Commit
·
d56baf3
1
Parent(s):
d752ca3
Training in progress, step 31000
Browse files- adapter_model.bin +1 -1
- {checkpoint-27000 → checkpoint-30000/adapter_model}/README.md +0 -0
- {checkpoint-27000 → checkpoint-30000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-27000 → checkpoint-30000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-27000/adapter_model → checkpoint-31000}/README.md +0 -0
- {checkpoint-27000/adapter_model → checkpoint-31000}/adapter_config.json +0 -0
- {checkpoint-27000/adapter_model → checkpoint-31000}/adapter_model.bin +1 -1
- {checkpoint-27000 → checkpoint-31000}/optimizer.pt +1 -1
- {checkpoint-27000 → checkpoint-31000}/rng_state.pth +1 -1
- {checkpoint-27000 → checkpoint-31000}/scheduler.pt +1 -1
- {checkpoint-27000 → checkpoint-31000}/trainer_state.json +277 -5
- {checkpoint-27000 → checkpoint-31000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aab37ed994bf1726713cdc1d84b9560a915246c4a4028a868388ed725770a29f
|
| 3 |
size 500897101
|
{checkpoint-27000 → checkpoint-30000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-27000 → checkpoint-30000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-27000 → checkpoint-30000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9537b3a7e6055deff50c06679b380e3a915c90d647696e0b31fc0265c367cbbc
|
| 3 |
size 500897101
|
{checkpoint-27000/adapter_model → checkpoint-31000}/README.md
RENAMED
|
File without changes
|
{checkpoint-27000/adapter_model → checkpoint-31000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-27000/adapter_model → checkpoint-31000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aab37ed994bf1726713cdc1d84b9560a915246c4a4028a868388ed725770a29f
|
| 3 |
size 500897101
|
{checkpoint-27000 → checkpoint-31000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6744b3d4cba5318614d3585f0a827cdfe30350a5eee8ea7ce4ae59a497f8b7d4
|
| 3 |
size 1001723453
|
{checkpoint-27000 → checkpoint-31000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d969796cab52300f6ffd6656a35cf7978e5664a3693183be68bcb0ae74d67a0
|
| 3 |
size 14575
|
{checkpoint-27000 → checkpoint-31000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:328b81c248c0e7203f70f9754413682ae0e11fa74f4beec41dcb9e104d993f01
|
| 3 |
size 627
|
{checkpoint-27000 → checkpoint-31000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1842,11 +1842,283 @@
|
|
| 1842 |
"eval_samples_per_second": 0.412,
|
| 1843 |
"eval_steps_per_second": 0.412,
|
| 1844 |
"step": 27000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1845 |
}
|
| 1846 |
],
|
| 1847 |
"max_steps": 80463,
|
| 1848 |
"num_train_epochs": 3,
|
| 1849 |
-
"total_flos":
|
| 1850 |
"trial_name": null,
|
| 1851 |
"trial_params": null
|
| 1852 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5811149477958679,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-31000",
|
| 4 |
+
"epoch": 1.1558107453115096,
|
| 5 |
+
"global_step": 31000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1842 |
"eval_samples_per_second": 0.412,
|
| 1843 |
"eval_steps_per_second": 0.412,
|
| 1844 |
"step": 27000
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.01,
|
| 1848 |
+
"learning_rate": 0.00014907623379770263,
|
| 1849 |
+
"loss": 0.5743,
|
| 1850 |
+
"step": 27100
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 1.01,
|
| 1854 |
+
"learning_rate": 0.00014873563090433547,
|
| 1855 |
+
"loss": 0.5095,
|
| 1856 |
+
"step": 27200
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 1.02,
|
| 1860 |
+
"learning_rate": 0.00014839428488694706,
|
| 1861 |
+
"loss": 0.5391,
|
| 1862 |
+
"step": 27300
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 1.02,
|
| 1866 |
+
"learning_rate": 0.00014805220095040334,
|
| 1867 |
+
"loss": 0.5532,
|
| 1868 |
+
"step": 27400
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 1.03,
|
| 1872 |
+
"learning_rate": 0.00014770938431082212,
|
| 1873 |
+
"loss": 0.536,
|
| 1874 |
+
"step": 27500
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 1.03,
|
| 1878 |
+
"learning_rate": 0.00014736584019549342,
|
| 1879 |
+
"loss": 0.5204,
|
| 1880 |
+
"step": 27600
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 1.03,
|
| 1884 |
+
"learning_rate": 0.00014702157384279997,
|
| 1885 |
+
"loss": 0.5026,
|
| 1886 |
+
"step": 27700
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.04,
|
| 1890 |
+
"learning_rate": 0.0001466765905021371,
|
| 1891 |
+
"loss": 0.5319,
|
| 1892 |
+
"step": 27800
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 1.04,
|
| 1896 |
+
"learning_rate": 0.00014633089543383295,
|
| 1897 |
+
"loss": 0.5112,
|
| 1898 |
+
"step": 27900
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 1.04,
|
| 1902 |
+
"learning_rate": 0.00014598449390906804,
|
| 1903 |
+
"loss": 0.5146,
|
| 1904 |
+
"step": 28000
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 1.04,
|
| 1908 |
+
"eval_loss": 0.5959522128105164,
|
| 1909 |
+
"eval_runtime": 1288.6066,
|
| 1910 |
+
"eval_samples_per_second": 0.421,
|
| 1911 |
+
"eval_steps_per_second": 0.421,
|
| 1912 |
+
"step": 28000
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 1.05,
|
| 1916 |
+
"learning_rate": 0.00014563739120979497,
|
| 1917 |
+
"loss": 0.5262,
|
| 1918 |
+
"step": 28100
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 1.05,
|
| 1922 |
+
"learning_rate": 0.00014528959262865798,
|
| 1923 |
+
"loss": 0.5082,
|
| 1924 |
+
"step": 28200
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 1.06,
|
| 1928 |
+
"learning_rate": 0.00014494110346891206,
|
| 1929 |
+
"loss": 0.5094,
|
| 1930 |
+
"step": 28300
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 1.06,
|
| 1934 |
+
"learning_rate": 0.00014459192904434226,
|
| 1935 |
+
"loss": 0.5012,
|
| 1936 |
+
"step": 28400
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 1.06,
|
| 1940 |
+
"learning_rate": 0.0001442420746791826,
|
| 1941 |
+
"loss": 0.4946,
|
| 1942 |
+
"step": 28500
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.07,
|
| 1946 |
+
"learning_rate": 0.00014389154570803477,
|
| 1947 |
+
"loss": 0.5138,
|
| 1948 |
+
"step": 28600
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 1.07,
|
| 1952 |
+
"learning_rate": 0.000143540347475787,
|
| 1953 |
+
"loss": 0.5082,
|
| 1954 |
+
"step": 28700
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 1.07,
|
| 1958 |
+
"learning_rate": 0.0001431884853375325,
|
| 1959 |
+
"loss": 0.4842,
|
| 1960 |
+
"step": 28800
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 1.08,
|
| 1964 |
+
"learning_rate": 0.0001428359646584876,
|
| 1965 |
+
"loss": 0.5143,
|
| 1966 |
+
"step": 28900
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 1.08,
|
| 1970 |
+
"learning_rate": 0.00014248279081391022,
|
| 1971 |
+
"loss": 0.5029,
|
| 1972 |
+
"step": 29000
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 1.08,
|
| 1976 |
+
"eval_loss": 0.5910914540290833,
|
| 1977 |
+
"eval_runtime": 1278.8257,
|
| 1978 |
+
"eval_samples_per_second": 0.424,
|
| 1979 |
+
"eval_steps_per_second": 0.424,
|
| 1980 |
+
"step": 29000
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 1.08,
|
| 1984 |
+
"learning_rate": 0.00014212896918901774,
|
| 1985 |
+
"loss": 0.5003,
|
| 1986 |
+
"step": 29100
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 1.09,
|
| 1990 |
+
"learning_rate": 0.00014177450517890503,
|
| 1991 |
+
"loss": 0.5102,
|
| 1992 |
+
"step": 29200
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.09,
|
| 1996 |
+
"learning_rate": 0.0001414194041884619,
|
| 1997 |
+
"loss": 0.524,
|
| 1998 |
+
"step": 29300
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.1,
|
| 2002 |
+
"learning_rate": 0.0001410636716322911,
|
| 2003 |
+
"loss": 0.5168,
|
| 2004 |
+
"step": 29400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 1.1,
|
| 2008 |
+
"learning_rate": 0.0001407073129346254,
|
| 2009 |
+
"loss": 0.514,
|
| 2010 |
+
"step": 29500
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 1.1,
|
| 2014 |
+
"learning_rate": 0.00014035033352924502,
|
| 2015 |
+
"loss": 0.5084,
|
| 2016 |
+
"step": 29600
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 1.11,
|
| 2020 |
+
"learning_rate": 0.0001399927388593948,
|
| 2021 |
+
"loss": 0.5203,
|
| 2022 |
+
"step": 29700
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 1.11,
|
| 2026 |
+
"learning_rate": 0.00013963453437770119,
|
| 2027 |
+
"loss": 0.5226,
|
| 2028 |
+
"step": 29800
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 1.11,
|
| 2032 |
+
"learning_rate": 0.000139275725546089,
|
| 2033 |
+
"loss": 0.5055,
|
| 2034 |
+
"step": 29900
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 1.12,
|
| 2038 |
+
"learning_rate": 0.00013891631783569838,
|
| 2039 |
+
"loss": 0.5303,
|
| 2040 |
+
"step": 30000
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.12,
|
| 2044 |
+
"eval_loss": 0.5869857668876648,
|
| 2045 |
+
"eval_runtime": 1272.0282,
|
| 2046 |
+
"eval_samples_per_second": 0.426,
|
| 2047 |
+
"eval_steps_per_second": 0.426,
|
| 2048 |
+
"step": 30000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.12,
|
| 2052 |
+
"learning_rate": 0.00013855631672680106,
|
| 2053 |
+
"loss": 0.5243,
|
| 2054 |
+
"step": 30100
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.13,
|
| 2058 |
+
"learning_rate": 0.00013819572770871702,
|
| 2059 |
+
"loss": 0.5148,
|
| 2060 |
+
"step": 30200
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 1.13,
|
| 2064 |
+
"learning_rate": 0.00013783455627973062,
|
| 2065 |
+
"loss": 0.522,
|
| 2066 |
+
"step": 30300
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 1.13,
|
| 2070 |
+
"learning_rate": 0.00013747280794700707,
|
| 2071 |
+
"loss": 0.5289,
|
| 2072 |
+
"step": 30400
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.14,
|
| 2076 |
+
"learning_rate": 0.00013711048822650802,
|
| 2077 |
+
"loss": 0.4996,
|
| 2078 |
+
"step": 30500
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 1.14,
|
| 2082 |
+
"learning_rate": 0.00013674760264290785,
|
| 2083 |
+
"loss": 0.5099,
|
| 2084 |
+
"step": 30600
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 1.14,
|
| 2088 |
+
"learning_rate": 0.0001363841567295091,
|
| 2089 |
+
"loss": 0.5219,
|
| 2090 |
+
"step": 30700
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.15,
|
| 2094 |
+
"learning_rate": 0.00013602015602815837,
|
| 2095 |
+
"loss": 0.5297,
|
| 2096 |
+
"step": 30800
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.15,
|
| 2100 |
+
"learning_rate": 0.00013565560608916165,
|
| 2101 |
+
"loss": 0.5029,
|
| 2102 |
+
"step": 30900
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 1.16,
|
| 2106 |
+
"learning_rate": 0.0001352905124711998,
|
| 2107 |
+
"loss": 0.5266,
|
| 2108 |
+
"step": 31000
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 1.16,
|
| 2112 |
+
"eval_loss": 0.5811149477958679,
|
| 2113 |
+
"eval_runtime": 1300.1475,
|
| 2114 |
+
"eval_samples_per_second": 0.417,
|
| 2115 |
+
"eval_steps_per_second": 0.417,
|
| 2116 |
+
"step": 31000
|
| 2117 |
}
|
| 2118 |
],
|
| 2119 |
"max_steps": 80463,
|
| 2120 |
"num_train_epochs": 3,
|
| 2121 |
+
"total_flos": 8.693892403678986e+18,
|
| 2122 |
"trial_name": null,
|
| 2123 |
"trial_params": null
|
| 2124 |
}
|
{checkpoint-27000 → checkpoint-31000}/training_args.bin
RENAMED
|
File without changes
|