval_steps_per_second": 16.116,
|
1799 |
+
"step": 2500
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 0.79,
|
1803 |
+
"grad_norm": 0.002932826289907098,
|
1804 |
+
"learning_rate": 1.1066666666666669e-05,
|
1805 |
+
"loss": 0.0785,
|
1806 |
+
"step": 2510
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 0.79,
|
1810 |
+
"grad_norm": 0.0028327910695225,
|
1811 |
+
"learning_rate": 1.1022222222222224e-05,
|
1812 |
+
"loss": 0.0283,
|
1813 |
+
"step": 2520
|
1814 |
+
},
|
1815 |
+
{
|
1816 |
+
"epoch": 0.79,
|
1817 |
+
"grad_norm": 1.9044318199157715,
|
1818 |
+
"learning_rate": 1.0977777777777779e-05,
|
1819 |
+
"loss": 0.1637,
|
1820 |
+
"step": 2530
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 0.8,
|
1824 |
+
"grad_norm": 0.0019232028862461448,
|
1825 |
+
"learning_rate": 1.0933333333333334e-05,
|
1826 |
+
"loss": 0.004,
|
1827 |
+
"step": 2540
|
1828 |
+
},
|
1829 |
+
{
|
1830 |
+
"epoch": 0.8,
|
1831 |
+
"grad_norm": 0.002454034984111786,
|
1832 |
+
"learning_rate": 1.088888888888889e-05,
|
1833 |
+
"loss": 0.0091,
|
1834 |
+
"step": 2550
|
1835 |
+
},
|
1836 |
+
{
|
1837 |
+
"epoch": 0.8,
|
1838 |
+
"grad_norm": 0.0015948887448757887,
|
1839 |
+
"learning_rate": 1.0844444444444446e-05,
|
1840 |
+
"loss": 0.0276,
|
1841 |
+
"step": 2560
|
1842 |
+
},
|
1843 |
+
{
|
1844 |
+
"epoch": 0.81,
|
1845 |
+
"grad_norm": 1.5870649814605713,
|
1846 |
+
"learning_rate": 1.0800000000000002e-05,
|
1847 |
+
"loss": 0.0262,
|
1848 |
+
"step": 2570
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 0.81,
|
1852 |
+
"grad_norm": 2.2778539657592773,
|
1853 |
+
"learning_rate": 1.0755555555555557e-05,
|
1854 |
+
"loss": 0.0424,
|
1855 |
+
"step": 2580
|
1856 |
+
},
|
1857 |
+
{
|
1858 |
+
"epoch": 0.81,
|
1859 |
+
"grad_norm": 2.6451878547668457,
|
1860 |
+
"learning_rate": 1.0711111111111112e-05,
|
1861 |
+
"loss": 0.0679,
|
1862 |
+
"step": 2590
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 0.81,
|
1866 |
+
"grad_norm": 0.001484246808104217,
|
1867 |
+
"learning_rate": 1.0666666666666667e-05,
|
1868 |
+
"loss": 0.0373,
|
1869 |
+
"step": 2600
|
1870 |
+
},
|
1871 |
+
{
|
1872 |
+
"epoch": 0.82,
|
1873 |
+
"grad_norm": 0.001308325445279479,
|
1874 |
+
"learning_rate": 1.0622222222222223e-05,
|
1875 |
+
"loss": 0.0191,
|
1876 |
+
"step": 2610
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 0.82,
|
1880 |
+
"grad_norm": 0.0027939951978623867,
|
1881 |
+
"learning_rate": 1.0577777777777778e-05,
|
1882 |
+
"loss": 0.0459,
|
1883 |
+
"step": 2620
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 0.82,
|
1887 |
+
"grad_norm": 0.009125540032982826,
|
1888 |
+
"learning_rate": 1.0533333333333333e-05,
|
1889 |
+
"loss": 0.0704,
|
1890 |
+
"step": 2630
|
1891 |
+
},
|
1892 |
+
{
|
1893 |
+
"epoch": 0.83,
|
1894 |
+
"grad_norm": 0.0012461950536817312,
|
1895 |
+
"learning_rate": 1.048888888888889e-05,
|
1896 |
+
"loss": 0.0463,
|
1897 |
+
"step": 2640
|
1898 |
+
},
|
1899 |
+
{
|
1900 |
+
"epoch": 0.83,
|
1901 |
+
"grad_norm": 2.906261920928955,
|
1902 |
+
"learning_rate": 1.0444444444444445e-05,
|
1903 |
+
"loss": 0.0923,
|
1904 |
+
"step": 2650
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 0.83,
|
1908 |
+
"grad_norm": 3.149310827255249,
|
1909 |
+
"learning_rate": 1.04e-05,
|
1910 |
+
"loss": 0.0558,
|
1911 |
+
"step": 2660
|
1912 |
+
},
|
1913 |
+
{
|
1914 |
+
"epoch": 0.84,
|
1915 |
+
"grad_norm": 0.07791195064783096,
|
1916 |
+
"learning_rate": 1.0355555555555557e-05,
|
1917 |
+
"loss": 0.0148,
|
1918 |
+
"step": 2670
|
1919 |
+
},
|
1920 |
+
{
|
1921 |
+
"epoch": 0.84,
|
1922 |
+
"grad_norm": 0.0020532067865133286,
|
1923 |
+
"learning_rate": 1.0311111111111113e-05,
|
1924 |
+
"loss": 0.062,
|
1925 |
+
"step": 2680
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 0.84,
|
1929 |
+
"grad_norm": 0.0010498109040781856,
|
1930 |
+
"learning_rate": 1.0266666666666668e-05,
|
1931 |
+
"loss": 0.0577,
|
1932 |
+
"step": 2690
|
1933 |
+
},
|
1934 |
+
{
|
1935 |
+
"epoch": 0.85,
|
1936 |
+
"grad_norm": 0.1568715125322342,
|
1937 |
+
"learning_rate": 1.0222222222222223e-05,
|
1938 |
+
"loss": 0.0025,
|
1939 |
+
"step": 2700
|
1940 |
+
},
|
1941 |
+
{
|
1942 |
+
"epoch": 0.85,
|
1943 |
+
"grad_norm": 0.4488503932952881,
|
1944 |
+
"learning_rate": 1.0177777777777778e-05,
|
1945 |
+
"loss": 0.0395,
|
1946 |
+
"step": 2710
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 0.85,
|
1950 |
+
"grad_norm": 1.2777860164642334,
|
1951 |
+
"learning_rate": 1.0133333333333335e-05,
|
1952 |
+
"loss": 0.0293,
|
1953 |
+
"step": 2720
|
1954 |
+
},
|
1955 |
+
{
|
1956 |
+
"epoch": 0.86,
|
1957 |
+
"grad_norm": 0.08910083770751953,
|
1958 |
+
"learning_rate": 1.008888888888889e-05,
|
1959 |
+
"loss": 0.0319,
|
1960 |
+
"step": 2730
|
1961 |
+
},
|
1962 |
+
{
|
1963 |
+
"epoch": 0.86,
|
1964 |
+
"grad_norm": 0.0010414342395961285,
|
1965 |
+
"learning_rate": 1.0044444444444446e-05,
|
1966 |
+
"loss": 0.0082,
|
1967 |
+
"step": 2740
|
1968 |
+
},
|
1969 |
+
{
|
1970 |
+
"epoch": 0.86,
|
1971 |
+
"grad_norm": 0.0031285579316318035,
|
1972 |
+
"learning_rate": 1e-05,
|
1973 |
+
"loss": 0.0415,
|
1974 |
+
"step": 2750
|
1975 |
+
},
|
1976 |
+
{
|
1977 |
+
"epoch": 0.86,
|
1978 |
+
"grad_norm": 0.005843820981681347,
|
1979 |
+
"learning_rate": 9.955555555555556e-06,
|
1980 |
+
"loss": 0.0557,
|
1981 |
+
"step": 2760
|
1982 |
+
},
|
1983 |
+
{
|
1984 |
+
"epoch": 0.87,
|
1985 |
+
"grad_norm": 0.0012017801636829972,
|
1986 |
+
"learning_rate": 9.911111111111113e-06,
|
1987 |
+
"loss": 0.052,
|
1988 |
+
"step": 2770
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 0.87,
|
1992 |
+
"grad_norm": 0.001439902582205832,
|
1993 |
+
"learning_rate": 9.866666666666668e-06,
|
1994 |
+
"loss": 0.0853,
|
1995 |
+
"step": 2780
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 0.87,
|
1999 |
+
"grad_norm": 0.0035338387824594975,
|
2000 |
+
"learning_rate": 9.822222222222223e-06,
|
2001 |
+
"loss": 0.0227,
|
2002 |
+
"step": 2790
|
2003 |
+
},
|
2004 |
+
{
|
2005 |
+
"epoch": 0.88,
|
2006 |
+
"grad_norm": 0.0015976278809830546,
|
2007 |
+
"learning_rate": 9.777777777777779e-06,
|
2008 |
+
"loss": 0.0647,
|
2009 |
+
"step": 2800
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 0.88,
|
2013 |
+
"grad_norm": 0.0012226419057697058,
|
2014 |
+
"learning_rate": 9.733333333333334e-06,
|
2015 |
+
"loss": 0.1104,
|
2016 |
+
"step": 2810
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 0.88,
|
2020 |
+
"grad_norm": 0.5588458776473999,
|
2021 |
+
"learning_rate": 9.688888888888889e-06,
|
2022 |
+
"loss": 0.0732,
|
2023 |
+
"step": 2820
|
2024 |
+
},
|
2025 |
+
{
|
2026 |
+
"epoch": 0.89,
|
2027 |
+
"grad_norm": 0.45813611149787903,
|
2028 |
+
"learning_rate": 9.644444444444444e-06,
|
2029 |
+
"loss": 0.0565,
|
2030 |
+
"step": 2830
|
2031 |
+
},
|
2032 |
+
{
|
2033 |
+
"epoch": 0.89,
|
2034 |
+
"grad_norm": 0.5053278207778931,
|
2035 |
+
"learning_rate": 9.600000000000001e-06,
|
2036 |
+
"loss": 0.0676,
|
2037 |
+
"step": 2840
|
2038 |
+
},
|
2039 |
+
{
|
2040 |
+
"epoch": 0.89,
|
2041 |
+
"grad_norm": 0.002007542410865426,
|
2042 |
+
"learning_rate": 9.555555555555556e-06,
|
2043 |
+
"loss": 0.0256,
|
2044 |
+
"step": 2850
|
2045 |
+
},
|
2046 |
+
{
|
2047 |
+
"epoch": 0.9,
|
2048 |
+
"grad_norm": 0.0025233286432921886,
|
2049 |
+
"learning_rate": 9.511111111111112e-06,
|
2050 |
+
"loss": 0.057,
|
2051 |
+
"step": 2860
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 0.9,
|
2055 |
+
"grad_norm": 0.0035814358852803707,
|
2056 |
+
"learning_rate": 9.466666666666667e-06,
|
2057 |
+
"loss": 0.1282,
|
2058 |
+
"step": 2870
|
2059 |
+
},
|
2060 |
+
{
|
2061 |
+
"epoch": 0.9,
|
2062 |
+
"grad_norm": 0.0031847648788243532,
|
2063 |
+
"learning_rate": 9.422222222222222e-06,
|
2064 |
+
"loss": 0.0126,
|
2065 |
+
"step": 2880
|
2066 |
+
},
|
2067 |
+
{
|
2068 |
+
"epoch": 0.91,
|
2069 |
+
"grad_norm": 0.0034481678158044815,
|
2070 |
+
"learning_rate": 9.377777777777779e-06,
|
2071 |
+
"loss": 0.0627,
|
2072 |
+
"step": 2890
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 0.91,
|
2076 |
+
"grad_norm": 0.18580594658851624,
|
2077 |
+
"learning_rate": 9.333333333333334e-06,
|
2078 |
+
"loss": 0.035,
|
2079 |
+
"step": 2900
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 0.91,
|
2083 |
+
"grad_norm": 0.0014439361402764916,
|
2084 |
+
"learning_rate": 9.28888888888889e-06,
|
2085 |
+
"loss": 0.0766,
|
2086 |
+
"step": 2910
|
2087 |
+
},
|
2088 |
+
{
|
2089 |
+
"epoch": 0.91,
|
2090 |
+
"grad_norm": 0.001558059360831976,
|
2091 |
+
"learning_rate": 9.244444444444445e-06,
|
2092 |
+
"loss": 0.023,
|
2093 |
+
"step": 2920
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 0.92,
|
2097 |
+
"grad_norm": 0.0029459816869348288,
|
2098 |
+
"learning_rate": 9.200000000000002e-06,
|
2099 |
+
"loss": 0.041,
|
2100 |
+
"step": 2930
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 0.92,
|
2104 |
+
"grad_norm": 0.0017032440518960357,
|
2105 |
+
"learning_rate": 9.155555555555557e-06,
|
2106 |
+
"loss": 0.0191,
|
2107 |
+
"step": 2940
|
2108 |
+
},
|
2109 |
+
{
|
2110 |
+
"epoch": 0.92,
|
2111 |
+
"grad_norm": 0.002062348648905754,
|
2112 |
+
"learning_rate": 9.111111111111112e-06,
|
2113 |
+
"loss": 0.0263,
|
2114 |
+
"step": 2950
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 0.93,
|
2118 |
+
"grad_norm": 0.001492173527367413,
|
2119 |
+
"learning_rate": 9.066666666666667e-06,
|
2120 |
+
"loss": 0.0419,
|
2121 |
+
"step": 2960
|
2122 |
+
},
|
2123 |
+
{
|
2124 |
+
"epoch": 0.93,
|
2125 |
+
"grad_norm": 0.903357982635498,
|
2126 |
+
"learning_rate": 9.022222222222223e-06,
|
2127 |
+
"loss": 0.0557,
|
2128 |
+
"step": 2970
|
2129 |
+
},
|
2130 |
+
{
|
2131 |
+
"epoch": 0.93,
|
2132 |
+
"grad_norm": 0.0011335255112498999,
|
2133 |
+
"learning_rate": 8.977777777777778e-06,
|
2134 |
+
"loss": 0.0065,
|
2135 |
+
"step": 2980
|
2136 |
+
},
|
2137 |
+
{
|
2138 |
+
"epoch": 0.94,
|
2139 |
+
"grad_norm": 0.0009899291908368468,
|
2140 |
+
"learning_rate": 8.933333333333333e-06,
|
2141 |
+
"loss": 0.0339,
|
2142 |
+
"step": 2990
|
2143 |
+
},
|
2144 |
+
{
|
2145 |
+
"epoch": 0.94,
|
2146 |
+
"grad_norm": 0.0012059457367286086,
|
2147 |
+
"learning_rate": 8.888888888888888e-06,
|
2148 |
+
"loss": 0.0374,
|
2149 |
+
"step": 3000
|
2150 |
+
},
|
2151 |
+
{
|
2152 |
+
"epoch": 0.94,
|
2153 |
+
"eval_loss": 0.022419294342398643,
|
2154 |
+
"eval_runtime": 62.1056,
|
2155 |
+
"eval_samples_per_second": 16.102,
|
2156 |
+
"eval_steps_per_second": 16.102,
|
2157 |
+
"step": 3000
|
2158 |
}
|
2159 |
],
|
2160 |
"logging_steps": 10,
|
|
|
2162 |
"num_input_tokens_seen": 0,
|
2163 |
"num_train_epochs": 2,
|
2164 |
"save_steps": 1000,
|
2165 |
+
"total_flos": 4.8306377981952e+16,
|
2166 |
"train_batch_size": 1,
|
2167 |
"trial_name": null,
|
2168 |
"trial_params": null
|