Commit
•
db29ebd
1
Parent(s):
a024d53
Training in progress, step 33000
Browse files- adapter_model.bin +1 -1
- {checkpoint-29000 → checkpoint-32000/adapter_model}/README.md +0 -0
- {checkpoint-29000 → checkpoint-32000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-29000 → checkpoint-32000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-29000/adapter_model → checkpoint-33000}/README.md +0 -0
- {checkpoint-29000/adapter_model → checkpoint-33000}/adapter_config.json +0 -0
- {checkpoint-29000/adapter_model → checkpoint-33000}/adapter_model.bin +1 -1
- {checkpoint-29000 → checkpoint-33000}/optimizer.pt +1 -1
- {checkpoint-29000 → checkpoint-33000}/rng_state.pth +1 -1
- {checkpoint-29000 → checkpoint-33000}/scheduler.pt +1 -1
- {checkpoint-29000 → checkpoint-33000}/trainer_state.json +277 -5
- {checkpoint-29000 → checkpoint-33000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d42cbea41e45d2d0b755f9666264a178377e0d657920d99d2ac36083b14a3fce
|
| 3 |
size 500897101
|
{checkpoint-29000 → checkpoint-32000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-29000 → checkpoint-32000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-29000 → checkpoint-32000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cd53296eb8a21ba14a04ebafc8d98535866046120ab70e3a5896c0601a199c0
|
| 3 |
size 500897101
|
{checkpoint-29000/adapter_model → checkpoint-33000}/README.md
RENAMED
|
File without changes
|
{checkpoint-29000/adapter_model → checkpoint-33000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-29000/adapter_model → checkpoint-33000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d42cbea41e45d2d0b755f9666264a178377e0d657920d99d2ac36083b14a3fce
|
| 3 |
size 500897101
|
{checkpoint-29000 → checkpoint-33000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d118e7ceed7e898d194344dd82e1f57237c110a54ae13d845e1624b5ac6a2be7
|
| 3 |
size 1001723453
|
{checkpoint-29000 → checkpoint-33000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:169992729d29d9e2e5dcb82f149024cae6d9e402f3075f5eab6b5fc3fbaa107d
|
| 3 |
size 14575
|
{checkpoint-29000 → checkpoint-33000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0f09eabe27444782208394bfcda4067efa0ba31e368befa219e0b566bccb1c
|
| 3 |
size 627
|
{checkpoint-29000 → checkpoint-33000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1978,11 +1978,283 @@
|
|
| 1978 |
"eval_samples_per_second": 0.424,
|
| 1979 |
"eval_steps_per_second": 0.424,
|
| 1980 |
"step": 29000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1981 |
}
|
| 1982 |
],
|
| 1983 |
"max_steps": 80463,
|
| 1984 |
"num_train_epochs": 3,
|
| 1985 |
-
"total_flos":
|
| 1986 |
"trial_name": null,
|
| 1987 |
"trial_params": null
|
| 1988 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5719351172447205,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-33000",
|
| 4 |
+
"epoch": 1.2303791804928974,
|
| 5 |
+
"global_step": 33000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1978 |
"eval_samples_per_second": 0.424,
|
| 1979 |
"eval_steps_per_second": 0.424,
|
| 1980 |
"step": 29000
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 1.08,
|
| 1984 |
+
"learning_rate": 0.00014212896918901774,
|
| 1985 |
+
"loss": 0.5003,
|
| 1986 |
+
"step": 29100
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 1.09,
|
| 1990 |
+
"learning_rate": 0.00014177450517890503,
|
| 1991 |
+
"loss": 0.5102,
|
| 1992 |
+
"step": 29200
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 1.09,
|
| 1996 |
+
"learning_rate": 0.0001414194041884619,
|
| 1997 |
+
"loss": 0.524,
|
| 1998 |
+
"step": 29300
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.1,
|
| 2002 |
+
"learning_rate": 0.0001410636716322911,
|
| 2003 |
+
"loss": 0.5168,
|
| 2004 |
+
"step": 29400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 1.1,
|
| 2008 |
+
"learning_rate": 0.0001407073129346254,
|
| 2009 |
+
"loss": 0.514,
|
| 2010 |
+
"step": 29500
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 1.1,
|
| 2014 |
+
"learning_rate": 0.00014035033352924502,
|
| 2015 |
+
"loss": 0.5084,
|
| 2016 |
+
"step": 29600
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 1.11,
|
| 2020 |
+
"learning_rate": 0.0001399927388593948,
|
| 2021 |
+
"loss": 0.5203,
|
| 2022 |
+
"step": 29700
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 1.11,
|
| 2026 |
+
"learning_rate": 0.00013963453437770119,
|
| 2027 |
+
"loss": 0.5226,
|
| 2028 |
+
"step": 29800
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 1.11,
|
| 2032 |
+
"learning_rate": 0.000139275725546089,
|
| 2033 |
+
"loss": 0.5055,
|
| 2034 |
+
"step": 29900
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 1.12,
|
| 2038 |
+
"learning_rate": 0.00013891631783569838,
|
| 2039 |
+
"loss": 0.5303,
|
| 2040 |
+
"step": 30000
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.12,
|
| 2044 |
+
"eval_loss": 0.5869857668876648,
|
| 2045 |
+
"eval_runtime": 1272.0282,
|
| 2046 |
+
"eval_samples_per_second": 0.426,
|
| 2047 |
+
"eval_steps_per_second": 0.426,
|
| 2048 |
+
"step": 30000
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.12,
|
| 2052 |
+
"learning_rate": 0.00013855631672680106,
|
| 2053 |
+
"loss": 0.5243,
|
| 2054 |
+
"step": 30100
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.13,
|
| 2058 |
+
"learning_rate": 0.00013819572770871702,
|
| 2059 |
+
"loss": 0.5148,
|
| 2060 |
+
"step": 30200
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 1.13,
|
| 2064 |
+
"learning_rate": 0.00013783455627973062,
|
| 2065 |
+
"loss": 0.522,
|
| 2066 |
+
"step": 30300
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 1.13,
|
| 2070 |
+
"learning_rate": 0.00013747280794700707,
|
| 2071 |
+
"loss": 0.5289,
|
| 2072 |
+
"step": 30400
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.14,
|
| 2076 |
+
"learning_rate": 0.00013711048822650802,
|
| 2077 |
+
"loss": 0.4996,
|
| 2078 |
+
"step": 30500
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 1.14,
|
| 2082 |
+
"learning_rate": 0.00013674760264290785,
|
| 2083 |
+
"loss": 0.5099,
|
| 2084 |
+
"step": 30600
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 1.14,
|
| 2088 |
+
"learning_rate": 0.0001363841567295091,
|
| 2089 |
+
"loss": 0.5219,
|
| 2090 |
+
"step": 30700
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.15,
|
| 2094 |
+
"learning_rate": 0.00013602015602815837,
|
| 2095 |
+
"loss": 0.5297,
|
| 2096 |
+
"step": 30800
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.15,
|
| 2100 |
+
"learning_rate": 0.00013565560608916165,
|
| 2101 |
+
"loss": 0.5029,
|
| 2102 |
+
"step": 30900
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 1.16,
|
| 2106 |
+
"learning_rate": 0.0001352905124711998,
|
| 2107 |
+
"loss": 0.5266,
|
| 2108 |
+
"step": 31000
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 1.16,
|
| 2112 |
+
"eval_loss": 0.5811149477958679,
|
| 2113 |
+
"eval_runtime": 1300.1475,
|
| 2114 |
+
"eval_samples_per_second": 0.417,
|
| 2115 |
+
"eval_steps_per_second": 0.417,
|
| 2116 |
+
"step": 31000
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 1.16,
|
| 2120 |
+
"learning_rate": 0.00013492488074124366,
|
| 2121 |
+
"loss": 0.5295,
|
| 2122 |
+
"step": 31100
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 1.16,
|
| 2126 |
+
"learning_rate": 0.00013455871647446923,
|
| 2127 |
+
"loss": 0.539,
|
| 2128 |
+
"step": 31200
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 1.17,
|
| 2132 |
+
"learning_rate": 0.00013419202525417277,
|
| 2133 |
+
"loss": 0.5217,
|
| 2134 |
+
"step": 31300
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 1.17,
|
| 2138 |
+
"learning_rate": 0.0001338248126716854,
|
| 2139 |
+
"loss": 0.5197,
|
| 2140 |
+
"step": 31400
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 1.17,
|
| 2144 |
+
"learning_rate": 0.00013345708432628824,
|
| 2145 |
+
"loss": 0.4991,
|
| 2146 |
+
"step": 31500
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 1.18,
|
| 2150 |
+
"learning_rate": 0.00013308884582512647,
|
| 2151 |
+
"loss": 0.5239,
|
| 2152 |
+
"step": 31600
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.18,
|
| 2156 |
+
"learning_rate": 0.00013272010278312453,
|
| 2157 |
+
"loss": 0.4899,
|
| 2158 |
+
"step": 31700
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 1.19,
|
| 2162 |
+
"learning_rate": 0.00013235086082289977,
|
| 2163 |
+
"loss": 0.5088,
|
| 2164 |
+
"step": 31800
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 1.19,
|
| 2168 |
+
"learning_rate": 0.00013198112557467732,
|
| 2169 |
+
"loss": 0.5497,
|
| 2170 |
+
"step": 31900
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 1.19,
|
| 2174 |
+
"learning_rate": 0.00013161090267620396,
|
| 2175 |
+
"loss": 0.5024,
|
| 2176 |
+
"step": 32000
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 1.19,
|
| 2180 |
+
"eval_loss": 0.5758991241455078,
|
| 2181 |
+
"eval_runtime": 1292.3362,
|
| 2182 |
+
"eval_samples_per_second": 0.419,
|
| 2183 |
+
"eval_steps_per_second": 0.419,
|
| 2184 |
+
"step": 32000
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 1.2,
|
| 2188 |
+
"learning_rate": 0.0001312401977726621,
|
| 2189 |
+
"loss": 0.534,
|
| 2190 |
+
"step": 32100
|
| 2191 |
+
},
|
| 2192 |
+
{
|
| 2193 |
+
"epoch": 1.2,
|
| 2194 |
+
"learning_rate": 0.0001308690165165839,
|
| 2195 |
+
"loss": 0.4936,
|
| 2196 |
+
"step": 32200
|
| 2197 |
+
},
|
| 2198 |
+
{
|
| 2199 |
+
"epoch": 1.2,
|
| 2200 |
+
"learning_rate": 0.00013049736456776485,
|
| 2201 |
+
"loss": 0.4999,
|
| 2202 |
+
"step": 32300
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 1.21,
|
| 2206 |
+
"learning_rate": 0.00013012524759317774,
|
| 2207 |
+
"loss": 0.5238,
|
| 2208 |
+
"step": 32400
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.21,
|
| 2212 |
+
"learning_rate": 0.000129752671266886,
|
| 2213 |
+
"loss": 0.4959,
|
| 2214 |
+
"step": 32500
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 1.22,
|
| 2218 |
+
"learning_rate": 0.00012937964126995727,
|
| 2219 |
+
"loss": 0.514,
|
| 2220 |
+
"step": 32600
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 1.22,
|
| 2224 |
+
"learning_rate": 0.00012900616329037694,
|
| 2225 |
+
"loss": 0.4964,
|
| 2226 |
+
"step": 32700
|
| 2227 |
+
},
|
| 2228 |
+
{
|
| 2229 |
+
"epoch": 1.22,
|
| 2230 |
+
"learning_rate": 0.00012863224302296107,
|
| 2231 |
+
"loss": 0.5054,
|
| 2232 |
+
"step": 32800
|
| 2233 |
+
},
|
| 2234 |
+
{
|
| 2235 |
+
"epoch": 1.23,
|
| 2236 |
+
"learning_rate": 0.0001282578861692699,
|
| 2237 |
+
"loss": 0.5079,
|
| 2238 |
+
"step": 32900
|
| 2239 |
+
},
|
| 2240 |
+
{
|
| 2241 |
+
"epoch": 1.23,
|
| 2242 |
+
"learning_rate": 0.0001278830984375206,
|
| 2243 |
+
"loss": 0.4929,
|
| 2244 |
+
"step": 33000
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 1.23,
|
| 2248 |
+
"eval_loss": 0.5719351172447205,
|
| 2249 |
+
"eval_runtime": 1267.7603,
|
| 2250 |
+
"eval_samples_per_second": 0.428,
|
| 2251 |
+
"eval_steps_per_second": 0.428,
|
| 2252 |
+
"step": 33000
|
| 2253 |
}
|
| 2254 |
],
|
| 2255 |
"max_steps": 80463,
|
| 2256 |
"num_train_epochs": 3,
|
| 2257 |
+
"total_flos": 9.255791880986788e+18,
|
| 2258 |
"trial_name": null,
|
| 2259 |
"trial_params": null
|
| 2260 |
}
|
{checkpoint-29000 → checkpoint-33000}/training_args.bin
RENAMED
|
File without changes
|