Training in progress, step 14000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1153648
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9216b19dc0f8593bf284a70d61788e7406f09da8e0b4c454ee0889333090eddc
|
| 3 |
size 1153648
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2334411
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40fbaf79a92b6172a58fee88bd2b464b4f73ef68dffc512e6ba22ee739613ad9
|
| 3 |
size 2334411
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbc5308bd88acd797c0a1037ac3340a3b807d807264c2e87c429d54dbd5e8a06
|
| 3 |
size 14645
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cddad71864d5787392d991db759adcbe200d31fc395cc4d0062373a7d3ae259c
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb99a0255375d4bc6605503ddd9dde21cdf2d72720db893e888ff4c5c03f6639
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -1688,6 +1688,286 @@
|
|
| 1688 |
"learning_rate": 0.000858258608905443,
|
| 1689 |
"loss": 5.078074951171875,
|
| 1690 |
"step": 12000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1691 |
}
|
| 1692 |
],
|
| 1693 |
"logging_steps": 50,
|
|
@@ -1707,7 +1987,7 @@
|
|
| 1707 |
"attributes": {}
|
| 1708 |
}
|
| 1709 |
},
|
| 1710 |
-
"total_flos":
|
| 1711 |
"train_batch_size": 16,
|
| 1712 |
"trial_name": null,
|
| 1713 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.2867148620696717,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 14000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1688 |
"learning_rate": 0.000858258608905443,
|
| 1689 |
"loss": 5.078074951171875,
|
| 1690 |
"step": 12000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.24677957770996745,
|
| 1694 |
+
"grad_norm": 0.5470142364501953,
|
| 1695 |
+
"learning_rate": 0.0008571347399925167,
|
| 1696 |
+
"loss": 5.07035888671875,
|
| 1697 |
+
"step": 12050
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.24780355936021625,
|
| 1701 |
+
"grad_norm": 0.8870255351066589,
|
| 1702 |
+
"learning_rate": 0.0008560071752169917,
|
| 1703 |
+
"loss": 5.051436462402344,
|
| 1704 |
+
"step": 12100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.2488275410104651,
|
| 1708 |
+
"grad_norm": 0.6577278971672058,
|
| 1709 |
+
"learning_rate": 0.0008548759262476418,
|
| 1710 |
+
"loss": 5.07404541015625,
|
| 1711 |
+
"step": 12150
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.24985152266071392,
|
| 1715 |
+
"grad_norm": 0.6292082071304321,
|
| 1716 |
+
"learning_rate": 0.0008537410047913672,
|
| 1717 |
+
"loss": 5.095699768066407,
|
| 1718 |
+
"step": 12200
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.2508755043109627,
|
| 1722 |
+
"grad_norm": 0.6083301305770874,
|
| 1723 |
+
"learning_rate": 0.0008526024225930733,
|
| 1724 |
+
"loss": 5.072946472167969,
|
| 1725 |
+
"step": 12250
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.2518994859612116,
|
| 1729 |
+
"grad_norm": 0.6643879413604736,
|
| 1730 |
+
"learning_rate": 0.0008514601914355496,
|
| 1731 |
+
"loss": 5.0981512451171875,
|
| 1732 |
+
"step": 12300
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.2529234676114604,
|
| 1736 |
+
"grad_norm": 0.7181370854377747,
|
| 1737 |
+
"learning_rate": 0.0008503143231393466,
|
| 1738 |
+
"loss": 5.097729187011719,
|
| 1739 |
+
"step": 12350
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.25394744926170926,
|
| 1743 |
+
"grad_norm": 0.6699804663658142,
|
| 1744 |
+
"learning_rate": 0.000849164829562655,
|
| 1745 |
+
"loss": 5.069935302734375,
|
| 1746 |
+
"step": 12400
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.25497143091195806,
|
| 1750 |
+
"grad_norm": 0.8463541269302368,
|
| 1751 |
+
"learning_rate": 0.0008480117226011824,
|
| 1752 |
+
"loss": 5.076376953125,
|
| 1753 |
+
"step": 12450
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.25599541256220687,
|
| 1757 |
+
"grad_norm": 0.6894559860229492,
|
| 1758 |
+
"learning_rate": 0.0008468550141880292,
|
| 1759 |
+
"loss": 5.071536560058593,
|
| 1760 |
+
"step": 12500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.25701939421245573,
|
| 1764 |
+
"grad_norm": 0.8314017653465271,
|
| 1765 |
+
"learning_rate": 0.0008456947162935669,
|
| 1766 |
+
"loss": 5.073755798339843,
|
| 1767 |
+
"step": 12550
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.25804337586270454,
|
| 1771 |
+
"grad_norm": 0.6796386241912842,
|
| 1772 |
+
"learning_rate": 0.0008445308409253126,
|
| 1773 |
+
"loss": 5.091232604980469,
|
| 1774 |
+
"step": 12600
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.25906735751295334,
|
| 1778 |
+
"grad_norm": 0.8205164670944214,
|
| 1779 |
+
"learning_rate": 0.0008433634001278056,
|
| 1780 |
+
"loss": 5.0807037353515625,
|
| 1781 |
+
"step": 12650
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.2600913391632022,
|
| 1785 |
+
"grad_norm": 0.6622213125228882,
|
| 1786 |
+
"learning_rate": 0.0008421924059824832,
|
| 1787 |
+
"loss": 5.079286193847656,
|
| 1788 |
+
"step": 12700
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.261115320813451,
|
| 1792 |
+
"grad_norm": 0.752617359161377,
|
| 1793 |
+
"learning_rate": 0.0008410178706075538,
|
| 1794 |
+
"loss": 5.060631103515625,
|
| 1795 |
+
"step": 12750
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.2621393024636999,
|
| 1799 |
+
"grad_norm": 0.5829367637634277,
|
| 1800 |
+
"learning_rate": 0.0008398398061578741,
|
| 1801 |
+
"loss": 5.079273681640625,
|
| 1802 |
+
"step": 12800
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.2631632841139487,
|
| 1806 |
+
"grad_norm": 0.7534170746803284,
|
| 1807 |
+
"learning_rate": 0.0008386582248248209,
|
| 1808 |
+
"loss": 5.0737002563476565,
|
| 1809 |
+
"step": 12850
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.2641872657641975,
|
| 1813 |
+
"grad_norm": 0.8280327320098877,
|
| 1814 |
+
"learning_rate": 0.0008374731388361666,
|
| 1815 |
+
"loss": 5.078436584472656,
|
| 1816 |
+
"step": 12900
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.26521124741444635,
|
| 1820 |
+
"grad_norm": 0.7851305603981018,
|
| 1821 |
+
"learning_rate": 0.0008362845604559519,
|
| 1822 |
+
"loss": 5.08459716796875,
|
| 1823 |
+
"step": 12950
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.26623522906469516,
|
| 1827 |
+
"grad_norm": 0.8221784234046936,
|
| 1828 |
+
"learning_rate": 0.0008350925019843591,
|
| 1829 |
+
"loss": 5.072332763671875,
|
| 1830 |
+
"step": 13000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.26725921071494396,
|
| 1834 |
+
"grad_norm": 0.8254167437553406,
|
| 1835 |
+
"learning_rate": 0.0008338969757575845,
|
| 1836 |
+
"loss": 5.1181182861328125,
|
| 1837 |
+
"step": 13050
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.2682831923651928,
|
| 1841 |
+
"grad_norm": 1.3413949012756348,
|
| 1842 |
+
"learning_rate": 0.0008326979941477112,
|
| 1843 |
+
"loss": 5.084407043457031,
|
| 1844 |
+
"step": 13100
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.26930717401544163,
|
| 1848 |
+
"grad_norm": 0.8049229979515076,
|
| 1849 |
+
"learning_rate": 0.0008314955695625807,
|
| 1850 |
+
"loss": 5.13783935546875,
|
| 1851 |
+
"step": 13150
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.2703311556656905,
|
| 1855 |
+
"grad_norm": 0.6545600295066833,
|
| 1856 |
+
"learning_rate": 0.0008302897144456653,
|
| 1857 |
+
"loss": 5.092869262695313,
|
| 1858 |
+
"step": 13200
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.2713551373159393,
|
| 1862 |
+
"grad_norm": 0.7590045928955078,
|
| 1863 |
+
"learning_rate": 0.0008290804412759375,
|
| 1864 |
+
"loss": 5.100916442871093,
|
| 1865 |
+
"step": 13250
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.2723791189661881,
|
| 1869 |
+
"grad_norm": 0.7717688083648682,
|
| 1870 |
+
"learning_rate": 0.000827867762567743,
|
| 1871 |
+
"loss": 5.128551635742188,
|
| 1872 |
+
"step": 13300
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.27340310061643697,
|
| 1876 |
+
"grad_norm": 1.0009087324142456,
|
| 1877 |
+
"learning_rate": 0.0008266516908706698,
|
| 1878 |
+
"loss": 5.10110595703125,
|
| 1879 |
+
"step": 13350
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.2744270822666858,
|
| 1883 |
+
"grad_norm": 0.644097089767456,
|
| 1884 |
+
"learning_rate": 0.0008254322387694189,
|
| 1885 |
+
"loss": 5.056725769042969,
|
| 1886 |
+
"step": 13400
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.27545106391693464,
|
| 1890 |
+
"grad_norm": 1.2674827575683594,
|
| 1891 |
+
"learning_rate": 0.0008242094188836737,
|
| 1892 |
+
"loss": 5.067946472167969,
|
| 1893 |
+
"step": 13450
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.27647504556718344,
|
| 1897 |
+
"grad_norm": 0.5991284847259521,
|
| 1898 |
+
"learning_rate": 0.0008229832438679701,
|
| 1899 |
+
"loss": 5.096033325195313,
|
| 1900 |
+
"step": 13500
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.27749902721743225,
|
| 1904 |
+
"grad_norm": 1.0202022790908813,
|
| 1905 |
+
"learning_rate": 0.0008217537264115645,
|
| 1906 |
+
"loss": 5.067895812988281,
|
| 1907 |
+
"step": 13550
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.2785230088676811,
|
| 1911 |
+
"grad_norm": 0.7392842769622803,
|
| 1912 |
+
"learning_rate": 0.0008205208792383035,
|
| 1913 |
+
"loss": 5.080252990722657,
|
| 1914 |
+
"step": 13600
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.2795469905179299,
|
| 1918 |
+
"grad_norm": 1.0701810121536255,
|
| 1919 |
+
"learning_rate": 0.0008192847151064915,
|
| 1920 |
+
"loss": 5.084961547851562,
|
| 1921 |
+
"step": 13650
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.2805709721681787,
|
| 1925 |
+
"grad_norm": 0.6204320192337036,
|
| 1926 |
+
"learning_rate": 0.0008180452468087592,
|
| 1927 |
+
"loss": 5.101495666503906,
|
| 1928 |
+
"step": 13700
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.2815949538184276,
|
| 1932 |
+
"grad_norm": 0.9073652029037476,
|
| 1933 |
+
"learning_rate": 0.0008168024871719304,
|
| 1934 |
+
"loss": 5.0699679565429685,
|
| 1935 |
+
"step": 13750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.2826189354686764,
|
| 1939 |
+
"grad_norm": 0.8042837977409363,
|
| 1940 |
+
"learning_rate": 0.0008155564490568906,
|
| 1941 |
+
"loss": 5.077220153808594,
|
| 1942 |
+
"step": 13800
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.28364291711892525,
|
| 1946 |
+
"grad_norm": 0.80221027135849,
|
| 1947 |
+
"learning_rate": 0.0008143071453584525,
|
| 1948 |
+
"loss": 5.073089599609375,
|
| 1949 |
+
"step": 13850
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.28466689876917406,
|
| 1953 |
+
"grad_norm": 1.1368046998977661,
|
| 1954 |
+
"learning_rate": 0.0008130545890052235,
|
| 1955 |
+
"loss": 5.071932678222656,
|
| 1956 |
+
"step": 13900
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.28569088041942287,
|
| 1960 |
+
"grad_norm": 0.7341012954711914,
|
| 1961 |
+
"learning_rate": 0.0008117987929594717,
|
| 1962 |
+
"loss": 5.065098876953125,
|
| 1963 |
+
"step": 13950
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.2867148620696717,
|
| 1967 |
+
"grad_norm": 0.671726644039154,
|
| 1968 |
+
"learning_rate": 0.0008105397702169912,
|
| 1969 |
+
"loss": 5.07258056640625,
|
| 1970 |
+
"step": 14000
|
| 1971 |
}
|
| 1972 |
],
|
| 1973 |
"logging_steps": 50,
|
|
|
|
| 1987 |
"attributes": {}
|
| 1988 |
}
|
| 1989 |
},
|
| 1990 |
+
"total_flos": 182458515456000.0,
|
| 1991 |
"train_batch_size": 16,
|
| 1992 |
"trial_name": null,
|
| 1993 |
"trial_params": null
|