Training in progress, step 300, 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 671149168
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e15ef1b2bbee32ac04522adc417ba407344082877bc9ca36612b8eb6d43eff8
|
| 3 |
size 671149168
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1fdfe774153fb7963570000ecb1f70fc959893fbcb43aaea3b43e34a28f12bee
|
| 3 |
+
size 341314644
|
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 sha256:4343b94b7458e71a9f147846a12ccdf8fd946dddf2a18693cc8327019b845b59
|
| 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 sha256:43da0688aca60835f4e18fa7e0f3cc099504828f82fd5dd994118be26b760a0f
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 2.
|
| 3 |
-
"best_model_checkpoint": "miner_id_24/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1805,6 +1805,364 @@
|
|
| 1805 |
"eval_samples_per_second": 11.125,
|
| 1806 |
"eval_steps_per_second": 2.781,
|
| 1807 |
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1808 |
}
|
| 1809 |
],
|
| 1810 |
"logging_steps": 1,
|
|
@@ -1833,7 +2191,7 @@
|
|
| 1833 |
"attributes": {}
|
| 1834 |
}
|
| 1835 |
},
|
| 1836 |
-
"total_flos":
|
| 1837 |
"train_batch_size": 8,
|
| 1838 |
"trial_name": null,
|
| 1839 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.094221830368042,
|
| 3 |
+
"best_model_checkpoint": "miner_id_24/checkpoint-300",
|
| 4 |
+
"epoch": 0.3725551071095933,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1805 |
"eval_samples_per_second": 11.125,
|
| 1806 |
"eval_steps_per_second": 2.781,
|
| 1807 |
"step": 250
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 0.31170443961502636,
|
| 1811 |
+
"grad_norm": 16.239639282226562,
|
| 1812 |
+
"learning_rate": 3.189374007355917e-05,
|
| 1813 |
+
"loss": 9.3269,
|
| 1814 |
+
"step": 251
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.31294628997205837,
|
| 1818 |
+
"grad_norm": 20.803409576416016,
|
| 1819 |
+
"learning_rate": 3.151889938850445e-05,
|
| 1820 |
+
"loss": 9.9335,
|
| 1821 |
+
"step": 252
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.31418814032909037,
|
| 1825 |
+
"grad_norm": 19.762643814086914,
|
| 1826 |
+
"learning_rate": 3.114525791558398e-05,
|
| 1827 |
+
"loss": 10.5991,
|
| 1828 |
+
"step": 253
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.3154299906861223,
|
| 1832 |
+
"grad_norm": 20.731128692626953,
|
| 1833 |
+
"learning_rate": 3.0772839899857464e-05,
|
| 1834 |
+
"loss": 10.1559,
|
| 1835 |
+
"step": 254
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 0.3166718410431543,
|
| 1839 |
+
"grad_norm": 14.596222877502441,
|
| 1840 |
+
"learning_rate": 3.0401669506996256e-05,
|
| 1841 |
+
"loss": 10.3318,
|
| 1842 |
+
"step": 255
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.31791369140018627,
|
| 1846 |
+
"grad_norm": 15.722493171691895,
|
| 1847 |
+
"learning_rate": 3.003177082171523e-05,
|
| 1848 |
+
"loss": 9.81,
|
| 1849 |
+
"step": 256
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.3191555417572183,
|
| 1853 |
+
"grad_norm": 12.498101234436035,
|
| 1854 |
+
"learning_rate": 2.9663167846209998e-05,
|
| 1855 |
+
"loss": 10.9766,
|
| 1856 |
+
"step": 257
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.3203973921142502,
|
| 1860 |
+
"grad_norm": 21.327978134155273,
|
| 1861 |
+
"learning_rate": 2.9295884498599414e-05,
|
| 1862 |
+
"loss": 9.3725,
|
| 1863 |
+
"step": 258
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 0.3216392424712822,
|
| 1867 |
+
"grad_norm": 12.107992172241211,
|
| 1868 |
+
"learning_rate": 2.8929944611373554e-05,
|
| 1869 |
+
"loss": 8.5133,
|
| 1870 |
+
"step": 259
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.3228810928283142,
|
| 1874 |
+
"grad_norm": 17.02812957763672,
|
| 1875 |
+
"learning_rate": 2.8565371929847284e-05,
|
| 1876 |
+
"loss": 10.7682,
|
| 1877 |
+
"step": 260
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.3241229431853462,
|
| 1881 |
+
"grad_norm": 11.257403373718262,
|
| 1882 |
+
"learning_rate": 2.8202190110619493e-05,
|
| 1883 |
+
"loss": 10.3188,
|
| 1884 |
+
"step": 261
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.32536479354237813,
|
| 1888 |
+
"grad_norm": 13.443743705749512,
|
| 1889 |
+
"learning_rate": 2.784042272003794e-05,
|
| 1890 |
+
"loss": 10.1404,
|
| 1891 |
+
"step": 262
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.32660664389941013,
|
| 1895 |
+
"grad_norm": 13.53212833404541,
|
| 1896 |
+
"learning_rate": 2.7480093232670158e-05,
|
| 1897 |
+
"loss": 9.7965,
|
| 1898 |
+
"step": 263
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.3278484942564421,
|
| 1902 |
+
"grad_norm": 14.074053764343262,
|
| 1903 |
+
"learning_rate": 2.712122502978024e-05,
|
| 1904 |
+
"loss": 9.5673,
|
| 1905 |
+
"step": 264
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 0.3290903446134741,
|
| 1909 |
+
"grad_norm": 14.366068840026855,
|
| 1910 |
+
"learning_rate": 2.6763841397811573e-05,
|
| 1911 |
+
"loss": 9.0527,
|
| 1912 |
+
"step": 265
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 0.33033219497050603,
|
| 1916 |
+
"grad_norm": 13.641874313354492,
|
| 1917 |
+
"learning_rate": 2.64079655268759e-05,
|
| 1918 |
+
"loss": 9.7353,
|
| 1919 |
+
"step": 266
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.33157404532753804,
|
| 1923 |
+
"grad_norm": 14.171212196350098,
|
| 1924 |
+
"learning_rate": 2.605362050924848e-05,
|
| 1925 |
+
"loss": 9.4741,
|
| 1926 |
+
"step": 267
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.33281589568457,
|
| 1930 |
+
"grad_norm": 14.457368850708008,
|
| 1931 |
+
"learning_rate": 2.57008293378697e-05,
|
| 1932 |
+
"loss": 10.0971,
|
| 1933 |
+
"step": 268
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.334057746041602,
|
| 1937 |
+
"grad_norm": 16.59000015258789,
|
| 1938 |
+
"learning_rate": 2.534961490485313e-05,
|
| 1939 |
+
"loss": 8.6236,
|
| 1940 |
+
"step": 269
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.335299596398634,
|
| 1944 |
+
"grad_norm": 19.911338806152344,
|
| 1945 |
+
"learning_rate": 2.500000000000001e-05,
|
| 1946 |
+
"loss": 9.7805,
|
| 1947 |
+
"step": 270
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 0.33654144675566594,
|
| 1951 |
+
"grad_norm": 15.089736938476562,
|
| 1952 |
+
"learning_rate": 2.4652007309320498e-05,
|
| 1953 |
+
"loss": 10.0094,
|
| 1954 |
+
"step": 271
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.33778329711269794,
|
| 1958 |
+
"grad_norm": 17.253883361816406,
|
| 1959 |
+
"learning_rate": 2.430565941356157e-05,
|
| 1960 |
+
"loss": 9.644,
|
| 1961 |
+
"step": 272
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.3390251474697299,
|
| 1965 |
+
"grad_norm": 14.608976364135742,
|
| 1966 |
+
"learning_rate": 2.3960978786741877e-05,
|
| 1967 |
+
"loss": 9.3195,
|
| 1968 |
+
"step": 273
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.3402669978267619,
|
| 1972 |
+
"grad_norm": 19.036563873291016,
|
| 1973 |
+
"learning_rate": 2.361798779469336e-05,
|
| 1974 |
+
"loss": 9.0081,
|
| 1975 |
+
"step": 274
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.34150884818379384,
|
| 1979 |
+
"grad_norm": 13.774943351745605,
|
| 1980 |
+
"learning_rate": 2.3276708693609943e-05,
|
| 1981 |
+
"loss": 8.7058,
|
| 1982 |
+
"step": 275
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.34275069854082585,
|
| 1986 |
+
"grad_norm": 15.643052101135254,
|
| 1987 |
+
"learning_rate": 2.2937163628603435e-05,
|
| 1988 |
+
"loss": 8.5943,
|
| 1989 |
+
"step": 276
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.3439925488978578,
|
| 1993 |
+
"grad_norm": 18.12767219543457,
|
| 1994 |
+
"learning_rate": 2.259937463226651e-05,
|
| 1995 |
+
"loss": 9.9232,
|
| 1996 |
+
"step": 277
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.3452343992548898,
|
| 2000 |
+
"grad_norm": 13.298846244812012,
|
| 2001 |
+
"learning_rate": 2.2263363623243054e-05,
|
| 2002 |
+
"loss": 9.3315,
|
| 2003 |
+
"step": 278
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 0.34647624961192175,
|
| 2007 |
+
"grad_norm": 19.000255584716797,
|
| 2008 |
+
"learning_rate": 2.192915240480596e-05,
|
| 2009 |
+
"loss": 9.2574,
|
| 2010 |
+
"step": 279
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.34771809996895375,
|
| 2014 |
+
"grad_norm": 18.62431526184082,
|
| 2015 |
+
"learning_rate": 2.1596762663442218e-05,
|
| 2016 |
+
"loss": 9.4321,
|
| 2017 |
+
"step": 280
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.3489599503259857,
|
| 2021 |
+
"grad_norm": 16.0582332611084,
|
| 2022 |
+
"learning_rate": 2.1266215967445824e-05,
|
| 2023 |
+
"loss": 9.6428,
|
| 2024 |
+
"step": 281
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 0.3502018006830177,
|
| 2028 |
+
"grad_norm": 16.208553314208984,
|
| 2029 |
+
"learning_rate": 2.0937533765518187e-05,
|
| 2030 |
+
"loss": 8.0977,
|
| 2031 |
+
"step": 282
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 0.35144365104004965,
|
| 2035 |
+
"grad_norm": 16.514801025390625,
|
| 2036 |
+
"learning_rate": 2.061073738537635e-05,
|
| 2037 |
+
"loss": 8.221,
|
| 2038 |
+
"step": 283
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 0.35268550139708166,
|
| 2042 |
+
"grad_norm": 14.571998596191406,
|
| 2043 |
+
"learning_rate": 2.0285848032369137e-05,
|
| 2044 |
+
"loss": 7.623,
|
| 2045 |
+
"step": 284
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 0.3539273517541136,
|
| 2049 |
+
"grad_norm": 17.91294288635254,
|
| 2050 |
+
"learning_rate": 1.996288678810105e-05,
|
| 2051 |
+
"loss": 9.8574,
|
| 2052 |
+
"step": 285
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.3551692021111456,
|
| 2056 |
+
"grad_norm": 16.300058364868164,
|
| 2057 |
+
"learning_rate": 1.9641874609064443e-05,
|
| 2058 |
+
"loss": 8.4222,
|
| 2059 |
+
"step": 286
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 0.35641105246817756,
|
| 2063 |
+
"grad_norm": 18.36516761779785,
|
| 2064 |
+
"learning_rate": 1.932283232527956e-05,
|
| 2065 |
+
"loss": 7.5621,
|
| 2066 |
+
"step": 287
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.35765290282520956,
|
| 2070 |
+
"grad_norm": 14.960530281066895,
|
| 2071 |
+
"learning_rate": 1.9005780638942982e-05,
|
| 2072 |
+
"loss": 8.2772,
|
| 2073 |
+
"step": 288
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 0.35889475318224157,
|
| 2077 |
+
"grad_norm": 18.4453182220459,
|
| 2078 |
+
"learning_rate": 1.8690740123084316e-05,
|
| 2079 |
+
"loss": 7.9102,
|
| 2080 |
+
"step": 289
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.3601366035392735,
|
| 2084 |
+
"grad_norm": 16.127838134765625,
|
| 2085 |
+
"learning_rate": 1.837773122023114e-05,
|
| 2086 |
+
"loss": 8.938,
|
| 2087 |
+
"step": 290
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 0.3613784538963055,
|
| 2091 |
+
"grad_norm": 15.636646270751953,
|
| 2092 |
+
"learning_rate": 1.8066774241082612e-05,
|
| 2093 |
+
"loss": 7.8458,
|
| 2094 |
+
"step": 291
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 0.36262030425333747,
|
| 2098 |
+
"grad_norm": 20.232114791870117,
|
| 2099 |
+
"learning_rate": 1.7757889363191483e-05,
|
| 2100 |
+
"loss": 7.0922,
|
| 2101 |
+
"step": 292
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.36386215461036947,
|
| 2105 |
+
"grad_norm": 15.795831680297852,
|
| 2106 |
+
"learning_rate": 1.745109662965481e-05,
|
| 2107 |
+
"loss": 7.8668,
|
| 2108 |
+
"step": 293
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 0.3651040049674014,
|
| 2112 |
+
"grad_norm": 18.210309982299805,
|
| 2113 |
+
"learning_rate": 1.714641594781347e-05,
|
| 2114 |
+
"loss": 7.5169,
|
| 2115 |
+
"step": 294
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 0.3663458553244334,
|
| 2119 |
+
"grad_norm": 16.791549682617188,
|
| 2120 |
+
"learning_rate": 1.684386708796025e-05,
|
| 2121 |
+
"loss": 6.6066,
|
| 2122 |
+
"step": 295
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 0.36758770568146537,
|
| 2126 |
+
"grad_norm": 17.71488380432129,
|
| 2127 |
+
"learning_rate": 1.6543469682057106e-05,
|
| 2128 |
+
"loss": 6.4865,
|
| 2129 |
+
"step": 296
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 0.3688295560384974,
|
| 2133 |
+
"grad_norm": 15.17342758178711,
|
| 2134 |
+
"learning_rate": 1.62452432224612e-05,
|
| 2135 |
+
"loss": 7.7187,
|
| 2136 |
+
"step": 297
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 0.3700714063955293,
|
| 2140 |
+
"grad_norm": 18.377864837646484,
|
| 2141 |
+
"learning_rate": 1.5949207060660138e-05,
|
| 2142 |
+
"loss": 7.2307,
|
| 2143 |
+
"step": 298
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.3713132567525613,
|
| 2147 |
+
"grad_norm": 21.42231559753418,
|
| 2148 |
+
"learning_rate": 1.5655380406016235e-05,
|
| 2149 |
+
"loss": 6.8554,
|
| 2150 |
+
"step": 299
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.3725551071095933,
|
| 2154 |
+
"grad_norm": 19.9051513671875,
|
| 2155 |
+
"learning_rate": 1.536378232452003e-05,
|
| 2156 |
+
"loss": 7.1716,
|
| 2157 |
+
"step": 300
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 0.3725551071095933,
|
| 2161 |
+
"eval_loss": 2.094221830368042,
|
| 2162 |
+
"eval_runtime": 121.6383,
|
| 2163 |
+
"eval_samples_per_second": 11.148,
|
| 2164 |
+
"eval_steps_per_second": 2.787,
|
| 2165 |
+
"step": 300
|
| 2166 |
}
|
| 2167 |
],
|
| 2168 |
"logging_steps": 1,
|
|
|
|
| 2191 |
"attributes": {}
|
| 2192 |
}
|
| 2193 |
},
|
| 2194 |
+
"total_flos": 5.248185197962199e+17,
|
| 2195 |
"train_batch_size": 8,
|
| 2196 |
"trial_name": null,
|
| 2197 |
"trial_params": null
|