Commit
•
35e0f32
1
Parent(s):
cd23378
Training in progress, step 35000
Browse files- adapter_model.bin +1 -1
- {checkpoint-31000 → checkpoint-34000/adapter_model}/README.md +0 -0
- {checkpoint-31000 → checkpoint-34000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-31000 → checkpoint-34000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-31000/adapter_model → checkpoint-35000}/README.md +0 -0
- {checkpoint-31000/adapter_model → checkpoint-35000}/adapter_config.json +0 -0
- {checkpoint-31000/adapter_model → checkpoint-35000}/adapter_model.bin +1 -1
- {checkpoint-31000 → checkpoint-35000}/optimizer.pt +1 -1
- {checkpoint-31000 → checkpoint-35000}/rng_state.pth +1 -1
- {checkpoint-31000 → checkpoint-35000}/scheduler.pt +1 -1
- {checkpoint-31000 → checkpoint-35000}/trainer_state.json +277 -5
- {checkpoint-31000 → checkpoint-35000}/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 sha256:3bcaf4cf9f9bc20718f7e19c06c7145a78a3a5c6c1ca1a845231e7f1741b7f0e
|
| 3 |
size 500897101
|
{checkpoint-31000 → checkpoint-34000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-31000 → checkpoint-34000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-31000 → checkpoint-34000/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 sha256:7511e65b55d29a7e86c8b6b83356365ea688292ba79d0f3a40289fa0d34e9a8c
|
| 3 |
size 500897101
|
{checkpoint-31000/adapter_model → checkpoint-35000}/README.md
RENAMED
|
File without changes
|
{checkpoint-31000/adapter_model → checkpoint-35000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-31000/adapter_model → checkpoint-35000}/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 sha256:3bcaf4cf9f9bc20718f7e19c06c7145a78a3a5c6c1ca1a845231e7f1741b7f0e
|
| 3 |
size 500897101
|
{checkpoint-31000 → checkpoint-35000}/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 sha256:021299a59b6ef922c41c510f3f8b93b7333417d0f6d05edbc3721b651b76908a
|
| 3 |
size 1001723453
|
{checkpoint-31000 → checkpoint-35000}/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:e07a8bcb98397480852e842446a8ddbd7b1c726fe43695a8f26d8722c832cf30
|
| 3 |
size 14575
|
{checkpoint-31000 → checkpoint-35000}/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 sha256:70880cdf77014daa95aad41e2ce4c4dffd35eef0882e2600a10e5ef00bca6a86
|
| 3 |
size 627
|
{checkpoint-31000 → checkpoint-35000}/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,
|
|
@@ -2114,11 +2114,283 @@
|
|
| 2114 |
"eval_samples_per_second": 0.417,
|
| 2115 |
"eval_steps_per_second": 0.417,
|
| 2116 |
"step": 31000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2117 |
}
|
| 2118 |
],
|
| 2119 |
"max_steps": 80463,
|
| 2120 |
"num_train_epochs": 3,
|
| 2121 |
-
"total_flos":
|
| 2122 |
"trial_name": null,
|
| 2123 |
"trial_params": null
|
| 2124 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.562954843044281,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-35000",
|
| 4 |
+
"epoch": 1.3049476156742852,
|
| 5 |
+
"global_step": 35000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 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 |
+
"epoch": 1.27,
|
| 2324 |
+
"learning_rate": 0.00012373363438847117,
|
| 2325 |
+
"loss": 0.5135,
|
| 2326 |
+
"step": 34100
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 1.28,
|
| 2330 |
+
"learning_rate": 0.00012335412366505324,
|
| 2331 |
+
"loss": 0.5065,
|
| 2332 |
+
"step": 34200
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 1.28,
|
| 2336 |
+
"learning_rate": 0.000122974256836462,
|
| 2337 |
+
"loss": 0.5223,
|
| 2338 |
+
"step": 34300
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 1.28,
|
| 2342 |
+
"learning_rate": 0.00012259403969493114,
|
| 2343 |
+
"loss": 0.4946,
|
| 2344 |
+
"step": 34400
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"epoch": 1.29,
|
| 2348 |
+
"learning_rate": 0.00012221347803803605,
|
| 2349 |
+
"loss": 0.5105,
|
| 2350 |
+
"step": 34500
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 1.29,
|
| 2354 |
+
"learning_rate": 0.00012183257766860514,
|
| 2355 |
+
"loss": 0.4812,
|
| 2356 |
+
"step": 34600
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.29,
|
| 2360 |
+
"learning_rate": 0.00012145134439463178,
|
| 2361 |
+
"loss": 0.4981,
|
| 2362 |
+
"step": 34700
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.3,
|
| 2366 |
+
"learning_rate": 0.0001210697840291852,
|
| 2367 |
+
"loss": 0.5038,
|
| 2368 |
+
"step": 34800
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 1.3,
|
| 2372 |
+
"learning_rate": 0.00012068790239032241,
|
| 2373 |
+
"loss": 0.5551,
|
| 2374 |
+
"step": 34900
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 1.3,
|
| 2378 |
+
"learning_rate": 0.00012030570530099902,
|
| 2379 |
+
"loss": 0.4964,
|
| 2380 |
+
"step": 35000
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 1.3,
|
| 2384 |
+
"eval_loss": 0.562954843044281,
|
| 2385 |
+
"eval_runtime": 1252.1434,
|
| 2386 |
+
"eval_samples_per_second": 0.433,
|
| 2387 |
+
"eval_steps_per_second": 0.433,
|
| 2388 |
+
"step": 35000
|
| 2389 |
}
|
| 2390 |
],
|
| 2391 |
"max_steps": 80463,
|
| 2392 |
"num_train_epochs": 3,
|
| 2393 |
+
"total_flos": 9.815104047716352e+18,
|
| 2394 |
"trial_name": null,
|
| 2395 |
"trial_params": null
|
| 2396 |
}
|
{checkpoint-31000 → checkpoint-35000}/training_args.bin
RENAMED
|
File without changes
|