Training in progress, step 2500, 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 536223056
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6323430495422b2f5e9b7844076fda108d2adbed4d0037f47d6e99938d8fca29
|
| 3 |
size 536223056
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1072594443
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23e6291f28b6db5850e454b2320b2900c167fefe5276101f07b3b0cce8757420
|
| 3 |
size 1072594443
|
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 sha256:37b2e328f1145450725e3266f16e300be997471c44b08eae4fb08a4a11d9367a
|
| 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,
|
|
@@ -1808,6 +1808,456 @@
|
|
| 1808 |
"mean_token_accuracy": 0.7935750424861908,
|
| 1809 |
"num_tokens": 2208982.0,
|
| 1810 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1811 |
}
|
| 1812 |
],
|
| 1813 |
"logging_steps": 10,
|
|
@@ -1827,7 +2277,7 @@
|
|
| 1827 |
"attributes": {}
|
| 1828 |
}
|
| 1829 |
},
|
| 1830 |
-
"total_flos":
|
| 1831 |
"train_batch_size": 8,
|
| 1832 |
"trial_name": null,
|
| 1833 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.5037275841225065,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 2500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1808 |
"mean_token_accuracy": 0.7935750424861908,
|
| 1809 |
"num_tokens": 2208982.0,
|
| 1810 |
"step": 2000
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.40499697763449527,
|
| 1814 |
+
"grad_norm": 17.875,
|
| 1815 |
+
"learning_rate": 1.7301363422661027e-05,
|
| 1816 |
+
"loss": 0.9803,
|
| 1817 |
+
"mean_token_accuracy": 0.7665176451206207,
|
| 1818 |
+
"num_tokens": 2219728.0,
|
| 1819 |
+
"step": 2010
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 0.4070118879709853,
|
| 1823 |
+
"grad_norm": 12.0,
|
| 1824 |
+
"learning_rate": 1.7287930687084426e-05,
|
| 1825 |
+
"loss": 0.9941,
|
| 1826 |
+
"mean_token_accuracy": 0.7633516311645507,
|
| 1827 |
+
"num_tokens": 2230087.0,
|
| 1828 |
+
"step": 2020
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.4090267983074753,
|
| 1832 |
+
"grad_norm": 11.0625,
|
| 1833 |
+
"learning_rate": 1.7274497951507828e-05,
|
| 1834 |
+
"loss": 0.8946,
|
| 1835 |
+
"mean_token_accuracy": 0.786914736032486,
|
| 1836 |
+
"num_tokens": 2240723.0,
|
| 1837 |
+
"step": 2030
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.41104170864396533,
|
| 1841 |
+
"grad_norm": 11.0,
|
| 1842 |
+
"learning_rate": 1.7261065215931227e-05,
|
| 1843 |
+
"loss": 0.8631,
|
| 1844 |
+
"mean_token_accuracy": 0.7922929883003235,
|
| 1845 |
+
"num_tokens": 2250999.0,
|
| 1846 |
+
"step": 2040
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 0.41305661898045537,
|
| 1850 |
+
"grad_norm": 12.0625,
|
| 1851 |
+
"learning_rate": 1.7247632480354626e-05,
|
| 1852 |
+
"loss": 0.8731,
|
| 1853 |
+
"mean_token_accuracy": 0.7846612274646759,
|
| 1854 |
+
"num_tokens": 2261138.0,
|
| 1855 |
+
"step": 2050
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 0.4150715293169454,
|
| 1859 |
+
"grad_norm": 10.125,
|
| 1860 |
+
"learning_rate": 1.7234199744778028e-05,
|
| 1861 |
+
"loss": 0.885,
|
| 1862 |
+
"mean_token_accuracy": 0.7879790186882019,
|
| 1863 |
+
"num_tokens": 2271083.0,
|
| 1864 |
+
"step": 2060
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.41708643965343545,
|
| 1868 |
+
"grad_norm": 11.375,
|
| 1869 |
+
"learning_rate": 1.7220767009201423e-05,
|
| 1870 |
+
"loss": 0.9247,
|
| 1871 |
+
"mean_token_accuracy": 0.7802027463912964,
|
| 1872 |
+
"num_tokens": 2282490.0,
|
| 1873 |
+
"step": 2070
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.41910134998992543,
|
| 1877 |
+
"grad_norm": 10.625,
|
| 1878 |
+
"learning_rate": 1.7207334273624825e-05,
|
| 1879 |
+
"loss": 0.953,
|
| 1880 |
+
"mean_token_accuracy": 0.7730051100254058,
|
| 1881 |
+
"num_tokens": 2293586.0,
|
| 1882 |
+
"step": 2080
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 0.42111626032641547,
|
| 1886 |
+
"grad_norm": 11.625,
|
| 1887 |
+
"learning_rate": 1.7193901538048224e-05,
|
| 1888 |
+
"loss": 1.0067,
|
| 1889 |
+
"mean_token_accuracy": 0.7572938621044158,
|
| 1890 |
+
"num_tokens": 2304802.0,
|
| 1891 |
+
"step": 2090
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.4231311706629055,
|
| 1895 |
+
"grad_norm": 12.3125,
|
| 1896 |
+
"learning_rate": 1.7180468802471623e-05,
|
| 1897 |
+
"loss": 0.9289,
|
| 1898 |
+
"mean_token_accuracy": 0.7806627154350281,
|
| 1899 |
+
"num_tokens": 2315105.0,
|
| 1900 |
+
"step": 2100
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.42514608099939555,
|
| 1904 |
+
"grad_norm": 12.9375,
|
| 1905 |
+
"learning_rate": 1.7167036066895025e-05,
|
| 1906 |
+
"loss": 0.8419,
|
| 1907 |
+
"mean_token_accuracy": 0.7955414175987243,
|
| 1908 |
+
"num_tokens": 2326101.0,
|
| 1909 |
+
"step": 2110
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 0.42716099133588553,
|
| 1913 |
+
"grad_norm": 11.1875,
|
| 1914 |
+
"learning_rate": 1.7153603331318424e-05,
|
| 1915 |
+
"loss": 0.9516,
|
| 1916 |
+
"mean_token_accuracy": 0.7715274512767791,
|
| 1917 |
+
"num_tokens": 2337205.0,
|
| 1918 |
+
"step": 2120
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 0.42917590167237557,
|
| 1922 |
+
"grad_norm": 11.8125,
|
| 1923 |
+
"learning_rate": 1.7140170595741826e-05,
|
| 1924 |
+
"loss": 0.8855,
|
| 1925 |
+
"mean_token_accuracy": 0.7864199817180634,
|
| 1926 |
+
"num_tokens": 2349394.0,
|
| 1927 |
+
"step": 2130
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.4311908120088656,
|
| 1931 |
+
"grad_norm": 11.1875,
|
| 1932 |
+
"learning_rate": 1.7126737860165222e-05,
|
| 1933 |
+
"loss": 0.9328,
|
| 1934 |
+
"mean_token_accuracy": 0.7836083233356476,
|
| 1935 |
+
"num_tokens": 2360556.0,
|
| 1936 |
+
"step": 2140
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.43320572234535565,
|
| 1940 |
+
"grad_norm": 11.4375,
|
| 1941 |
+
"learning_rate": 1.7113305124588624e-05,
|
| 1942 |
+
"loss": 0.9885,
|
| 1943 |
+
"mean_token_accuracy": 0.7649979829788208,
|
| 1944 |
+
"num_tokens": 2371734.0,
|
| 1945 |
+
"step": 2150
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 0.43522063268184563,
|
| 1949 |
+
"grad_norm": 12.8125,
|
| 1950 |
+
"learning_rate": 1.7099872389012023e-05,
|
| 1951 |
+
"loss": 0.9407,
|
| 1952 |
+
"mean_token_accuracy": 0.7679962277412414,
|
| 1953 |
+
"num_tokens": 2382510.0,
|
| 1954 |
+
"step": 2160
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.43723554301833567,
|
| 1958 |
+
"grad_norm": 12.625,
|
| 1959 |
+
"learning_rate": 1.708643965343542e-05,
|
| 1960 |
+
"loss": 0.9365,
|
| 1961 |
+
"mean_token_accuracy": 0.7730210840702056,
|
| 1962 |
+
"num_tokens": 2395814.0,
|
| 1963 |
+
"step": 2170
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.4392504533548257,
|
| 1967 |
+
"grad_norm": 14.4375,
|
| 1968 |
+
"learning_rate": 1.7073006917858824e-05,
|
| 1969 |
+
"loss": 0.9407,
|
| 1970 |
+
"mean_token_accuracy": 0.773440134525299,
|
| 1971 |
+
"num_tokens": 2407124.0,
|
| 1972 |
+
"step": 2180
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 0.44126536369131575,
|
| 1976 |
+
"grad_norm": 11.375,
|
| 1977 |
+
"learning_rate": 1.7059574182282223e-05,
|
| 1978 |
+
"loss": 0.9601,
|
| 1979 |
+
"mean_token_accuracy": 0.7624564170837402,
|
| 1980 |
+
"num_tokens": 2418793.0,
|
| 1981 |
+
"step": 2190
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 0.4432802740278058,
|
| 1985 |
+
"grad_norm": 10.4375,
|
| 1986 |
+
"learning_rate": 1.7046141446705625e-05,
|
| 1987 |
+
"loss": 0.8615,
|
| 1988 |
+
"mean_token_accuracy": 0.7828535497188568,
|
| 1989 |
+
"num_tokens": 2430231.0,
|
| 1990 |
+
"step": 2200
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.44529518436429577,
|
| 1994 |
+
"grad_norm": 13.0625,
|
| 1995 |
+
"learning_rate": 1.7032708711129024e-05,
|
| 1996 |
+
"loss": 0.994,
|
| 1997 |
+
"mean_token_accuracy": 0.7661273539066314,
|
| 1998 |
+
"num_tokens": 2441407.0,
|
| 1999 |
+
"step": 2210
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.4473100947007858,
|
| 2003 |
+
"grad_norm": 14.4375,
|
| 2004 |
+
"learning_rate": 1.7019275975552423e-05,
|
| 2005 |
+
"loss": 0.975,
|
| 2006 |
+
"mean_token_accuracy": 0.7622927308082581,
|
| 2007 |
+
"num_tokens": 2453109.0,
|
| 2008 |
+
"step": 2220
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 0.44932500503727585,
|
| 2012 |
+
"grad_norm": 8.875,
|
| 2013 |
+
"learning_rate": 1.7005843239975825e-05,
|
| 2014 |
+
"loss": 0.9016,
|
| 2015 |
+
"mean_token_accuracy": 0.7804525554180145,
|
| 2016 |
+
"num_tokens": 2463823.0,
|
| 2017 |
+
"step": 2230
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.4513399153737659,
|
| 2021 |
+
"grad_norm": 13.8125,
|
| 2022 |
+
"learning_rate": 1.699241050439922e-05,
|
| 2023 |
+
"loss": 0.8886,
|
| 2024 |
+
"mean_token_accuracy": 0.7791055798530578,
|
| 2025 |
+
"num_tokens": 2474062.0,
|
| 2026 |
+
"step": 2240
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.45335482571025587,
|
| 2030 |
+
"grad_norm": 12.5625,
|
| 2031 |
+
"learning_rate": 1.6978977768822623e-05,
|
| 2032 |
+
"loss": 0.8376,
|
| 2033 |
+
"mean_token_accuracy": 0.791073453426361,
|
| 2034 |
+
"num_tokens": 2485012.0,
|
| 2035 |
+
"step": 2250
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 0.4553697360467459,
|
| 2039 |
+
"grad_norm": 12.4375,
|
| 2040 |
+
"learning_rate": 1.696554503324602e-05,
|
| 2041 |
+
"loss": 0.9526,
|
| 2042 |
+
"mean_token_accuracy": 0.7676692366600036,
|
| 2043 |
+
"num_tokens": 2497094.0,
|
| 2044 |
+
"step": 2260
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 0.45738464638323595,
|
| 2048 |
+
"grad_norm": 10.75,
|
| 2049 |
+
"learning_rate": 1.695211229766942e-05,
|
| 2050 |
+
"loss": 1.0348,
|
| 2051 |
+
"mean_token_accuracy": 0.7595704078674317,
|
| 2052 |
+
"num_tokens": 2509251.0,
|
| 2053 |
+
"step": 2270
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.459399556719726,
|
| 2057 |
+
"grad_norm": 11.375,
|
| 2058 |
+
"learning_rate": 1.6938679562092822e-05,
|
| 2059 |
+
"loss": 0.8975,
|
| 2060 |
+
"mean_token_accuracy": 0.7786314010620117,
|
| 2061 |
+
"num_tokens": 2519167.0,
|
| 2062 |
+
"step": 2280
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.46141446705621597,
|
| 2066 |
+
"grad_norm": 10.5,
|
| 2067 |
+
"learning_rate": 1.692524682651622e-05,
|
| 2068 |
+
"loss": 0.931,
|
| 2069 |
+
"mean_token_accuracy": 0.780303293466568,
|
| 2070 |
+
"num_tokens": 2530095.0,
|
| 2071 |
+
"step": 2290
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 0.463429377392706,
|
| 2075 |
+
"grad_norm": 10.0,
|
| 2076 |
+
"learning_rate": 1.6911814090939623e-05,
|
| 2077 |
+
"loss": 0.9055,
|
| 2078 |
+
"mean_token_accuracy": 0.7792095363140106,
|
| 2079 |
+
"num_tokens": 2542753.0,
|
| 2080 |
+
"step": 2300
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.46544428772919605,
|
| 2084 |
+
"grad_norm": 12.625,
|
| 2085 |
+
"learning_rate": 1.689838135536302e-05,
|
| 2086 |
+
"loss": 0.8634,
|
| 2087 |
+
"mean_token_accuracy": 0.7921158850193024,
|
| 2088 |
+
"num_tokens": 2553761.0,
|
| 2089 |
+
"step": 2310
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.4674591980656861,
|
| 2093 |
+
"grad_norm": 10.75,
|
| 2094 |
+
"learning_rate": 1.688494861978642e-05,
|
| 2095 |
+
"loss": 0.8504,
|
| 2096 |
+
"mean_token_accuracy": 0.7890695691108703,
|
| 2097 |
+
"num_tokens": 2564639.0,
|
| 2098 |
+
"step": 2320
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 0.4694741084021761,
|
| 2102 |
+
"grad_norm": 13.4375,
|
| 2103 |
+
"learning_rate": 1.687151588420982e-05,
|
| 2104 |
+
"loss": 0.9293,
|
| 2105 |
+
"mean_token_accuracy": 0.7696837067604065,
|
| 2106 |
+
"num_tokens": 2576449.0,
|
| 2107 |
+
"step": 2330
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 0.4714890187386661,
|
| 2111 |
+
"grad_norm": 12.375,
|
| 2112 |
+
"learning_rate": 1.685808314863322e-05,
|
| 2113 |
+
"loss": 0.867,
|
| 2114 |
+
"mean_token_accuracy": 0.7936202645301819,
|
| 2115 |
+
"num_tokens": 2588236.0,
|
| 2116 |
+
"step": 2340
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.47350392907515615,
|
| 2120 |
+
"grad_norm": 12.875,
|
| 2121 |
+
"learning_rate": 1.684465041305662e-05,
|
| 2122 |
+
"loss": 0.8259,
|
| 2123 |
+
"mean_token_accuracy": 0.7961400330066681,
|
| 2124 |
+
"num_tokens": 2599720.0,
|
| 2125 |
+
"step": 2350
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.4755188394116462,
|
| 2129 |
+
"grad_norm": 12.875,
|
| 2130 |
+
"learning_rate": 1.683121767748002e-05,
|
| 2131 |
+
"loss": 0.9056,
|
| 2132 |
+
"mean_token_accuracy": 0.780539608001709,
|
| 2133 |
+
"num_tokens": 2609396.0,
|
| 2134 |
+
"step": 2360
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 0.4775337497481362,
|
| 2138 |
+
"grad_norm": 11.625,
|
| 2139 |
+
"learning_rate": 1.681778494190342e-05,
|
| 2140 |
+
"loss": 0.9019,
|
| 2141 |
+
"mean_token_accuracy": 0.7818064391613007,
|
| 2142 |
+
"num_tokens": 2621392.0,
|
| 2143 |
+
"step": 2370
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.4795486600846262,
|
| 2147 |
+
"grad_norm": 10.75,
|
| 2148 |
+
"learning_rate": 1.680435220632682e-05,
|
| 2149 |
+
"loss": 0.7993,
|
| 2150 |
+
"mean_token_accuracy": 0.8025827884674073,
|
| 2151 |
+
"num_tokens": 2633038.0,
|
| 2152 |
+
"step": 2380
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.48156357042111625,
|
| 2156 |
+
"grad_norm": 11.4375,
|
| 2157 |
+
"learning_rate": 1.679091947075022e-05,
|
| 2158 |
+
"loss": 0.9316,
|
| 2159 |
+
"mean_token_accuracy": 0.7733452200889588,
|
| 2160 |
+
"num_tokens": 2644078.0,
|
| 2161 |
+
"step": 2390
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 0.4835784807576063,
|
| 2165 |
+
"grad_norm": 9.5625,
|
| 2166 |
+
"learning_rate": 1.6777486735173622e-05,
|
| 2167 |
+
"loss": 0.8044,
|
| 2168 |
+
"mean_token_accuracy": 0.8011213660240173,
|
| 2169 |
+
"num_tokens": 2655374.0,
|
| 2170 |
+
"step": 2400
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 0.4855933910940963,
|
| 2174 |
+
"grad_norm": 11.8125,
|
| 2175 |
+
"learning_rate": 1.6764053999597017e-05,
|
| 2176 |
+
"loss": 0.8751,
|
| 2177 |
+
"mean_token_accuracy": 0.7866755127906799,
|
| 2178 |
+
"num_tokens": 2665838.0,
|
| 2179 |
+
"step": 2410
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.48760830143058637,
|
| 2183 |
+
"grad_norm": 10.125,
|
| 2184 |
+
"learning_rate": 1.675062126402042e-05,
|
| 2185 |
+
"loss": 0.827,
|
| 2186 |
+
"mean_token_accuracy": 0.7927514970302582,
|
| 2187 |
+
"num_tokens": 2675934.0,
|
| 2188 |
+
"step": 2420
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 0.48962321176707635,
|
| 2192 |
+
"grad_norm": 12.8125,
|
| 2193 |
+
"learning_rate": 1.673718852844382e-05,
|
| 2194 |
+
"loss": 0.9346,
|
| 2195 |
+
"mean_token_accuracy": 0.7792349219322204,
|
| 2196 |
+
"num_tokens": 2687584.0,
|
| 2197 |
+
"step": 2430
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"epoch": 0.4916381221035664,
|
| 2201 |
+
"grad_norm": 11.875,
|
| 2202 |
+
"learning_rate": 1.6723755792867217e-05,
|
| 2203 |
+
"loss": 0.8867,
|
| 2204 |
+
"mean_token_accuracy": 0.7851879954338074,
|
| 2205 |
+
"num_tokens": 2697927.0,
|
| 2206 |
+
"step": 2440
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 0.4936530324400564,
|
| 2210 |
+
"grad_norm": 11.75,
|
| 2211 |
+
"learning_rate": 1.671032305729062e-05,
|
| 2212 |
+
"loss": 0.8585,
|
| 2213 |
+
"mean_token_accuracy": 0.7973346531391143,
|
| 2214 |
+
"num_tokens": 2708092.0,
|
| 2215 |
+
"step": 2450
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.49566794277654647,
|
| 2219 |
+
"grad_norm": 14.25,
|
| 2220 |
+
"learning_rate": 1.669689032171402e-05,
|
| 2221 |
+
"loss": 0.9075,
|
| 2222 |
+
"mean_token_accuracy": 0.7788807570934295,
|
| 2223 |
+
"num_tokens": 2719293.0,
|
| 2224 |
+
"step": 2460
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 0.49768285311303645,
|
| 2228 |
+
"grad_norm": 9.875,
|
| 2229 |
+
"learning_rate": 1.668345758613742e-05,
|
| 2230 |
+
"loss": 0.8931,
|
| 2231 |
+
"mean_token_accuracy": 0.7861545145511627,
|
| 2232 |
+
"num_tokens": 2730531.0,
|
| 2233 |
+
"step": 2470
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 0.4996977634495265,
|
| 2237 |
+
"grad_norm": 10.4375,
|
| 2238 |
+
"learning_rate": 1.6670024850560816e-05,
|
| 2239 |
+
"loss": 0.9438,
|
| 2240 |
+
"mean_token_accuracy": 0.7664293229579926,
|
| 2241 |
+
"num_tokens": 2741732.0,
|
| 2242 |
+
"step": 2480
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.5017126737860165,
|
| 2246 |
+
"grad_norm": 11.125,
|
| 2247 |
+
"learning_rate": 1.6656592114984218e-05,
|
| 2248 |
+
"loss": 0.8719,
|
| 2249 |
+
"mean_token_accuracy": 0.7909434497356415,
|
| 2250 |
+
"num_tokens": 2753005.0,
|
| 2251 |
+
"step": 2490
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 0.5037275841225065,
|
| 2255 |
+
"grad_norm": 11.25,
|
| 2256 |
+
"learning_rate": 1.6643159379407617e-05,
|
| 2257 |
+
"loss": 0.8739,
|
| 2258 |
+
"mean_token_accuracy": 0.7839685261249543,
|
| 2259 |
+
"num_tokens": 2765568.0,
|
| 2260 |
+
"step": 2500
|
| 2261 |
}
|
| 2262 |
],
|
| 2263 |
"logging_steps": 10,
|
|
|
|
| 2277 |
"attributes": {}
|
| 2278 |
}
|
| 2279 |
},
|
| 2280 |
+
"total_flos": 3352572806252544.0,
|
| 2281 |
"train_batch_size": 8,
|
| 2282 |
"trial_name": null,
|
| 2283 |
"trial_params": null
|