Training in progress, step 351, checkpoint
Browse files
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1720044440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:076af47d7078327fdcddbb6d8fc9040b2f10aea9193be908e14f3d5b5ca2d65c
|
| 3 |
size 1720044440
|
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:fe62285d37cb7ae2f76e17d237cec158f486f0d7fd385eb64625216b53c53bd6
|
| 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 sha256:6cfc2996c76931e741c35f64ea4489062ec8d7b0dac2fcacee62f28292fc1fdc
|
| 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": 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,
|
|
@@ -1680,6 +1680,790 @@
|
|
| 1680 |
"learning_rate": 4.159163858762254e-05,
|
| 1681 |
"loss": 0.0,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +2478,12 @@
|
|
| 1694 |
"should_evaluate": false,
|
| 1695 |
"should_log": false,
|
| 1696 |
"should_save": true,
|
| 1697 |
-
"should_training_stop":
|
| 1698 |
},
|
| 1699 |
"attributes": {}
|
| 1700 |
}
|
| 1701 |
},
|
| 1702 |
-
"total_flos": 1.
|
| 1703 |
"train_batch_size": 4,
|
| 1704 |
"trial_name": null,
|
| 1705 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9985775248933144,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 351,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 4.159163858762254e-05,
|
| 1681 |
"loss": 0.0,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6827880512091038,
|
| 1686 |
+
"grad_norm": NaN,
|
| 1687 |
+
"learning_rate": 4.097541059439698e-05,
|
| 1688 |
+
"loss": 0.0,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6856330014224751,
|
| 1693 |
+
"grad_norm": NaN,
|
| 1694 |
+
"learning_rate": 4.036059635594578e-05,
|
| 1695 |
+
"loss": 0.0,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6884779516358464,
|
| 1700 |
+
"grad_norm": NaN,
|
| 1701 |
+
"learning_rate": 3.9747292186519456e-05,
|
| 1702 |
+
"loss": 0.0,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6913229018492176,
|
| 1707 |
+
"grad_norm": NaN,
|
| 1708 |
+
"learning_rate": 3.913559416380743e-05,
|
| 1709 |
+
"loss": 0.0,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6941678520625889,
|
| 1714 |
+
"grad_norm": NaN,
|
| 1715 |
+
"learning_rate": 3.8525598113886755e-05,
|
| 1716 |
+
"loss": 0.0,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6970128022759602,
|
| 1721 |
+
"grad_norm": NaN,
|
| 1722 |
+
"learning_rate": 3.791739959621054e-05,
|
| 1723 |
+
"loss": 0.0,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6998577524893315,
|
| 1728 |
+
"grad_norm": NaN,
|
| 1729 |
+
"learning_rate": 3.73110938886379e-05,
|
| 1730 |
+
"loss": 0.0,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7027027027027027,
|
| 1735 |
+
"grad_norm": NaN,
|
| 1736 |
+
"learning_rate": 3.670677597250819e-05,
|
| 1737 |
+
"loss": 0.0,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.705547652916074,
|
| 1742 |
+
"grad_norm": NaN,
|
| 1743 |
+
"learning_rate": 3.610454051776159e-05,
|
| 1744 |
+
"loss": 0.0,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7083926031294452,
|
| 1749 |
+
"grad_norm": NaN,
|
| 1750 |
+
"learning_rate": 3.5504481868108496e-05,
|
| 1751 |
+
"loss": 0.0,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7112375533428165,
|
| 1756 |
+
"grad_norm": NaN,
|
| 1757 |
+
"learning_rate": 3.490669402625007e-05,
|
| 1758 |
+
"loss": 0.0,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7140825035561877,
|
| 1763 |
+
"grad_norm": NaN,
|
| 1764 |
+
"learning_rate": 3.4311270639152125e-05,
|
| 1765 |
+
"loss": 0.0,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.716927453769559,
|
| 1770 |
+
"grad_norm": NaN,
|
| 1771 |
+
"learning_rate": 3.371830498337475e-05,
|
| 1772 |
+
"loss": 0.0,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7197724039829303,
|
| 1777 |
+
"grad_norm": NaN,
|
| 1778 |
+
"learning_rate": 3.31278899504601e-05,
|
| 1779 |
+
"loss": 0.0,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7226173541963016,
|
| 1784 |
+
"grad_norm": NaN,
|
| 1785 |
+
"learning_rate": 3.254011803238026e-05,
|
| 1786 |
+
"loss": 0.0,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7254623044096729,
|
| 1791 |
+
"grad_norm": NaN,
|
| 1792 |
+
"learning_rate": 3.195508130704795e-05,
|
| 1793 |
+
"loss": 0.0,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7283072546230441,
|
| 1798 |
+
"grad_norm": NaN,
|
| 1799 |
+
"learning_rate": 3.137287142389189e-05,
|
| 1800 |
+
"loss": 0.0,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7311522048364154,
|
| 1805 |
+
"grad_norm": NaN,
|
| 1806 |
+
"learning_rate": 3.079357958949946e-05,
|
| 1807 |
+
"loss": 0.0,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7339971550497866,
|
| 1812 |
+
"grad_norm": NaN,
|
| 1813 |
+
"learning_rate": 3.0217296553328578e-05,
|
| 1814 |
+
"loss": 0.0,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7368421052631579,
|
| 1819 |
+
"grad_norm": NaN,
|
| 1820 |
+
"learning_rate": 2.9644112593491313e-05,
|
| 1821 |
+
"loss": 0.0,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7396870554765291,
|
| 1826 |
+
"grad_norm": NaN,
|
| 1827 |
+
"learning_rate": 2.90741175026113e-05,
|
| 1828 |
+
"loss": 0.0,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7425320056899004,
|
| 1833 |
+
"grad_norm": NaN,
|
| 1834 |
+
"learning_rate": 2.8507400573757158e-05,
|
| 1835 |
+
"loss": 0.0,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7453769559032717,
|
| 1840 |
+
"grad_norm": NaN,
|
| 1841 |
+
"learning_rate": 2.7944050586454214e-05,
|
| 1842 |
+
"loss": 0.0,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.748221906116643,
|
| 1847 |
+
"grad_norm": NaN,
|
| 1848 |
+
"learning_rate": 2.738415579277672e-05,
|
| 1849 |
+
"loss": 0.0,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7510668563300142,
|
| 1854 |
+
"grad_norm": NaN,
|
| 1855 |
+
"learning_rate": 2.682780390352262e-05,
|
| 1856 |
+
"loss": 0.0,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7539118065433855,
|
| 1861 |
+
"grad_norm": NaN,
|
| 1862 |
+
"learning_rate": 2.6275082074473077e-05,
|
| 1863 |
+
"loss": 0.0,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7567567567567568,
|
| 1868 |
+
"grad_norm": NaN,
|
| 1869 |
+
"learning_rate": 2.5726076892739125e-05,
|
| 1870 |
+
"loss": 0.0,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7596017069701281,
|
| 1875 |
+
"grad_norm": NaN,
|
| 1876 |
+
"learning_rate": 2.5180874363197215e-05,
|
| 1877 |
+
"loss": 0.0,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7624466571834992,
|
| 1882 |
+
"grad_norm": NaN,
|
| 1883 |
+
"learning_rate": 2.4639559895016068e-05,
|
| 1884 |
+
"loss": 0.0,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7652916073968705,
|
| 1889 |
+
"grad_norm": NaN,
|
| 1890 |
+
"learning_rate": 2.41022182882768e-05,
|
| 1891 |
+
"loss": 0.0,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7681365576102418,
|
| 1896 |
+
"grad_norm": NaN,
|
| 1897 |
+
"learning_rate": 2.3568933720688545e-05,
|
| 1898 |
+
"loss": 0.0,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7709815078236131,
|
| 1903 |
+
"grad_norm": NaN,
|
| 1904 |
+
"learning_rate": 2.3039789734401522e-05,
|
| 1905 |
+
"loss": 0.0,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7738264580369844,
|
| 1910 |
+
"grad_norm": NaN,
|
| 1911 |
+
"learning_rate": 2.2514869222919572e-05,
|
| 1912 |
+
"loss": 0.0,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7766714082503556,
|
| 1917 |
+
"grad_norm": NaN,
|
| 1918 |
+
"learning_rate": 2.1994254418114522e-05,
|
| 1919 |
+
"loss": 0.0,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7795163584637269,
|
| 1924 |
+
"grad_norm": NaN,
|
| 1925 |
+
"learning_rate": 2.1478026877344087e-05,
|
| 1926 |
+
"loss": 0.0,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7823613086770982,
|
| 1931 |
+
"grad_norm": NaN,
|
| 1932 |
+
"learning_rate": 2.0966267470675273e-05,
|
| 1933 |
+
"loss": 0.0,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7852062588904695,
|
| 1938 |
+
"grad_norm": NaN,
|
| 1939 |
+
"learning_rate": 2.0459056368215785e-05,
|
| 1940 |
+
"loss": 0.0,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7880512091038406,
|
| 1945 |
+
"grad_norm": NaN,
|
| 1946 |
+
"learning_rate": 1.9956473027554846e-05,
|
| 1947 |
+
"loss": 0.0,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7908961593172119,
|
| 1952 |
+
"grad_norm": NaN,
|
| 1953 |
+
"learning_rate": 1.945859618131564e-05,
|
| 1954 |
+
"loss": 0.0,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7937411095305832,
|
| 1959 |
+
"grad_norm": NaN,
|
| 1960 |
+
"learning_rate": 1.8965503824821495e-05,
|
| 1961 |
+
"loss": 0.0,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7965860597439545,
|
| 1966 |
+
"grad_norm": NaN,
|
| 1967 |
+
"learning_rate": 1.8477273203877398e-05,
|
| 1968 |
+
"loss": 0.0,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7994310099573257,
|
| 1973 |
+
"grad_norm": NaN,
|
| 1974 |
+
"learning_rate": 1.7993980802668946e-05,
|
| 1975 |
+
"loss": 0.0,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.802275960170697,
|
| 1980 |
+
"grad_norm": NaN,
|
| 1981 |
+
"learning_rate": 1.7515702331780753e-05,
|
| 1982 |
+
"loss": 0.0,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8051209103840683,
|
| 1987 |
+
"grad_norm": NaN,
|
| 1988 |
+
"learning_rate": 1.7042512716335873e-05,
|
| 1989 |
+
"loss": 0.0,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8079658605974396,
|
| 1994 |
+
"grad_norm": NaN,
|
| 1995 |
+
"learning_rate": 1.6574486084258366e-05,
|
| 1996 |
+
"loss": 0.0,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8108108108108109,
|
| 2001 |
+
"grad_norm": NaN,
|
| 2002 |
+
"learning_rate": 1.6111695754660667e-05,
|
| 2003 |
+
"loss": 0.0,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.813655761024182,
|
| 2008 |
+
"grad_norm": NaN,
|
| 2009 |
+
"learning_rate": 1.565421422635782e-05,
|
| 2010 |
+
"loss": 0.0,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8165007112375533,
|
| 2015 |
+
"grad_norm": NaN,
|
| 2016 |
+
"learning_rate": 1.5202113166510057e-05,
|
| 2017 |
+
"loss": 0.0,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8193456614509246,
|
| 2022 |
+
"grad_norm": NaN,
|
| 2023 |
+
"learning_rate": 1.475546339939568e-05,
|
| 2024 |
+
"loss": 0.0,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8221906116642959,
|
| 2029 |
+
"grad_norm": NaN,
|
| 2030 |
+
"learning_rate": 1.4314334895316094e-05,
|
| 2031 |
+
"loss": 0.0,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8250355618776671,
|
| 2036 |
+
"grad_norm": NaN,
|
| 2037 |
+
"learning_rate": 1.3878796759634544e-05,
|
| 2038 |
+
"loss": 0.0,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8278805120910384,
|
| 2043 |
+
"grad_norm": NaN,
|
| 2044 |
+
"learning_rate": 1.3448917221950263e-05,
|
| 2045 |
+
"loss": 0.0,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8307254623044097,
|
| 2050 |
+
"grad_norm": NaN,
|
| 2051 |
+
"learning_rate": 1.3024763625410024e-05,
|
| 2052 |
+
"loss": 0.0,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.833570412517781,
|
| 2057 |
+
"grad_norm": NaN,
|
| 2058 |
+
"learning_rate": 1.2606402416158392e-05,
|
| 2059 |
+
"loss": 0.0,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8364153627311522,
|
| 2064 |
+
"grad_norm": NaN,
|
| 2065 |
+
"learning_rate": 1.2193899132928537e-05,
|
| 2066 |
+
"loss": 0.0,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8392603129445235,
|
| 2071 |
+
"grad_norm": NaN,
|
| 2072 |
+
"learning_rate": 1.1787318396775188e-05,
|
| 2073 |
+
"loss": 0.0,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8421052631578947,
|
| 2078 |
+
"grad_norm": NaN,
|
| 2079 |
+
"learning_rate": 1.138672390095143e-05,
|
| 2080 |
+
"loss": 0.0,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.844950213371266,
|
| 2085 |
+
"grad_norm": NaN,
|
| 2086 |
+
"learning_rate": 1.0992178400930753e-05,
|
| 2087 |
+
"loss": 0.0,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8477951635846372,
|
| 2092 |
+
"grad_norm": NaN,
|
| 2093 |
+
"learning_rate": 1.060374370457599e-05,
|
| 2094 |
+
"loss": 0.0,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8506401137980085,
|
| 2099 |
+
"grad_norm": NaN,
|
| 2100 |
+
"learning_rate": 1.0221480662456845e-05,
|
| 2101 |
+
"loss": 0.0,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8534850640113798,
|
| 2106 |
+
"grad_norm": NaN,
|
| 2107 |
+
"learning_rate": 9.845449158317215e-06,
|
| 2108 |
+
"loss": 0.0,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8563300142247511,
|
| 2113 |
+
"grad_norm": NaN,
|
| 2114 |
+
"learning_rate": 9.475708099694124e-06,
|
| 2115 |
+
"loss": 0.0,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8591749644381224,
|
| 2120 |
+
"grad_norm": NaN,
|
| 2121 |
+
"learning_rate": 9.112315408689414e-06,
|
| 2122 |
+
"loss": 0.0,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8620199146514936,
|
| 2127 |
+
"grad_norm": NaN,
|
| 2128 |
+
"learning_rate": 8.755328012896003e-06,
|
| 2129 |
+
"loss": 0.0,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8648648648648649,
|
| 2134 |
+
"grad_norm": NaN,
|
| 2135 |
+
"learning_rate": 8.404801836479808e-06,
|
| 2136 |
+
"loss": 0.0,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8677098150782361,
|
| 2141 |
+
"grad_norm": NaN,
|
| 2142 |
+
"learning_rate": 8.060791791418886e-06,
|
| 2143 |
+
"loss": 0.0,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8705547652916074,
|
| 2148 |
+
"grad_norm": NaN,
|
| 2149 |
+
"learning_rate": 7.723351768901172e-06,
|
| 2150 |
+
"loss": 0.0,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8733997155049786,
|
| 2155 |
+
"grad_norm": NaN,
|
| 2156 |
+
"learning_rate": 7.392534630882092e-06,
|
| 2157 |
+
"loss": 0.0,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8762446657183499,
|
| 2162 |
+
"grad_norm": NaN,
|
| 2163 |
+
"learning_rate": 7.06839220180342e-06,
|
| 2164 |
+
"loss": 0.0,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8790896159317212,
|
| 2169 |
+
"grad_norm": NaN,
|
| 2170 |
+
"learning_rate": 6.750975260474718e-06,
|
| 2171 |
+
"loss": 0.0,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8819345661450925,
|
| 2176 |
+
"grad_norm": NaN,
|
| 2177 |
+
"learning_rate": 6.440333532118503e-06,
|
| 2178 |
+
"loss": 0.0,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8847795163584637,
|
| 2183 |
+
"grad_norm": NaN,
|
| 2184 |
+
"learning_rate": 6.136515680580479e-06,
|
| 2185 |
+
"loss": 0.0,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.887624466571835,
|
| 2190 |
+
"grad_norm": NaN,
|
| 2191 |
+
"learning_rate": 5.839569300706127e-06,
|
| 2192 |
+
"loss": 0.0,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8904694167852063,
|
| 2197 |
+
"grad_norm": NaN,
|
| 2198 |
+
"learning_rate": 5.549540910884648e-06,
|
| 2199 |
+
"loss": 0.0,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8933143669985776,
|
| 2204 |
+
"grad_norm": NaN,
|
| 2205 |
+
"learning_rate": 5.266475945761562e-06,
|
| 2206 |
+
"loss": 0.0,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8961593172119487,
|
| 2211 |
+
"grad_norm": NaN,
|
| 2212 |
+
"learning_rate": 4.990418749121178e-06,
|
| 2213 |
+
"loss": 0.0,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.89900426742532,
|
| 2218 |
+
"grad_norm": NaN,
|
| 2219 |
+
"learning_rate": 4.721412566939804e-06,
|
| 2220 |
+
"loss": 0.0,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9018492176386913,
|
| 2225 |
+
"grad_norm": NaN,
|
| 2226 |
+
"learning_rate": 4.459499540611078e-06,
|
| 2227 |
+
"loss": 0.0,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9046941678520626,
|
| 2232 |
+
"grad_norm": NaN,
|
| 2233 |
+
"learning_rate": 4.2047207003442e-06,
|
| 2234 |
+
"loss": 0.0,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9075391180654339,
|
| 2239 |
+
"grad_norm": NaN,
|
| 2240 |
+
"learning_rate": 3.9571159587363734e-06,
|
| 2241 |
+
"loss": 0.0,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9103840682788051,
|
| 2246 |
+
"grad_norm": NaN,
|
| 2247 |
+
"learning_rate": 3.7167241045202473e-06,
|
| 2248 |
+
"loss": 0.0,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9132290184921764,
|
| 2253 |
+
"grad_norm": NaN,
|
| 2254 |
+
"learning_rate": 3.4835827964873945e-06,
|
| 2255 |
+
"loss": 0.0,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9160739687055477,
|
| 2260 |
+
"grad_norm": NaN,
|
| 2261 |
+
"learning_rate": 3.2577285575889017e-06,
|
| 2262 |
+
"loss": 0.0,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.918918918918919,
|
| 2267 |
+
"grad_norm": NaN,
|
| 2268 |
+
"learning_rate": 3.039196769213787e-06,
|
| 2269 |
+
"loss": 0.0,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9217638691322901,
|
| 2274 |
+
"grad_norm": NaN,
|
| 2275 |
+
"learning_rate": 2.8280216656463408e-06,
|
| 2276 |
+
"loss": 0.0,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9246088193456614,
|
| 2281 |
+
"grad_norm": NaN,
|
| 2282 |
+
"learning_rate": 2.6242363287030613e-06,
|
| 2283 |
+
"loss": 0.0,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9274537695590327,
|
| 2288 |
+
"grad_norm": NaN,
|
| 2289 |
+
"learning_rate": 2.4278726825502697e-06,
|
| 2290 |
+
"loss": 0.0,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.930298719772404,
|
| 2295 |
+
"grad_norm": NaN,
|
| 2296 |
+
"learning_rate": 2.2389614887029566e-06,
|
| 2297 |
+
"loss": 0.0,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9331436699857752,
|
| 2302 |
+
"grad_norm": NaN,
|
| 2303 |
+
"learning_rate": 2.0575323412058033e-06,
|
| 2304 |
+
"loss": 0.0,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9359886201991465,
|
| 2309 |
+
"grad_norm": NaN,
|
| 2310 |
+
"learning_rate": 1.8836136619971466e-06,
|
| 2311 |
+
"loss": 0.0,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9388335704125178,
|
| 2316 |
+
"grad_norm": NaN,
|
| 2317 |
+
"learning_rate": 1.7172326964564778e-06,
|
| 2318 |
+
"loss": 0.0,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9416785206258891,
|
| 2323 |
+
"grad_norm": NaN,
|
| 2324 |
+
"learning_rate": 1.5584155091362906e-06,
|
| 2325 |
+
"loss": 0.0,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9445234708392604,
|
| 2330 |
+
"grad_norm": NaN,
|
| 2331 |
+
"learning_rate": 1.4071869796789428e-06,
|
| 2332 |
+
"loss": 0.0,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9473684210526315,
|
| 2337 |
+
"grad_norm": NaN,
|
| 2338 |
+
"learning_rate": 1.263570798919106e-06,
|
| 2339 |
+
"loss": 0.0,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.9502133712660028,
|
| 2344 |
+
"grad_norm": NaN,
|
| 2345 |
+
"learning_rate": 1.1275894651724517e-06,
|
| 2346 |
+
"loss": 0.0,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.9530583214793741,
|
| 2351 |
+
"grad_norm": NaN,
|
| 2352 |
+
"learning_rate": 9.992642807111485e-07,
|
| 2353 |
+
"loss": 0.0,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9559032716927454,
|
| 2358 |
+
"grad_norm": NaN,
|
| 2359 |
+
"learning_rate": 8.786153484267589e-07,
|
| 2360 |
+
"loss": 0.0,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9587482219061166,
|
| 2365 |
+
"grad_norm": NaN,
|
| 2366 |
+
"learning_rate": 7.656615686809976e-07,
|
| 2367 |
+
"loss": 0.0,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9615931721194879,
|
| 2372 |
+
"grad_norm": NaN,
|
| 2373 |
+
"learning_rate": 6.604206363448661e-07,
|
| 2374 |
+
"loss": 0.0,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9644381223328592,
|
| 2379 |
+
"grad_norm": NaN,
|
| 2380 |
+
"learning_rate": 5.629090380266544e-07,
|
| 2381 |
+
"loss": 0.0,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9672830725462305,
|
| 2386 |
+
"grad_norm": NaN,
|
| 2387 |
+
"learning_rate": 4.7314204948923356e-07,
|
| 2388 |
+
"loss": 0.0,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9701280227596017,
|
| 2393 |
+
"grad_norm": NaN,
|
| 2394 |
+
"learning_rate": 3.9113373325698754e-07,
|
| 2395 |
+
"loss": 0.0,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.972972972972973,
|
| 2400 |
+
"grad_norm": NaN,
|
| 2401 |
+
"learning_rate": 3.168969364128527e-07,
|
| 2402 |
+
"loss": 0.0,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9758179231863442,
|
| 2407 |
+
"grad_norm": NaN,
|
| 2408 |
+
"learning_rate": 2.5044328858576106e-07,
|
| 2409 |
+
"loss": 0.0,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9786628733997155,
|
| 2414 |
+
"grad_norm": NaN,
|
| 2415 |
+
"learning_rate": 1.917832001287645e-07,
|
| 2416 |
+
"loss": 0.0,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9815078236130867,
|
| 2421 |
+
"grad_norm": NaN,
|
| 2422 |
+
"learning_rate": 1.4092586048820578e-07,
|
| 2423 |
+
"loss": 0.0,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.984352773826458,
|
| 2428 |
+
"grad_norm": NaN,
|
| 2429 |
+
"learning_rate": 9.787923676414235e-08,
|
| 2430 |
+
"loss": 0.0,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9871977240398293,
|
| 2435 |
+
"grad_norm": NaN,
|
| 2436 |
+
"learning_rate": 6.265007246223364e-08,
|
| 2437 |
+
"loss": 0.0,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9900426742532006,
|
| 2442 |
+
"grad_norm": NaN,
|
| 2443 |
+
"learning_rate": 3.524388643736387e-08,
|
| 2444 |
+
"loss": 0.0,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9928876244665719,
|
| 2449 |
+
"grad_norm": NaN,
|
| 2450 |
+
"learning_rate": 1.566497202904471e-08,
|
| 2451 |
+
"loss": 0.0,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9957325746799431,
|
| 2456 |
+
"grad_norm": NaN,
|
| 2457 |
+
"learning_rate": 3.91639638886998e-09,
|
| 2458 |
+
"loss": 0.0,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9985775248933144,
|
| 2463 |
+
"grad_norm": NaN,
|
| 2464 |
+
"learning_rate": 0.0,
|
| 2465 |
+
"loss": 0.0,
|
| 2466 |
+
"step": 351
|
| 2467 |
}
|
| 2468 |
],
|
| 2469 |
"logging_steps": 1,
|
|
|
|
| 2478 |
"should_evaluate": false,
|
| 2479 |
"should_log": false,
|
| 2480 |
"should_save": true,
|
| 2481 |
+
"should_training_stop": true
|
| 2482 |
},
|
| 2483 |
"attributes": {}
|
| 2484 |
}
|
| 2485 |
},
|
| 2486 |
+
"total_flos": 1.583402480220242e+17,
|
| 2487 |
"train_batch_size": 4,
|
| 2488 |
"trial_name": null,
|
| 2489 |
"trial_params": null
|