Training in progress, step 301, 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:c59e4672ef7cc36f9f5487ce55f2ccfa0a52617841525b4bb1ac622d5b4aa80f
|
| 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 sha256:00570a8c2ef93f07116ec0c50f9481945aab0c0c66ec1e76db067833c5ce46f6
|
| 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 sha256:5d35b3a9dd0a1f8f9486755fe2563ec30d796d72743863e4be5bb2ab47e364b5
|
| 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 sha256:207ebf8b22c8abb89ae63b0d6979b5f27d625d140e3311a16b44653eec5ce343
|
| 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":
|
| 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,
|
|
@@ -1603,6 +1603,525 @@
|
|
| 1603 |
"learning_rate": 1.427141579677374e-05,
|
| 1604 |
"loss": 0.5648,
|
| 1605 |
"step": 228
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1606 |
}
|
| 1607 |
],
|
| 1608 |
"logging_steps": 1,
|
|
@@ -1617,12 +2136,12 @@
|
|
| 1617 |
"should_evaluate": false,
|
| 1618 |
"should_log": false,
|
| 1619 |
"should_save": true,
|
| 1620 |
-
"should_training_stop":
|
| 1621 |
},
|
| 1622 |
"attributes": {}
|
| 1623 |
}
|
| 1624 |
},
|
| 1625 |
-
"total_flos":
|
| 1626 |
"train_batch_size": 4,
|
| 1627 |
"trial_name": null,
|
| 1628 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0016638935108153,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 301,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1603 |
"learning_rate": 1.427141579677374e-05,
|
| 1604 |
"loss": 0.5648,
|
| 1605 |
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.762063227953411,
|
| 1609 |
+
"grad_norm": 0.5607278943061829,
|
| 1610 |
+
"learning_rate": 1.3902195302273779e-05,
|
| 1611 |
+
"loss": 0.3263,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7653910149750416,
|
| 1616 |
+
"grad_norm": 0.5801076889038086,
|
| 1617 |
+
"learning_rate": 1.3537041048046695e-05,
|
| 1618 |
+
"loss": 0.4707,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7687188019966722,
|
| 1623 |
+
"grad_norm": 0.7465184926986694,
|
| 1624 |
+
"learning_rate": 1.3175994166924394e-05,
|
| 1625 |
+
"loss": 0.4633,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7720465890183028,
|
| 1630 |
+
"grad_norm": 0.893689751625061,
|
| 1631 |
+
"learning_rate": 1.2819095329063469e-05,
|
| 1632 |
+
"loss": 0.4621,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7753743760399334,
|
| 1637 |
+
"grad_norm": 0.6686207056045532,
|
| 1638 |
+
"learning_rate": 1.246638473736378e-05,
|
| 1639 |
+
"loss": 0.5619,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.778702163061564,
|
| 1644 |
+
"grad_norm": 0.8450759053230286,
|
| 1645 |
+
"learning_rate": 1.2117902122939861e-05,
|
| 1646 |
+
"loss": 0.578,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7820299500831946,
|
| 1651 |
+
"grad_norm": 0.7605635523796082,
|
| 1652 |
+
"learning_rate": 1.1773686740645384e-05,
|
| 1653 |
+
"loss": 0.6534,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7853577371048253,
|
| 1658 |
+
"grad_norm": 0.9870882630348206,
|
| 1659 |
+
"learning_rate": 1.1433777364651271e-05,
|
| 1660 |
+
"loss": 0.8853,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7886855241264559,
|
| 1665 |
+
"grad_norm": 0.5999244451522827,
|
| 1666 |
+
"learning_rate": 1.1098212284078036e-05,
|
| 1667 |
+
"loss": 0.3816,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7920133111480865,
|
| 1672 |
+
"grad_norm": 0.8197599053382874,
|
| 1673 |
+
"learning_rate": 1.076702929868264e-05,
|
| 1674 |
+
"loss": 0.8135,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7953410981697171,
|
| 1679 |
+
"grad_norm": 0.7237629294395447,
|
| 1680 |
+
"learning_rate": 1.0440265714600572e-05,
|
| 1681 |
+
"loss": 0.6637,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7986688851913477,
|
| 1686 |
+
"grad_norm": 0.8603866696357727,
|
| 1687 |
+
"learning_rate": 1.0117958340143507e-05,
|
| 1688 |
+
"loss": 0.8529,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8019966722129783,
|
| 1693 |
+
"grad_norm": 0.7199721336364746,
|
| 1694 |
+
"learning_rate": 9.800143481652979e-06,
|
| 1695 |
+
"loss": 0.8565,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8053244592346089,
|
| 1700 |
+
"grad_norm": 0.6758659482002258,
|
| 1701 |
+
"learning_rate": 9.48685693941067e-06,
|
| 1702 |
+
"loss": 0.794,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8086522462562395,
|
| 1707 |
+
"grad_norm": 0.7372802495956421,
|
| 1708 |
+
"learning_rate": 9.17813400360572e-06,
|
| 1709 |
+
"loss": 0.8753,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8119800332778702,
|
| 1714 |
+
"grad_norm": 0.6816331148147583,
|
| 1715 |
+
"learning_rate": 8.874009450359427e-06,
|
| 1716 |
+
"loss": 0.7685,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8153078202995009,
|
| 1721 |
+
"grad_norm": 0.6845401525497437,
|
| 1722 |
+
"learning_rate": 8.574517537807897e-06,
|
| 1723 |
+
"loss": 0.7908,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8186356073211315,
|
| 1728 |
+
"grad_norm": 0.7325373291969299,
|
| 1729 |
+
"learning_rate": 8.279692002243027e-06,
|
| 1730 |
+
"loss": 0.9694,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8219633943427621,
|
| 1735 |
+
"grad_norm": 0.7457829713821411,
|
| 1736 |
+
"learning_rate": 7.989566054312287e-06,
|
| 1737 |
+
"loss": 0.6347,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8252911813643927,
|
| 1742 |
+
"grad_norm": 0.7450778484344482,
|
| 1743 |
+
"learning_rate": 7.704172375277691e-06,
|
| 1744 |
+
"loss": 0.8447,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8286189683860233,
|
| 1749 |
+
"grad_norm": 0.7165175676345825,
|
| 1750 |
+
"learning_rate": 7.423543113334436e-06,
|
| 1751 |
+
"loss": 0.786,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.831946755407654,
|
| 1756 |
+
"grad_norm": 1.0344496965408325,
|
| 1757 |
+
"learning_rate": 7.14770987998954e-06,
|
| 1758 |
+
"loss": 0.8206,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8352745424292846,
|
| 1763 |
+
"grad_norm": 0.6772722005844116,
|
| 1764 |
+
"learning_rate": 6.876703746500984e-06,
|
| 1765 |
+
"loss": 0.4412,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8386023294509152,
|
| 1770 |
+
"grad_norm": 0.6309685111045837,
|
| 1771 |
+
"learning_rate": 6.610555240377652e-06,
|
| 1772 |
+
"loss": 0.4023,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8419301164725458,
|
| 1777 |
+
"grad_norm": 0.8043044805526733,
|
| 1778 |
+
"learning_rate": 6.349294341940593e-06,
|
| 1779 |
+
"loss": 0.4782,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8452579034941764,
|
| 1784 |
+
"grad_norm": 0.6318528056144714,
|
| 1785 |
+
"learning_rate": 6.092950480945897e-06,
|
| 1786 |
+
"loss": 0.4712,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.848585690515807,
|
| 1791 |
+
"grad_norm": 0.5549326539039612,
|
| 1792 |
+
"learning_rate": 5.841552533269534e-06,
|
| 1793 |
+
"loss": 0.3927,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8519134775374376,
|
| 1798 |
+
"grad_norm": 0.5588014721870422,
|
| 1799 |
+
"learning_rate": 5.595128817654638e-06,
|
| 1800 |
+
"loss": 0.3666,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8552412645590682,
|
| 1805 |
+
"grad_norm": 0.6045800447463989,
|
| 1806 |
+
"learning_rate": 5.353707092521582e-06,
|
| 1807 |
+
"loss": 0.5723,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8585690515806988,
|
| 1812 |
+
"grad_norm": 0.6416806578636169,
|
| 1813 |
+
"learning_rate": 5.117314552841052e-06,
|
| 1814 |
+
"loss": 0.4304,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8618968386023295,
|
| 1819 |
+
"grad_norm": 0.5945389866828918,
|
| 1820 |
+
"learning_rate": 4.885977827070748e-06,
|
| 1821 |
+
"loss": 0.343,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8652246256239601,
|
| 1826 |
+
"grad_norm": 0.6209045052528381,
|
| 1827 |
+
"learning_rate": 4.659722974155767e-06,
|
| 1828 |
+
"loss": 0.3872,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8685524126455907,
|
| 1833 |
+
"grad_norm": 0.627796471118927,
|
| 1834 |
+
"learning_rate": 4.43857548059321e-06,
|
| 1835 |
+
"loss": 0.3544,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8718801996672213,
|
| 1840 |
+
"grad_norm": 0.5913266539573669,
|
| 1841 |
+
"learning_rate": 4.2225602575612755e-06,
|
| 1842 |
+
"loss": 0.4773,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8752079866888519,
|
| 1847 |
+
"grad_norm": 0.6416683793067932,
|
| 1848 |
+
"learning_rate": 4.011701638113063e-06,
|
| 1849 |
+
"loss": 0.412,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8785357737104825,
|
| 1854 |
+
"grad_norm": 0.8090274930000305,
|
| 1855 |
+
"learning_rate": 3.8060233744356633e-06,
|
| 1856 |
+
"loss": 0.5336,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8818635607321131,
|
| 1861 |
+
"grad_norm": 0.5056114792823792,
|
| 1862 |
+
"learning_rate": 3.605548635174533e-06,
|
| 1863 |
+
"loss": 0.2786,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8851913477537438,
|
| 1868 |
+
"grad_norm": 0.566896915435791,
|
| 1869 |
+
"learning_rate": 3.410300002823691e-06,
|
| 1870 |
+
"loss": 0.3718,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8885191347753744,
|
| 1875 |
+
"grad_norm": 0.5743393898010254,
|
| 1876 |
+
"learning_rate": 3.220299471181898e-06,
|
| 1877 |
+
"loss": 0.4044,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.891846921797005,
|
| 1882 |
+
"grad_norm": 0.634122908115387,
|
| 1883 |
+
"learning_rate": 3.035568442875136e-06,
|
| 1884 |
+
"loss": 0.6042,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8951747088186356,
|
| 1889 |
+
"grad_norm": 0.39157751202583313,
|
| 1890 |
+
"learning_rate": 2.85612772694579e-06,
|
| 1891 |
+
"loss": 0.153,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8985024958402662,
|
| 1896 |
+
"grad_norm": 0.5736538171768188,
|
| 1897 |
+
"learning_rate": 2.6819975365085237e-06,
|
| 1898 |
+
"loss": 0.3097,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9018302828618968,
|
| 1903 |
+
"grad_norm": 0.5841122269630432,
|
| 1904 |
+
"learning_rate": 2.5131974864734066e-06,
|
| 1905 |
+
"loss": 0.4648,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9051580698835274,
|
| 1910 |
+
"grad_norm": 0.5904918313026428,
|
| 1911 |
+
"learning_rate": 2.349746591336405e-06,
|
| 1912 |
+
"loss": 0.5159,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.908485856905158,
|
| 1917 |
+
"grad_norm": 0.5497831702232361,
|
| 1918 |
+
"learning_rate": 2.191663263037458e-06,
|
| 1919 |
+
"loss": 0.3491,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9118136439267887,
|
| 1924 |
+
"grad_norm": 0.6325455904006958,
|
| 1925 |
+
"learning_rate": 2.0389653088865036e-06,
|
| 1926 |
+
"loss": 0.4999,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9151414309484193,
|
| 1931 |
+
"grad_norm": 0.5889260172843933,
|
| 1932 |
+
"learning_rate": 1.8916699295575324e-06,
|
| 1933 |
+
"loss": 0.4644,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9184692179700499,
|
| 1938 |
+
"grad_norm": 0.5664623379707336,
|
| 1939 |
+
"learning_rate": 1.7497937171510547e-06,
|
| 1940 |
+
"loss": 0.3335,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9217970049916805,
|
| 1945 |
+
"grad_norm": 0.49772658944129944,
|
| 1946 |
+
"learning_rate": 1.6133526533250565e-06,
|
| 1947 |
+
"loss": 0.3306,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9251247920133111,
|
| 1952 |
+
"grad_norm": 0.4992808699607849,
|
| 1953 |
+
"learning_rate": 1.4823621074947503e-06,
|
| 1954 |
+
"loss": 0.354,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9284525790349417,
|
| 1959 |
+
"grad_norm": 0.4994637966156006,
|
| 1960 |
+
"learning_rate": 1.3568368351012717e-06,
|
| 1961 |
+
"loss": 0.3448,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9317803660565723,
|
| 1966 |
+
"grad_norm": 0.48505714535713196,
|
| 1967 |
+
"learning_rate": 1.236790975949592e-06,
|
| 1968 |
+
"loss": 0.3501,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9351081530782029,
|
| 1973 |
+
"grad_norm": 1.016489028930664,
|
| 1974 |
+
"learning_rate": 1.1222380526156928e-06,
|
| 1975 |
+
"loss": 0.4423,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9384359400998337,
|
| 1980 |
+
"grad_norm": 0.7452896237373352,
|
| 1981 |
+
"learning_rate": 1.0131909689233442e-06,
|
| 1982 |
+
"loss": 0.6174,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9417637271214643,
|
| 1987 |
+
"grad_norm": 1.0553852319717407,
|
| 1988 |
+
"learning_rate": 9.096620084905472e-07,
|
| 1989 |
+
"loss": 0.7014,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9450915141430949,
|
| 1994 |
+
"grad_norm": 0.7587735652923584,
|
| 1995 |
+
"learning_rate": 8.11662833345822e-07,
|
| 1996 |
+
"loss": 0.6485,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9484193011647255,
|
| 2001 |
+
"grad_norm": 0.7494546175003052,
|
| 2002 |
+
"learning_rate": 7.192044826145771e-07,
|
| 2003 |
+
"loss": 0.7379,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9517470881863561,
|
| 2008 |
+
"grad_norm": 0.8453697562217712,
|
| 2009 |
+
"learning_rate": 6.322973712755697e-07,
|
| 2010 |
+
"loss": 0.8027,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9550748752079867,
|
| 2015 |
+
"grad_norm": 0.6284394860267639,
|
| 2016 |
+
"learning_rate": 5.509512889877333e-07,
|
| 2017 |
+
"loss": 0.5964,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9584026622296173,
|
| 2022 |
+
"grad_norm": 0.8004696369171143,
|
| 2023 |
+
"learning_rate": 4.7517539898741524e-07,
|
| 2024 |
+
"loss": 0.6745,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.961730449251248,
|
| 2029 |
+
"grad_norm": 0.7253815531730652,
|
| 2030 |
+
"learning_rate": 4.049782370561583e-07,
|
| 2031 |
+
"loss": 0.7672,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9650582362728786,
|
| 2036 |
+
"grad_norm": 0.7120758891105652,
|
| 2037 |
+
"learning_rate": 3.4036771055923066e-07,
|
| 2038 |
+
"loss": 0.6682,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9683860232945092,
|
| 2043 |
+
"grad_norm": 0.7899070978164673,
|
| 2044 |
+
"learning_rate": 2.813510975548772e-07,
|
| 2045 |
+
"loss": 0.7882,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9717138103161398,
|
| 2050 |
+
"grad_norm": 0.8241245150566101,
|
| 2051 |
+
"learning_rate": 2.2793504597447002e-07,
|
| 2052 |
+
"loss": 0.8099,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9750415973377704,
|
| 2057 |
+
"grad_norm": 0.7875584363937378,
|
| 2058 |
+
"learning_rate": 1.8012557287367392e-07,
|
| 2059 |
+
"loss": 1.0536,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.978369384359401,
|
| 2064 |
+
"grad_norm": 0.8683494329452515,
|
| 2065 |
+
"learning_rate": 1.379280637546443e-07,
|
| 2066 |
+
"loss": 0.8868,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9816971713810316,
|
| 2071 |
+
"grad_norm": 0.7692094445228577,
|
| 2072 |
+
"learning_rate": 1.0134727195937333e-07,
|
| 2073 |
+
"loss": 0.7747,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9850249584026622,
|
| 2078 |
+
"grad_norm": 0.7598888278007507,
|
| 2079 |
+
"learning_rate": 7.038731813426291e-08,
|
| 2080 |
+
"loss": 0.7562,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9883527454242929,
|
| 2085 |
+
"grad_norm": 0.759955108165741,
|
| 2086 |
+
"learning_rate": 4.5051689765929214e-08,
|
| 2087 |
+
"loss": 0.775,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9916805324459235,
|
| 2092 |
+
"grad_norm": 0.764261782169342,
|
| 2093 |
+
"learning_rate": 2.534324078837802e-08,
|
| 2094 |
+
"loss": 0.6648,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9950083194675541,
|
| 2099 |
+
"grad_norm": 0.8436802625656128,
|
| 2100 |
+
"learning_rate": 1.1264191261528557e-08,
|
| 2101 |
+
"loss": 0.848,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9983361064891847,
|
| 2106 |
+
"grad_norm": 0.9775305390357971,
|
| 2107 |
+
"learning_rate": 2.8161271211024633e-09,
|
| 2108 |
+
"loss": 0.7154,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9983361064891847,
|
| 2113 |
+
"eval_loss": 0.49492859840393066,
|
| 2114 |
+
"eval_runtime": 26.7842,
|
| 2115 |
+
"eval_samples_per_second": 9.446,
|
| 2116 |
+
"eval_steps_per_second": 2.389,
|
| 2117 |
+
"step": 300
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.0016638935108153,
|
| 2121 |
+
"grad_norm": 0.9796708226203918,
|
| 2122 |
+
"learning_rate": 0.0,
|
| 2123 |
+
"loss": 0.4979,
|
| 2124 |
+
"step": 301
|
| 2125 |
}
|
| 2126 |
],
|
| 2127 |
"logging_steps": 1,
|
|
|
|
| 2136 |
"should_evaluate": false,
|
| 2137 |
"should_log": false,
|
| 2138 |
"should_save": true,
|
| 2139 |
+
"should_training_stop": true
|
| 2140 |
},
|
| 2141 |
"attributes": {}
|
| 2142 |
}
|
| 2143 |
},
|
| 2144 |
+
"total_flos": 3.93702389745451e+17,
|
| 2145 |
"train_batch_size": 4,
|
| 2146 |
"trial_name": null,
|
| 2147 |
"trial_params": null
|