End of training
Browse files- README.md +6 -4
- all_results.json +11 -11
- eval_results.json +7 -7
- runs/Jan16_14-53-45_vorace/events.out.tfevents.1705423620.vorace.510473.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +1201 -213
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: mit
|
| 3 |
base_model: facebook/w2v-bert-2.0
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- wer
|
|
@@ -15,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# wav2vec2-bert-CV16-en-libri
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Wer: 0.
|
| 22 |
-
- Cer: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 2 |
license: mit
|
| 3 |
base_model: facebook/w2v-bert-2.0
|
| 4 |
tags:
|
| 5 |
+
- automatic-speech-recognition
|
| 6 |
+
- librispeech_asr
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- wer
|
|
|
|
| 17 |
|
| 18 |
# wav2vec2-bert-CV16-en-libri
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the LIBRISPEECH_ASR - CLEAN dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.1035
|
| 23 |
+
- Wer: 0.0708
|
| 24 |
+
- Cer: 0.0194
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 59.
|
| 6 |
"eval_samples": 2528,
|
| 7 |
-
"eval_samples_per_second": 42.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 0.
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 28538,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.99,
|
| 3 |
+
"eval_cer": 0.019377383733700754,
|
| 4 |
+
"eval_loss": 0.10354145616292953,
|
| 5 |
+
"eval_runtime": 59.6491,
|
| 6 |
"eval_samples": 2528,
|
| 7 |
+
"eval_samples_per_second": 42.381,
|
| 8 |
+
"eval_steps_per_second": 1.19,
|
| 9 |
+
"eval_wer": 0.07078875870266599,
|
| 10 |
+
"train_loss": 0.05688602840165039,
|
| 11 |
+
"train_runtime": 6445.258,
|
| 12 |
"train_samples": 28538,
|
| 13 |
+
"train_samples_per_second": 30.994,
|
| 14 |
+
"train_steps_per_second": 0.43
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 59.
|
| 6 |
"eval_samples": 2528,
|
| 7 |
-
"eval_samples_per_second": 42.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 0.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.99,
|
| 3 |
+
"eval_cer": 0.019377383733700754,
|
| 4 |
+
"eval_loss": 0.10354145616292953,
|
| 5 |
+
"eval_runtime": 59.6491,
|
| 6 |
"eval_samples": 2528,
|
| 7 |
+
"eval_samples_per_second": 42.381,
|
| 8 |
+
"eval_steps_per_second": 1.19,
|
| 9 |
+
"eval_wer": 0.07078875870266599
|
| 10 |
}
|
runs/Jan16_14-53-45_vorace/events.out.tfevents.1705423620.vorace.510473.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95954bbf24073745de4bd78f01796d56e41c36f6ab40add343e5fee7d55cc556
|
| 3 |
+
size 405
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 28538,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.99,
|
| 3 |
+
"train_loss": 0.05688602840165039,
|
| 4 |
+
"train_runtime": 6445.258,
|
| 5 |
"train_samples": 28538,
|
| 6 |
+
"train_samples_per_second": 30.994,
|
| 7 |
+
"train_steps_per_second": 0.43
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 250,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1871,605 +1871,1593 @@
|
|
| 1871 |
{
|
| 1872 |
"epoch": 3.8,
|
| 1873 |
"learning_rate": 4.497e-06,
|
| 1874 |
-
"loss": 0.
|
| 1875 |
"step": 1505
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
"epoch": 3.81,
|
| 1879 |
-
"learning_rate": 4.
|
| 1880 |
-
"loss": 0.
|
| 1881 |
"step": 1510
|
| 1882 |
},
|
| 1883 |
{
|
| 1884 |
"epoch": 3.82,
|
| 1885 |
-
"learning_rate": 4.
|
| 1886 |
-
"loss": 0.
|
| 1887 |
"step": 1515
|
| 1888 |
},
|
| 1889 |
{
|
| 1890 |
"epoch": 3.84,
|
| 1891 |
-
"learning_rate": 4.
|
| 1892 |
-
"loss": 0.
|
| 1893 |
"step": 1520
|
| 1894 |
},
|
| 1895 |
{
|
| 1896 |
"epoch": 3.85,
|
| 1897 |
-
"learning_rate": 4.
|
| 1898 |
-
"loss": 0.
|
| 1899 |
"step": 1525
|
| 1900 |
},
|
| 1901 |
{
|
| 1902 |
"epoch": 3.86,
|
| 1903 |
-
"learning_rate": 4.
|
| 1904 |
-
"loss": 0.
|
| 1905 |
"step": 1530
|
| 1906 |
},
|
| 1907 |
{
|
| 1908 |
-
"epoch": 3.
|
| 1909 |
-
"learning_rate": 4.
|
| 1910 |
-
"loss": 0.
|
| 1911 |
"step": 1535
|
| 1912 |
},
|
| 1913 |
{
|
| 1914 |
"epoch": 3.89,
|
| 1915 |
"learning_rate": 4.5989999999999995e-06,
|
| 1916 |
-
"loss": 0.
|
| 1917 |
"step": 1540
|
| 1918 |
},
|
| 1919 |
{
|
| 1920 |
"epoch": 3.9,
|
| 1921 |
"learning_rate": 4.614e-06,
|
| 1922 |
-
"loss": 0.
|
| 1923 |
"step": 1545
|
| 1924 |
},
|
| 1925 |
{
|
| 1926 |
"epoch": 3.91,
|
| 1927 |
"learning_rate": 4.629e-06,
|
| 1928 |
-
"loss": 0.
|
| 1929 |
"step": 1550
|
| 1930 |
},
|
| 1931 |
{
|
| 1932 |
-
"epoch": 3.
|
| 1933 |
"learning_rate": 4.644e-06,
|
| 1934 |
-
"loss": 0.
|
| 1935 |
"step": 1555
|
| 1936 |
},
|
| 1937 |
{
|
| 1938 |
"epoch": 3.94,
|
| 1939 |
-
"learning_rate": 4.
|
| 1940 |
-
"loss": 0.
|
| 1941 |
"step": 1560
|
| 1942 |
},
|
| 1943 |
{
|
| 1944 |
"epoch": 3.95,
|
| 1945 |
-
"learning_rate": 4.
|
| 1946 |
-
"loss": 0.
|
| 1947 |
"step": 1565
|
| 1948 |
},
|
| 1949 |
{
|
| 1950 |
"epoch": 3.96,
|
| 1951 |
-
"learning_rate": 4.
|
| 1952 |
-
"loss": 0.
|
| 1953 |
"step": 1570
|
| 1954 |
},
|
| 1955 |
{
|
| 1956 |
-
"epoch": 3.
|
| 1957 |
-
"learning_rate": 4.
|
| 1958 |
-
"loss": 0.
|
| 1959 |
"step": 1575
|
| 1960 |
},
|
| 1961 |
{
|
| 1962 |
"epoch": 3.99,
|
| 1963 |
-
"learning_rate": 4.
|
| 1964 |
-
"loss": 0.
|
| 1965 |
"step": 1580
|
| 1966 |
},
|
| 1967 |
{
|
| 1968 |
"epoch": 4.0,
|
| 1969 |
-
"learning_rate": 4.
|
| 1970 |
-
"loss": 0.
|
| 1971 |
"step": 1585
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"epoch": 4.01,
|
| 1975 |
-
"learning_rate": 4.
|
| 1976 |
-
"loss": 0.
|
| 1977 |
"step": 1590
|
| 1978 |
},
|
| 1979 |
{
|
| 1980 |
"epoch": 4.03,
|
| 1981 |
-
"learning_rate": 4.
|
| 1982 |
-
"loss": 0.
|
| 1983 |
"step": 1595
|
| 1984 |
},
|
| 1985 |
{
|
| 1986 |
"epoch": 4.04,
|
| 1987 |
-
"learning_rate": 4.
|
| 1988 |
-
"loss": 0.
|
| 1989 |
"step": 1600
|
| 1990 |
},
|
| 1991 |
{
|
| 1992 |
"epoch": 4.05,
|
| 1993 |
-
"learning_rate": 4.
|
| 1994 |
-
"loss": 0.
|
| 1995 |
"step": 1605
|
| 1996 |
},
|
| 1997 |
{
|
| 1998 |
"epoch": 4.06,
|
| 1999 |
-
"learning_rate": 4.
|
| 2000 |
-
"loss": 0.
|
| 2001 |
"step": 1610
|
| 2002 |
},
|
| 2003 |
{
|
| 2004 |
"epoch": 4.08,
|
| 2005 |
-
"learning_rate": 4.
|
| 2006 |
-
"loss": 0.
|
| 2007 |
"step": 1615
|
| 2008 |
},
|
| 2009 |
{
|
| 2010 |
"epoch": 4.09,
|
| 2011 |
-
"learning_rate": 4.
|
| 2012 |
-
"loss": 0.
|
| 2013 |
"step": 1620
|
| 2014 |
},
|
| 2015 |
{
|
| 2016 |
"epoch": 4.1,
|
| 2017 |
-
"learning_rate": 4.
|
| 2018 |
-
"loss": 0.
|
| 2019 |
"step": 1625
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
"epoch": 4.11,
|
| 2023 |
-
"learning_rate": 4.
|
| 2024 |
-
"loss": 0.
|
| 2025 |
"step": 1630
|
| 2026 |
},
|
| 2027 |
{
|
| 2028 |
"epoch": 4.13,
|
| 2029 |
-
"learning_rate": 4.
|
| 2030 |
-
"loss": 0.
|
| 2031 |
"step": 1635
|
| 2032 |
},
|
| 2033 |
{
|
| 2034 |
"epoch": 4.14,
|
| 2035 |
-
"learning_rate": 4.
|
| 2036 |
-
"loss": 0.
|
| 2037 |
"step": 1640
|
| 2038 |
},
|
| 2039 |
{
|
| 2040 |
"epoch": 4.15,
|
| 2041 |
-
"learning_rate": 4.
|
| 2042 |
-
"loss": 0.
|
| 2043 |
"step": 1645
|
| 2044 |
},
|
| 2045 |
{
|
| 2046 |
-
"epoch": 4.
|
| 2047 |
-
"learning_rate": 4.
|
| 2048 |
-
"loss": 0.
|
| 2049 |
"step": 1650
|
| 2050 |
},
|
| 2051 |
{
|
| 2052 |
"epoch": 4.18,
|
| 2053 |
-
"learning_rate": 4.
|
| 2054 |
-
"loss": 0.
|
| 2055 |
"step": 1655
|
| 2056 |
},
|
| 2057 |
{
|
| 2058 |
"epoch": 4.19,
|
| 2059 |
-
"learning_rate": 4.
|
| 2060 |
-
"loss": 0.
|
| 2061 |
"step": 1660
|
| 2062 |
},
|
| 2063 |
{
|
| 2064 |
"epoch": 4.2,
|
| 2065 |
-
"learning_rate": 4.
|
| 2066 |
-
"loss": 0.
|
| 2067 |
"step": 1665
|
| 2068 |
},
|
| 2069 |
{
|
| 2070 |
-
"epoch": 4.
|
| 2071 |
-
"learning_rate": 4.
|
| 2072 |
-
"loss": 0.
|
| 2073 |
"step": 1670
|
| 2074 |
},
|
| 2075 |
{
|
| 2076 |
"epoch": 4.23,
|
| 2077 |
-
"learning_rate": 5.
|
| 2078 |
-
"loss": 0.
|
| 2079 |
"step": 1675
|
| 2080 |
},
|
| 2081 |
{
|
| 2082 |
"epoch": 4.24,
|
| 2083 |
-
"learning_rate": 5.
|
| 2084 |
-
"loss": 0.
|
| 2085 |
"step": 1680
|
| 2086 |
},
|
| 2087 |
{
|
| 2088 |
"epoch": 4.25,
|
| 2089 |
-
"learning_rate": 5.
|
| 2090 |
-
"loss": 0.
|
| 2091 |
"step": 1685
|
| 2092 |
},
|
| 2093 |
{
|
| 2094 |
-
"epoch": 4.
|
| 2095 |
-
"learning_rate": 5.
|
| 2096 |
-
"loss": 0.
|
| 2097 |
"step": 1690
|
| 2098 |
},
|
| 2099 |
{
|
| 2100 |
"epoch": 4.28,
|
| 2101 |
-
"learning_rate": 5.
|
| 2102 |
-
"loss": 0.
|
| 2103 |
"step": 1695
|
| 2104 |
},
|
| 2105 |
{
|
| 2106 |
"epoch": 4.29,
|
| 2107 |
-
"learning_rate": 5.
|
| 2108 |
-
"loss": 0.
|
| 2109 |
"step": 1700
|
| 2110 |
},
|
| 2111 |
{
|
| 2112 |
"epoch": 4.3,
|
| 2113 |
-
"learning_rate": 5.
|
| 2114 |
-
"loss": 0.
|
| 2115 |
"step": 1705
|
| 2116 |
},
|
| 2117 |
{
|
| 2118 |
"epoch": 4.32,
|
| 2119 |
-
"learning_rate": 5.
|
| 2120 |
-
"loss": 0.
|
| 2121 |
"step": 1710
|
| 2122 |
},
|
| 2123 |
{
|
| 2124 |
"epoch": 4.33,
|
| 2125 |
-
"learning_rate": 5.
|
| 2126 |
-
"loss": 0.
|
| 2127 |
"step": 1715
|
| 2128 |
},
|
| 2129 |
{
|
| 2130 |
"epoch": 4.34,
|
| 2131 |
-
"learning_rate": 5.
|
| 2132 |
-
"loss": 0.
|
| 2133 |
"step": 1720
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"epoch": 4.35,
|
| 2137 |
-
"learning_rate": 5.
|
| 2138 |
-
"loss": 0.
|
| 2139 |
"step": 1725
|
| 2140 |
},
|
| 2141 |
{
|
| 2142 |
"epoch": 4.37,
|
| 2143 |
-
"learning_rate": 5.
|
| 2144 |
-
"loss": 0.
|
| 2145 |
"step": 1730
|
| 2146 |
},
|
| 2147 |
{
|
| 2148 |
"epoch": 4.38,
|
| 2149 |
-
"learning_rate": 5.
|
| 2150 |
-
"loss": 0.
|
| 2151 |
"step": 1735
|
| 2152 |
},
|
| 2153 |
{
|
| 2154 |
"epoch": 4.39,
|
| 2155 |
-
"learning_rate": 5.
|
| 2156 |
-
"loss": 0.
|
| 2157 |
"step": 1740
|
| 2158 |
},
|
| 2159 |
{
|
| 2160 |
"epoch": 4.4,
|
| 2161 |
-
"learning_rate": 5.
|
| 2162 |
-
"loss": 0.
|
| 2163 |
"step": 1745
|
| 2164 |
},
|
| 2165 |
{
|
| 2166 |
"epoch": 4.42,
|
| 2167 |
-
"learning_rate": 5.
|
| 2168 |
-
"loss": 0.
|
| 2169 |
"step": 1750
|
| 2170 |
},
|
| 2171 |
{
|
| 2172 |
"epoch": 4.42,
|
| 2173 |
-
"eval_cer": 0.
|
| 2174 |
-
"eval_loss": 0.
|
| 2175 |
-
"eval_runtime": 60.
|
| 2176 |
-
"eval_samples_per_second": 41.
|
| 2177 |
-
"eval_steps_per_second": 1.
|
| 2178 |
-
"eval_wer": 0.
|
| 2179 |
"step": 1750
|
| 2180 |
},
|
| 2181 |
{
|
| 2182 |
"epoch": 4.43,
|
| 2183 |
-
"learning_rate": 5.
|
| 2184 |
-
"loss": 0.
|
| 2185 |
"step": 1755
|
| 2186 |
},
|
| 2187 |
{
|
| 2188 |
"epoch": 4.44,
|
| 2189 |
-
"learning_rate": 5.
|
| 2190 |
"loss": 0.1444,
|
| 2191 |
"step": 1760
|
| 2192 |
},
|
| 2193 |
{
|
| 2194 |
-
"epoch": 4.
|
| 2195 |
-
"learning_rate": 5.
|
| 2196 |
-
"loss": 0.
|
| 2197 |
"step": 1765
|
| 2198 |
},
|
| 2199 |
{
|
| 2200 |
"epoch": 4.47,
|
| 2201 |
-
"learning_rate": 5.
|
| 2202 |
-
"loss": 0.
|
| 2203 |
"step": 1770
|
| 2204 |
},
|
| 2205 |
{
|
| 2206 |
"epoch": 4.48,
|
| 2207 |
-
"learning_rate": 5.
|
| 2208 |
-
"loss": 0.
|
| 2209 |
"step": 1775
|
| 2210 |
},
|
| 2211 |
{
|
| 2212 |
"epoch": 4.49,
|
| 2213 |
-
"learning_rate": 5.
|
| 2214 |
-
"loss": 0.
|
| 2215 |
"step": 1780
|
| 2216 |
},
|
| 2217 |
{
|
| 2218 |
-
"epoch": 4.
|
| 2219 |
-
"learning_rate": 5.
|
| 2220 |
-
"loss": 0.
|
| 2221 |
"step": 1785
|
| 2222 |
},
|
| 2223 |
{
|
| 2224 |
"epoch": 4.52,
|
| 2225 |
-
"learning_rate": 5.
|
| 2226 |
-
"loss": 0.
|
| 2227 |
"step": 1790
|
| 2228 |
},
|
| 2229 |
{
|
| 2230 |
"epoch": 4.53,
|
| 2231 |
-
"learning_rate": 5.
|
| 2232 |
-
"loss": 0.
|
| 2233 |
"step": 1795
|
| 2234 |
},
|
| 2235 |
{
|
| 2236 |
"epoch": 4.54,
|
| 2237 |
-
"learning_rate": 5.
|
| 2238 |
-
"loss": 0.
|
| 2239 |
"step": 1800
|
| 2240 |
},
|
| 2241 |
{
|
| 2242 |
-
"epoch": 4.
|
| 2243 |
-
"learning_rate": 5.
|
| 2244 |
-
"loss": 0.
|
| 2245 |
"step": 1805
|
| 2246 |
},
|
| 2247 |
{
|
| 2248 |
"epoch": 4.57,
|
| 2249 |
-
"learning_rate": 5.
|
| 2250 |
-
"loss": 0.
|
| 2251 |
"step": 1810
|
| 2252 |
},
|
| 2253 |
{
|
| 2254 |
"epoch": 4.58,
|
| 2255 |
-
"learning_rate": 5.
|
| 2256 |
-
"loss": 0.
|
| 2257 |
"step": 1815
|
| 2258 |
},
|
| 2259 |
{
|
| 2260 |
"epoch": 4.59,
|
| 2261 |
-
"learning_rate": 5.
|
| 2262 |
-
"loss": 0.
|
| 2263 |
"step": 1820
|
| 2264 |
},
|
| 2265 |
{
|
| 2266 |
"epoch": 4.61,
|
| 2267 |
-
"learning_rate": 5.
|
| 2268 |
-
"loss": 0.
|
| 2269 |
"step": 1825
|
| 2270 |
},
|
| 2271 |
{
|
| 2272 |
"epoch": 4.62,
|
| 2273 |
-
"learning_rate": 5.
|
| 2274 |
-
"loss": 0.
|
| 2275 |
"step": 1830
|
| 2276 |
},
|
| 2277 |
{
|
| 2278 |
"epoch": 4.63,
|
| 2279 |
-
"learning_rate": 5.
|
| 2280 |
-
"loss": 0.
|
| 2281 |
"step": 1835
|
| 2282 |
},
|
| 2283 |
{
|
| 2284 |
"epoch": 4.64,
|
| 2285 |
-
"learning_rate": 5.
|
| 2286 |
-
"loss": 0.
|
| 2287 |
"step": 1840
|
| 2288 |
},
|
| 2289 |
{
|
| 2290 |
"epoch": 4.66,
|
| 2291 |
-
"learning_rate": 5.
|
| 2292 |
-
"loss": 0.
|
| 2293 |
"step": 1845
|
| 2294 |
},
|
| 2295 |
{
|
| 2296 |
"epoch": 4.67,
|
| 2297 |
-
"learning_rate": 5.
|
| 2298 |
-
"loss": 0.
|
| 2299 |
"step": 1850
|
| 2300 |
},
|
| 2301 |
{
|
| 2302 |
"epoch": 4.68,
|
| 2303 |
-
"learning_rate": 5.
|
| 2304 |
-
"loss": 0.
|
| 2305 |
"step": 1855
|
| 2306 |
},
|
| 2307 |
{
|
| 2308 |
"epoch": 4.69,
|
| 2309 |
-
"learning_rate": 5.
|
| 2310 |
-
"loss": 0.
|
| 2311 |
"step": 1860
|
| 2312 |
},
|
| 2313 |
{
|
| 2314 |
"epoch": 4.71,
|
| 2315 |
-
"learning_rate": 5.
|
| 2316 |
-
"loss": 0.
|
| 2317 |
"step": 1865
|
| 2318 |
},
|
| 2319 |
{
|
| 2320 |
"epoch": 4.72,
|
| 2321 |
-
"learning_rate": 5.
|
| 2322 |
-
"loss": 0.
|
| 2323 |
"step": 1870
|
| 2324 |
},
|
| 2325 |
{
|
| 2326 |
"epoch": 4.73,
|
| 2327 |
-
"learning_rate": 5.
|
| 2328 |
-
"loss": 0.
|
| 2329 |
"step": 1875
|
| 2330 |
},
|
| 2331 |
{
|
| 2332 |
-
"epoch": 4.
|
| 2333 |
-
"learning_rate": 5.
|
| 2334 |
-
"loss": 0.
|
| 2335 |
"step": 1880
|
| 2336 |
},
|
| 2337 |
{
|
| 2338 |
"epoch": 4.76,
|
| 2339 |
-
"learning_rate": 5.
|
| 2340 |
-
"loss": 0.
|
| 2341 |
"step": 1885
|
| 2342 |
},
|
| 2343 |
{
|
| 2344 |
"epoch": 4.77,
|
| 2345 |
-
"learning_rate": 5.
|
| 2346 |
-
"loss": 0.
|
| 2347 |
"step": 1890
|
| 2348 |
},
|
| 2349 |
{
|
| 2350 |
"epoch": 4.78,
|
| 2351 |
-
"learning_rate": 5.
|
| 2352 |
-
"loss": 0.
|
| 2353 |
"step": 1895
|
| 2354 |
},
|
| 2355 |
{
|
| 2356 |
-
"epoch": 4.
|
| 2357 |
-
"learning_rate": 5.
|
| 2358 |
-
"loss": 0.
|
| 2359 |
"step": 1900
|
| 2360 |
},
|
| 2361 |
{
|
| 2362 |
"epoch": 4.81,
|
| 2363 |
-
"learning_rate": 5.
|
| 2364 |
-
"loss": 0.
|
| 2365 |
"step": 1905
|
| 2366 |
},
|
| 2367 |
{
|
| 2368 |
"epoch": 4.82,
|
| 2369 |
-
"learning_rate": 5.
|
| 2370 |
-
"loss": 0.
|
| 2371 |
"step": 1910
|
| 2372 |
},
|
| 2373 |
{
|
| 2374 |
"epoch": 4.83,
|
| 2375 |
-
"learning_rate": 5.
|
| 2376 |
-
"loss": 0.
|
| 2377 |
"step": 1915
|
| 2378 |
},
|
| 2379 |
{
|
| 2380 |
-
"epoch": 4.
|
| 2381 |
-
"learning_rate": 5.
|
| 2382 |
-
"loss": 0.
|
| 2383 |
"step": 1920
|
| 2384 |
},
|
| 2385 |
{
|
| 2386 |
"epoch": 4.86,
|
| 2387 |
-
"learning_rate": 5.
|
| 2388 |
-
"loss": 0.
|
| 2389 |
"step": 1925
|
| 2390 |
},
|
| 2391 |
{
|
| 2392 |
"epoch": 4.87,
|
| 2393 |
-
"learning_rate": 5.
|
| 2394 |
-
"loss": 0.
|
| 2395 |
"step": 1930
|
| 2396 |
},
|
| 2397 |
{
|
| 2398 |
"epoch": 4.88,
|
| 2399 |
-
"learning_rate": 5.
|
| 2400 |
-
"loss": 0.
|
| 2401 |
"step": 1935
|
| 2402 |
},
|
| 2403 |
{
|
| 2404 |
"epoch": 4.9,
|
| 2405 |
-
"learning_rate": 5.
|
| 2406 |
-
"loss": 0.
|
| 2407 |
"step": 1940
|
| 2408 |
},
|
| 2409 |
{
|
| 2410 |
"epoch": 4.91,
|
| 2411 |
-
"learning_rate": 5.
|
| 2412 |
-
"loss": 0.
|
| 2413 |
"step": 1945
|
| 2414 |
},
|
| 2415 |
{
|
| 2416 |
"epoch": 4.92,
|
| 2417 |
-
"learning_rate": 5.
|
| 2418 |
-
"loss": 0.
|
| 2419 |
"step": 1950
|
| 2420 |
},
|
| 2421 |
{
|
| 2422 |
"epoch": 4.93,
|
| 2423 |
-
"learning_rate": 5.
|
| 2424 |
-
"loss": 0.
|
| 2425 |
"step": 1955
|
| 2426 |
},
|
| 2427 |
{
|
| 2428 |
"epoch": 4.95,
|
| 2429 |
-
"learning_rate": 5.
|
| 2430 |
-
"loss": 0.
|
| 2431 |
"step": 1960
|
| 2432 |
},
|
| 2433 |
{
|
| 2434 |
"epoch": 4.96,
|
| 2435 |
-
"learning_rate": 5.
|
| 2436 |
-
"loss": 0.
|
| 2437 |
"step": 1965
|
| 2438 |
},
|
| 2439 |
{
|
| 2440 |
"epoch": 4.97,
|
| 2441 |
"learning_rate": 5.886000000000001e-06,
|
| 2442 |
-
"loss": 0.
|
| 2443 |
"step": 1970
|
| 2444 |
},
|
| 2445 |
{
|
| 2446 |
"epoch": 4.98,
|
| 2447 |
"learning_rate": 5.901000000000001e-06,
|
| 2448 |
-
"loss": 0.
|
| 2449 |
"step": 1975
|
| 2450 |
},
|
| 2451 |
{
|
| 2452 |
"epoch": 5.0,
|
| 2453 |
"learning_rate": 5.916e-06,
|
| 2454 |
-
"loss": 0.
|
| 2455 |
"step": 1980
|
| 2456 |
},
|
| 2457 |
{
|
| 2458 |
-
"epoch": 5.
|
| 2459 |
-
"
|
| 2460 |
-
"
|
| 2461 |
-
"
|
| 2462 |
-
|
| 2463 |
-
|
| 2464 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2465 |
}
|
| 2466 |
],
|
| 2467 |
"logging_steps": 5,
|
| 2468 |
-
"max_steps":
|
| 2469 |
"num_input_tokens_seen": 0,
|
| 2470 |
-
"num_train_epochs":
|
| 2471 |
"save_steps": 500,
|
| 2472 |
-
"total_flos":
|
| 2473 |
"train_batch_size": 12,
|
| 2474 |
"trial_name": null,
|
| 2475 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.994955863808323,
|
| 5 |
"eval_steps": 250,
|
| 6 |
+
"global_step": 2772,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1871 |
{
|
| 1872 |
"epoch": 3.8,
|
| 1873 |
"learning_rate": 4.497e-06,
|
| 1874 |
+
"loss": 0.2097,
|
| 1875 |
"step": 1505
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
"epoch": 3.81,
|
| 1879 |
+
"learning_rate": 4.512e-06,
|
| 1880 |
+
"loss": 0.1645,
|
| 1881 |
"step": 1510
|
| 1882 |
},
|
| 1883 |
{
|
| 1884 |
"epoch": 3.82,
|
| 1885 |
+
"learning_rate": 4.527e-06,
|
| 1886 |
+
"loss": 0.163,
|
| 1887 |
"step": 1515
|
| 1888 |
},
|
| 1889 |
{
|
| 1890 |
"epoch": 3.84,
|
| 1891 |
+
"learning_rate": 4.542000000000001e-06,
|
| 1892 |
+
"loss": 0.1676,
|
| 1893 |
"step": 1520
|
| 1894 |
},
|
| 1895 |
{
|
| 1896 |
"epoch": 3.85,
|
| 1897 |
+
"learning_rate": 4.557000000000001e-06,
|
| 1898 |
+
"loss": 0.1946,
|
| 1899 |
"step": 1525
|
| 1900 |
},
|
| 1901 |
{
|
| 1902 |
"epoch": 3.86,
|
| 1903 |
+
"learning_rate": 4.5720000000000004e-06,
|
| 1904 |
+
"loss": 0.1582,
|
| 1905 |
"step": 1530
|
| 1906 |
},
|
| 1907 |
{
|
| 1908 |
+
"epoch": 3.88,
|
| 1909 |
+
"learning_rate": 4.587e-06,
|
| 1910 |
+
"loss": 0.1588,
|
| 1911 |
"step": 1535
|
| 1912 |
},
|
| 1913 |
{
|
| 1914 |
"epoch": 3.89,
|
| 1915 |
"learning_rate": 4.5989999999999995e-06,
|
| 1916 |
+
"loss": 0.1936,
|
| 1917 |
"step": 1540
|
| 1918 |
},
|
| 1919 |
{
|
| 1920 |
"epoch": 3.9,
|
| 1921 |
"learning_rate": 4.614e-06,
|
| 1922 |
+
"loss": 0.1525,
|
| 1923 |
"step": 1545
|
| 1924 |
},
|
| 1925 |
{
|
| 1926 |
"epoch": 3.91,
|
| 1927 |
"learning_rate": 4.629e-06,
|
| 1928 |
+
"loss": 0.1456,
|
| 1929 |
"step": 1550
|
| 1930 |
},
|
| 1931 |
{
|
| 1932 |
+
"epoch": 3.93,
|
| 1933 |
"learning_rate": 4.644e-06,
|
| 1934 |
+
"loss": 0.1883,
|
| 1935 |
"step": 1555
|
| 1936 |
},
|
| 1937 |
{
|
| 1938 |
"epoch": 3.94,
|
| 1939 |
+
"learning_rate": 4.656e-06,
|
| 1940 |
+
"loss": 0.1538,
|
| 1941 |
"step": 1560
|
| 1942 |
},
|
| 1943 |
{
|
| 1944 |
"epoch": 3.95,
|
| 1945 |
+
"learning_rate": 4.671000000000001e-06,
|
| 1946 |
+
"loss": 0.1621,
|
| 1947 |
"step": 1565
|
| 1948 |
},
|
| 1949 |
{
|
| 1950 |
"epoch": 3.96,
|
| 1951 |
+
"learning_rate": 4.6860000000000005e-06,
|
| 1952 |
+
"loss": 0.1677,
|
| 1953 |
"step": 1570
|
| 1954 |
},
|
| 1955 |
{
|
| 1956 |
+
"epoch": 3.98,
|
| 1957 |
+
"learning_rate": 4.701e-06,
|
| 1958 |
+
"loss": 0.1697,
|
| 1959 |
"step": 1575
|
| 1960 |
},
|
| 1961 |
{
|
| 1962 |
"epoch": 3.99,
|
| 1963 |
+
"learning_rate": 4.716e-06,
|
| 1964 |
+
"loss": 0.1669,
|
| 1965 |
"step": 1580
|
| 1966 |
},
|
| 1967 |
{
|
| 1968 |
"epoch": 4.0,
|
| 1969 |
+
"learning_rate": 4.731e-06,
|
| 1970 |
+
"loss": 0.1954,
|
| 1971 |
"step": 1585
|
| 1972 |
},
|
| 1973 |
{
|
| 1974 |
"epoch": 4.01,
|
| 1975 |
+
"learning_rate": 4.746e-06,
|
| 1976 |
+
"loss": 0.1518,
|
| 1977 |
"step": 1590
|
| 1978 |
},
|
| 1979 |
{
|
| 1980 |
"epoch": 4.03,
|
| 1981 |
+
"learning_rate": 4.761000000000001e-06,
|
| 1982 |
+
"loss": 0.153,
|
| 1983 |
"step": 1595
|
| 1984 |
},
|
| 1985 |
{
|
| 1986 |
"epoch": 4.04,
|
| 1987 |
+
"learning_rate": 4.7760000000000005e-06,
|
| 1988 |
+
"loss": 0.1736,
|
| 1989 |
"step": 1600
|
| 1990 |
},
|
| 1991 |
{
|
| 1992 |
"epoch": 4.05,
|
| 1993 |
+
"learning_rate": 4.791e-06,
|
| 1994 |
+
"loss": 0.1748,
|
| 1995 |
"step": 1605
|
| 1996 |
},
|
| 1997 |
{
|
| 1998 |
"epoch": 4.06,
|
| 1999 |
+
"learning_rate": 4.806e-06,
|
| 2000 |
+
"loss": 0.1545,
|
| 2001 |
"step": 1610
|
| 2002 |
},
|
| 2003 |
{
|
| 2004 |
"epoch": 4.08,
|
| 2005 |
+
"learning_rate": 4.821e-06,
|
| 2006 |
+
"loss": 0.143,
|
| 2007 |
"step": 1615
|
| 2008 |
},
|
| 2009 |
{
|
| 2010 |
"epoch": 4.09,
|
| 2011 |
+
"learning_rate": 4.836000000000001e-06,
|
| 2012 |
+
"loss": 0.1808,
|
| 2013 |
"step": 1620
|
| 2014 |
},
|
| 2015 |
{
|
| 2016 |
"epoch": 4.1,
|
| 2017 |
+
"learning_rate": 4.851000000000001e-06,
|
| 2018 |
+
"loss": 0.166,
|
| 2019 |
"step": 1625
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
"epoch": 4.11,
|
| 2023 |
+
"learning_rate": 4.8660000000000005e-06,
|
| 2024 |
+
"loss": 0.1432,
|
| 2025 |
"step": 1630
|
| 2026 |
},
|
| 2027 |
{
|
| 2028 |
"epoch": 4.13,
|
| 2029 |
+
"learning_rate": 4.881e-06,
|
| 2030 |
+
"loss": 0.1863,
|
| 2031 |
"step": 1635
|
| 2032 |
},
|
| 2033 |
{
|
| 2034 |
"epoch": 4.14,
|
| 2035 |
+
"learning_rate": 4.896e-06,
|
| 2036 |
+
"loss": 0.1423,
|
| 2037 |
"step": 1640
|
| 2038 |
},
|
| 2039 |
{
|
| 2040 |
"epoch": 4.15,
|
| 2041 |
+
"learning_rate": 4.911e-06,
|
| 2042 |
+
"loss": 0.1519,
|
| 2043 |
"step": 1645
|
| 2044 |
},
|
| 2045 |
{
|
| 2046 |
+
"epoch": 4.17,
|
| 2047 |
+
"learning_rate": 4.926000000000001e-06,
|
| 2048 |
+
"loss": 0.169,
|
| 2049 |
"step": 1650
|
| 2050 |
},
|
| 2051 |
{
|
| 2052 |
"epoch": 4.18,
|
| 2053 |
+
"learning_rate": 4.941000000000001e-06,
|
| 2054 |
+
"loss": 0.1729,
|
| 2055 |
"step": 1655
|
| 2056 |
},
|
| 2057 |
{
|
| 2058 |
"epoch": 4.19,
|
| 2059 |
+
"learning_rate": 4.9560000000000005e-06,
|
| 2060 |
+
"loss": 0.1585,
|
| 2061 |
"step": 1660
|
| 2062 |
},
|
| 2063 |
{
|
| 2064 |
"epoch": 4.2,
|
| 2065 |
+
"learning_rate": 4.9709999999999995e-06,
|
| 2066 |
+
"loss": 0.1413,
|
| 2067 |
"step": 1665
|
| 2068 |
},
|
| 2069 |
{
|
| 2070 |
+
"epoch": 4.22,
|
| 2071 |
+
"learning_rate": 4.985999999999999e-06,
|
| 2072 |
+
"loss": 0.1832,
|
| 2073 |
"step": 1670
|
| 2074 |
},
|
| 2075 |
{
|
| 2076 |
"epoch": 4.23,
|
| 2077 |
+
"learning_rate": 5.001e-06,
|
| 2078 |
+
"loss": 0.143,
|
| 2079 |
"step": 1675
|
| 2080 |
},
|
| 2081 |
{
|
| 2082 |
"epoch": 4.24,
|
| 2083 |
+
"learning_rate": 5.016e-06,
|
| 2084 |
+
"loss": 0.1527,
|
| 2085 |
"step": 1680
|
| 2086 |
},
|
| 2087 |
{
|
| 2088 |
"epoch": 4.25,
|
| 2089 |
+
"learning_rate": 5.031e-06,
|
| 2090 |
+
"loss": 0.1669,
|
| 2091 |
"step": 1685
|
| 2092 |
},
|
| 2093 |
{
|
| 2094 |
+
"epoch": 4.27,
|
| 2095 |
+
"learning_rate": 5.046e-06,
|
| 2096 |
+
"loss": 0.1313,
|
| 2097 |
"step": 1690
|
| 2098 |
},
|
| 2099 |
{
|
| 2100 |
"epoch": 4.28,
|
| 2101 |
+
"learning_rate": 5.0609999999999995e-06,
|
| 2102 |
+
"loss": 0.1273,
|
| 2103 |
"step": 1695
|
| 2104 |
},
|
| 2105 |
{
|
| 2106 |
"epoch": 4.29,
|
| 2107 |
+
"learning_rate": 5.076e-06,
|
| 2108 |
+
"loss": 0.1433,
|
| 2109 |
"step": 1700
|
| 2110 |
},
|
| 2111 |
{
|
| 2112 |
"epoch": 4.3,
|
| 2113 |
+
"learning_rate": 5.091e-06,
|
| 2114 |
+
"loss": 0.1579,
|
| 2115 |
"step": 1705
|
| 2116 |
},
|
| 2117 |
{
|
| 2118 |
"epoch": 4.32,
|
| 2119 |
+
"learning_rate": 5.106e-06,
|
| 2120 |
+
"loss": 0.1415,
|
| 2121 |
"step": 1710
|
| 2122 |
},
|
| 2123 |
{
|
| 2124 |
"epoch": 4.33,
|
| 2125 |
+
"learning_rate": 5.121e-06,
|
| 2126 |
+
"loss": 0.147,
|
| 2127 |
"step": 1715
|
| 2128 |
},
|
| 2129 |
{
|
| 2130 |
"epoch": 4.34,
|
| 2131 |
+
"learning_rate": 5.136e-06,
|
| 2132 |
+
"loss": 0.1682,
|
| 2133 |
"step": 1720
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
"epoch": 4.35,
|
| 2137 |
+
"learning_rate": 5.1509999999999995e-06,
|
| 2138 |
+
"loss": 0.1352,
|
| 2139 |
"step": 1725
|
| 2140 |
},
|
| 2141 |
{
|
| 2142 |
"epoch": 4.37,
|
| 2143 |
+
"learning_rate": 5.166e-06,
|
| 2144 |
+
"loss": 0.14,
|
| 2145 |
"step": 1730
|
| 2146 |
},
|
| 2147 |
{
|
| 2148 |
"epoch": 4.38,
|
| 2149 |
+
"learning_rate": 5.181e-06,
|
| 2150 |
+
"loss": 0.1724,
|
| 2151 |
"step": 1735
|
| 2152 |
},
|
| 2153 |
{
|
| 2154 |
"epoch": 4.39,
|
| 2155 |
+
"learning_rate": 5.196e-06,
|
| 2156 |
+
"loss": 0.1442,
|
| 2157 |
"step": 1740
|
| 2158 |
},
|
| 2159 |
{
|
| 2160 |
"epoch": 4.4,
|
| 2161 |
+
"learning_rate": 5.211e-06,
|
| 2162 |
+
"loss": 0.1316,
|
| 2163 |
"step": 1745
|
| 2164 |
},
|
| 2165 |
{
|
| 2166 |
"epoch": 4.42,
|
| 2167 |
+
"learning_rate": 5.226e-06,
|
| 2168 |
+
"loss": 0.1546,
|
| 2169 |
"step": 1750
|
| 2170 |
},
|
| 2171 |
{
|
| 2172 |
"epoch": 4.42,
|
| 2173 |
+
"eval_cer": 0.029413526627927875,
|
| 2174 |
+
"eval_loss": 0.14518263936042786,
|
| 2175 |
+
"eval_runtime": 60.4847,
|
| 2176 |
+
"eval_samples_per_second": 41.796,
|
| 2177 |
+
"eval_steps_per_second": 1.174,
|
| 2178 |
+
"eval_wer": 0.11381389030395653,
|
| 2179 |
"step": 1750
|
| 2180 |
},
|
| 2181 |
{
|
| 2182 |
"epoch": 4.43,
|
| 2183 |
+
"learning_rate": 5.241e-06,
|
| 2184 |
+
"loss": 0.1528,
|
| 2185 |
"step": 1755
|
| 2186 |
},
|
| 2187 |
{
|
| 2188 |
"epoch": 4.44,
|
| 2189 |
+
"learning_rate": 5.256e-06,
|
| 2190 |
"loss": 0.1444,
|
| 2191 |
"step": 1760
|
| 2192 |
},
|
| 2193 |
{
|
| 2194 |
+
"epoch": 4.46,
|
| 2195 |
+
"learning_rate": 5.271e-06,
|
| 2196 |
+
"loss": 0.1514,
|
| 2197 |
"step": 1765
|
| 2198 |
},
|
| 2199 |
{
|
| 2200 |
"epoch": 4.47,
|
| 2201 |
+
"learning_rate": 5.286e-06,
|
| 2202 |
+
"loss": 0.1924,
|
| 2203 |
"step": 1770
|
| 2204 |
},
|
| 2205 |
{
|
| 2206 |
"epoch": 4.48,
|
| 2207 |
+
"learning_rate": 5.301e-06,
|
| 2208 |
+
"loss": 0.1223,
|
| 2209 |
"step": 1775
|
| 2210 |
},
|
| 2211 |
{
|
| 2212 |
"epoch": 4.49,
|
| 2213 |
+
"learning_rate": 5.3160000000000004e-06,
|
| 2214 |
+
"loss": 0.1296,
|
| 2215 |
"step": 1780
|
| 2216 |
},
|
| 2217 |
{
|
| 2218 |
+
"epoch": 4.51,
|
| 2219 |
+
"learning_rate": 5.331e-06,
|
| 2220 |
+
"loss": 0.1668,
|
| 2221 |
"step": 1785
|
| 2222 |
},
|
| 2223 |
{
|
| 2224 |
"epoch": 4.52,
|
| 2225 |
+
"learning_rate": 5.346e-06,
|
| 2226 |
+
"loss": 0.137,
|
| 2227 |
"step": 1790
|
| 2228 |
},
|
| 2229 |
{
|
| 2230 |
"epoch": 4.53,
|
| 2231 |
+
"learning_rate": 5.361e-06,
|
| 2232 |
+
"loss": 0.1365,
|
| 2233 |
"step": 1795
|
| 2234 |
},
|
| 2235 |
{
|
| 2236 |
"epoch": 4.54,
|
| 2237 |
+
"learning_rate": 5.376e-06,
|
| 2238 |
+
"loss": 0.1565,
|
| 2239 |
"step": 1800
|
| 2240 |
},
|
| 2241 |
{
|
| 2242 |
+
"epoch": 4.56,
|
| 2243 |
+
"learning_rate": 5.391e-06,
|
| 2244 |
+
"loss": 0.1421,
|
| 2245 |
"step": 1805
|
| 2246 |
},
|
| 2247 |
{
|
| 2248 |
"epoch": 4.57,
|
| 2249 |
+
"learning_rate": 5.406e-06,
|
| 2250 |
+
"loss": 0.1418,
|
| 2251 |
"step": 1810
|
| 2252 |
},
|
| 2253 |
{
|
| 2254 |
"epoch": 4.58,
|
| 2255 |
+
"learning_rate": 5.421e-06,
|
| 2256 |
+
"loss": 0.1415,
|
| 2257 |
"step": 1815
|
| 2258 |
},
|
| 2259 |
{
|
| 2260 |
"epoch": 4.59,
|
| 2261 |
+
"learning_rate": 5.436e-06,
|
| 2262 |
+
"loss": 0.1851,
|
| 2263 |
"step": 1820
|
| 2264 |
},
|
| 2265 |
{
|
| 2266 |
"epoch": 4.61,
|
| 2267 |
+
"learning_rate": 5.451e-06,
|
| 2268 |
+
"loss": 0.1354,
|
| 2269 |
"step": 1825
|
| 2270 |
},
|
| 2271 |
{
|
| 2272 |
"epoch": 4.62,
|
| 2273 |
+
"learning_rate": 5.466e-06,
|
| 2274 |
+
"loss": 0.1344,
|
| 2275 |
"step": 1830
|
| 2276 |
},
|
| 2277 |
{
|
| 2278 |
"epoch": 4.63,
|
| 2279 |
+
"learning_rate": 5.4810000000000005e-06,
|
| 2280 |
+
"loss": 0.1761,
|
| 2281 |
"step": 1835
|
| 2282 |
},
|
| 2283 |
{
|
| 2284 |
"epoch": 4.64,
|
| 2285 |
+
"learning_rate": 5.496e-06,
|
| 2286 |
+
"loss": 0.1257,
|
| 2287 |
"step": 1840
|
| 2288 |
},
|
| 2289 |
{
|
| 2290 |
"epoch": 4.66,
|
| 2291 |
+
"learning_rate": 5.511e-06,
|
| 2292 |
+
"loss": 0.1291,
|
| 2293 |
"step": 1845
|
| 2294 |
},
|
| 2295 |
{
|
| 2296 |
"epoch": 4.67,
|
| 2297 |
+
"learning_rate": 5.526e-06,
|
| 2298 |
+
"loss": 0.1458,
|
| 2299 |
"step": 1850
|
| 2300 |
},
|
| 2301 |
{
|
| 2302 |
"epoch": 4.68,
|
| 2303 |
+
"learning_rate": 5.541e-06,
|
| 2304 |
+
"loss": 0.1613,
|
| 2305 |
"step": 1855
|
| 2306 |
},
|
| 2307 |
{
|
| 2308 |
"epoch": 4.69,
|
| 2309 |
+
"learning_rate": 5.556e-06,
|
| 2310 |
+
"loss": 0.12,
|
| 2311 |
"step": 1860
|
| 2312 |
},
|
| 2313 |
{
|
| 2314 |
"epoch": 4.71,
|
| 2315 |
+
"learning_rate": 5.5710000000000005e-06,
|
| 2316 |
+
"loss": 0.1333,
|
| 2317 |
"step": 1865
|
| 2318 |
},
|
| 2319 |
{
|
| 2320 |
"epoch": 4.72,
|
| 2321 |
+
"learning_rate": 5.586e-06,
|
| 2322 |
+
"loss": 0.1639,
|
| 2323 |
"step": 1870
|
| 2324 |
},
|
| 2325 |
{
|
| 2326 |
"epoch": 4.73,
|
| 2327 |
+
"learning_rate": 5.601e-06,
|
| 2328 |
+
"loss": 0.1361,
|
| 2329 |
"step": 1875
|
| 2330 |
},
|
| 2331 |
{
|
| 2332 |
+
"epoch": 4.75,
|
| 2333 |
+
"learning_rate": 5.616e-06,
|
| 2334 |
+
"loss": 0.1319,
|
| 2335 |
"step": 1880
|
| 2336 |
},
|
| 2337 |
{
|
| 2338 |
"epoch": 4.76,
|
| 2339 |
+
"learning_rate": 5.631e-06,
|
| 2340 |
+
"loss": 0.1608,
|
| 2341 |
"step": 1885
|
| 2342 |
},
|
| 2343 |
{
|
| 2344 |
"epoch": 4.77,
|
| 2345 |
+
"learning_rate": 5.646000000000001e-06,
|
| 2346 |
+
"loss": 0.123,
|
| 2347 |
"step": 1890
|
| 2348 |
},
|
| 2349 |
{
|
| 2350 |
"epoch": 4.78,
|
| 2351 |
+
"learning_rate": 5.6610000000000005e-06,
|
| 2352 |
+
"loss": 0.1208,
|
| 2353 |
"step": 1895
|
| 2354 |
},
|
| 2355 |
{
|
| 2356 |
+
"epoch": 4.8,
|
| 2357 |
+
"learning_rate": 5.676e-06,
|
| 2358 |
+
"loss": 0.1335,
|
| 2359 |
"step": 1900
|
| 2360 |
},
|
| 2361 |
{
|
| 2362 |
"epoch": 4.81,
|
| 2363 |
+
"learning_rate": 5.691e-06,
|
| 2364 |
+
"loss": 0.1401,
|
| 2365 |
"step": 1905
|
| 2366 |
},
|
| 2367 |
{
|
| 2368 |
"epoch": 4.82,
|
| 2369 |
+
"learning_rate": 5.706e-06,
|
| 2370 |
+
"loss": 0.1159,
|
| 2371 |
"step": 1910
|
| 2372 |
},
|
| 2373 |
{
|
| 2374 |
"epoch": 4.83,
|
| 2375 |
+
"learning_rate": 5.721000000000001e-06,
|
| 2376 |
+
"loss": 0.1388,
|
| 2377 |
"step": 1915
|
| 2378 |
},
|
| 2379 |
{
|
| 2380 |
+
"epoch": 4.85,
|
| 2381 |
+
"learning_rate": 5.736000000000001e-06,
|
| 2382 |
+
"loss": 0.1646,
|
| 2383 |
"step": 1920
|
| 2384 |
},
|
| 2385 |
{
|
| 2386 |
"epoch": 4.86,
|
| 2387 |
+
"learning_rate": 5.7510000000000005e-06,
|
| 2388 |
+
"loss": 0.1265,
|
| 2389 |
"step": 1925
|
| 2390 |
},
|
| 2391 |
{
|
| 2392 |
"epoch": 4.87,
|
| 2393 |
+
"learning_rate": 5.766e-06,
|
| 2394 |
+
"loss": 0.1182,
|
| 2395 |
"step": 1930
|
| 2396 |
},
|
| 2397 |
{
|
| 2398 |
"epoch": 4.88,
|
| 2399 |
+
"learning_rate": 5.781e-06,
|
| 2400 |
+
"loss": 0.1615,
|
| 2401 |
"step": 1935
|
| 2402 |
},
|
| 2403 |
{
|
| 2404 |
"epoch": 4.9,
|
| 2405 |
+
"learning_rate": 5.796e-06,
|
| 2406 |
+
"loss": 0.1244,
|
| 2407 |
"step": 1940
|
| 2408 |
},
|
| 2409 |
{
|
| 2410 |
"epoch": 4.91,
|
| 2411 |
+
"learning_rate": 5.811000000000001e-06,
|
| 2412 |
+
"loss": 0.1281,
|
| 2413 |
"step": 1945
|
| 2414 |
},
|
| 2415 |
{
|
| 2416 |
"epoch": 4.92,
|
| 2417 |
+
"learning_rate": 5.826000000000001e-06,
|
| 2418 |
+
"loss": 0.1366,
|
| 2419 |
"step": 1950
|
| 2420 |
},
|
| 2421 |
{
|
| 2422 |
"epoch": 4.93,
|
| 2423 |
+
"learning_rate": 5.8410000000000005e-06,
|
| 2424 |
+
"loss": 0.1452,
|
| 2425 |
"step": 1955
|
| 2426 |
},
|
| 2427 |
{
|
| 2428 |
"epoch": 4.95,
|
| 2429 |
+
"learning_rate": 5.856e-06,
|
| 2430 |
+
"loss": 0.1169,
|
| 2431 |
"step": 1960
|
| 2432 |
},
|
| 2433 |
{
|
| 2434 |
"epoch": 4.96,
|
| 2435 |
+
"learning_rate": 5.871e-06,
|
| 2436 |
+
"loss": 0.1341,
|
| 2437 |
"step": 1965
|
| 2438 |
},
|
| 2439 |
{
|
| 2440 |
"epoch": 4.97,
|
| 2441 |
"learning_rate": 5.886000000000001e-06,
|
| 2442 |
+
"loss": 0.1373,
|
| 2443 |
"step": 1970
|
| 2444 |
},
|
| 2445 |
{
|
| 2446 |
"epoch": 4.98,
|
| 2447 |
"learning_rate": 5.901000000000001e-06,
|
| 2448 |
+
"loss": 0.1185,
|
| 2449 |
"step": 1975
|
| 2450 |
},
|
| 2451 |
{
|
| 2452 |
"epoch": 5.0,
|
| 2453 |
"learning_rate": 5.916e-06,
|
| 2454 |
+
"loss": 0.132,
|
| 2455 |
"step": 1980
|
| 2456 |
},
|
| 2457 |
{
|
| 2458 |
+
"epoch": 5.01,
|
| 2459 |
+
"learning_rate": 5.931e-06,
|
| 2460 |
+
"loss": 0.1279,
|
| 2461 |
+
"step": 1985
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 5.02,
|
| 2465 |
+
"learning_rate": 5.9459999999999995e-06,
|
| 2466 |
+
"loss": 0.1163,
|
| 2467 |
+
"step": 1990
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 5.04,
|
| 2471 |
+
"learning_rate": 5.961e-06,
|
| 2472 |
+
"loss": 0.121,
|
| 2473 |
+
"step": 1995
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 5.05,
|
| 2477 |
+
"learning_rate": 5.976e-06,
|
| 2478 |
+
"loss": 0.1245,
|
| 2479 |
+
"step": 2000
|
| 2480 |
+
},
|
| 2481 |
+
{
|
| 2482 |
+
"epoch": 5.05,
|
| 2483 |
+
"eval_cer": 0.025970965873919206,
|
| 2484 |
+
"eval_loss": 0.1316203624010086,
|
| 2485 |
+
"eval_runtime": 59.578,
|
| 2486 |
+
"eval_samples_per_second": 42.432,
|
| 2487 |
+
"eval_steps_per_second": 1.192,
|
| 2488 |
+
"eval_wer": 0.09734250297164204,
|
| 2489 |
+
"step": 2000
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 5.06,
|
| 2493 |
+
"learning_rate": 5.991e-06,
|
| 2494 |
+
"loss": 0.117,
|
| 2495 |
+
"step": 2005
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 5.07,
|
| 2499 |
+
"learning_rate": 6.006e-06,
|
| 2500 |
+
"loss": 0.115,
|
| 2501 |
+
"step": 2010
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 5.09,
|
| 2505 |
+
"learning_rate": 6.021e-06,
|
| 2506 |
+
"loss": 0.1416,
|
| 2507 |
+
"step": 2015
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 5.1,
|
| 2511 |
+
"learning_rate": 6.0359999999999995e-06,
|
| 2512 |
+
"loss": 0.1174,
|
| 2513 |
+
"step": 2020
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 5.11,
|
| 2517 |
+
"learning_rate": 6.051e-06,
|
| 2518 |
+
"loss": 0.1158,
|
| 2519 |
+
"step": 2025
|
| 2520 |
+
},
|
| 2521 |
+
{
|
| 2522 |
+
"epoch": 5.12,
|
| 2523 |
+
"learning_rate": 6.066e-06,
|
| 2524 |
+
"loss": 0.1249,
|
| 2525 |
+
"step": 2030
|
| 2526 |
+
},
|
| 2527 |
+
{
|
| 2528 |
+
"epoch": 5.14,
|
| 2529 |
+
"learning_rate": 6.081e-06,
|
| 2530 |
+
"loss": 0.1219,
|
| 2531 |
+
"step": 2035
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 5.15,
|
| 2535 |
+
"learning_rate": 6.096e-06,
|
| 2536 |
+
"loss": 0.1154,
|
| 2537 |
+
"step": 2040
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 5.16,
|
| 2541 |
+
"learning_rate": 6.111e-06,
|
| 2542 |
+
"loss": 0.1172,
|
| 2543 |
+
"step": 2045
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 5.17,
|
| 2547 |
+
"learning_rate": 6.126e-06,
|
| 2548 |
+
"loss": 0.1442,
|
| 2549 |
+
"step": 2050
|
| 2550 |
+
},
|
| 2551 |
+
{
|
| 2552 |
+
"epoch": 5.19,
|
| 2553 |
+
"learning_rate": 6.141e-06,
|
| 2554 |
+
"loss": 0.1095,
|
| 2555 |
+
"step": 2055
|
| 2556 |
+
},
|
| 2557 |
+
{
|
| 2558 |
+
"epoch": 5.2,
|
| 2559 |
+
"learning_rate": 6.156e-06,
|
| 2560 |
+
"loss": 0.1127,
|
| 2561 |
+
"step": 2060
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 5.21,
|
| 2565 |
+
"learning_rate": 6.171e-06,
|
| 2566 |
+
"loss": 0.1494,
|
| 2567 |
+
"step": 2065
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 5.22,
|
| 2571 |
+
"learning_rate": 6.186e-06,
|
| 2572 |
+
"loss": 0.111,
|
| 2573 |
+
"step": 2070
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 5.24,
|
| 2577 |
+
"learning_rate": 6.201e-06,
|
| 2578 |
+
"loss": 0.106,
|
| 2579 |
+
"step": 2075
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 5.25,
|
| 2583 |
+
"learning_rate": 6.216e-06,
|
| 2584 |
+
"loss": 0.1269,
|
| 2585 |
+
"step": 2080
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 5.26,
|
| 2589 |
+
"learning_rate": 6.231e-06,
|
| 2590 |
+
"loss": 0.1254,
|
| 2591 |
+
"step": 2085
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 5.27,
|
| 2595 |
+
"learning_rate": 6.246e-06,
|
| 2596 |
+
"loss": 0.1137,
|
| 2597 |
+
"step": 2090
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 5.29,
|
| 2601 |
+
"learning_rate": 6.261e-06,
|
| 2602 |
+
"loss": 0.1229,
|
| 2603 |
+
"step": 2095
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 5.3,
|
| 2607 |
+
"learning_rate": 6.276e-06,
|
| 2608 |
+
"loss": 0.1341,
|
| 2609 |
+
"step": 2100
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 5.31,
|
| 2613 |
+
"learning_rate": 6.291e-06,
|
| 2614 |
+
"loss": 0.1174,
|
| 2615 |
+
"step": 2105
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 5.33,
|
| 2619 |
+
"learning_rate": 6.306e-06,
|
| 2620 |
+
"loss": 0.1136,
|
| 2621 |
+
"step": 2110
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 5.34,
|
| 2625 |
+
"learning_rate": 6.321e-06,
|
| 2626 |
+
"loss": 0.1498,
|
| 2627 |
+
"step": 2115
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 5.35,
|
| 2631 |
+
"learning_rate": 6.336e-06,
|
| 2632 |
+
"loss": 0.1035,
|
| 2633 |
+
"step": 2120
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 5.36,
|
| 2637 |
+
"learning_rate": 6.351e-06,
|
| 2638 |
+
"loss": 0.1065,
|
| 2639 |
+
"step": 2125
|
| 2640 |
+
},
|
| 2641 |
+
{
|
| 2642 |
+
"epoch": 5.38,
|
| 2643 |
+
"learning_rate": 6.3660000000000005e-06,
|
| 2644 |
+
"loss": 0.1331,
|
| 2645 |
+
"step": 2130
|
| 2646 |
+
},
|
| 2647 |
+
{
|
| 2648 |
+
"epoch": 5.39,
|
| 2649 |
+
"learning_rate": 6.381e-06,
|
| 2650 |
+
"loss": 0.1261,
|
| 2651 |
+
"step": 2135
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 5.4,
|
| 2655 |
+
"learning_rate": 6.396e-06,
|
| 2656 |
+
"loss": 0.105,
|
| 2657 |
+
"step": 2140
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 5.41,
|
| 2661 |
+
"learning_rate": 6.411e-06,
|
| 2662 |
+
"loss": 0.1193,
|
| 2663 |
+
"step": 2145
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 5.43,
|
| 2667 |
+
"learning_rate": 6.426e-06,
|
| 2668 |
+
"loss": 0.1445,
|
| 2669 |
+
"step": 2150
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 5.44,
|
| 2673 |
+
"learning_rate": 6.441e-06,
|
| 2674 |
+
"loss": 0.1145,
|
| 2675 |
+
"step": 2155
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 5.45,
|
| 2679 |
+
"learning_rate": 6.4560000000000005e-06,
|
| 2680 |
+
"loss": 0.1091,
|
| 2681 |
+
"step": 2160
|
| 2682 |
+
},
|
| 2683 |
+
{
|
| 2684 |
+
"epoch": 5.46,
|
| 2685 |
+
"learning_rate": 6.471e-06,
|
| 2686 |
+
"loss": 0.154,
|
| 2687 |
+
"step": 2165
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"epoch": 5.48,
|
| 2691 |
+
"learning_rate": 6.486e-06,
|
| 2692 |
+
"loss": 0.1198,
|
| 2693 |
+
"step": 2170
|
| 2694 |
+
},
|
| 2695 |
+
{
|
| 2696 |
+
"epoch": 5.49,
|
| 2697 |
+
"learning_rate": 6.501e-06,
|
| 2698 |
+
"loss": 0.0978,
|
| 2699 |
+
"step": 2175
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 5.5,
|
| 2703 |
+
"learning_rate": 6.516e-06,
|
| 2704 |
+
"loss": 0.114,
|
| 2705 |
+
"step": 2180
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 5.51,
|
| 2709 |
+
"learning_rate": 6.531000000000001e-06,
|
| 2710 |
+
"loss": 0.1289,
|
| 2711 |
+
"step": 2185
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 5.53,
|
| 2715 |
+
"learning_rate": 6.5460000000000005e-06,
|
| 2716 |
+
"loss": 0.1059,
|
| 2717 |
+
"step": 2190
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 5.54,
|
| 2721 |
+
"learning_rate": 6.561e-06,
|
| 2722 |
+
"loss": 0.1077,
|
| 2723 |
+
"step": 2195
|
| 2724 |
+
},
|
| 2725 |
+
{
|
| 2726 |
+
"epoch": 5.55,
|
| 2727 |
+
"learning_rate": 6.576e-06,
|
| 2728 |
+
"loss": 0.1407,
|
| 2729 |
+
"step": 2200
|
| 2730 |
+
},
|
| 2731 |
+
{
|
| 2732 |
+
"epoch": 5.56,
|
| 2733 |
+
"learning_rate": 6.591e-06,
|
| 2734 |
+
"loss": 0.1077,
|
| 2735 |
+
"step": 2205
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 5.58,
|
| 2739 |
+
"learning_rate": 6.606000000000001e-06,
|
| 2740 |
+
"loss": 0.1089,
|
| 2741 |
+
"step": 2210
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 5.59,
|
| 2745 |
+
"learning_rate": 6.621000000000001e-06,
|
| 2746 |
+
"loss": 0.1244,
|
| 2747 |
+
"step": 2215
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 5.6,
|
| 2751 |
+
"learning_rate": 6.6360000000000005e-06,
|
| 2752 |
+
"loss": 0.1006,
|
| 2753 |
+
"step": 2220
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 5.62,
|
| 2757 |
+
"learning_rate": 6.651e-06,
|
| 2758 |
+
"loss": 0.1124,
|
| 2759 |
+
"step": 2225
|
| 2760 |
+
},
|
| 2761 |
+
{
|
| 2762 |
+
"epoch": 5.63,
|
| 2763 |
+
"learning_rate": 6.666e-06,
|
| 2764 |
+
"loss": 0.1202,
|
| 2765 |
+
"step": 2230
|
| 2766 |
+
},
|
| 2767 |
+
{
|
| 2768 |
+
"epoch": 5.64,
|
| 2769 |
+
"learning_rate": 6.681e-06,
|
| 2770 |
+
"loss": 0.1208,
|
| 2771 |
+
"step": 2235
|
| 2772 |
+
},
|
| 2773 |
+
{
|
| 2774 |
+
"epoch": 5.65,
|
| 2775 |
+
"learning_rate": 6.696000000000001e-06,
|
| 2776 |
+
"loss": 0.0993,
|
| 2777 |
+
"step": 2240
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 5.67,
|
| 2781 |
+
"learning_rate": 6.711000000000001e-06,
|
| 2782 |
+
"loss": 0.1111,
|
| 2783 |
+
"step": 2245
|
| 2784 |
+
},
|
| 2785 |
+
{
|
| 2786 |
+
"epoch": 5.68,
|
| 2787 |
+
"learning_rate": 6.7260000000000005e-06,
|
| 2788 |
+
"loss": 0.1341,
|
| 2789 |
+
"step": 2250
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 5.68,
|
| 2793 |
+
"eval_cer": 0.023351105253699156,
|
| 2794 |
+
"eval_loss": 0.11955570429563522,
|
| 2795 |
+
"eval_runtime": 60.3253,
|
| 2796 |
+
"eval_samples_per_second": 41.906,
|
| 2797 |
+
"eval_steps_per_second": 1.177,
|
| 2798 |
+
"eval_wer": 0.08670826965528952,
|
| 2799 |
+
"step": 2250
|
| 2800 |
+
},
|
| 2801 |
+
{
|
| 2802 |
+
"epoch": 5.69,
|
| 2803 |
+
"learning_rate": 6.741e-06,
|
| 2804 |
+
"loss": 0.0995,
|
| 2805 |
+
"step": 2255
|
| 2806 |
+
},
|
| 2807 |
+
{
|
| 2808 |
+
"epoch": 5.7,
|
| 2809 |
+
"learning_rate": 6.756e-06,
|
| 2810 |
+
"loss": 0.1081,
|
| 2811 |
+
"step": 2260
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 5.72,
|
| 2815 |
+
"learning_rate": 6.771000000000001e-06,
|
| 2816 |
+
"loss": 0.1339,
|
| 2817 |
+
"step": 2265
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 5.73,
|
| 2821 |
+
"learning_rate": 6.786000000000001e-06,
|
| 2822 |
+
"loss": 0.1044,
|
| 2823 |
+
"step": 2270
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 5.74,
|
| 2827 |
+
"learning_rate": 6.801000000000001e-06,
|
| 2828 |
+
"loss": 0.1029,
|
| 2829 |
+
"step": 2275
|
| 2830 |
+
},
|
| 2831 |
+
{
|
| 2832 |
+
"epoch": 5.75,
|
| 2833 |
+
"learning_rate": 6.8160000000000005e-06,
|
| 2834 |
+
"loss": 0.1184,
|
| 2835 |
+
"step": 2280
|
| 2836 |
+
},
|
| 2837 |
+
{
|
| 2838 |
+
"epoch": 5.77,
|
| 2839 |
+
"learning_rate": 6.831e-06,
|
| 2840 |
+
"loss": 0.1163,
|
| 2841 |
+
"step": 2285
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 5.78,
|
| 2845 |
+
"learning_rate": 6.845999999999999e-06,
|
| 2846 |
+
"loss": 0.1023,
|
| 2847 |
+
"step": 2290
|
| 2848 |
+
},
|
| 2849 |
+
{
|
| 2850 |
+
"epoch": 5.79,
|
| 2851 |
+
"learning_rate": 6.861e-06,
|
| 2852 |
+
"loss": 0.1078,
|
| 2853 |
+
"step": 2295
|
| 2854 |
+
},
|
| 2855 |
+
{
|
| 2856 |
+
"epoch": 5.8,
|
| 2857 |
+
"learning_rate": 6.876e-06,
|
| 2858 |
+
"loss": 0.1242,
|
| 2859 |
+
"step": 2300
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 5.82,
|
| 2863 |
+
"learning_rate": 6.891e-06,
|
| 2864 |
+
"loss": 0.1036,
|
| 2865 |
+
"step": 2305
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 5.83,
|
| 2869 |
+
"learning_rate": 6.906e-06,
|
| 2870 |
+
"loss": 0.109,
|
| 2871 |
+
"step": 2310
|
| 2872 |
+
},
|
| 2873 |
+
{
|
| 2874 |
+
"epoch": 5.84,
|
| 2875 |
+
"learning_rate": 6.9209999999999995e-06,
|
| 2876 |
+
"loss": 0.1163,
|
| 2877 |
+
"step": 2315
|
| 2878 |
+
},
|
| 2879 |
+
{
|
| 2880 |
+
"epoch": 5.85,
|
| 2881 |
+
"learning_rate": 6.936e-06,
|
| 2882 |
+
"loss": 0.0939,
|
| 2883 |
+
"step": 2320
|
| 2884 |
+
},
|
| 2885 |
+
{
|
| 2886 |
+
"epoch": 5.87,
|
| 2887 |
+
"learning_rate": 6.951e-06,
|
| 2888 |
+
"loss": 0.1015,
|
| 2889 |
+
"step": 2325
|
| 2890 |
+
},
|
| 2891 |
+
{
|
| 2892 |
+
"epoch": 5.88,
|
| 2893 |
+
"learning_rate": 6.966e-06,
|
| 2894 |
+
"loss": 0.112,
|
| 2895 |
+
"step": 2330
|
| 2896 |
+
},
|
| 2897 |
+
{
|
| 2898 |
+
"epoch": 5.89,
|
| 2899 |
+
"learning_rate": 6.981e-06,
|
| 2900 |
+
"loss": 0.1072,
|
| 2901 |
+
"step": 2335
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 5.91,
|
| 2905 |
+
"learning_rate": 6.996e-06,
|
| 2906 |
+
"loss": 0.109,
|
| 2907 |
+
"step": 2340
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 5.92,
|
| 2911 |
+
"learning_rate": 7.011e-06,
|
| 2912 |
+
"loss": 0.1115,
|
| 2913 |
+
"step": 2345
|
| 2914 |
+
},
|
| 2915 |
+
{
|
| 2916 |
+
"epoch": 5.93,
|
| 2917 |
+
"learning_rate": 7.026e-06,
|
| 2918 |
+
"loss": 0.1295,
|
| 2919 |
+
"step": 2350
|
| 2920 |
+
},
|
| 2921 |
+
{
|
| 2922 |
+
"epoch": 5.94,
|
| 2923 |
+
"learning_rate": 7.041e-06,
|
| 2924 |
+
"loss": 0.094,
|
| 2925 |
+
"step": 2355
|
| 2926 |
+
},
|
| 2927 |
+
{
|
| 2928 |
+
"epoch": 5.96,
|
| 2929 |
+
"learning_rate": 7.056e-06,
|
| 2930 |
+
"loss": 0.1087,
|
| 2931 |
+
"step": 2360
|
| 2932 |
+
},
|
| 2933 |
+
{
|
| 2934 |
+
"epoch": 5.97,
|
| 2935 |
+
"learning_rate": 7.071e-06,
|
| 2936 |
+
"loss": 0.1491,
|
| 2937 |
+
"step": 2365
|
| 2938 |
+
},
|
| 2939 |
+
{
|
| 2940 |
+
"epoch": 5.98,
|
| 2941 |
+
"learning_rate": 7.086e-06,
|
| 2942 |
+
"loss": 0.1048,
|
| 2943 |
+
"step": 2370
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 5.99,
|
| 2947 |
+
"learning_rate": 7.101e-06,
|
| 2948 |
+
"loss": 0.0986,
|
| 2949 |
+
"step": 2375
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 6.01,
|
| 2953 |
+
"learning_rate": 7.116e-06,
|
| 2954 |
+
"loss": 0.1152,
|
| 2955 |
+
"step": 2380
|
| 2956 |
+
},
|
| 2957 |
+
{
|
| 2958 |
+
"epoch": 6.02,
|
| 2959 |
+
"learning_rate": 7.131e-06,
|
| 2960 |
+
"loss": 0.0956,
|
| 2961 |
+
"step": 2385
|
| 2962 |
+
},
|
| 2963 |
+
{
|
| 2964 |
+
"epoch": 6.03,
|
| 2965 |
+
"learning_rate": 7.146e-06,
|
| 2966 |
+
"loss": 0.0967,
|
| 2967 |
+
"step": 2390
|
| 2968 |
+
},
|
| 2969 |
+
{
|
| 2970 |
+
"epoch": 6.04,
|
| 2971 |
+
"learning_rate": 7.161e-06,
|
| 2972 |
+
"loss": 0.1229,
|
| 2973 |
+
"step": 2395
|
| 2974 |
+
},
|
| 2975 |
+
{
|
| 2976 |
+
"epoch": 6.06,
|
| 2977 |
+
"learning_rate": 7.176e-06,
|
| 2978 |
+
"loss": 0.0931,
|
| 2979 |
+
"step": 2400
|
| 2980 |
+
},
|
| 2981 |
+
{
|
| 2982 |
+
"epoch": 6.07,
|
| 2983 |
+
"learning_rate": 7.191e-06,
|
| 2984 |
+
"loss": 0.0938,
|
| 2985 |
+
"step": 2405
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 6.08,
|
| 2989 |
+
"learning_rate": 7.206e-06,
|
| 2990 |
+
"loss": 0.1209,
|
| 2991 |
+
"step": 2410
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 6.09,
|
| 2995 |
+
"learning_rate": 7.221e-06,
|
| 2996 |
+
"loss": 0.1081,
|
| 2997 |
+
"step": 2415
|
| 2998 |
+
},
|
| 2999 |
+
{
|
| 3000 |
+
"epoch": 6.11,
|
| 3001 |
+
"learning_rate": 7.236e-06,
|
| 3002 |
+
"loss": 0.0918,
|
| 3003 |
+
"step": 2420
|
| 3004 |
+
},
|
| 3005 |
+
{
|
| 3006 |
+
"epoch": 6.12,
|
| 3007 |
+
"learning_rate": 7.2510000000000005e-06,
|
| 3008 |
+
"loss": 0.0979,
|
| 3009 |
+
"step": 2425
|
| 3010 |
+
},
|
| 3011 |
+
{
|
| 3012 |
+
"epoch": 6.13,
|
| 3013 |
+
"learning_rate": 7.266e-06,
|
| 3014 |
+
"loss": 0.1218,
|
| 3015 |
+
"step": 2430
|
| 3016 |
+
},
|
| 3017 |
+
{
|
| 3018 |
+
"epoch": 6.15,
|
| 3019 |
+
"learning_rate": 7.281e-06,
|
| 3020 |
+
"loss": 0.0971,
|
| 3021 |
+
"step": 2435
|
| 3022 |
+
},
|
| 3023 |
+
{
|
| 3024 |
+
"epoch": 6.16,
|
| 3025 |
+
"learning_rate": 7.296e-06,
|
| 3026 |
+
"loss": 0.0977,
|
| 3027 |
+
"step": 2440
|
| 3028 |
+
},
|
| 3029 |
+
{
|
| 3030 |
+
"epoch": 6.17,
|
| 3031 |
+
"learning_rate": 7.311e-06,
|
| 3032 |
+
"loss": 0.1212,
|
| 3033 |
+
"step": 2445
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 6.18,
|
| 3037 |
+
"learning_rate": 7.326e-06,
|
| 3038 |
+
"loss": 0.0886,
|
| 3039 |
+
"step": 2450
|
| 3040 |
+
},
|
| 3041 |
+
{
|
| 3042 |
+
"epoch": 6.2,
|
| 3043 |
+
"learning_rate": 7.3410000000000005e-06,
|
| 3044 |
+
"loss": 0.0943,
|
| 3045 |
+
"step": 2455
|
| 3046 |
+
},
|
| 3047 |
+
{
|
| 3048 |
+
"epoch": 6.21,
|
| 3049 |
+
"learning_rate": 7.356e-06,
|
| 3050 |
+
"loss": 0.1049,
|
| 3051 |
+
"step": 2460
|
| 3052 |
+
},
|
| 3053 |
+
{
|
| 3054 |
+
"epoch": 6.22,
|
| 3055 |
+
"learning_rate": 7.371e-06,
|
| 3056 |
+
"loss": 0.0964,
|
| 3057 |
+
"step": 2465
|
| 3058 |
+
},
|
| 3059 |
+
{
|
| 3060 |
+
"epoch": 6.23,
|
| 3061 |
+
"learning_rate": 7.386e-06,
|
| 3062 |
+
"loss": 0.0978,
|
| 3063 |
+
"step": 2470
|
| 3064 |
+
},
|
| 3065 |
+
{
|
| 3066 |
+
"epoch": 6.25,
|
| 3067 |
+
"learning_rate": 7.401e-06,
|
| 3068 |
+
"loss": 0.1037,
|
| 3069 |
+
"step": 2475
|
| 3070 |
+
},
|
| 3071 |
+
{
|
| 3072 |
+
"epoch": 6.26,
|
| 3073 |
+
"learning_rate": 7.416000000000001e-06,
|
| 3074 |
+
"loss": 0.1043,
|
| 3075 |
+
"step": 2480
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 6.27,
|
| 3079 |
+
"learning_rate": 7.4310000000000005e-06,
|
| 3080 |
+
"loss": 0.0984,
|
| 3081 |
+
"step": 2485
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 6.28,
|
| 3085 |
+
"learning_rate": 7.446e-06,
|
| 3086 |
+
"loss": 0.0903,
|
| 3087 |
+
"step": 2490
|
| 3088 |
+
},
|
| 3089 |
+
{
|
| 3090 |
+
"epoch": 6.3,
|
| 3091 |
+
"learning_rate": 7.4579999999999996e-06,
|
| 3092 |
+
"loss": 0.1169,
|
| 3093 |
+
"step": 2495
|
| 3094 |
+
},
|
| 3095 |
+
{
|
| 3096 |
+
"epoch": 6.31,
|
| 3097 |
+
"learning_rate": 7.4729999999999994e-06,
|
| 3098 |
+
"loss": 0.0942,
|
| 3099 |
+
"step": 2500
|
| 3100 |
+
},
|
| 3101 |
+
{
|
| 3102 |
+
"epoch": 6.31,
|
| 3103 |
+
"eval_cer": 0.02133029812895625,
|
| 3104 |
+
"eval_loss": 0.11278611421585083,
|
| 3105 |
+
"eval_runtime": 59.9521,
|
| 3106 |
+
"eval_samples_per_second": 42.167,
|
| 3107 |
+
"eval_steps_per_second": 1.184,
|
| 3108 |
+
"eval_wer": 0.0794277466462897,
|
| 3109 |
+
"step": 2500
|
| 3110 |
+
},
|
| 3111 |
+
{
|
| 3112 |
+
"epoch": 6.32,
|
| 3113 |
+
"learning_rate": 7.488e-06,
|
| 3114 |
+
"loss": 0.0928,
|
| 3115 |
+
"step": 2505
|
| 3116 |
+
},
|
| 3117 |
+
{
|
| 3118 |
+
"epoch": 6.33,
|
| 3119 |
+
"learning_rate": 7.503e-06,
|
| 3120 |
+
"loss": 0.1041,
|
| 3121 |
+
"step": 2510
|
| 3122 |
+
},
|
| 3123 |
+
{
|
| 3124 |
+
"epoch": 6.35,
|
| 3125 |
+
"learning_rate": 7.518e-06,
|
| 3126 |
+
"loss": 0.1044,
|
| 3127 |
+
"step": 2515
|
| 3128 |
+
},
|
| 3129 |
+
{
|
| 3130 |
+
"epoch": 6.36,
|
| 3131 |
+
"learning_rate": 7.533e-06,
|
| 3132 |
+
"loss": 0.0916,
|
| 3133 |
+
"step": 2520
|
| 3134 |
+
},
|
| 3135 |
+
{
|
| 3136 |
+
"epoch": 6.37,
|
| 3137 |
+
"learning_rate": 7.5479999999999996e-06,
|
| 3138 |
+
"loss": 0.0963,
|
| 3139 |
+
"step": 2525
|
| 3140 |
+
},
|
| 3141 |
+
{
|
| 3142 |
+
"epoch": 6.38,
|
| 3143 |
+
"learning_rate": 7.563e-06,
|
| 3144 |
+
"loss": 0.1231,
|
| 3145 |
+
"step": 2530
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 6.4,
|
| 3149 |
+
"learning_rate": 7.578e-06,
|
| 3150 |
+
"loss": 0.09,
|
| 3151 |
+
"step": 2535
|
| 3152 |
+
},
|
| 3153 |
+
{
|
| 3154 |
+
"epoch": 6.41,
|
| 3155 |
+
"learning_rate": 7.593e-06,
|
| 3156 |
+
"loss": 0.0903,
|
| 3157 |
+
"step": 2540
|
| 3158 |
+
},
|
| 3159 |
+
{
|
| 3160 |
+
"epoch": 6.42,
|
| 3161 |
+
"learning_rate": 7.608e-06,
|
| 3162 |
+
"loss": 0.1099,
|
| 3163 |
+
"step": 2545
|
| 3164 |
+
},
|
| 3165 |
+
{
|
| 3166 |
+
"epoch": 6.44,
|
| 3167 |
+
"learning_rate": 7.623e-06,
|
| 3168 |
+
"loss": 0.0868,
|
| 3169 |
+
"step": 2550
|
| 3170 |
+
},
|
| 3171 |
+
{
|
| 3172 |
+
"epoch": 6.45,
|
| 3173 |
+
"learning_rate": 7.638e-06,
|
| 3174 |
+
"loss": 0.086,
|
| 3175 |
+
"step": 2555
|
| 3176 |
+
},
|
| 3177 |
+
{
|
| 3178 |
+
"epoch": 6.46,
|
| 3179 |
+
"learning_rate": 7.653e-06,
|
| 3180 |
+
"loss": 0.1063,
|
| 3181 |
+
"step": 2560
|
| 3182 |
+
},
|
| 3183 |
+
{
|
| 3184 |
+
"epoch": 6.47,
|
| 3185 |
+
"learning_rate": 7.668e-06,
|
| 3186 |
+
"loss": 0.1004,
|
| 3187 |
+
"step": 2565
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 6.49,
|
| 3191 |
+
"learning_rate": 7.683e-06,
|
| 3192 |
+
"loss": 0.0893,
|
| 3193 |
+
"step": 2570
|
| 3194 |
+
},
|
| 3195 |
+
{
|
| 3196 |
+
"epoch": 6.5,
|
| 3197 |
+
"learning_rate": 7.698e-06,
|
| 3198 |
+
"loss": 0.0942,
|
| 3199 |
+
"step": 2575
|
| 3200 |
+
},
|
| 3201 |
+
{
|
| 3202 |
+
"epoch": 6.51,
|
| 3203 |
+
"learning_rate": 7.713e-06,
|
| 3204 |
+
"loss": 0.1092,
|
| 3205 |
+
"step": 2580
|
| 3206 |
+
},
|
| 3207 |
+
{
|
| 3208 |
+
"epoch": 6.52,
|
| 3209 |
+
"learning_rate": 7.728e-06,
|
| 3210 |
+
"loss": 0.0904,
|
| 3211 |
+
"step": 2585
|
| 3212 |
+
},
|
| 3213 |
+
{
|
| 3214 |
+
"epoch": 6.54,
|
| 3215 |
+
"learning_rate": 7.743e-06,
|
| 3216 |
+
"loss": 0.087,
|
| 3217 |
+
"step": 2590
|
| 3218 |
+
},
|
| 3219 |
+
{
|
| 3220 |
+
"epoch": 6.55,
|
| 3221 |
+
"learning_rate": 7.758000000000001e-06,
|
| 3222 |
+
"loss": 0.1143,
|
| 3223 |
+
"step": 2595
|
| 3224 |
+
},
|
| 3225 |
+
{
|
| 3226 |
+
"epoch": 6.56,
|
| 3227 |
+
"learning_rate": 7.773e-06,
|
| 3228 |
+
"loss": 0.0878,
|
| 3229 |
+
"step": 2600
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 6.57,
|
| 3233 |
+
"learning_rate": 7.788e-06,
|
| 3234 |
+
"loss": 0.0854,
|
| 3235 |
+
"step": 2605
|
| 3236 |
+
},
|
| 3237 |
+
{
|
| 3238 |
+
"epoch": 6.59,
|
| 3239 |
+
"learning_rate": 7.803e-06,
|
| 3240 |
+
"loss": 0.1021,
|
| 3241 |
+
"step": 2610
|
| 3242 |
+
},
|
| 3243 |
+
{
|
| 3244 |
+
"epoch": 6.6,
|
| 3245 |
+
"learning_rate": 7.818e-06,
|
| 3246 |
+
"loss": 0.1077,
|
| 3247 |
+
"step": 2615
|
| 3248 |
+
},
|
| 3249 |
+
{
|
| 3250 |
+
"epoch": 6.61,
|
| 3251 |
+
"learning_rate": 7.833e-06,
|
| 3252 |
+
"loss": 0.0884,
|
| 3253 |
+
"step": 2620
|
| 3254 |
+
},
|
| 3255 |
+
{
|
| 3256 |
+
"epoch": 6.62,
|
| 3257 |
+
"learning_rate": 7.848e-06,
|
| 3258 |
+
"loss": 0.09,
|
| 3259 |
+
"step": 2625
|
| 3260 |
+
},
|
| 3261 |
+
{
|
| 3262 |
+
"epoch": 6.64,
|
| 3263 |
+
"learning_rate": 7.863e-06,
|
| 3264 |
+
"loss": 0.1044,
|
| 3265 |
+
"step": 2630
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 6.65,
|
| 3269 |
+
"learning_rate": 7.878e-06,
|
| 3270 |
+
"loss": 0.083,
|
| 3271 |
+
"step": 2635
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 6.66,
|
| 3275 |
+
"learning_rate": 7.893e-06,
|
| 3276 |
+
"loss": 0.0884,
|
| 3277 |
+
"step": 2640
|
| 3278 |
+
},
|
| 3279 |
+
{
|
| 3280 |
+
"epoch": 6.67,
|
| 3281 |
+
"learning_rate": 7.908e-06,
|
| 3282 |
+
"loss": 0.1213,
|
| 3283 |
+
"step": 2645
|
| 3284 |
+
},
|
| 3285 |
+
{
|
| 3286 |
+
"epoch": 6.69,
|
| 3287 |
+
"learning_rate": 7.923e-06,
|
| 3288 |
+
"loss": 0.0983,
|
| 3289 |
+
"step": 2650
|
| 3290 |
+
},
|
| 3291 |
+
{
|
| 3292 |
+
"epoch": 6.7,
|
| 3293 |
+
"learning_rate": 7.938000000000001e-06,
|
| 3294 |
+
"loss": 0.0923,
|
| 3295 |
+
"step": 2655
|
| 3296 |
+
},
|
| 3297 |
+
{
|
| 3298 |
+
"epoch": 6.71,
|
| 3299 |
+
"learning_rate": 7.953e-06,
|
| 3300 |
+
"loss": 0.1014,
|
| 3301 |
+
"step": 2660
|
| 3302 |
+
},
|
| 3303 |
+
{
|
| 3304 |
+
"epoch": 6.73,
|
| 3305 |
+
"learning_rate": 7.968e-06,
|
| 3306 |
+
"loss": 0.0897,
|
| 3307 |
+
"step": 2665
|
| 3308 |
+
},
|
| 3309 |
+
{
|
| 3310 |
+
"epoch": 6.74,
|
| 3311 |
+
"learning_rate": 7.983e-06,
|
| 3312 |
+
"loss": 0.0831,
|
| 3313 |
+
"step": 2670
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 6.75,
|
| 3317 |
+
"learning_rate": 7.998e-06,
|
| 3318 |
+
"loss": 0.0856,
|
| 3319 |
+
"step": 2675
|
| 3320 |
+
},
|
| 3321 |
+
{
|
| 3322 |
+
"epoch": 6.76,
|
| 3323 |
+
"learning_rate": 8.013000000000001e-06,
|
| 3324 |
+
"loss": 0.1153,
|
| 3325 |
+
"step": 2680
|
| 3326 |
+
},
|
| 3327 |
+
{
|
| 3328 |
+
"epoch": 6.78,
|
| 3329 |
+
"learning_rate": 8.028e-06,
|
| 3330 |
+
"loss": 0.0824,
|
| 3331 |
+
"step": 2685
|
| 3332 |
+
},
|
| 3333 |
+
{
|
| 3334 |
+
"epoch": 6.79,
|
| 3335 |
+
"learning_rate": 8.043e-06,
|
| 3336 |
+
"loss": 0.0869,
|
| 3337 |
+
"step": 2690
|
| 3338 |
+
},
|
| 3339 |
+
{
|
| 3340 |
+
"epoch": 6.8,
|
| 3341 |
+
"learning_rate": 8.058e-06,
|
| 3342 |
+
"loss": 0.1197,
|
| 3343 |
+
"step": 2695
|
| 3344 |
+
},
|
| 3345 |
+
{
|
| 3346 |
+
"epoch": 6.81,
|
| 3347 |
+
"learning_rate": 8.073e-06,
|
| 3348 |
+
"loss": 0.0872,
|
| 3349 |
+
"step": 2700
|
| 3350 |
+
},
|
| 3351 |
+
{
|
| 3352 |
+
"epoch": 6.83,
|
| 3353 |
+
"learning_rate": 8.088000000000001e-06,
|
| 3354 |
+
"loss": 0.0873,
|
| 3355 |
+
"step": 2705
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 6.84,
|
| 3359 |
+
"learning_rate": 8.103e-06,
|
| 3360 |
+
"loss": 0.1031,
|
| 3361 |
+
"step": 2710
|
| 3362 |
+
},
|
| 3363 |
+
{
|
| 3364 |
+
"epoch": 6.85,
|
| 3365 |
+
"learning_rate": 8.118000000000001e-06,
|
| 3366 |
+
"loss": 0.0902,
|
| 3367 |
+
"step": 2715
|
| 3368 |
+
},
|
| 3369 |
+
{
|
| 3370 |
+
"epoch": 6.86,
|
| 3371 |
+
"learning_rate": 8.133e-06,
|
| 3372 |
+
"loss": 0.0878,
|
| 3373 |
+
"step": 2720
|
| 3374 |
+
},
|
| 3375 |
+
{
|
| 3376 |
+
"epoch": 6.88,
|
| 3377 |
+
"learning_rate": 8.148e-06,
|
| 3378 |
+
"loss": 0.099,
|
| 3379 |
+
"step": 2725
|
| 3380 |
+
},
|
| 3381 |
+
{
|
| 3382 |
+
"epoch": 6.89,
|
| 3383 |
+
"learning_rate": 8.163000000000001e-06,
|
| 3384 |
+
"loss": 0.1037,
|
| 3385 |
+
"step": 2730
|
| 3386 |
+
},
|
| 3387 |
+
{
|
| 3388 |
+
"epoch": 6.9,
|
| 3389 |
+
"learning_rate": 8.178e-06,
|
| 3390 |
+
"loss": 0.0897,
|
| 3391 |
+
"step": 2735
|
| 3392 |
+
},
|
| 3393 |
+
{
|
| 3394 |
+
"epoch": 6.91,
|
| 3395 |
+
"learning_rate": 8.193000000000001e-06,
|
| 3396 |
+
"loss": 0.0888,
|
| 3397 |
+
"step": 2740
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 6.93,
|
| 3401 |
+
"learning_rate": 8.208e-06,
|
| 3402 |
+
"loss": 0.1109,
|
| 3403 |
+
"step": 2745
|
| 3404 |
+
},
|
| 3405 |
+
{
|
| 3406 |
+
"epoch": 6.94,
|
| 3407 |
+
"learning_rate": 8.223e-06,
|
| 3408 |
+
"loss": 0.0848,
|
| 3409 |
+
"step": 2750
|
| 3410 |
+
},
|
| 3411 |
+
{
|
| 3412 |
+
"epoch": 6.94,
|
| 3413 |
+
"eval_cer": 0.01967291679146948,
|
| 3414 |
+
"eval_loss": 0.10771536827087402,
|
| 3415 |
+
"eval_runtime": 59.6954,
|
| 3416 |
+
"eval_samples_per_second": 42.348,
|
| 3417 |
+
"eval_steps_per_second": 1.189,
|
| 3418 |
+
"eval_wer": 0.07172270334521991,
|
| 3419 |
+
"step": 2750
|
| 3420 |
+
},
|
| 3421 |
+
{
|
| 3422 |
+
"epoch": 6.95,
|
| 3423 |
+
"learning_rate": 8.238e-06,
|
| 3424 |
+
"loss": 0.0871,
|
| 3425 |
+
"step": 2755
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 6.96,
|
| 3429 |
+
"learning_rate": 8.253e-06,
|
| 3430 |
+
"loss": 0.1006,
|
| 3431 |
+
"step": 2760
|
| 3432 |
+
},
|
| 3433 |
+
{
|
| 3434 |
+
"epoch": 6.98,
|
| 3435 |
+
"learning_rate": 8.268000000000001e-06,
|
| 3436 |
+
"loss": 0.1059,
|
| 3437 |
+
"step": 2765
|
| 3438 |
+
},
|
| 3439 |
+
{
|
| 3440 |
+
"epoch": 6.99,
|
| 3441 |
+
"learning_rate": 8.283e-06,
|
| 3442 |
+
"loss": 0.0973,
|
| 3443 |
+
"step": 2770
|
| 3444 |
+
},
|
| 3445 |
+
{
|
| 3446 |
+
"epoch": 6.99,
|
| 3447 |
+
"step": 2772,
|
| 3448 |
+
"total_flos": 7.402330898771948e+19,
|
| 3449 |
+
"train_loss": 0.05688602840165039,
|
| 3450 |
+
"train_runtime": 6445.258,
|
| 3451 |
+
"train_samples_per_second": 30.994,
|
| 3452 |
+
"train_steps_per_second": 0.43
|
| 3453 |
}
|
| 3454 |
],
|
| 3455 |
"logging_steps": 5,
|
| 3456 |
+
"max_steps": 2772,
|
| 3457 |
"num_input_tokens_seen": 0,
|
| 3458 |
+
"num_train_epochs": 7,
|
| 3459 |
"save_steps": 500,
|
| 3460 |
+
"total_flos": 7.402330898771948e+19,
|
| 3461 |
"train_batch_size": 12,
|
| 3462 |
"trial_name": null,
|
| 3463 |
"trial_params": null
|