Commit
•
cd23378
1
Parent(s):
db29ebd
Training in progress, step 34000
Browse files- adapter_model.bin +1 -1
- {checkpoint-30000 → checkpoint-33000/adapter_model}/README.md +0 -0
- {checkpoint-30000 → checkpoint-33000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-30000 → checkpoint-33000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-30000/adapter_model → checkpoint-34000}/README.md +0 -0
- {checkpoint-30000/adapter_model → checkpoint-34000}/adapter_config.json +0 -0
- {checkpoint-30000/adapter_model → checkpoint-34000}/adapter_model.bin +1 -1
- {checkpoint-30000 → checkpoint-34000}/optimizer.pt +1 -1
- {checkpoint-30000 → checkpoint-34000}/rng_state.pth +1 -1
- {checkpoint-30000 → checkpoint-34000}/scheduler.pt +1 -1
- {checkpoint-30000 → checkpoint-34000}/trainer_state.json +277 -5
- {checkpoint-30000 → checkpoint-34000}/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 sha256:7511e65b55d29a7e86c8b6b83356365ea688292ba79d0f3a40289fa0d34e9a8c
|
| 3 |
size 500897101
|
{checkpoint-30000 → checkpoint-33000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-30000 → checkpoint-33000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-30000 → checkpoint-33000/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:d42cbea41e45d2d0b755f9666264a178377e0d657920d99d2ac36083b14a3fce
|
| 3 |
size 500897101
|
{checkpoint-30000/adapter_model → checkpoint-34000}/README.md
RENAMED
|
File without changes
|
{checkpoint-30000/adapter_model → checkpoint-34000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-30000/adapter_model → checkpoint-34000}/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 sha256:7511e65b55d29a7e86c8b6b83356365ea688292ba79d0f3a40289fa0d34e9a8c
|
| 3 |
size 500897101
|
{checkpoint-30000 → checkpoint-34000}/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:ed98569c7d57f25cd04cdbe01503f350ab8d1c53424912e6e1c8f906e815dfd9
|
| 3 |
size 1001723453
|
{checkpoint-30000 → checkpoint-34000}/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:c78019e6be9dcb9854fd190539da62f7f53c85d3563e7512184a250af881b96a
|
| 3 |
size 14575
|
{checkpoint-30000 → checkpoint-34000}/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:f1f76b1b4ddd5a5fd1441715b84be8e888cf97db173bbe93e49ec15bc19296ed
|
| 3 |
size 627
|
{checkpoint-30000 → checkpoint-34000}/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,
|
|
@@ -2046,11 +2046,283 @@
|
|
| 2046 |
"eval_samples_per_second": 0.426,
|
| 2047 |
"eval_steps_per_second": 0.426,
|
| 2048 |
"step": 30000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2049 |
}
|
| 2050 |
],
|
| 2051 |
"max_steps": 80463,
|
| 2052 |
"num_train_epochs": 3,
|
| 2053 |
-
"total_flos":
|
| 2054 |
"trial_name": null,
|
| 2055 |
"trial_params": null
|
| 2056 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5665221810340881,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-34000",
|
| 4 |
+
"epoch": 1.2676633980835912,
|
| 5 |
+
"global_step": 34000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 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 |
+
"epoch": 1.23,
|
| 2256 |
+
"learning_rate": 0.0001275078855425007,
|
| 2257 |
+
"loss": 0.4971,
|
| 2258 |
+
"step": 33100
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 1.24,
|
| 2262 |
+
"learning_rate": 0.0001271322532054803,
|
| 2263 |
+
"loss": 0.4977,
|
| 2264 |
+
"step": 33200
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.24,
|
| 2268 |
+
"learning_rate": 0.0001267562071541254,
|
| 2269 |
+
"loss": 0.499,
|
| 2270 |
+
"step": 33300
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 1.25,
|
| 2274 |
+
"learning_rate": 0.00012637975312241022,
|
| 2275 |
+
"loss": 0.5044,
|
| 2276 |
+
"step": 33400
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 1.25,
|
| 2280 |
+
"learning_rate": 0.00012600289685052996,
|
| 2281 |
+
"loss": 0.5019,
|
| 2282 |
+
"step": 33500
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 1.25,
|
| 2286 |
+
"learning_rate": 0.00012562564408481327,
|
| 2287 |
+
"loss": 0.5225,
|
| 2288 |
+
"step": 33600
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 1.26,
|
| 2292 |
+
"learning_rate": 0.00012524800057763438,
|
| 2293 |
+
"loss": 0.5503,
|
| 2294 |
+
"step": 33700
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 1.26,
|
| 2298 |
+
"learning_rate": 0.00012486997208732573,
|
| 2299 |
+
"loss": 0.5025,
|
| 2300 |
+
"step": 33800
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 1.26,
|
| 2304 |
+
"learning_rate": 0.0001244915643780899,
|
| 2305 |
+
"loss": 0.5187,
|
| 2306 |
+
"step": 33900
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.27,
|
| 2310 |
+
"learning_rate": 0.00012411278321991195,
|
| 2311 |
+
"loss": 0.5199,
|
| 2312 |
+
"step": 34000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 1.27,
|
| 2316 |
+
"eval_loss": 0.5665221810340881,
|
| 2317 |
+
"eval_runtime": 1263.0264,
|
| 2318 |
+
"eval_samples_per_second": 0.429,
|
| 2319 |
+
"eval_steps_per_second": 0.429,
|
| 2320 |
+
"step": 34000
|
| 2321 |
}
|
| 2322 |
],
|
| 2323 |
"max_steps": 80463,
|
| 2324 |
"num_train_epochs": 3,
|
| 2325 |
+
"total_flos": 9.535353877436006e+18,
|
| 2326 |
"trial_name": null,
|
| 2327 |
"trial_params": null
|
| 2328 |
}
|
{checkpoint-30000 → checkpoint-34000}/training_args.bin
RENAMED
|
File without changes
|