Training in progress, step 7000, 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 223144592
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:091774848cf933bf5f5fc46ece65f027a383f1e21eb863c4a6ee2196c3220b2a
|
| 3 |
size 223144592
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 440235130
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:155509552fc0634827843af40e8311f79d74bbfd6abac6c020ec8a903e6fa055
|
| 3 |
size 440235130
|
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 sha256:8ab75688d4ff919c8094f56bc168299a67ac20801a7146ef66e60584aec195e2
|
| 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:bf4fea072617c07f655d6ce962b9373f4ef22170badd6ae1c2242759ff04e1cf
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": Infinity,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1741,6 +1741,295 @@
|
|
| 1741 |
"eval_steps_per_second": 1.269,
|
| 1742 |
"eval_wer": Infinity,
|
| 1743 |
"step": 6000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1744 |
}
|
| 1745 |
],
|
| 1746 |
"logging_steps": 25,
|
|
@@ -1760,7 +2049,7 @@
|
|
| 1760 |
"attributes": {}
|
| 1761 |
}
|
| 1762 |
},
|
| 1763 |
-
"total_flos": 1.
|
| 1764 |
"train_batch_size": 12,
|
| 1765 |
"trial_name": null,
|
| 1766 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": Infinity,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 11.551610239471511,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 7000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1741 |
"eval_steps_per_second": 1.269,
|
| 1742 |
"eval_wer": Infinity,
|
| 1743 |
"step": 6000
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 9.943022295623452,
|
| 1747 |
+
"grad_norm": 5.562705993652344,
|
| 1748 |
+
"learning_rate": 8.372631578947368e-06,
|
| 1749 |
+
"loss": 0.3252,
|
| 1750 |
+
"step": 6025
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 9.984310487200661,
|
| 1754 |
+
"grad_norm": 7.461434364318848,
|
| 1755 |
+
"learning_rate": 8.32e-06,
|
| 1756 |
+
"loss": 0.344,
|
| 1757 |
+
"step": 6050
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 10.024772914946325,
|
| 1761 |
+
"grad_norm": 5.992921352386475,
|
| 1762 |
+
"learning_rate": 8.267368421052632e-06,
|
| 1763 |
+
"loss": 0.2875,
|
| 1764 |
+
"step": 6075
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 10.066061106523534,
|
| 1768 |
+
"grad_norm": 4.986021995544434,
|
| 1769 |
+
"learning_rate": 8.214736842105265e-06,
|
| 1770 |
+
"loss": 0.2773,
|
| 1771 |
+
"step": 6100
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 10.107349298100743,
|
| 1775 |
+
"grad_norm": 4.8151092529296875,
|
| 1776 |
+
"learning_rate": 8.162105263157895e-06,
|
| 1777 |
+
"loss": 0.2886,
|
| 1778 |
+
"step": 6125
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 10.148637489677952,
|
| 1782 |
+
"grad_norm": 4.7961883544921875,
|
| 1783 |
+
"learning_rate": 8.109473684210527e-06,
|
| 1784 |
+
"loss": 0.3002,
|
| 1785 |
+
"step": 6150
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 10.189925681255161,
|
| 1789 |
+
"grad_norm": 5.7180399894714355,
|
| 1790 |
+
"learning_rate": 8.056842105263158e-06,
|
| 1791 |
+
"loss": 0.2995,
|
| 1792 |
+
"step": 6175
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 10.23121387283237,
|
| 1796 |
+
"grad_norm": 5.287256717681885,
|
| 1797 |
+
"learning_rate": 8.00421052631579e-06,
|
| 1798 |
+
"loss": 0.2723,
|
| 1799 |
+
"step": 6200
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 10.27250206440958,
|
| 1803 |
+
"grad_norm": 6.107348442077637,
|
| 1804 |
+
"learning_rate": 7.951578947368421e-06,
|
| 1805 |
+
"loss": 0.2909,
|
| 1806 |
+
"step": 6225
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 10.313790255986788,
|
| 1810 |
+
"grad_norm": 5.607096195220947,
|
| 1811 |
+
"learning_rate": 7.898947368421053e-06,
|
| 1812 |
+
"loss": 0.3019,
|
| 1813 |
+
"step": 6250
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 10.355078447563997,
|
| 1817 |
+
"grad_norm": 5.941197872161865,
|
| 1818 |
+
"learning_rate": 7.846315789473685e-06,
|
| 1819 |
+
"loss": 0.29,
|
| 1820 |
+
"step": 6275
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 10.396366639141206,
|
| 1824 |
+
"grad_norm": 5.736889362335205,
|
| 1825 |
+
"learning_rate": 7.793684210526316e-06,
|
| 1826 |
+
"loss": 0.3138,
|
| 1827 |
+
"step": 6300
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 10.437654830718415,
|
| 1831 |
+
"grad_norm": 5.2913126945495605,
|
| 1832 |
+
"learning_rate": 7.741052631578948e-06,
|
| 1833 |
+
"loss": 0.28,
|
| 1834 |
+
"step": 6325
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 10.478943022295624,
|
| 1838 |
+
"grad_norm": 5.760861396789551,
|
| 1839 |
+
"learning_rate": 7.68842105263158e-06,
|
| 1840 |
+
"loss": 0.3069,
|
| 1841 |
+
"step": 6350
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 10.520231213872833,
|
| 1845 |
+
"grad_norm": 4.886830806732178,
|
| 1846 |
+
"learning_rate": 7.635789473684211e-06,
|
| 1847 |
+
"loss": 0.2785,
|
| 1848 |
+
"step": 6375
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 10.561519405450042,
|
| 1852 |
+
"grad_norm": 5.5294389724731445,
|
| 1853 |
+
"learning_rate": 7.583157894736842e-06,
|
| 1854 |
+
"loss": 0.2862,
|
| 1855 |
+
"step": 6400
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 10.602807597027251,
|
| 1859 |
+
"grad_norm": 5.370204448699951,
|
| 1860 |
+
"learning_rate": 7.5305263157894745e-06,
|
| 1861 |
+
"loss": 0.2794,
|
| 1862 |
+
"step": 6425
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 10.64409578860446,
|
| 1866 |
+
"grad_norm": 4.194797039031982,
|
| 1867 |
+
"learning_rate": 7.477894736842106e-06,
|
| 1868 |
+
"loss": 0.3064,
|
| 1869 |
+
"step": 6450
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 10.685383980181667,
|
| 1873 |
+
"grad_norm": 6.214608669281006,
|
| 1874 |
+
"learning_rate": 7.425263157894738e-06,
|
| 1875 |
+
"loss": 0.3177,
|
| 1876 |
+
"step": 6475
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 10.726672171758876,
|
| 1880 |
+
"grad_norm": 5.77274227142334,
|
| 1881 |
+
"learning_rate": 7.3726315789473694e-06,
|
| 1882 |
+
"loss": 0.31,
|
| 1883 |
+
"step": 6500
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 10.767960363336085,
|
| 1887 |
+
"grad_norm": 6.4990105628967285,
|
| 1888 |
+
"learning_rate": 7.32e-06,
|
| 1889 |
+
"loss": 0.2716,
|
| 1890 |
+
"step": 6525
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 10.809248554913294,
|
| 1894 |
+
"grad_norm": 5.706321716308594,
|
| 1895 |
+
"learning_rate": 7.267368421052632e-06,
|
| 1896 |
+
"loss": 0.2983,
|
| 1897 |
+
"step": 6550
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 10.850536746490503,
|
| 1901 |
+
"grad_norm": 6.846646308898926,
|
| 1902 |
+
"learning_rate": 7.2147368421052635e-06,
|
| 1903 |
+
"loss": 0.2678,
|
| 1904 |
+
"step": 6575
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 10.891824938067712,
|
| 1908 |
+
"grad_norm": 5.539628982543945,
|
| 1909 |
+
"learning_rate": 7.162105263157896e-06,
|
| 1910 |
+
"loss": 0.2809,
|
| 1911 |
+
"step": 6600
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 10.933113129644921,
|
| 1915 |
+
"grad_norm": 5.272496700286865,
|
| 1916 |
+
"learning_rate": 7.109473684210528e-06,
|
| 1917 |
+
"loss": 0.2473,
|
| 1918 |
+
"step": 6625
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 10.97440132122213,
|
| 1922 |
+
"grad_norm": 5.260607719421387,
|
| 1923 |
+
"learning_rate": 7.056842105263158e-06,
|
| 1924 |
+
"loss": 0.2931,
|
| 1925 |
+
"step": 6650
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 11.014863748967795,
|
| 1929 |
+
"grad_norm": 5.241586685180664,
|
| 1930 |
+
"learning_rate": 7.00421052631579e-06,
|
| 1931 |
+
"loss": 0.2771,
|
| 1932 |
+
"step": 6675
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"epoch": 11.056151940545003,
|
| 1936 |
+
"grad_norm": 4.1120781898498535,
|
| 1937 |
+
"learning_rate": 6.951578947368422e-06,
|
| 1938 |
+
"loss": 0.2511,
|
| 1939 |
+
"step": 6700
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 11.097440132122212,
|
| 1943 |
+
"grad_norm": 5.060145378112793,
|
| 1944 |
+
"learning_rate": 6.8989473684210525e-06,
|
| 1945 |
+
"loss": 0.2502,
|
| 1946 |
+
"step": 6725
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 11.138728323699421,
|
| 1950 |
+
"grad_norm": 6.010138988494873,
|
| 1951 |
+
"learning_rate": 6.846315789473684e-06,
|
| 1952 |
+
"loss": 0.2442,
|
| 1953 |
+
"step": 6750
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 11.18001651527663,
|
| 1957 |
+
"grad_norm": 5.282010078430176,
|
| 1958 |
+
"learning_rate": 6.793684210526317e-06,
|
| 1959 |
+
"loss": 0.2485,
|
| 1960 |
+
"step": 6775
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 11.22130470685384,
|
| 1964 |
+
"grad_norm": 4.690225601196289,
|
| 1965 |
+
"learning_rate": 6.741052631578948e-06,
|
| 1966 |
+
"loss": 0.2506,
|
| 1967 |
+
"step": 6800
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 11.262592898431048,
|
| 1971 |
+
"grad_norm": 3.9595465660095215,
|
| 1972 |
+
"learning_rate": 6.68842105263158e-06,
|
| 1973 |
+
"loss": 0.2325,
|
| 1974 |
+
"step": 6825
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 11.303881090008257,
|
| 1978 |
+
"grad_norm": 5.889693737030029,
|
| 1979 |
+
"learning_rate": 6.635789473684211e-06,
|
| 1980 |
+
"loss": 0.2527,
|
| 1981 |
+
"step": 6850
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 11.345169281585466,
|
| 1985 |
+
"grad_norm": 5.33756685256958,
|
| 1986 |
+
"learning_rate": 6.583157894736842e-06,
|
| 1987 |
+
"loss": 0.2488,
|
| 1988 |
+
"step": 6875
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 11.386457473162675,
|
| 1992 |
+
"grad_norm": 5.049898147583008,
|
| 1993 |
+
"learning_rate": 6.530526315789474e-06,
|
| 1994 |
+
"loss": 0.2355,
|
| 1995 |
+
"step": 6900
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 11.427745664739884,
|
| 1999 |
+
"grad_norm": 5.452274799346924,
|
| 2000 |
+
"learning_rate": 6.477894736842106e-06,
|
| 2001 |
+
"loss": 0.249,
|
| 2002 |
+
"step": 6925
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 11.469033856317093,
|
| 2006 |
+
"grad_norm": 4.464880466461182,
|
| 2007 |
+
"learning_rate": 6.425263157894738e-06,
|
| 2008 |
+
"loss": 0.2523,
|
| 2009 |
+
"step": 6950
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 11.510322047894302,
|
| 2013 |
+
"grad_norm": 5.923793792724609,
|
| 2014 |
+
"learning_rate": 6.372631578947369e-06,
|
| 2015 |
+
"loss": 0.2564,
|
| 2016 |
+
"step": 6975
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 11.551610239471511,
|
| 2020 |
+
"grad_norm": 4.821681976318359,
|
| 2021 |
+
"learning_rate": 6.3200000000000005e-06,
|
| 2022 |
+
"loss": 0.2322,
|
| 2023 |
+
"step": 7000
|
| 2024 |
+
},
|
| 2025 |
+
{
|
| 2026 |
+
"epoch": 11.551610239471511,
|
| 2027 |
+
"eval_loss": 1.0933067798614502,
|
| 2028 |
+
"eval_runtime": 721.5402,
|
| 2029 |
+
"eval_samples_per_second": 14.664,
|
| 2030 |
+
"eval_steps_per_second": 1.222,
|
| 2031 |
+
"eval_wer": Infinity,
|
| 2032 |
+
"step": 7000
|
| 2033 |
}
|
| 2034 |
],
|
| 2035 |
"logging_steps": 25,
|
|
|
|
| 2049 |
"attributes": {}
|
| 2050 |
}
|
| 2051 |
},
|
| 2052 |
+
"total_flos": 1.364431611101184e+19,
|
| 2053 |
"train_batch_size": 12,
|
| 2054 |
"trial_name": null,
|
| 2055 |
"trial_params": null
|