Training in progress, step 350, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 838904832
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:420b8a326986f8e830cd838bbf34a7080a1e6d4ec3ce8653b2e4eebadcfa2c77
|
3 |
size 838904832
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 421458386
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b7c1879fc8f1886e9f801904c72d7dbe4bf99bc2fd9a6c759591a82e966b8031
|
3 |
size 421458386
|
last-checkpoint/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65a3ee4d233a8121789677494e2ecd1a75477024c32056ca4bbd20461a9ed78a
|
3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:32cd83ea7b3c3d26f4b1d83df062610c973ecff83b1326ee086a5f9f50324c14
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 1.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -1957,13 +1957,163 @@
|
|
1957 |
"learning_rate": 0.0002,
|
1958 |
"loss": 0.4449,
|
1959 |
"step": 325
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1960 |
}
|
1961 |
],
|
1962 |
"logging_steps": 1,
|
1963 |
"max_steps": 500,
|
1964 |
"num_train_epochs": 2,
|
1965 |
"save_steps": 25,
|
1966 |
-
"total_flos":
|
1967 |
"trial_name": null,
|
1968 |
"trial_params": null
|
1969 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.4,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 350,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
1957 |
"learning_rate": 0.0002,
|
1958 |
"loss": 0.4449,
|
1959 |
"step": 325
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 1.3,
|
1963 |
+
"learning_rate": 0.0002,
|
1964 |
+
"loss": 0.3912,
|
1965 |
+
"step": 326
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 1.31,
|
1969 |
+
"learning_rate": 0.0002,
|
1970 |
+
"loss": 0.4508,
|
1971 |
+
"step": 327
|
1972 |
+
},
|
1973 |
+
{
|
1974 |
+
"epoch": 1.31,
|
1975 |
+
"learning_rate": 0.0002,
|
1976 |
+
"loss": 0.4124,
|
1977 |
+
"step": 328
|
1978 |
+
},
|
1979 |
+
{
|
1980 |
+
"epoch": 1.32,
|
1981 |
+
"learning_rate": 0.0002,
|
1982 |
+
"loss": 0.4305,
|
1983 |
+
"step": 329
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 1.32,
|
1987 |
+
"learning_rate": 0.0002,
|
1988 |
+
"loss": 0.4207,
|
1989 |
+
"step": 330
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 1.32,
|
1993 |
+
"learning_rate": 0.0002,
|
1994 |
+
"loss": 0.3785,
|
1995 |
+
"step": 331
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 1.33,
|
1999 |
+
"learning_rate": 0.0002,
|
2000 |
+
"loss": 0.428,
|
2001 |
+
"step": 332
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 1.33,
|
2005 |
+
"learning_rate": 0.0002,
|
2006 |
+
"loss": 0.3683,
|
2007 |
+
"step": 333
|
2008 |
+
},
|
2009 |
+
{
|
2010 |
+
"epoch": 1.34,
|
2011 |
+
"learning_rate": 0.0002,
|
2012 |
+
"loss": 0.3742,
|
2013 |
+
"step": 334
|
2014 |
+
},
|
2015 |
+
{
|
2016 |
+
"epoch": 1.34,
|
2017 |
+
"learning_rate": 0.0002,
|
2018 |
+
"loss": 0.3734,
|
2019 |
+
"step": 335
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 1.34,
|
2023 |
+
"learning_rate": 0.0002,
|
2024 |
+
"loss": 0.3748,
|
2025 |
+
"step": 336
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 1.35,
|
2029 |
+
"learning_rate": 0.0002,
|
2030 |
+
"loss": 0.4496,
|
2031 |
+
"step": 337
|
2032 |
+
},
|
2033 |
+
{
|
2034 |
+
"epoch": 1.35,
|
2035 |
+
"learning_rate": 0.0002,
|
2036 |
+
"loss": 0.3368,
|
2037 |
+
"step": 338
|
2038 |
+
},
|
2039 |
+
{
|
2040 |
+
"epoch": 1.36,
|
2041 |
+
"learning_rate": 0.0002,
|
2042 |
+
"loss": 0.3801,
|
2043 |
+
"step": 339
|
2044 |
+
},
|
2045 |
+
{
|
2046 |
+
"epoch": 1.36,
|
2047 |
+
"learning_rate": 0.0002,
|
2048 |
+
"loss": 0.3133,
|
2049 |
+
"step": 340
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 1.36,
|
2053 |
+
"learning_rate": 0.0002,
|
2054 |
+
"loss": 0.3159,
|
2055 |
+
"step": 341
|
2056 |
+
},
|
2057 |
+
{
|
2058 |
+
"epoch": 1.37,
|
2059 |
+
"learning_rate": 0.0002,
|
2060 |
+
"loss": 0.3227,
|
2061 |
+
"step": 342
|
2062 |
+
},
|
2063 |
+
{
|
2064 |
+
"epoch": 1.37,
|
2065 |
+
"learning_rate": 0.0002,
|
2066 |
+
"loss": 0.3268,
|
2067 |
+
"step": 343
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 1.38,
|
2071 |
+
"learning_rate": 0.0002,
|
2072 |
+
"loss": 0.402,
|
2073 |
+
"step": 344
|
2074 |
+
},
|
2075 |
+
{
|
2076 |
+
"epoch": 1.38,
|
2077 |
+
"learning_rate": 0.0002,
|
2078 |
+
"loss": 0.327,
|
2079 |
+
"step": 345
|
2080 |
+
},
|
2081 |
+
{
|
2082 |
+
"epoch": 1.38,
|
2083 |
+
"learning_rate": 0.0002,
|
2084 |
+
"loss": 0.2898,
|
2085 |
+
"step": 346
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 1.39,
|
2089 |
+
"learning_rate": 0.0002,
|
2090 |
+
"loss": 0.3437,
|
2091 |
+
"step": 347
|
2092 |
+
},
|
2093 |
+
{
|
2094 |
+
"epoch": 1.39,
|
2095 |
+
"learning_rate": 0.0002,
|
2096 |
+
"loss": 0.3099,
|
2097 |
+
"step": 348
|
2098 |
+
},
|
2099 |
+
{
|
2100 |
+
"epoch": 1.4,
|
2101 |
+
"learning_rate": 0.0002,
|
2102 |
+
"loss": 0.2742,
|
2103 |
+
"step": 349
|
2104 |
+
},
|
2105 |
+
{
|
2106 |
+
"epoch": 1.4,
|
2107 |
+
"learning_rate": 0.0002,
|
2108 |
+
"loss": 0.3122,
|
2109 |
+
"step": 350
|
2110 |
}
|
2111 |
],
|
2112 |
"logging_steps": 1,
|
2113 |
"max_steps": 500,
|
2114 |
"num_train_epochs": 2,
|
2115 |
"save_steps": 25,
|
2116 |
+
"total_flos": 4.187771329769472e+16,
|
2117 |
"trial_name": null,
|
2118 |
"trial_params": null
|
2119 |
}
|