Training in progress, step 100000
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +203 -3
- pytorch_model.bin +1 -1
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 50044689
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2756bd704606cd6c5c35ea5f45a21e975a94f7ac54bd0802ebe8750dfbd1eba8
|
3 |
size 50044689
|
last-checkpoint/pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0fc703713a0fad50a78f7cc73423f660e122486f1451ea9412d49c8df9646af6
|
3 |
size 25761253
|
last-checkpoint/rng_state_0.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_1.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_2.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_3.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_4.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_5.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_6.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/rng_state_7.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f94f403b594a29ecb6816cee93c65f5e3a0566d5747151b3697716adac4e9951
|
3 |
size 14503
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d85ea74361bfabc4dca40ed2a4dec24f25124d91f625a1176acad7044d70175
|
3 |
size 623
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 4.
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -1806,11 +1806,211 @@
|
|
1806 |
"eval_samples_per_second": 1053.586,
|
1807 |
"eval_steps_per_second": 16.512,
|
1808 |
"step": 90000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1809 |
}
|
1810 |
],
|
1811 |
"max_steps": 250000,
|
1812 |
"num_train_epochs": 12,
|
1813 |
-
"total_flos": 1.
|
1814 |
"trial_name": null,
|
1815 |
"trial_params": null
|
1816 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 4.461099214846538,
|
5 |
+
"global_step": 100000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
1806 |
"eval_samples_per_second": 1053.586,
|
1807 |
"eval_steps_per_second": 16.512,
|
1808 |
"step": 90000
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 4.04,
|
1812 |
+
"learning_rate": 0.00045642490670472436,
|
1813 |
+
"loss": 0.4084,
|
1814 |
+
"step": 90500
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 4.06,
|
1818 |
+
"learning_rate": 0.0004547471653280225,
|
1819 |
+
"loss": 0.4078,
|
1820 |
+
"step": 91000
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 4.06,
|
1824 |
+
"eval_loss": 0.37994059920310974,
|
1825 |
+
"eval_runtime": 2.2046,
|
1826 |
+
"eval_samples_per_second": 1041.911,
|
1827 |
+
"eval_steps_per_second": 16.329,
|
1828 |
+
"step": 91000
|
1829 |
+
},
|
1830 |
+
{
|
1831 |
+
"epoch": 4.08,
|
1832 |
+
"learning_rate": 0.00045306287352519543,
|
1833 |
+
"loss": 0.407,
|
1834 |
+
"step": 91500
|
1835 |
+
},
|
1836 |
+
{
|
1837 |
+
"epoch": 4.1,
|
1838 |
+
"learning_rate": 0.00045137210497262333,
|
1839 |
+
"loss": 0.4068,
|
1840 |
+
"step": 92000
|
1841 |
+
},
|
1842 |
+
{
|
1843 |
+
"epoch": 4.1,
|
1844 |
+
"eval_loss": 0.3793868124485016,
|
1845 |
+
"eval_runtime": 2.2708,
|
1846 |
+
"eval_samples_per_second": 1011.534,
|
1847 |
+
"eval_steps_per_second": 15.853,
|
1848 |
+
"step": 92000
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 4.13,
|
1852 |
+
"learning_rate": 0.0004496749336299999,
|
1853 |
+
"loss": 0.4063,
|
1854 |
+
"step": 92500
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 4.15,
|
1858 |
+
"learning_rate": 0.0004479714337370977,
|
1859 |
+
"loss": 0.4057,
|
1860 |
+
"step": 93000
|
1861 |
+
},
|
1862 |
+
{
|
1863 |
+
"epoch": 4.15,
|
1864 |
+
"eval_loss": 0.37835967540740967,
|
1865 |
+
"eval_runtime": 2.237,
|
1866 |
+
"eval_samples_per_second": 1026.834,
|
1867 |
+
"eval_steps_per_second": 16.093,
|
1868 |
+
"step": 93000
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 4.17,
|
1872 |
+
"learning_rate": 0.00044626167981052036,
|
1873 |
+
"loss": 0.4052,
|
1874 |
+
"step": 93500
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 4.19,
|
1878 |
+
"learning_rate": 0.00044454574664044404,
|
1879 |
+
"loss": 0.4047,
|
1880 |
+
"step": 94000
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 4.19,
|
1884 |
+
"eval_loss": 0.37884432077407837,
|
1885 |
+
"eval_runtime": 2.2678,
|
1886 |
+
"eval_samples_per_second": 1012.88,
|
1887 |
+
"eval_steps_per_second": 15.874,
|
1888 |
+
"step": 94000
|
1889 |
+
},
|
1890 |
+
{
|
1891 |
+
"epoch": 4.22,
|
1892 |
+
"learning_rate": 0.000442823709287344,
|
1893 |
+
"loss": 0.4044,
|
1894 |
+
"step": 94500
|
1895 |
+
},
|
1896 |
+
{
|
1897 |
+
"epoch": 4.24,
|
1898 |
+
"learning_rate": 0.0004410956430787129,
|
1899 |
+
"loss": 0.4047,
|
1900 |
+
"step": 95000
|
1901 |
+
},
|
1902 |
+
{
|
1903 |
+
"epoch": 4.24,
|
1904 |
+
"eval_loss": 0.37695789337158203,
|
1905 |
+
"eval_runtime": 2.1458,
|
1906 |
+
"eval_samples_per_second": 1070.466,
|
1907 |
+
"eval_steps_per_second": 16.777,
|
1908 |
+
"step": 95000
|
1909 |
+
},
|
1910 |
+
{
|
1911 |
+
"epoch": 4.26,
|
1912 |
+
"learning_rate": 0.0004393616236057647,
|
1913 |
+
"loss": 0.4036,
|
1914 |
+
"step": 95500
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 4.28,
|
1918 |
+
"learning_rate": 0.00043762172672012875,
|
1919 |
+
"loss": 0.4029,
|
1920 |
+
"step": 96000
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 4.28,
|
1924 |
+
"eval_loss": 0.37500157952308655,
|
1925 |
+
"eval_runtime": 2.2301,
|
1926 |
+
"eval_samples_per_second": 1029.999,
|
1927 |
+
"eval_steps_per_second": 16.143,
|
1928 |
+
"step": 96000
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 4.3,
|
1932 |
+
"learning_rate": 0.0004358760285305312,
|
1933 |
+
"loss": 0.4025,
|
1934 |
+
"step": 96500
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 4.33,
|
1938 |
+
"learning_rate": 0.0004341246053994663,
|
1939 |
+
"loss": 0.4022,
|
1940 |
+
"step": 97000
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 4.33,
|
1944 |
+
"eval_loss": 0.37471264600753784,
|
1945 |
+
"eval_runtime": 2.2149,
|
1946 |
+
"eval_samples_per_second": 1037.069,
|
1947 |
+
"eval_steps_per_second": 16.254,
|
1948 |
+
"step": 97000
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 4.35,
|
1952 |
+
"learning_rate": 0.00043236753393985534,
|
1953 |
+
"loss": 0.4019,
|
1954 |
+
"step": 97500
|
1955 |
+
},
|
1956 |
+
{
|
1957 |
+
"epoch": 4.37,
|
1958 |
+
"learning_rate": 0.0004306048910116964,
|
1959 |
+
"loss": 0.4015,
|
1960 |
+
"step": 98000
|
1961 |
+
},
|
1962 |
+
{
|
1963 |
+
"epoch": 4.37,
|
1964 |
+
"eval_loss": 0.3735538125038147,
|
1965 |
+
"eval_runtime": 2.2361,
|
1966 |
+
"eval_samples_per_second": 1027.233,
|
1967 |
+
"eval_steps_per_second": 16.099,
|
1968 |
+
"step": 98000
|
1969 |
+
},
|
1970 |
+
{
|
1971 |
+
"epoch": 4.39,
|
1972 |
+
"learning_rate": 0.0004288367537187012,
|
1973 |
+
"loss": 0.4011,
|
1974 |
+
"step": 98500
|
1975 |
+
},
|
1976 |
+
{
|
1977 |
+
"epoch": 4.42,
|
1978 |
+
"learning_rate": 0.00042706319940492284,
|
1979 |
+
"loss": 0.4007,
|
1980 |
+
"step": 99000
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 4.42,
|
1984 |
+
"eval_loss": 0.3751888573169708,
|
1985 |
+
"eval_runtime": 2.2595,
|
1986 |
+
"eval_samples_per_second": 1016.613,
|
1987 |
+
"eval_steps_per_second": 15.933,
|
1988 |
+
"step": 99000
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 4.44,
|
1992 |
+
"learning_rate": 0.00042528430565137254,
|
1993 |
+
"loss": 0.4003,
|
1994 |
+
"step": 99500
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 4.46,
|
1998 |
+
"learning_rate": 0.00042350015027262593,
|
1999 |
+
"loss": 0.4,
|
2000 |
+
"step": 100000
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 4.46,
|
2004 |
+
"eval_loss": 0.37434616684913635,
|
2005 |
+
"eval_runtime": 2.2061,
|
2006 |
+
"eval_samples_per_second": 1041.22,
|
2007 |
+
"eval_steps_per_second": 16.319,
|
2008 |
+
"step": 100000
|
2009 |
}
|
2010 |
],
|
2011 |
"max_steps": 250000,
|
2012 |
"num_train_epochs": 12,
|
2013 |
+
"total_flos": 1.6016515007293466e+21,
|
2014 |
"trial_name": null,
|
2015 |
"trial_params": null
|
2016 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0fc703713a0fad50a78f7cc73423f660e122486f1451ea9412d49c8df9646af6
|
3 |
size 25761253
|