Training in progress, step 298, 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 1154870440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:892b19e9b1c6ec8c931ec31eaaea64c1f54d229ad9301361d06c23010fe58615
|
| 3 |
size 1154870440
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2309999768
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a6cbc8b8810e660d48486c9cac79327a7e9d0bfe0336dab2b0b2be950dafd55
|
| 3 |
size 2309999768
|
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:ec20f60ea398d86da91aa0036eb29c67b6647d2b936f7f70a003e8aecb18b498
|
| 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 sha256:7299d6d674e2331f7fef6ad21b861c3a91e48af7b3ecf10af405659140b65185
|
| 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 +1680,419 @@
|
|
| 1680 |
"learning_rate": 2.0354603547267985e-05,
|
| 1681 |
"loss": 0.4991,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +2107,12 @@
|
|
| 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": 0.9991617770326907,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 298,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 2.0354603547267985e-05,
|
| 1681 |
"loss": 0.4991,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8046940486169321,
|
| 1686 |
+
"grad_norm": 15.609128952026367,
|
| 1687 |
+
"learning_rate": 1.9719515643116674e-05,
|
| 1688 |
+
"loss": 0.885,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8080469404861693,
|
| 1693 |
+
"grad_norm": 14.140607833862305,
|
| 1694 |
+
"learning_rate": 1.9092050688969738e-05,
|
| 1695 |
+
"loss": 0.6477,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8113998323554066,
|
| 1700 |
+
"grad_norm": 8.173004150390625,
|
| 1701 |
+
"learning_rate": 1.847236664577389e-05,
|
| 1702 |
+
"loss": 0.3195,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8147527242246437,
|
| 1707 |
+
"grad_norm": 12.263628959655762,
|
| 1708 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 1709 |
+
"loss": 0.3825,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.818105616093881,
|
| 1714 |
+
"grad_norm": 9.158538818359375,
|
| 1715 |
+
"learning_rate": 1.725696330273575e-05,
|
| 1716 |
+
"loss": 0.5394,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8214585079631181,
|
| 1721 |
+
"grad_norm": 11.503657341003418,
|
| 1722 |
+
"learning_rate": 1.6661549974185424e-05,
|
| 1723 |
+
"loss": 0.5017,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8248113998323554,
|
| 1728 |
+
"grad_norm": 11.45801830291748,
|
| 1729 |
+
"learning_rate": 1.60745294221434e-05,
|
| 1730 |
+
"loss": 0.5327,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8281642917015927,
|
| 1735 |
+
"grad_norm": 14.59035873413086,
|
| 1736 |
+
"learning_rate": 1.549604942589441e-05,
|
| 1737 |
+
"loss": 0.6128,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8315171835708298,
|
| 1742 |
+
"grad_norm": 11.337044715881348,
|
| 1743 |
+
"learning_rate": 1.4926255614683932e-05,
|
| 1744 |
+
"loss": 0.4564,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8348700754400671,
|
| 1749 |
+
"grad_norm": 9.969059944152832,
|
| 1750 |
+
"learning_rate": 1.4365291431056871e-05,
|
| 1751 |
+
"loss": 0.7156,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8382229673093042,
|
| 1756 |
+
"grad_norm": 10.448427200317383,
|
| 1757 |
+
"learning_rate": 1.3813298094746491e-05,
|
| 1758 |
+
"loss": 0.5685,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8415758591785415,
|
| 1763 |
+
"grad_norm": 11.464609146118164,
|
| 1764 |
+
"learning_rate": 1.327041456712334e-05,
|
| 1765 |
+
"loss": 0.6357,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8449287510477788,
|
| 1770 |
+
"grad_norm": 12.060856819152832,
|
| 1771 |
+
"learning_rate": 1.2736777516212266e-05,
|
| 1772 |
+
"loss": 0.7539,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8482816429170159,
|
| 1777 |
+
"grad_norm": 12.243000984191895,
|
| 1778 |
+
"learning_rate": 1.2212521282287092e-05,
|
| 1779 |
+
"loss": 0.7645,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8516345347862532,
|
| 1784 |
+
"grad_norm": 11.656434059143066,
|
| 1785 |
+
"learning_rate": 1.1697777844051105e-05,
|
| 1786 |
+
"loss": 0.708,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8549874266554903,
|
| 1791 |
+
"grad_norm": 12.392927169799805,
|
| 1792 |
+
"learning_rate": 1.1192676785412154e-05,
|
| 1793 |
+
"loss": 0.7513,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8583403185247276,
|
| 1798 |
+
"grad_norm": 9.757222175598145,
|
| 1799 |
+
"learning_rate": 1.0697345262860636e-05,
|
| 1800 |
+
"loss": 0.6079,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8616932103939648,
|
| 1805 |
+
"grad_norm": 8.159704208374023,
|
| 1806 |
+
"learning_rate": 1.021190797345839e-05,
|
| 1807 |
+
"loss": 0.2301,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.865046102263202,
|
| 1812 |
+
"grad_norm": 7.381707668304443,
|
| 1813 |
+
"learning_rate": 9.73648712344707e-06,
|
| 1814 |
+
"loss": 0.337,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8683989941324393,
|
| 1819 |
+
"grad_norm": 9.870096206665039,
|
| 1820 |
+
"learning_rate": 9.271202397483215e-06,
|
| 1821 |
+
"loss": 0.2997,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8717518860016764,
|
| 1826 |
+
"grad_norm": 8.603653907775879,
|
| 1827 |
+
"learning_rate": 8.816170928508365e-06,
|
| 1828 |
+
"loss": 0.2499,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8751047778709137,
|
| 1833 |
+
"grad_norm": 6.676860809326172,
|
| 1834 |
+
"learning_rate": 8.371507268261437e-06,
|
| 1835 |
+
"loss": 0.3058,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8784576697401508,
|
| 1840 |
+
"grad_norm": 8.418916702270508,
|
| 1841 |
+
"learning_rate": 7.937323358440935e-06,
|
| 1842 |
+
"loss": 0.2895,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8818105616093881,
|
| 1847 |
+
"grad_norm": 10.64456558227539,
|
| 1848 |
+
"learning_rate": 7.513728502524286e-06,
|
| 1849 |
+
"loss": 0.3863,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8851634534786254,
|
| 1854 |
+
"grad_norm": 7.481652736663818,
|
| 1855 |
+
"learning_rate": 7.100829338251147e-06,
|
| 1856 |
+
"loss": 0.2319,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8885163453478625,
|
| 1861 |
+
"grad_norm": 7.484499454498291,
|
| 1862 |
+
"learning_rate": 6.698729810778065e-06,
|
| 1863 |
+
"loss": 0.1861,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8918692372170998,
|
| 1868 |
+
"grad_norm": 6.579418659210205,
|
| 1869 |
+
"learning_rate": 6.3075311465107535e-06,
|
| 1870 |
+
"loss": 0.2557,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8952221290863369,
|
| 1875 |
+
"grad_norm": 6.663677215576172,
|
| 1876 |
+
"learning_rate": 5.927331827620903e-06,
|
| 1877 |
+
"loss": 0.3708,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8985750209555742,
|
| 1882 |
+
"grad_norm": 9.156440734863281,
|
| 1883 |
+
"learning_rate": 5.558227567253832e-06,
|
| 1884 |
+
"loss": 0.3865,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9019279128248114,
|
| 1889 |
+
"grad_norm": 5.36343240737915,
|
| 1890 |
+
"learning_rate": 5.200311285433213e-06,
|
| 1891 |
+
"loss": 0.2776,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9052808046940486,
|
| 1896 |
+
"grad_norm": 7.757360935211182,
|
| 1897 |
+
"learning_rate": 4.853673085668947e-06,
|
| 1898 |
+
"loss": 0.335,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9086336965632859,
|
| 1903 |
+
"grad_norm": 12.774531364440918,
|
| 1904 |
+
"learning_rate": 4.5184002322740785e-06,
|
| 1905 |
+
"loss": 0.3653,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.911986588432523,
|
| 1910 |
+
"grad_norm": 8.137080192565918,
|
| 1911 |
+
"learning_rate": 4.19457712839652e-06,
|
| 1912 |
+
"loss": 0.392,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9153394803017603,
|
| 1917 |
+
"grad_norm": 9.441322326660156,
|
| 1918 |
+
"learning_rate": 3.8822852947709375e-06,
|
| 1919 |
+
"loss": 0.3059,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9186923721709975,
|
| 1924 |
+
"grad_norm": 5.269786834716797,
|
| 1925 |
+
"learning_rate": 3.581603349196372e-06,
|
| 1926 |
+
"loss": 0.2275,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9220452640402347,
|
| 1931 |
+
"grad_norm": 9.277483940124512,
|
| 1932 |
+
"learning_rate": 3.2926069867446675e-06,
|
| 1933 |
+
"loss": 0.335,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9253981559094719,
|
| 1938 |
+
"grad_norm": 9.3627290725708,
|
| 1939 |
+
"learning_rate": 3.0153689607045845e-06,
|
| 1940 |
+
"loss": 0.4954,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9287510477787091,
|
| 1945 |
+
"grad_norm": 7.629518508911133,
|
| 1946 |
+
"learning_rate": 2.7499590642665774e-06,
|
| 1947 |
+
"loss": 0.2234,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9321039396479464,
|
| 1952 |
+
"grad_norm": 5.571048736572266,
|
| 1953 |
+
"learning_rate": 2.496444112952734e-06,
|
| 1954 |
+
"loss": 0.3485,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9354568315171836,
|
| 1959 |
+
"grad_norm": 18.15276336669922,
|
| 1960 |
+
"learning_rate": 2.2548879277963064e-06,
|
| 1961 |
+
"loss": 0.3434,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9388097233864208,
|
| 1966 |
+
"grad_norm": 14.836206436157227,
|
| 1967 |
+
"learning_rate": 2.0253513192751373e-06,
|
| 1968 |
+
"loss": 0.3263,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.942162615255658,
|
| 1973 |
+
"grad_norm": 13.923564910888672,
|
| 1974 |
+
"learning_rate": 1.807892072002898e-06,
|
| 1975 |
+
"loss": 0.4678,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9455155071248952,
|
| 1980 |
+
"grad_norm": 15.387225151062012,
|
| 1981 |
+
"learning_rate": 1.6025649301821876e-06,
|
| 1982 |
+
"loss": 0.4785,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9488683989941324,
|
| 1987 |
+
"grad_norm": 13.055351257324219,
|
| 1988 |
+
"learning_rate": 1.4094215838229176e-06,
|
| 1989 |
+
"loss": 0.563,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9522212908633697,
|
| 1994 |
+
"grad_norm": 17.376829147338867,
|
| 1995 |
+
"learning_rate": 1.2285106557296477e-06,
|
| 1996 |
+
"loss": 0.3891,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9555741827326069,
|
| 2001 |
+
"grad_norm": 17.277982711791992,
|
| 2002 |
+
"learning_rate": 1.0598776892610685e-06,
|
| 2003 |
+
"loss": 0.5198,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9589270746018441,
|
| 2008 |
+
"grad_norm": 11.45787239074707,
|
| 2009 |
+
"learning_rate": 9.035651368646648e-07,
|
| 2010 |
+
"loss": 0.4912,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9622799664710813,
|
| 2015 |
+
"grad_norm": 12.024084091186523,
|
| 2016 |
+
"learning_rate": 7.596123493895991e-07,
|
| 2017 |
+
"loss": 0.4365,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9656328583403185,
|
| 2022 |
+
"grad_norm": 15.519173622131348,
|
| 2023 |
+
"learning_rate": 6.280555661802856e-07,
|
| 2024 |
+
"loss": 0.6457,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9689857502095558,
|
| 2029 |
+
"grad_norm": 15.936901092529297,
|
| 2030 |
+
"learning_rate": 5.089279059533658e-07,
|
| 2031 |
+
"loss": 0.7402,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.972338642078793,
|
| 2036 |
+
"grad_norm": 14.855793952941895,
|
| 2037 |
+
"learning_rate": 4.02259358460233e-07,
|
| 2038 |
+
"loss": 0.5426,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9756915339480302,
|
| 2043 |
+
"grad_norm": 17.52886199951172,
|
| 2044 |
+
"learning_rate": 3.080767769372939e-07,
|
| 2045 |
+
"loss": 0.5786,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9790444258172674,
|
| 2050 |
+
"grad_norm": 14.614376068115234,
|
| 2051 |
+
"learning_rate": 2.2640387134577058e-07,
|
| 2052 |
+
"loss": 0.4134,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9823973176865046,
|
| 2057 |
+
"grad_norm": 12.858384132385254,
|
| 2058 |
+
"learning_rate": 1.5726120240288634e-07,
|
| 2059 |
+
"loss": 0.9686,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9857502095557418,
|
| 2064 |
+
"grad_norm": 14.443262100219727,
|
| 2065 |
+
"learning_rate": 1.0066617640578368e-07,
|
| 2066 |
+
"loss": 0.5455,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.989103101424979,
|
| 2071 |
+
"grad_norm": 14.196720123291016,
|
| 2072 |
+
"learning_rate": 5.663304084960186e-08,
|
| 2073 |
+
"loss": 0.4852,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9924559932942163,
|
| 2078 |
+
"grad_norm": 13.345819473266602,
|
| 2079 |
+
"learning_rate": 2.5172880840745873e-08,
|
| 2080 |
+
"loss": 0.4879,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9958088851634534,
|
| 2085 |
+
"grad_norm": 4.6811842918396,
|
| 2086 |
+
"learning_rate": 6.293616306246586e-09,
|
| 2087 |
+
"loss": 0.2391,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9991617770326907,
|
| 2092 |
+
"grad_norm": 7.340142250061035,
|
| 2093 |
+
"learning_rate": 0.0,
|
| 2094 |
+
"loss": 0.4579,
|
| 2095 |
+
"step": 298
|
| 2096 |
}
|
| 2097 |
],
|
| 2098 |
"logging_steps": 1,
|
|
|
|
| 2107 |
"should_evaluate": false,
|
| 2108 |
"should_log": false,
|
| 2109 |
"should_save": true,
|
| 2110 |
+
"should_training_stop": true
|
| 2111 |
},
|
| 2112 |
"attributes": {}
|
| 2113 |
}
|
| 2114 |
},
|
| 2115 |
+
"total_flos": 1.998363176038564e+17,
|
| 2116 |
"train_batch_size": 4,
|
| 2117 |
"trial_name": null,
|
| 2118 |
"trial_params": null
|