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 479005064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c9e4dd816674f656017c8a2ed41209cd2c297c41f55fae0279c467dc4d8041e
|
| 3 |
size 479005064
|
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 sha256:3eac352bf4a411d3fd08e9ea4170ef4fe6642e77dd87eadf5a63f336287b9b85
|
| 3 |
+
size 243802996
|
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 sha256:3341cd00b63d6a7a00c7b3a77f57159c6ca21f09bf4c7cab59461e7b2bf3b187
|
| 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 sha256:0d93420319c4318ff13366855f16b6ec61d99b866bdf2a20293a1621b040b36f
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 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": 7.598,
|
| 1806 |
"eval_steps_per_second": 7.598,
|
| 1807 |
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1808 |
}
|
| 1809 |
],
|
| 1810 |
"logging_steps": 1,
|
|
@@ -1819,12 +2177,12 @@
|
|
| 1819 |
"should_evaluate": false,
|
| 1820 |
"should_log": false,
|
| 1821 |
"should_save": true,
|
| 1822 |
-
"should_training_stop":
|
| 1823 |
},
|
| 1824 |
"attributes": {}
|
| 1825 |
}
|
| 1826 |
},
|
| 1827 |
-
"total_flos":
|
| 1828 |
"train_batch_size": 1,
|
| 1829 |
"trial_name": null,
|
| 1830 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.014986980061071943,
|
| 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": 7.598,
|
| 1806 |
"eval_steps_per_second": 7.598,
|
| 1807 |
"step": 250
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 0.01253910665109686,
|
| 1811 |
+
"grad_norm": 1.913027048110962,
|
| 1812 |
+
"learning_rate": 2.0641226378639715e-05,
|
| 1813 |
+
"loss": 0.9081,
|
| 1814 |
+
"step": 251
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.012589063251300433,
|
| 1818 |
+
"grad_norm": 1.6886483430862427,
|
| 1819 |
+
"learning_rate": 1.9826210008049785e-05,
|
| 1820 |
+
"loss": 0.8158,
|
| 1821 |
+
"step": 252
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.012639019851504006,
|
| 1825 |
+
"grad_norm": 1.481825590133667,
|
| 1826 |
+
"learning_rate": 1.902647010887655e-05,
|
| 1827 |
+
"loss": 0.7513,
|
| 1828 |
+
"step": 253
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.012688976451707579,
|
| 1832 |
+
"grad_norm": 1.1585659980773926,
|
| 1833 |
+
"learning_rate": 1.8242100534143062e-05,
|
| 1834 |
+
"loss": 0.5836,
|
| 1835 |
+
"step": 254
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 0.012738933051911152,
|
| 1839 |
+
"grad_norm": 1.1122362613677979,
|
| 1840 |
+
"learning_rate": 1.7473193333096575e-05,
|
| 1841 |
+
"loss": 0.3605,
|
| 1842 |
+
"step": 255
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.012788889652114725,
|
| 1846 |
+
"grad_norm": 1.0502725839614868,
|
| 1847 |
+
"learning_rate": 1.671983874040631e-05,
|
| 1848 |
+
"loss": 0.3302,
|
| 1849 |
+
"step": 256
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.0128388462523183,
|
| 1853 |
+
"grad_norm": 0.9571714997291565,
|
| 1854 |
+
"learning_rate": 1.598212516557394e-05,
|
| 1855 |
+
"loss": 0.3259,
|
| 1856 |
+
"step": 257
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.012888802852521872,
|
| 1860 |
+
"grad_norm": 0.7586105465888977,
|
| 1861 |
+
"learning_rate": 1.526013918255836e-05,
|
| 1862 |
+
"loss": 0.3028,
|
| 1863 |
+
"step": 258
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 0.012938759452725445,
|
| 1867 |
+
"grad_norm": 0.6112863421440125,
|
| 1868 |
+
"learning_rate": 1.4553965519615723e-05,
|
| 1869 |
+
"loss": 0.3124,
|
| 1870 |
+
"step": 259
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.012988716052929018,
|
| 1874 |
+
"grad_norm": 0.619231641292572,
|
| 1875 |
+
"learning_rate": 1.3863687049356464e-05,
|
| 1876 |
+
"loss": 0.3073,
|
| 1877 |
+
"step": 260
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.01303867265313259,
|
| 1881 |
+
"grad_norm": 0.5576855540275574,
|
| 1882 |
+
"learning_rate": 1.3189384779019535e-05,
|
| 1883 |
+
"loss": 0.32,
|
| 1884 |
+
"step": 261
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.013088629253336164,
|
| 1888 |
+
"grad_norm": 9.921064376831055,
|
| 1889 |
+
"learning_rate": 1.25311378409661e-05,
|
| 1890 |
+
"loss": 0.437,
|
| 1891 |
+
"step": 262
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.013138585853539737,
|
| 1895 |
+
"grad_norm": 0.4192136824131012,
|
| 1896 |
+
"learning_rate": 1.1889023483392879e-05,
|
| 1897 |
+
"loss": 0.2452,
|
| 1898 |
+
"step": 263
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.013188542453743311,
|
| 1902 |
+
"grad_norm": 0.47411757707595825,
|
| 1903 |
+
"learning_rate": 1.1263117061266675e-05,
|
| 1904 |
+
"loss": 0.26,
|
| 1905 |
+
"step": 264
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 0.013238499053946884,
|
| 1909 |
+
"grad_norm": 0.475243479013443,
|
| 1910 |
+
"learning_rate": 1.0653492027481286e-05,
|
| 1911 |
+
"loss": 0.2812,
|
| 1912 |
+
"step": 265
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 0.013288455654150457,
|
| 1916 |
+
"grad_norm": 0.41210004687309265,
|
| 1917 |
+
"learning_rate": 1.0060219924237379e-05,
|
| 1918 |
+
"loss": 0.2719,
|
| 1919 |
+
"step": 266
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.01333841225435403,
|
| 1923 |
+
"grad_norm": 0.39359334111213684,
|
| 1924 |
+
"learning_rate": 9.48337037464666e-06,
|
| 1925 |
+
"loss": 0.2401,
|
| 1926 |
+
"step": 267
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.013388368854557603,
|
| 1930 |
+
"grad_norm": 0.48420053720474243,
|
| 1931 |
+
"learning_rate": 8.923011074561404e-06,
|
| 1932 |
+
"loss": 0.2553,
|
| 1933 |
+
"step": 268
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.013438325454761176,
|
| 1937 |
+
"grad_norm": 0.4084879159927368,
|
| 1938 |
+
"learning_rate": 8.379207784630004e-06,
|
| 1939 |
+
"loss": 0.2873,
|
| 1940 |
+
"step": 269
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.013488282054964749,
|
| 1944 |
+
"grad_norm": 0.3936237692832947,
|
| 1945 |
+
"learning_rate": 7.852024322579648e-06,
|
| 1946 |
+
"loss": 0.25,
|
| 1947 |
+
"step": 270
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 0.013538238655168322,
|
| 1951 |
+
"grad_norm": 0.3661133944988251,
|
| 1952 |
+
"learning_rate": 7.34152255572697e-06,
|
| 1953 |
+
"loss": 0.2493,
|
| 1954 |
+
"step": 271
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.013588195255371896,
|
| 1958 |
+
"grad_norm": 0.6841079592704773,
|
| 1959 |
+
"learning_rate": 6.847762393717782e-06,
|
| 1960 |
+
"loss": 0.5608,
|
| 1961 |
+
"step": 272
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.01363815185557547,
|
| 1965 |
+
"grad_norm": 1.142737865447998,
|
| 1966 |
+
"learning_rate": 6.370801781496326e-06,
|
| 1967 |
+
"loss": 0.8011,
|
| 1968 |
+
"step": 273
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.013688108455779042,
|
| 1972 |
+
"grad_norm": 1.2466866970062256,
|
| 1973 |
+
"learning_rate": 5.910696692505201e-06,
|
| 1974 |
+
"loss": 0.651,
|
| 1975 |
+
"step": 274
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.013738065055982615,
|
| 1979 |
+
"grad_norm": 1.065708041191101,
|
| 1980 |
+
"learning_rate": 5.467501122116563e-06,
|
| 1981 |
+
"loss": 0.6143,
|
| 1982 |
+
"step": 275
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.013788021656186188,
|
| 1986 |
+
"grad_norm": 0.997288167476654,
|
| 1987 |
+
"learning_rate": 5.0412670812956465e-06,
|
| 1988 |
+
"loss": 0.5756,
|
| 1989 |
+
"step": 276
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.013837978256389761,
|
| 1993 |
+
"grad_norm": 1.138244390487671,
|
| 1994 |
+
"learning_rate": 4.6320445904969475e-06,
|
| 1995 |
+
"loss": 0.6112,
|
| 1996 |
+
"step": 277
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.013887934856593334,
|
| 2000 |
+
"grad_norm": 1.1244843006134033,
|
| 2001 |
+
"learning_rate": 4.239881673794165e-06,
|
| 2002 |
+
"loss": 0.656,
|
| 2003 |
+
"step": 278
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 0.013937891456796908,
|
| 2007 |
+
"grad_norm": 1.0220907926559448,
|
| 2008 |
+
"learning_rate": 3.864824353244367e-06,
|
| 2009 |
+
"loss": 0.5637,
|
| 2010 |
+
"step": 279
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.013987848057000481,
|
| 2014 |
+
"grad_norm": 0.9114366173744202,
|
| 2015 |
+
"learning_rate": 3.506916643487001e-06,
|
| 2016 |
+
"loss": 0.6459,
|
| 2017 |
+
"step": 280
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.014037804657204054,
|
| 2021 |
+
"grad_norm": 1.131927728652954,
|
| 2022 |
+
"learning_rate": 3.166200546578718e-06,
|
| 2023 |
+
"loss": 0.4715,
|
| 2024 |
+
"step": 281
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 0.014087761257407627,
|
| 2028 |
+
"grad_norm": 1.0005278587341309,
|
| 2029 |
+
"learning_rate": 2.8427160470641253e-06,
|
| 2030 |
+
"loss": 0.4945,
|
| 2031 |
+
"step": 282
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 0.0141377178576112,
|
| 2035 |
+
"grad_norm": 1.0136245489120483,
|
| 2036 |
+
"learning_rate": 2.5365011072835117e-06,
|
| 2037 |
+
"loss": 0.4782,
|
| 2038 |
+
"step": 283
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 0.014187674457814773,
|
| 2042 |
+
"grad_norm": 1.0716521739959717,
|
| 2043 |
+
"learning_rate": 2.2475916629177415e-06,
|
| 2044 |
+
"loss": 0.7236,
|
| 2045 |
+
"step": 284
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 0.014237631058018346,
|
| 2049 |
+
"grad_norm": 1.3851267099380493,
|
| 2050 |
+
"learning_rate": 1.9760216187710787e-06,
|
| 2051 |
+
"loss": 1.4572,
|
| 2052 |
+
"step": 285
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.014287587658221919,
|
| 2056 |
+
"grad_norm": 1.6684229373931885,
|
| 2057 |
+
"learning_rate": 1.7218228447922867e-06,
|
| 2058 |
+
"loss": 1.241,
|
| 2059 |
+
"step": 286
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 0.014337544258425493,
|
| 2063 |
+
"grad_norm": 1.824377417564392,
|
| 2064 |
+
"learning_rate": 1.4850251723345196e-06,
|
| 2065 |
+
"loss": 1.2239,
|
| 2066 |
+
"step": 287
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.014387500858629066,
|
| 2070 |
+
"grad_norm": 1.8666472434997559,
|
| 2071 |
+
"learning_rate": 1.2656563906545902e-06,
|
| 2072 |
+
"loss": 1.0805,
|
| 2073 |
+
"step": 288
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 0.01443745745883264,
|
| 2077 |
+
"grad_norm": 1.4249438047409058,
|
| 2078 |
+
"learning_rate": 1.0637422436516274e-06,
|
| 2079 |
+
"loss": 1.2095,
|
| 2080 |
+
"step": 289
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.014487414059036212,
|
| 2084 |
+
"grad_norm": 1.5718159675598145,
|
| 2085 |
+
"learning_rate": 8.793064268460604e-07,
|
| 2086 |
+
"loss": 1.1586,
|
| 2087 |
+
"step": 290
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 0.014537370659239785,
|
| 2091 |
+
"grad_norm": 1.645261526107788,
|
| 2092 |
+
"learning_rate": 7.123705845987093e-07,
|
| 2093 |
+
"loss": 1.36,
|
| 2094 |
+
"step": 291
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 0.014587327259443358,
|
| 2098 |
+
"grad_norm": 1.381696343421936,
|
| 2099 |
+
"learning_rate": 5.629543075708176e-07,
|
| 2100 |
+
"loss": 0.9215,
|
| 2101 |
+
"step": 292
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.014637283859646931,
|
| 2105 |
+
"grad_norm": 1.696902871131897,
|
| 2106 |
+
"learning_rate": 4.310751304249738e-07,
|
| 2107 |
+
"loss": 1.2197,
|
| 2108 |
+
"step": 293
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 0.014687240459850506,
|
| 2112 |
+
"grad_norm": 1.790809988975525,
|
| 2113 |
+
"learning_rate": 3.167485297673411e-07,
|
| 2114 |
+
"loss": 1.1144,
|
| 2115 |
+
"step": 294
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 0.014737197060054079,
|
| 2119 |
+
"grad_norm": 1.3841776847839355,
|
| 2120 |
+
"learning_rate": 2.1998792233142714e-07,
|
| 2121 |
+
"loss": 1.1018,
|
| 2122 |
+
"step": 295
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 0.014787153660257651,
|
| 2126 |
+
"grad_norm": 1.6458978652954102,
|
| 2127 |
+
"learning_rate": 1.4080466340349316e-07,
|
| 2128 |
+
"loss": 1.0377,
|
| 2129 |
+
"step": 296
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 0.014837110260461224,
|
| 2133 |
+
"grad_norm": 1.813490629196167,
|
| 2134 |
+
"learning_rate": 7.92080454900701e-08,
|
| 2135 |
+
"loss": 1.1065,
|
| 2136 |
+
"step": 297
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 0.014887066860664797,
|
| 2140 |
+
"grad_norm": 1.5537409782409668,
|
| 2141 |
+
"learning_rate": 3.5205297227380855e-08,
|
| 2142 |
+
"loss": 0.9795,
|
| 2143 |
+
"step": 298
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.01493702346086837,
|
| 2147 |
+
"grad_norm": 1.6343427896499634,
|
| 2148 |
+
"learning_rate": 8.801582533035644e-09,
|
| 2149 |
+
"loss": 1.0971,
|
| 2150 |
+
"step": 299
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.014986980061071943,
|
| 2154 |
+
"grad_norm": 2.027508497238159,
|
| 2155 |
+
"learning_rate": 0.0,
|
| 2156 |
+
"loss": 0.9164,
|
| 2157 |
+
"step": 300
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 0.014986980061071943,
|
| 2161 |
+
"eval_loss": 0.5926434993743896,
|
| 2162 |
+
"eval_runtime": 84.603,
|
| 2163 |
+
"eval_samples_per_second": 7.588,
|
| 2164 |
+
"eval_steps_per_second": 7.588,
|
| 2165 |
+
"step": 300
|
| 2166 |
}
|
| 2167 |
],
|
| 2168 |
"logging_steps": 1,
|
|
|
|
| 2177 |
"should_evaluate": false,
|
| 2178 |
"should_log": false,
|
| 2179 |
"should_save": true,
|
| 2180 |
+
"should_training_stop": true
|
| 2181 |
},
|
| 2182 |
"attributes": {}
|
| 2183 |
}
|
| 2184 |
},
|
| 2185 |
+
"total_flos": 4.26901495653335e+16,
|
| 2186 |
"train_batch_size": 1,
|
| 2187 |
"trial_name": null,
|
| 2188 |
"trial_params": null
|