Muhammad Khalifa
commited on
Commit
•
9e05026
1
Parent(s):
5a9e909
add instrucode llama-2 13b
Browse files- with_input/decomp_code_with_intermediates/llama-2-13b/best_model/adapter_model.bin +1 -1
- with_input/decomp_code_with_intermediates/llama-2-13b/best_model/optimizer.pt +1 -1
- with_input/decomp_code_with_intermediates/llama-2-13b/best_model/rng_state.pth +1 -1
- with_input/decomp_code_with_intermediates/llama-2-13b/best_model/scheduler.pt +1 -1
- with_input/decomp_code_with_intermediates/llama-2-13b/best_model/trainer_state.json +645 -5
with_input/decomp_code_with_intermediates/llama-2-13b/best_model/adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 104973389
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a4671c2d6162e00bd0679f6a17788fe6ca996a105b9eec0605c113d55d51046e
|
| 3 |
size 104973389
|
with_input/decomp_code_with_intermediates/llama-2-13b/best_model/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 209984517
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0b33193e2eaa387f1f2594dcd74e6f6c7f8d0bb6ccad541fdc73f56d8806f54
|
| 3 |
size 209984517
|
with_input/decomp_code_with_intermediates/llama-2-13b/best_model/rng_state.pth
CHANGED
|
@@ -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:c7a3940c13988eef2d142987af977c8946a726c9931362bfe39e6700c5381106
|
| 3 |
size 14575
|
with_input/decomp_code_with_intermediates/llama-2-13b/best_model/scheduler.pt
CHANGED
|
@@ -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 sha256:1a28d59beefca59c5677feb27231298b8d4e00afdda714d30a1a974e09b38f41
|
| 3 |
size 627
|
with_input/decomp_code_with_intermediates/llama-2-13b/best_model/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "checkpoints/instrucode/with_input/decomp_code_with_intermediates/llama-2-13b/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 200,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1927,13 +1927,653 @@
|
|
| 1927 |
"eval_samples_per_second": 2.461,
|
| 1928 |
"eval_steps_per_second": 0.308,
|
| 1929 |
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1930 |
}
|
| 1931 |
],
|
| 1932 |
"logging_steps": 10,
|
| 1933 |
"max_steps": 7775,
|
| 1934 |
"num_train_epochs": 5,
|
| 1935 |
"save_steps": 200,
|
| 1936 |
-
"total_flos":
|
| 1937 |
"trial_name": null,
|
| 1938 |
"trial_params": null
|
| 1939 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.2936367690563202,
|
| 3 |
+
"best_model_checkpoint": "checkpoints/instrucode/with_input/decomp_code_with_intermediates/llama-2-13b/checkpoint-4000",
|
| 4 |
+
"epoch": 2.572347266881029,
|
| 5 |
"eval_steps": 200,
|
| 6 |
+
"global_step": 4000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1927 |
"eval_samples_per_second": 2.461,
|
| 1928 |
"eval_steps_per_second": 0.308,
|
| 1929 |
"step": 3000
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 1.94,
|
| 1933 |
+
"learning_rate": 0.00018633224755700326,
|
| 1934 |
+
"loss": 0.2842,
|
| 1935 |
+
"step": 3010
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.94,
|
| 1939 |
+
"learning_rate": 0.00018594136807817586,
|
| 1940 |
+
"loss": 0.2872,
|
| 1941 |
+
"step": 3020
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 1.95,
|
| 1945 |
+
"learning_rate": 0.00018555048859934852,
|
| 1946 |
+
"loss": 0.2872,
|
| 1947 |
+
"step": 3030
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 1.95,
|
| 1951 |
+
"learning_rate": 0.00018515960912052115,
|
| 1952 |
+
"loss": 0.2781,
|
| 1953 |
+
"step": 3040
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 1.96,
|
| 1957 |
+
"learning_rate": 0.0001847687296416938,
|
| 1958 |
+
"loss": 0.2933,
|
| 1959 |
+
"step": 3050
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 1.97,
|
| 1963 |
+
"learning_rate": 0.00018437785016286646,
|
| 1964 |
+
"loss": 0.2964,
|
| 1965 |
+
"step": 3060
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 1.97,
|
| 1969 |
+
"learning_rate": 0.00018398697068403906,
|
| 1970 |
+
"loss": 0.2787,
|
| 1971 |
+
"step": 3070
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.98,
|
| 1975 |
+
"learning_rate": 0.0001835960912052117,
|
| 1976 |
+
"loss": 0.2891,
|
| 1977 |
+
"step": 3080
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.99,
|
| 1981 |
+
"learning_rate": 0.00018320521172638434,
|
| 1982 |
+
"loss": 0.2875,
|
| 1983 |
+
"step": 3090
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 1.99,
|
| 1987 |
+
"learning_rate": 0.000182814332247557,
|
| 1988 |
+
"loss": 0.294,
|
| 1989 |
+
"step": 3100
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 2.0,
|
| 1993 |
+
"learning_rate": 0.00018242345276872965,
|
| 1994 |
+
"loss": 0.2793,
|
| 1995 |
+
"step": 3110
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 2.01,
|
| 1999 |
+
"learning_rate": 0.00018203257328990225,
|
| 2000 |
+
"loss": 0.2884,
|
| 2001 |
+
"step": 3120
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 2.01,
|
| 2005 |
+
"learning_rate": 0.0001816416938110749,
|
| 2006 |
+
"loss": 0.2795,
|
| 2007 |
+
"step": 3130
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 2.02,
|
| 2011 |
+
"learning_rate": 0.00018125081433224754,
|
| 2012 |
+
"loss": 0.2722,
|
| 2013 |
+
"step": 3140
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 2.03,
|
| 2017 |
+
"learning_rate": 0.0001808599348534202,
|
| 2018 |
+
"loss": 0.2653,
|
| 2019 |
+
"step": 3150
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.03,
|
| 2023 |
+
"learning_rate": 0.00018046905537459282,
|
| 2024 |
+
"loss": 0.2664,
|
| 2025 |
+
"step": 3160
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.04,
|
| 2029 |
+
"learning_rate": 0.00018007817589576545,
|
| 2030 |
+
"loss": 0.2785,
|
| 2031 |
+
"step": 3170
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 2.05,
|
| 2035 |
+
"learning_rate": 0.00017968729641693808,
|
| 2036 |
+
"loss": 0.2728,
|
| 2037 |
+
"step": 3180
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 2.05,
|
| 2041 |
+
"learning_rate": 0.00017929641693811073,
|
| 2042 |
+
"loss": 0.2675,
|
| 2043 |
+
"step": 3190
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 2.06,
|
| 2047 |
+
"learning_rate": 0.00017890553745928339,
|
| 2048 |
+
"loss": 0.2607,
|
| 2049 |
+
"step": 3200
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 2.06,
|
| 2053 |
+
"eval_loss": 0.2949482798576355,
|
| 2054 |
+
"eval_runtime": 812.7173,
|
| 2055 |
+
"eval_samples_per_second": 2.461,
|
| 2056 |
+
"eval_steps_per_second": 0.308,
|
| 2057 |
+
"step": 3200
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 2.06,
|
| 2061 |
+
"learning_rate": 0.00017851465798045601,
|
| 2062 |
+
"loss": 0.2773,
|
| 2063 |
+
"step": 3210
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 2.07,
|
| 2067 |
+
"learning_rate": 0.00017812377850162864,
|
| 2068 |
+
"loss": 0.2684,
|
| 2069 |
+
"step": 3220
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.08,
|
| 2073 |
+
"learning_rate": 0.00017773289902280127,
|
| 2074 |
+
"loss": 0.2656,
|
| 2075 |
+
"step": 3230
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.08,
|
| 2079 |
+
"learning_rate": 0.00017734201954397393,
|
| 2080 |
+
"loss": 0.287,
|
| 2081 |
+
"step": 3240
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.09,
|
| 2085 |
+
"learning_rate": 0.00017695114006514658,
|
| 2086 |
+
"loss": 0.2694,
|
| 2087 |
+
"step": 3250
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 2.1,
|
| 2091 |
+
"learning_rate": 0.0001765602605863192,
|
| 2092 |
+
"loss": 0.2667,
|
| 2093 |
+
"step": 3260
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 2.1,
|
| 2097 |
+
"learning_rate": 0.00017616938110749184,
|
| 2098 |
+
"loss": 0.2744,
|
| 2099 |
+
"step": 3270
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 2.11,
|
| 2103 |
+
"learning_rate": 0.00017577850162866447,
|
| 2104 |
+
"loss": 0.2744,
|
| 2105 |
+
"step": 3280
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 2.12,
|
| 2109 |
+
"learning_rate": 0.00017538762214983712,
|
| 2110 |
+
"loss": 0.2615,
|
| 2111 |
+
"step": 3290
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 2.12,
|
| 2115 |
+
"learning_rate": 0.00017499674267100975,
|
| 2116 |
+
"loss": 0.2694,
|
| 2117 |
+
"step": 3300
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.13,
|
| 2121 |
+
"learning_rate": 0.0001746058631921824,
|
| 2122 |
+
"loss": 0.281,
|
| 2123 |
+
"step": 3310
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.14,
|
| 2127 |
+
"learning_rate": 0.00017421498371335506,
|
| 2128 |
+
"loss": 0.2744,
|
| 2129 |
+
"step": 3320
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 2.14,
|
| 2133 |
+
"learning_rate": 0.00017382410423452766,
|
| 2134 |
+
"loss": 0.268,
|
| 2135 |
+
"step": 3330
|
| 2136 |
+
},
|
| 2137 |
+
{
|
| 2138 |
+
"epoch": 2.15,
|
| 2139 |
+
"learning_rate": 0.00017343322475570032,
|
| 2140 |
+
"loss": 0.2859,
|
| 2141 |
+
"step": 3340
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 2.15,
|
| 2145 |
+
"learning_rate": 0.00017304234527687294,
|
| 2146 |
+
"loss": 0.2631,
|
| 2147 |
+
"step": 3350
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 2.16,
|
| 2151 |
+
"learning_rate": 0.0001726514657980456,
|
| 2152 |
+
"loss": 0.2791,
|
| 2153 |
+
"step": 3360
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 2.17,
|
| 2157 |
+
"learning_rate": 0.00017226058631921825,
|
| 2158 |
+
"loss": 0.2873,
|
| 2159 |
+
"step": 3370
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.17,
|
| 2163 |
+
"learning_rate": 0.00017186970684039086,
|
| 2164 |
+
"loss": 0.2712,
|
| 2165 |
+
"step": 3380
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.18,
|
| 2169 |
+
"learning_rate": 0.0001714788273615635,
|
| 2170 |
+
"loss": 0.2843,
|
| 2171 |
+
"step": 3390
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 2.19,
|
| 2175 |
+
"learning_rate": 0.00017108794788273614,
|
| 2176 |
+
"loss": 0.267,
|
| 2177 |
+
"step": 3400
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 2.19,
|
| 2181 |
+
"eval_loss": 0.29497140645980835,
|
| 2182 |
+
"eval_runtime": 812.7448,
|
| 2183 |
+
"eval_samples_per_second": 2.461,
|
| 2184 |
+
"eval_steps_per_second": 0.308,
|
| 2185 |
+
"step": 3400
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 2.19,
|
| 2189 |
+
"learning_rate": 0.0001706970684039088,
|
| 2190 |
+
"loss": 0.2697,
|
| 2191 |
+
"step": 3410
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 2.2,
|
| 2195 |
+
"learning_rate": 0.00017030618892508142,
|
| 2196 |
+
"loss": 0.2788,
|
| 2197 |
+
"step": 3420
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"epoch": 2.21,
|
| 2201 |
+
"learning_rate": 0.00016991530944625405,
|
| 2202 |
+
"loss": 0.2775,
|
| 2203 |
+
"step": 3430
|
| 2204 |
+
},
|
| 2205 |
+
{
|
| 2206 |
+
"epoch": 2.21,
|
| 2207 |
+
"learning_rate": 0.00016952442996742668,
|
| 2208 |
+
"loss": 0.274,
|
| 2209 |
+
"step": 3440
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 2.22,
|
| 2213 |
+
"learning_rate": 0.00016913355048859933,
|
| 2214 |
+
"loss": 0.2733,
|
| 2215 |
+
"step": 3450
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.23,
|
| 2219 |
+
"learning_rate": 0.000168742671009772,
|
| 2220 |
+
"loss": 0.2675,
|
| 2221 |
+
"step": 3460
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 2.23,
|
| 2225 |
+
"learning_rate": 0.00016835179153094462,
|
| 2226 |
+
"loss": 0.2715,
|
| 2227 |
+
"step": 3470
|
| 2228 |
+
},
|
| 2229 |
+
{
|
| 2230 |
+
"epoch": 2.24,
|
| 2231 |
+
"learning_rate": 0.00016796091205211725,
|
| 2232 |
+
"loss": 0.2658,
|
| 2233 |
+
"step": 3480
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 2.24,
|
| 2237 |
+
"learning_rate": 0.00016757003257328987,
|
| 2238 |
+
"loss": 0.2797,
|
| 2239 |
+
"step": 3490
|
| 2240 |
+
},
|
| 2241 |
+
{
|
| 2242 |
+
"epoch": 2.25,
|
| 2243 |
+
"learning_rate": 0.00016717915309446253,
|
| 2244 |
+
"loss": 0.2671,
|
| 2245 |
+
"step": 3500
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 2.26,
|
| 2249 |
+
"learning_rate": 0.00016678827361563518,
|
| 2250 |
+
"loss": 0.279,
|
| 2251 |
+
"step": 3510
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 2.26,
|
| 2255 |
+
"learning_rate": 0.0001663973941368078,
|
| 2256 |
+
"loss": 0.2697,
|
| 2257 |
+
"step": 3520
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.27,
|
| 2261 |
+
"learning_rate": 0.00016600651465798044,
|
| 2262 |
+
"loss": 0.2689,
|
| 2263 |
+
"step": 3530
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 2.28,
|
| 2267 |
+
"learning_rate": 0.00016561563517915307,
|
| 2268 |
+
"loss": 0.2613,
|
| 2269 |
+
"step": 3540
|
| 2270 |
+
},
|
| 2271 |
+
{
|
| 2272 |
+
"epoch": 2.28,
|
| 2273 |
+
"learning_rate": 0.00016522475570032572,
|
| 2274 |
+
"loss": 0.2719,
|
| 2275 |
+
"step": 3550
|
| 2276 |
+
},
|
| 2277 |
+
{
|
| 2278 |
+
"epoch": 2.29,
|
| 2279 |
+
"learning_rate": 0.00016483387622149835,
|
| 2280 |
+
"loss": 0.2722,
|
| 2281 |
+
"step": 3560
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 2.3,
|
| 2285 |
+
"learning_rate": 0.000164442996742671,
|
| 2286 |
+
"loss": 0.2714,
|
| 2287 |
+
"step": 3570
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 2.3,
|
| 2291 |
+
"learning_rate": 0.0001640521172638436,
|
| 2292 |
+
"loss": 0.2677,
|
| 2293 |
+
"step": 3580
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 2.31,
|
| 2297 |
+
"learning_rate": 0.00016366123778501626,
|
| 2298 |
+
"loss": 0.2599,
|
| 2299 |
+
"step": 3590
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.32,
|
| 2303 |
+
"learning_rate": 0.00016327035830618892,
|
| 2304 |
+
"loss": 0.2765,
|
| 2305 |
+
"step": 3600
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 2.32,
|
| 2309 |
+
"eval_loss": 0.294181764125824,
|
| 2310 |
+
"eval_runtime": 814.8768,
|
| 2311 |
+
"eval_samples_per_second": 2.454,
|
| 2312 |
+
"eval_steps_per_second": 0.307,
|
| 2313 |
+
"step": 3600
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.32,
|
| 2317 |
+
"learning_rate": 0.00016287947882736155,
|
| 2318 |
+
"loss": 0.2608,
|
| 2319 |
+
"step": 3610
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 2.33,
|
| 2323 |
+
"learning_rate": 0.0001624885993485342,
|
| 2324 |
+
"loss": 0.2644,
|
| 2325 |
+
"step": 3620
|
| 2326 |
+
},
|
| 2327 |
+
{
|
| 2328 |
+
"epoch": 2.33,
|
| 2329 |
+
"learning_rate": 0.0001620977198697068,
|
| 2330 |
+
"loss": 0.2905,
|
| 2331 |
+
"step": 3630
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 2.34,
|
| 2335 |
+
"learning_rate": 0.00016170684039087946,
|
| 2336 |
+
"loss": 0.2722,
|
| 2337 |
+
"step": 3640
|
| 2338 |
+
},
|
| 2339 |
+
{
|
| 2340 |
+
"epoch": 2.35,
|
| 2341 |
+
"learning_rate": 0.0001613159609120521,
|
| 2342 |
+
"loss": 0.2679,
|
| 2343 |
+
"step": 3650
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 2.35,
|
| 2347 |
+
"learning_rate": 0.00016092508143322474,
|
| 2348 |
+
"loss": 0.274,
|
| 2349 |
+
"step": 3660
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 2.36,
|
| 2353 |
+
"learning_rate": 0.0001605342019543974,
|
| 2354 |
+
"loss": 0.2692,
|
| 2355 |
+
"step": 3670
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.37,
|
| 2359 |
+
"learning_rate": 0.00016014332247557003,
|
| 2360 |
+
"loss": 0.2663,
|
| 2361 |
+
"step": 3680
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 2.37,
|
| 2365 |
+
"learning_rate": 0.00015975244299674265,
|
| 2366 |
+
"loss": 0.2598,
|
| 2367 |
+
"step": 3690
|
| 2368 |
+
},
|
| 2369 |
+
{
|
| 2370 |
+
"epoch": 2.38,
|
| 2371 |
+
"learning_rate": 0.00015936156351791528,
|
| 2372 |
+
"loss": 0.2756,
|
| 2373 |
+
"step": 3700
|
| 2374 |
+
},
|
| 2375 |
+
{
|
| 2376 |
+
"epoch": 2.39,
|
| 2377 |
+
"learning_rate": 0.00015897068403908794,
|
| 2378 |
+
"loss": 0.2715,
|
| 2379 |
+
"step": 3710
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 2.39,
|
| 2383 |
+
"learning_rate": 0.0001585798045602606,
|
| 2384 |
+
"loss": 0.2741,
|
| 2385 |
+
"step": 3720
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 2.4,
|
| 2389 |
+
"learning_rate": 0.00015818892508143322,
|
| 2390 |
+
"loss": 0.2751,
|
| 2391 |
+
"step": 3730
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 2.41,
|
| 2395 |
+
"learning_rate": 0.00015779804560260585,
|
| 2396 |
+
"loss": 0.2727,
|
| 2397 |
+
"step": 3740
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.41,
|
| 2401 |
+
"learning_rate": 0.00015740716612377848,
|
| 2402 |
+
"loss": 0.262,
|
| 2403 |
+
"step": 3750
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 2.42,
|
| 2407 |
+
"learning_rate": 0.00015701628664495113,
|
| 2408 |
+
"loss": 0.2677,
|
| 2409 |
+
"step": 3760
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 2.42,
|
| 2413 |
+
"learning_rate": 0.00015662540716612376,
|
| 2414 |
+
"loss": 0.2723,
|
| 2415 |
+
"step": 3770
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 2.43,
|
| 2419 |
+
"learning_rate": 0.00015623452768729641,
|
| 2420 |
+
"loss": 0.2682,
|
| 2421 |
+
"step": 3780
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"epoch": 2.44,
|
| 2425 |
+
"learning_rate": 0.00015584364820846904,
|
| 2426 |
+
"loss": 0.2685,
|
| 2427 |
+
"step": 3790
|
| 2428 |
+
},
|
| 2429 |
+
{
|
| 2430 |
+
"epoch": 2.44,
|
| 2431 |
+
"learning_rate": 0.00015545276872964167,
|
| 2432 |
+
"loss": 0.2654,
|
| 2433 |
+
"step": 3800
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.44,
|
| 2437 |
+
"eval_loss": 0.2940288186073303,
|
| 2438 |
+
"eval_runtime": 812.6786,
|
| 2439 |
+
"eval_samples_per_second": 2.461,
|
| 2440 |
+
"eval_steps_per_second": 0.308,
|
| 2441 |
+
"step": 3800
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 2.45,
|
| 2445 |
+
"learning_rate": 0.00015506188925081433,
|
| 2446 |
+
"loss": 0.2652,
|
| 2447 |
+
"step": 3810
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.46,
|
| 2451 |
+
"learning_rate": 0.00015467100977198695,
|
| 2452 |
+
"loss": 0.2711,
|
| 2453 |
+
"step": 3820
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.46,
|
| 2457 |
+
"learning_rate": 0.0001542801302931596,
|
| 2458 |
+
"loss": 0.2719,
|
| 2459 |
+
"step": 3830
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 2.47,
|
| 2463 |
+
"learning_rate": 0.0001538892508143322,
|
| 2464 |
+
"loss": 0.2548,
|
| 2465 |
+
"step": 3840
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 2.48,
|
| 2469 |
+
"learning_rate": 0.00015349837133550487,
|
| 2470 |
+
"loss": 0.2743,
|
| 2471 |
+
"step": 3850
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 2.48,
|
| 2475 |
+
"learning_rate": 0.00015310749185667752,
|
| 2476 |
+
"loss": 0.2862,
|
| 2477 |
+
"step": 3860
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 2.49,
|
| 2481 |
+
"learning_rate": 0.00015271661237785015,
|
| 2482 |
+
"loss": 0.281,
|
| 2483 |
+
"step": 3870
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 2.5,
|
| 2487 |
+
"learning_rate": 0.0001523257328990228,
|
| 2488 |
+
"loss": 0.271,
|
| 2489 |
+
"step": 3880
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.5,
|
| 2493 |
+
"learning_rate": 0.0001519348534201954,
|
| 2494 |
+
"loss": 0.2677,
|
| 2495 |
+
"step": 3890
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.51,
|
| 2499 |
+
"learning_rate": 0.00015154397394136806,
|
| 2500 |
+
"loss": 0.2614,
|
| 2501 |
+
"step": 3900
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 2.51,
|
| 2505 |
+
"learning_rate": 0.0001511530944625407,
|
| 2506 |
+
"loss": 0.2719,
|
| 2507 |
+
"step": 3910
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 2.52,
|
| 2511 |
+
"learning_rate": 0.00015076221498371334,
|
| 2512 |
+
"loss": 0.2678,
|
| 2513 |
+
"step": 3920
|
| 2514 |
+
},
|
| 2515 |
+
{
|
| 2516 |
+
"epoch": 2.53,
|
| 2517 |
+
"learning_rate": 0.000150371335504886,
|
| 2518 |
+
"loss": 0.262,
|
| 2519 |
+
"step": 3930
|
| 2520 |
+
},
|
| 2521 |
+
{
|
| 2522 |
+
"epoch": 2.53,
|
| 2523 |
+
"learning_rate": 0.00014998045602605863,
|
| 2524 |
+
"loss": 0.2853,
|
| 2525 |
+
"step": 3940
|
| 2526 |
+
},
|
| 2527 |
+
{
|
| 2528 |
+
"epoch": 2.54,
|
| 2529 |
+
"learning_rate": 0.00014958957654723126,
|
| 2530 |
+
"loss": 0.2776,
|
| 2531 |
+
"step": 3950
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 2.55,
|
| 2535 |
+
"learning_rate": 0.00014919869706840388,
|
| 2536 |
+
"loss": 0.2768,
|
| 2537 |
+
"step": 3960
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 2.55,
|
| 2541 |
+
"learning_rate": 0.00014880781758957654,
|
| 2542 |
+
"loss": 0.2716,
|
| 2543 |
+
"step": 3970
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 2.56,
|
| 2547 |
+
"learning_rate": 0.00014841693811074917,
|
| 2548 |
+
"loss": 0.2617,
|
| 2549 |
+
"step": 3980
|
| 2550 |
+
},
|
| 2551 |
+
{
|
| 2552 |
+
"epoch": 2.57,
|
| 2553 |
+
"learning_rate": 0.00014802605863192182,
|
| 2554 |
+
"loss": 0.2742,
|
| 2555 |
+
"step": 3990
|
| 2556 |
+
},
|
| 2557 |
+
{
|
| 2558 |
+
"epoch": 2.57,
|
| 2559 |
+
"learning_rate": 0.00014763517915309445,
|
| 2560 |
+
"loss": 0.2673,
|
| 2561 |
+
"step": 4000
|
| 2562 |
+
},
|
| 2563 |
+
{
|
| 2564 |
+
"epoch": 2.57,
|
| 2565 |
+
"eval_loss": 0.2936367690563202,
|
| 2566 |
+
"eval_runtime": 812.6744,
|
| 2567 |
+
"eval_samples_per_second": 2.461,
|
| 2568 |
+
"eval_steps_per_second": 0.308,
|
| 2569 |
+
"step": 4000
|
| 2570 |
}
|
| 2571 |
],
|
| 2572 |
"logging_steps": 10,
|
| 2573 |
"max_steps": 7775,
|
| 2574 |
"num_train_epochs": 5,
|
| 2575 |
"save_steps": 200,
|
| 2576 |
+
"total_flos": 7.045037400998707e+18,
|
| 2577 |
"trial_name": null,
|
| 2578 |
"trial_params": null
|
| 2579 |
}
|