Wilsonwin commited on
Commit
3514c8c
·
verified ·
1 Parent(s): 4636f95

Training in progress, step 3000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eb64855cd768e7aa0f6f46d54fdfee34da708569cc56da521bb521f1101f672a
3
  size 328277848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0040a3a21f9641792c3f89f7d6dfca3f71efa2ba141950721bbcaee769ba9210
3
  size 328277848
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b80bf474cf78a89650cb9274c36cb3774d0d787508f786edf03b70cad49b53c
3
  size 318646859
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df243823a1856f3c4f2211ac8e77f155a41df2c74dca71e8073bb281d83f8fb0
3
  size 318646859
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e85fbbb21d9891a877eaefba7c40e5005f7303b4635375bbc6e0c808069fd11f
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abce81d7290a22f9b260f2e004a835c5fd7f98ca8d48012d38a32b582885319d
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5c5f5054d1fb89b5c064db193ff9ee8b30b57ffe17a11e00d28cfa91ea00081e
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69dd724b4433a1f1a09f5e5486cf5f9b72f5c8cfac628d8213b1ad6d68ac0c2b
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.42236864335191754,
6
  "eval_steps": 500,
7
- "global_step": 2500,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -1805,6 +1805,364 @@
1805
  "eval_samples_per_second": 272.384,
1806
  "eval_steps_per_second": 5.72,
1807
  "step": 2500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1808
  }
1809
  ],
1810
  "logging_steps": 10,
@@ -1824,7 +2182,7 @@
1824
  "attributes": {}
1825
  }
1826
  },
1827
- "total_flos": 8.361420521472e+16,
1828
  "train_batch_size": 48,
1829
  "trial_name": null,
1830
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.5068423720223011,
6
  "eval_steps": 500,
7
+ "global_step": 3000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
1805
  "eval_samples_per_second": 272.384,
1806
  "eval_steps_per_second": 5.72,
1807
  "step": 2500
1808
+ },
1809
+ {
1810
+ "epoch": 0.4240581179253252,
1811
+ "grad_norm": 0.7897567749023438,
1812
+ "learning_rate": 0.0002980229093636812,
1813
+ "loss": 5.12952766418457,
1814
+ "step": 2510
1815
+ },
1816
+ {
1817
+ "epoch": 0.4257475924987329,
1818
+ "grad_norm": 0.7485550045967102,
1819
+ "learning_rate": 0.00029794464059532426,
1820
+ "loss": 5.139109039306641,
1821
+ "step": 2520
1822
+ },
1823
+ {
1824
+ "epoch": 0.4274370670721406,
1825
+ "grad_norm": 0.6715283989906311,
1826
+ "learning_rate": 0.0002978648631891952,
1827
+ "loss": 5.1415565490722654,
1828
+ "step": 2530
1829
+ },
1830
+ {
1831
+ "epoch": 0.42912654164554825,
1832
+ "grad_norm": 0.7465823292732239,
1833
+ "learning_rate": 0.0002977835779588093,
1834
+ "loss": 5.108527374267578,
1835
+ "step": 2540
1836
+ },
1837
+ {
1838
+ "epoch": 0.4308160162189559,
1839
+ "grad_norm": 0.6686388850212097,
1840
+ "learning_rate": 0.0002977007857330575,
1841
+ "loss": 5.125405120849609,
1842
+ "step": 2550
1843
+ },
1844
+ {
1845
+ "epoch": 0.43250549079236356,
1846
+ "grad_norm": 0.6750932335853577,
1847
+ "learning_rate": 0.0002976164873561979,
1848
+ "loss": 5.0938568115234375,
1849
+ "step": 2560
1850
+ },
1851
+ {
1852
+ "epoch": 0.43419496536577123,
1853
+ "grad_norm": 0.680456280708313,
1854
+ "learning_rate": 0.0002975306836878474,
1855
+ "loss": 5.087078475952149,
1856
+ "step": 2570
1857
+ },
1858
+ {
1859
+ "epoch": 0.4358844399391789,
1860
+ "grad_norm": 0.710243821144104,
1861
+ "learning_rate": 0.000297443375602973,
1862
+ "loss": 5.109777450561523,
1863
+ "step": 2580
1864
+ },
1865
+ {
1866
+ "epoch": 0.4375739145125866,
1867
+ "grad_norm": 0.753292977809906,
1868
+ "learning_rate": 0.0002973545639918824,
1869
+ "loss": 5.111831283569336,
1870
+ "step": 2590
1871
+ },
1872
+ {
1873
+ "epoch": 0.43926338908599427,
1874
+ "grad_norm": 0.5997875928878784,
1875
+ "learning_rate": 0.00029726424976021543,
1876
+ "loss": 5.091217803955078,
1877
+ "step": 2600
1878
+ },
1879
+ {
1880
+ "epoch": 0.44095286365940195,
1881
+ "grad_norm": 0.679447591304779,
1882
+ "learning_rate": 0.0002971724338289346,
1883
+ "loss": 5.094500732421875,
1884
+ "step": 2610
1885
+ },
1886
+ {
1887
+ "epoch": 0.4426423382328096,
1888
+ "grad_norm": 0.6854575276374817,
1889
+ "learning_rate": 0.0002970791171343156,
1890
+ "loss": 5.108087158203125,
1891
+ "step": 2620
1892
+ },
1893
+ {
1894
+ "epoch": 0.44433181280621725,
1895
+ "grad_norm": 0.6901696920394897,
1896
+ "learning_rate": 0.000296984300627938,
1897
+ "loss": 5.0770519256591795,
1898
+ "step": 2630
1899
+ },
1900
+ {
1901
+ "epoch": 0.44602128737962493,
1902
+ "grad_norm": 0.6569661498069763,
1903
+ "learning_rate": 0.00029688798527667537,
1904
+ "loss": 5.08624496459961,
1905
+ "step": 2640
1906
+ },
1907
+ {
1908
+ "epoch": 0.4477107619530326,
1909
+ "grad_norm": 0.7255785465240479,
1910
+ "learning_rate": 0.00029679017206268545,
1911
+ "loss": 5.067557907104492,
1912
+ "step": 2650
1913
+ },
1914
+ {
1915
+ "epoch": 0.4494002365264403,
1916
+ "grad_norm": 0.6398410797119141,
1917
+ "learning_rate": 0.00029669086198340014,
1918
+ "loss": 5.078428268432617,
1919
+ "step": 2660
1920
+ },
1921
+ {
1922
+ "epoch": 0.45108971109984797,
1923
+ "grad_norm": 0.8819550275802612,
1924
+ "learning_rate": 0.0002965900560515155,
1925
+ "loss": 5.078685760498047,
1926
+ "step": 2670
1927
+ },
1928
+ {
1929
+ "epoch": 0.4527791856732556,
1930
+ "grad_norm": 0.7163764238357544,
1931
+ "learning_rate": 0.00029648775529498103,
1932
+ "loss": 5.066202545166016,
1933
+ "step": 2680
1934
+ },
1935
+ {
1936
+ "epoch": 0.45446866024666327,
1937
+ "grad_norm": 0.6640359163284302,
1938
+ "learning_rate": 0.00029638396075698953,
1939
+ "loss": 5.063289642333984,
1940
+ "step": 2690
1941
+ },
1942
+ {
1943
+ "epoch": 0.45615813482007095,
1944
+ "grad_norm": 0.6448007225990295,
1945
+ "learning_rate": 0.00029627867349596654,
1946
+ "loss": 5.0232593536376955,
1947
+ "step": 2700
1948
+ },
1949
+ {
1950
+ "epoch": 0.4578476093934786,
1951
+ "grad_norm": 0.683907151222229,
1952
+ "learning_rate": 0.000296171894585559,
1953
+ "loss": 5.0559242248535154,
1954
+ "step": 2710
1955
+ },
1956
+ {
1957
+ "epoch": 0.4595370839668863,
1958
+ "grad_norm": 0.6728157997131348,
1959
+ "learning_rate": 0.00029606362511462494,
1960
+ "loss": 5.069067001342773,
1961
+ "step": 2720
1962
+ },
1963
+ {
1964
+ "epoch": 0.461226558540294,
1965
+ "grad_norm": 0.6807008385658264,
1966
+ "learning_rate": 0.000295953866187222,
1967
+ "loss": 5.054449462890625,
1968
+ "step": 2730
1969
+ },
1970
+ {
1971
+ "epoch": 0.46291603311370166,
1972
+ "grad_norm": 0.708153247833252,
1973
+ "learning_rate": 0.00029584261892259627,
1974
+ "loss": 5.04737434387207,
1975
+ "step": 2740
1976
+ },
1977
+ {
1978
+ "epoch": 0.4646055076871093,
1979
+ "grad_norm": 0.6978984475135803,
1980
+ "learning_rate": 0.00029572988445517094,
1981
+ "loss": 5.033502578735352,
1982
+ "step": 2750
1983
+ },
1984
+ {
1985
+ "epoch": 0.46629498226051697,
1986
+ "grad_norm": 0.5966620445251465,
1987
+ "learning_rate": 0.0002956156639345346,
1988
+ "loss": 5.023481369018555,
1989
+ "step": 2760
1990
+ },
1991
+ {
1992
+ "epoch": 0.46798445683392464,
1993
+ "grad_norm": 0.6365967988967896,
1994
+ "learning_rate": 0.00029549995852542967,
1995
+ "loss": 5.014099502563477,
1996
+ "step": 2770
1997
+ },
1998
+ {
1999
+ "epoch": 0.4696739314073323,
2000
+ "grad_norm": 0.605812668800354,
2001
+ "learning_rate": 0.00029538276940774044,
2002
+ "loss": 5.030900192260742,
2003
+ "step": 2780
2004
+ },
2005
+ {
2006
+ "epoch": 0.47136340598074,
2007
+ "grad_norm": 0.5997155904769897,
2008
+ "learning_rate": 0.0002952640977764808,
2009
+ "loss": 5.025382232666016,
2010
+ "step": 2790
2011
+ },
2012
+ {
2013
+ "epoch": 0.4730528805541477,
2014
+ "grad_norm": 0.6811186075210571,
2015
+ "learning_rate": 0.00029514394484178266,
2016
+ "loss": 5.031198120117187,
2017
+ "step": 2800
2018
+ },
2019
+ {
2020
+ "epoch": 0.4747423551275553,
2021
+ "grad_norm": 0.6368680000305176,
2022
+ "learning_rate": 0.00029502231182888306,
2023
+ "loss": 5.021847152709961,
2024
+ "step": 2810
2025
+ },
2026
+ {
2027
+ "epoch": 0.476431829700963,
2028
+ "grad_norm": 0.62898188829422,
2029
+ "learning_rate": 0.0002948991999781118,
2030
+ "loss": 5.010713958740235,
2031
+ "step": 2820
2032
+ },
2033
+ {
2034
+ "epoch": 0.47812130427437066,
2035
+ "grad_norm": 0.5657271146774292,
2036
+ "learning_rate": 0.000294774610544879,
2037
+ "loss": 5.027156829833984,
2038
+ "step": 2830
2039
+ },
2040
+ {
2041
+ "epoch": 0.47981077884777834,
2042
+ "grad_norm": 0.7271172404289246,
2043
+ "learning_rate": 0.0002946485447996621,
2044
+ "loss": 5.0581409454345705,
2045
+ "step": 2840
2046
+ },
2047
+ {
2048
+ "epoch": 0.481500253421186,
2049
+ "grad_norm": 0.6273407936096191,
2050
+ "learning_rate": 0.0002945210040279928,
2051
+ "loss": 4.975908660888672,
2052
+ "step": 2850
2053
+ },
2054
+ {
2055
+ "epoch": 0.4831897279945937,
2056
+ "grad_norm": 0.6546807289123535,
2057
+ "learning_rate": 0.0002943919895304443,
2058
+ "loss": 4.990020751953125,
2059
+ "step": 2860
2060
+ },
2061
+ {
2062
+ "epoch": 0.4848792025680014,
2063
+ "grad_norm": 0.5948675274848938,
2064
+ "learning_rate": 0.0002942615026226179,
2065
+ "loss": 4.9899650573730465,
2066
+ "step": 2870
2067
+ },
2068
+ {
2069
+ "epoch": 0.486568677141409,
2070
+ "grad_norm": 0.5863410234451294,
2071
+ "learning_rate": 0.0002941295446351292,
2072
+ "loss": 4.98871955871582,
2073
+ "step": 2880
2074
+ },
2075
+ {
2076
+ "epoch": 0.4882581517148167,
2077
+ "grad_norm": 0.5854840278625488,
2078
+ "learning_rate": 0.00029399611691359527,
2079
+ "loss": 4.9739124298095705,
2080
+ "step": 2890
2081
+ },
2082
+ {
2083
+ "epoch": 0.48994762628822436,
2084
+ "grad_norm": 0.6988956928253174,
2085
+ "learning_rate": 0.0002938612208186202,
2086
+ "loss": 4.996525573730469,
2087
+ "step": 2900
2088
+ },
2089
+ {
2090
+ "epoch": 0.49163710086163204,
2091
+ "grad_norm": 0.6269127130508423,
2092
+ "learning_rate": 0.0002937248577257817,
2093
+ "loss": 5.007583236694336,
2094
+ "step": 2910
2095
+ },
2096
+ {
2097
+ "epoch": 0.4933265754350397,
2098
+ "grad_norm": 0.7633141875267029,
2099
+ "learning_rate": 0.0002935870290256169,
2100
+ "loss": 4.988922882080078,
2101
+ "step": 2920
2102
+ },
2103
+ {
2104
+ "epoch": 0.4950160500084474,
2105
+ "grad_norm": 0.6798629760742188,
2106
+ "learning_rate": 0.0002934477361236081,
2107
+ "loss": 4.977730178833008,
2108
+ "step": 2930
2109
+ },
2110
+ {
2111
+ "epoch": 0.496705524581855,
2112
+ "grad_norm": 0.6074113845825195,
2113
+ "learning_rate": 0.0002933069804401687,
2114
+ "loss": 4.997277450561524,
2115
+ "step": 2940
2116
+ },
2117
+ {
2118
+ "epoch": 0.4983949991552627,
2119
+ "grad_norm": 0.6577877402305603,
2120
+ "learning_rate": 0.0002931647634106282,
2121
+ "loss": 4.97070083618164,
2122
+ "step": 2950
2123
+ },
2124
+ {
2125
+ "epoch": 0.5000844737286704,
2126
+ "grad_norm": 0.5998643636703491,
2127
+ "learning_rate": 0.0002930210864852184,
2128
+ "loss": 4.9836891174316404,
2129
+ "step": 2960
2130
+ },
2131
+ {
2132
+ "epoch": 0.501773948302078,
2133
+ "grad_norm": 0.5889402031898499,
2134
+ "learning_rate": 0.00029287595112905773,
2135
+ "loss": 4.966540145874023,
2136
+ "step": 2970
2137
+ },
2138
+ {
2139
+ "epoch": 0.5034634228754857,
2140
+ "grad_norm": 0.627815306186676,
2141
+ "learning_rate": 0.00029272935882213675,
2142
+ "loss": 4.942932891845703,
2143
+ "step": 2980
2144
+ },
2145
+ {
2146
+ "epoch": 0.5051528974488934,
2147
+ "grad_norm": 0.6884767413139343,
2148
+ "learning_rate": 0.00029258131105930314,
2149
+ "loss": 4.950519561767578,
2150
+ "step": 2990
2151
+ },
2152
+ {
2153
+ "epoch": 0.5068423720223011,
2154
+ "grad_norm": 0.613629937171936,
2155
+ "learning_rate": 0.0002924318093502462,
2156
+ "loss": 4.950509643554687,
2157
+ "step": 3000
2158
+ },
2159
+ {
2160
+ "epoch": 0.5068423720223011,
2161
+ "eval_loss": 4.925691604614258,
2162
+ "eval_runtime": 3.6598,
2163
+ "eval_samples_per_second": 273.24,
2164
+ "eval_steps_per_second": 5.738,
2165
+ "step": 3000
2166
  }
2167
  ],
2168
  "logging_steps": 10,
 
2182
  "attributes": {}
2183
  }
2184
  },
2185
+ "total_flos": 1.00337046257664e+17,
2186
  "train_batch_size": 48,
2187
  "trial_name": null,
2188
  "trial_params": null