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:f3c59fb91a81c6ab98b4aabe0a7438b32a58451b9fb9ace348766af89cfda15d
|
| 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 1342555602
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c970254d3306a957828d5c26075003c01e620f518676c48a5253bf97a5dba93b
|
| 3 |
size 1342555602
|
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 sha256:38d169f080c63c9b73e3c1674fc796d1e4dce0c37250952b72a146639cbf7ec8
|
| 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:ec7445dd881aa0e896fdc6ef61cf9c541f4299feaa6850ba7cac238afd6649f3
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 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": 5.532,
|
| 1806 |
"eval_steps_per_second": 2.766,
|
| 1807 |
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1808 |
}
|
| 1809 |
],
|
| 1810 |
"logging_steps": 1,
|
|
@@ -1828,12 +2186,12 @@
|
|
| 1828 |
"should_evaluate": false,
|
| 1829 |
"should_log": false,
|
| 1830 |
"should_save": true,
|
| 1831 |
-
"should_training_stop":
|
| 1832 |
},
|
| 1833 |
"attributes": {}
|
| 1834 |
}
|
| 1835 |
},
|
| 1836 |
-
"total_flos":
|
| 1837 |
"train_batch_size": 8,
|
| 1838 |
"trial_name": null,
|
| 1839 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.530606210231781,
|
| 3 |
+
"best_model_checkpoint": "miner_id_24/checkpoint-300",
|
| 4 |
+
"epoch": 0.068575347162695,
|
| 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": 5.532,
|
| 1806 |
"eval_steps_per_second": 2.766,
|
| 1807 |
"step": 250
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 0.05737470712612149,
|
| 1811 |
+
"grad_norm": 31.539043426513672,
|
| 1812 |
+
"learning_rate": 7.908753070237123e-06,
|
| 1813 |
+
"loss": 4.1344,
|
| 1814 |
+
"step": 251
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.05760329161666381,
|
| 1818 |
+
"grad_norm": 10.60650634765625,
|
| 1819 |
+
"learning_rate": 7.597595192178702e-06,
|
| 1820 |
+
"loss": 3.037,
|
| 1821 |
+
"step": 252
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.05783187610720612,
|
| 1825 |
+
"grad_norm": 5.5060715675354,
|
| 1826 |
+
"learning_rate": 7.292177920973725e-06,
|
| 1827 |
+
"loss": 2.9713,
|
| 1828 |
+
"step": 253
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.05806046059774844,
|
| 1832 |
+
"grad_norm": 5.154425144195557,
|
| 1833 |
+
"learning_rate": 6.992542605231739e-06,
|
| 1834 |
+
"loss": 2.2515,
|
| 1835 |
+
"step": 254
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 0.05828904508829076,
|
| 1839 |
+
"grad_norm": 5.6335129737854,
|
| 1840 |
+
"learning_rate": 6.698729810778065e-06,
|
| 1841 |
+
"loss": 2.3608,
|
| 1842 |
+
"step": 255
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.05851762957883307,
|
| 1846 |
+
"grad_norm": 6.264554023742676,
|
| 1847 |
+
"learning_rate": 6.410779315161886e-06,
|
| 1848 |
+
"loss": 2.2597,
|
| 1849 |
+
"step": 256
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.05874621406937539,
|
| 1853 |
+
"grad_norm": 9.82129955291748,
|
| 1854 |
+
"learning_rate": 6.128730102270897e-06,
|
| 1855 |
+
"loss": 1.9233,
|
| 1856 |
+
"step": 257
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.05897479855991771,
|
| 1860 |
+
"grad_norm": 12.5000581741333,
|
| 1861 |
+
"learning_rate": 5.852620357053651e-06,
|
| 1862 |
+
"loss": 2.1487,
|
| 1863 |
+
"step": 258
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 0.05920338305046002,
|
| 1867 |
+
"grad_norm": 17.345787048339844,
|
| 1868 |
+
"learning_rate": 5.582487460349805e-06,
|
| 1869 |
+
"loss": 2.1851,
|
| 1870 |
+
"step": 259
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.05943196754100234,
|
| 1874 |
+
"grad_norm": 15.88868236541748,
|
| 1875 |
+
"learning_rate": 5.318367983829392e-06,
|
| 1876 |
+
"loss": 0.7703,
|
| 1877 |
+
"step": 260
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.05966055203154466,
|
| 1881 |
+
"grad_norm": 18.099214553833008,
|
| 1882 |
+
"learning_rate": 5.060297685041659e-06,
|
| 1883 |
+
"loss": 0.5869,
|
| 1884 |
+
"step": 261
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.05988913652208697,
|
| 1888 |
+
"grad_norm": 11.634132385253906,
|
| 1889 |
+
"learning_rate": 4.8083115025739756e-06,
|
| 1890 |
+
"loss": 0.375,
|
| 1891 |
+
"step": 262
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.06011772101262929,
|
| 1895 |
+
"grad_norm": 12.680456161499023,
|
| 1896 |
+
"learning_rate": 4.562443551321788e-06,
|
| 1897 |
+
"loss": 0.6281,
|
| 1898 |
+
"step": 263
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.06034630550317161,
|
| 1902 |
+
"grad_norm": 10.923900604248047,
|
| 1903 |
+
"learning_rate": 4.322727117869951e-06,
|
| 1904 |
+
"loss": 0.5998,
|
| 1905 |
+
"step": 264
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 0.06057488999371392,
|
| 1909 |
+
"grad_norm": 33.76637268066406,
|
| 1910 |
+
"learning_rate": 4.089194655986306e-06,
|
| 1911 |
+
"loss": 2.3875,
|
| 1912 |
+
"step": 265
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 0.06080347448425624,
|
| 1916 |
+
"grad_norm": 29.129016876220703,
|
| 1917 |
+
"learning_rate": 3.861877782227885e-06,
|
| 1918 |
+
"loss": 2.1657,
|
| 1919 |
+
"step": 266
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.06103205897479856,
|
| 1923 |
+
"grad_norm": 25.44561767578125,
|
| 1924 |
+
"learning_rate": 3.6408072716606346e-06,
|
| 1925 |
+
"loss": 2.1342,
|
| 1926 |
+
"step": 267
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.061260643465340874,
|
| 1930 |
+
"grad_norm": 29.16153335571289,
|
| 1931 |
+
"learning_rate": 3.426013053692878e-06,
|
| 1932 |
+
"loss": 1.9011,
|
| 1933 |
+
"step": 268
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.06148922795588319,
|
| 1937 |
+
"grad_norm": 27.153125762939453,
|
| 1938 |
+
"learning_rate": 3.2175242080234313e-06,
|
| 1939 |
+
"loss": 1.852,
|
| 1940 |
+
"step": 269
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.06171781244642551,
|
| 1944 |
+
"grad_norm": 23.434755325317383,
|
| 1945 |
+
"learning_rate": 3.0153689607045845e-06,
|
| 1946 |
+
"loss": 2.194,
|
| 1947 |
+
"step": 270
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 0.061946396936967824,
|
| 1951 |
+
"grad_norm": 31.765398025512695,
|
| 1952 |
+
"learning_rate": 2.8195746803208244e-06,
|
| 1953 |
+
"loss": 2.348,
|
| 1954 |
+
"step": 271
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.06217498142751014,
|
| 1958 |
+
"grad_norm": 20.402849197387695,
|
| 1959 |
+
"learning_rate": 2.63016787428354e-06,
|
| 1960 |
+
"loss": 2.2186,
|
| 1961 |
+
"step": 272
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.06240356591805246,
|
| 1965 |
+
"grad_norm": 43.326622009277344,
|
| 1966 |
+
"learning_rate": 2.4471741852423237e-06,
|
| 1967 |
+
"loss": 1.9633,
|
| 1968 |
+
"step": 273
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.06263215040859478,
|
| 1972 |
+
"grad_norm": 55.71512222290039,
|
| 1973 |
+
"learning_rate": 2.2706183876134045e-06,
|
| 1974 |
+
"loss": 2.5231,
|
| 1975 |
+
"step": 274
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.0628607348991371,
|
| 1979 |
+
"grad_norm": 34.0222282409668,
|
| 1980 |
+
"learning_rate": 2.100524384225555e-06,
|
| 1981 |
+
"loss": 2.5434,
|
| 1982 |
+
"step": 275
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.0630893193896794,
|
| 1986 |
+
"grad_norm": 30.28105354309082,
|
| 1987 |
+
"learning_rate": 1.9369152030840556e-06,
|
| 1988 |
+
"loss": 2.2744,
|
| 1989 |
+
"step": 276
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.06331790388022172,
|
| 1993 |
+
"grad_norm": 70.08159637451172,
|
| 1994 |
+
"learning_rate": 1.7798129942530551e-06,
|
| 1995 |
+
"loss": 2.8638,
|
| 1996 |
+
"step": 277
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.06354648837076404,
|
| 2000 |
+
"grad_norm": 30.30040740966797,
|
| 2001 |
+
"learning_rate": 1.6292390268568104e-06,
|
| 2002 |
+
"loss": 2.3145,
|
| 2003 |
+
"step": 278
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 0.06377507286130636,
|
| 2007 |
+
"grad_norm": 12.503889083862305,
|
| 2008 |
+
"learning_rate": 1.4852136862001764e-06,
|
| 2009 |
+
"loss": 1.9287,
|
| 2010 |
+
"step": 279
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.06400365735184868,
|
| 2014 |
+
"grad_norm": 10.321149826049805,
|
| 2015 |
+
"learning_rate": 1.3477564710088098e-06,
|
| 2016 |
+
"loss": 1.7086,
|
| 2017 |
+
"step": 280
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.064232241842391,
|
| 2021 |
+
"grad_norm": 13.305763244628906,
|
| 2022 |
+
"learning_rate": 1.2168859907892904e-06,
|
| 2023 |
+
"loss": 2.1641,
|
| 2024 |
+
"step": 281
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 0.0644608263329333,
|
| 2028 |
+
"grad_norm": 11.012710571289062,
|
| 2029 |
+
"learning_rate": 1.0926199633097157e-06,
|
| 2030 |
+
"loss": 2.1994,
|
| 2031 |
+
"step": 282
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 0.06468941082347562,
|
| 2035 |
+
"grad_norm": 11.165496826171875,
|
| 2036 |
+
"learning_rate": 9.749752122010346e-07,
|
| 2037 |
+
"loss": 1.8865,
|
| 2038 |
+
"step": 283
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 0.06491799531401794,
|
| 2042 |
+
"grad_norm": 8.645009994506836,
|
| 2043 |
+
"learning_rate": 8.639676646793382e-07,
|
| 2044 |
+
"loss": 1.4106,
|
| 2045 |
+
"step": 284
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 0.06514657980456026,
|
| 2049 |
+
"grad_norm": 12.307551383972168,
|
| 2050 |
+
"learning_rate": 7.596123493895991e-07,
|
| 2051 |
+
"loss": 1.9273,
|
| 2052 |
+
"step": 285
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.06537516429510258,
|
| 2056 |
+
"grad_norm": 11.830700874328613,
|
| 2057 |
+
"learning_rate": 6.61923394371039e-07,
|
| 2058 |
+
"loss": 2.3673,
|
| 2059 |
+
"step": 286
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 0.0656037487856449,
|
| 2063 |
+
"grad_norm": 10.462738037109375,
|
| 2064 |
+
"learning_rate": 5.7091402514442e-07,
|
| 2065 |
+
"loss": 1.6951,
|
| 2066 |
+
"step": 287
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.0658323332761872,
|
| 2070 |
+
"grad_norm": 11.972997665405273,
|
| 2071 |
+
"learning_rate": 4.865965629214819e-07,
|
| 2072 |
+
"loss": 2.0444,
|
| 2073 |
+
"step": 288
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 0.06606091776672952,
|
| 2077 |
+
"grad_norm": 18.008352279663086,
|
| 2078 |
+
"learning_rate": 4.089824229369155e-07,
|
| 2079 |
+
"loss": 2.4141,
|
| 2080 |
+
"step": 289
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.06628950225727184,
|
| 2084 |
+
"grad_norm": 11.071226119995117,
|
| 2085 |
+
"learning_rate": 3.380821129028489e-07,
|
| 2086 |
+
"loss": 1.8623,
|
| 2087 |
+
"step": 290
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 0.06651808674781416,
|
| 2091 |
+
"grad_norm": 15.886430740356445,
|
| 2092 |
+
"learning_rate": 2.7390523158633554e-07,
|
| 2093 |
+
"loss": 2.2248,
|
| 2094 |
+
"step": 291
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 0.06674667123835648,
|
| 2098 |
+
"grad_norm": 14.11369800567627,
|
| 2099 |
+
"learning_rate": 2.1646046750978254e-07,
|
| 2100 |
+
"loss": 3.1426,
|
| 2101 |
+
"step": 292
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.0669752557288988,
|
| 2105 |
+
"grad_norm": 17.461822509765625,
|
| 2106 |
+
"learning_rate": 1.657555977746972e-07,
|
| 2107 |
+
"loss": 3.738,
|
| 2108 |
+
"step": 293
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 0.0672038402194411,
|
| 2112 |
+
"grad_norm": 9.49361801147461,
|
| 2113 |
+
"learning_rate": 1.2179748700879012e-07,
|
| 2114 |
+
"loss": 3.9952,
|
| 2115 |
+
"step": 294
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 0.06743242470998342,
|
| 2119 |
+
"grad_norm": 11.396438598632812,
|
| 2120 |
+
"learning_rate": 8.459208643659122e-08,
|
| 2121 |
+
"loss": 4.4705,
|
| 2122 |
+
"step": 295
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 0.06766100920052574,
|
| 2126 |
+
"grad_norm": 10.36643123626709,
|
| 2127 |
+
"learning_rate": 5.4144433073771707e-08,
|
| 2128 |
+
"loss": 4.337,
|
| 2129 |
+
"step": 296
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 0.06788959369106806,
|
| 2133 |
+
"grad_norm": 11.684181213378906,
|
| 2134 |
+
"learning_rate": 3.04586490452119e-08,
|
| 2135 |
+
"loss": 4.1834,
|
| 2136 |
+
"step": 297
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 0.06811817818161038,
|
| 2140 |
+
"grad_norm": 8.283185958862305,
|
| 2141 |
+
"learning_rate": 1.3537941026914303e-08,
|
| 2142 |
+
"loss": 2.6821,
|
| 2143 |
+
"step": 298
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.0683467626721527,
|
| 2147 |
+
"grad_norm": 10.205315589904785,
|
| 2148 |
+
"learning_rate": 3.384599811889766e-09,
|
| 2149 |
+
"loss": 3.7379,
|
| 2150 |
+
"step": 299
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.068575347162695,
|
| 2154 |
+
"grad_norm": 11.798584938049316,
|
| 2155 |
+
"learning_rate": 0.0,
|
| 2156 |
+
"loss": 3.3263,
|
| 2157 |
+
"step": 300
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 0.068575347162695,
|
| 2161 |
+
"eval_loss": 0.530606210231781,
|
| 2162 |
+
"eval_runtime": 1331.9933,
|
| 2163 |
+
"eval_samples_per_second": 5.532,
|
| 2164 |
+
"eval_steps_per_second": 2.766,
|
| 2165 |
+
"step": 300
|
| 2166 |
}
|
| 2167 |
],
|
| 2168 |
"logging_steps": 1,
|
|
|
|
| 2186 |
"should_evaluate": false,
|
| 2187 |
"should_log": false,
|
| 2188 |
"should_save": true,
|
| 2189 |
+
"should_training_stop": true
|
| 2190 |
},
|
| 2191 |
"attributes": {}
|
| 2192 |
}
|
| 2193 |
},
|
| 2194 |
+
"total_flos": 6.686338196994785e+17,
|
| 2195 |
"train_batch_size": 8,
|
| 2196 |
"trial_name": null,
|
| 2197 |
"trial_params": null
|