Upload folder using huggingface_hub
Browse files- all_results.json +18 -18
- eval_results.json +14 -14
- pytorch_model.bin +2 -2
- train_results.json +5 -5
- trainer_state.json +534 -270
- training_args.bin +2 -2
all_results.json
CHANGED
|
@@ -1,26 +1,26 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_HasAns_exact":
|
| 4 |
-
"eval_HasAns_f1":
|
| 5 |
"eval_HasAns_total": 5928,
|
| 6 |
-
"eval_NoAns_exact": 79.
|
| 7 |
-
"eval_NoAns_f1": 79.
|
| 8 |
"eval_NoAns_total": 5945,
|
| 9 |
-
"eval_best_exact":
|
| 10 |
"eval_best_exact_thresh": 0.0,
|
| 11 |
-
"eval_best_f1":
|
| 12 |
"eval_best_f1_thresh": 0.0,
|
| 13 |
-
"eval_exact":
|
| 14 |
-
"eval_f1":
|
| 15 |
-
"eval_loss": 0.
|
| 16 |
-
"eval_runtime":
|
| 17 |
-
"eval_samples":
|
| 18 |
-
"eval_samples_per_second":
|
| 19 |
-
"eval_steps_per_second":
|
| 20 |
"eval_total": 11873,
|
| 21 |
-
"train_loss": 0.
|
| 22 |
-
"train_runtime":
|
| 23 |
"train_samples": 130319,
|
| 24 |
-
"train_samples_per_second":
|
| 25 |
-
"train_steps_per_second":
|
| 26 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_HasAns_exact": 75.97840755735493,
|
| 4 |
+
"eval_HasAns_f1": 81.70934619524567,
|
| 5 |
"eval_HasAns_total": 5928,
|
| 6 |
+
"eval_NoAns_exact": 79.84861227922624,
|
| 7 |
+
"eval_NoAns_f1": 79.84861227922624,
|
| 8 |
"eval_NoAns_total": 5945,
|
| 9 |
+
"eval_best_exact": 77.91628063673882,
|
| 10 |
"eval_best_exact_thresh": 0.0,
|
| 11 |
+
"eval_best_f1": 80.77764711912864,
|
| 12 |
"eval_best_f1_thresh": 0.0,
|
| 13 |
+
"eval_exact": 77.91628063673882,
|
| 14 |
+
"eval_f1": 80.77764711912879,
|
| 15 |
+
"eval_loss": 0.717133104801178,
|
| 16 |
+
"eval_runtime": 484.7669,
|
| 17 |
+
"eval_samples": 12309,
|
| 18 |
+
"eval_samples_per_second": 25.392,
|
| 19 |
+
"eval_steps_per_second": 3.175,
|
| 20 |
"eval_total": 11873,
|
| 21 |
+
"train_loss": 0.5451086903730735,
|
| 22 |
+
"train_runtime": 8999.8943,
|
| 23 |
"train_samples": 130319,
|
| 24 |
+
"train_samples_per_second": 14.48,
|
| 25 |
+
"train_steps_per_second": 7.24
|
| 26 |
}
|
eval_results.json
CHANGED
|
@@ -1,21 +1,21 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_HasAns_exact":
|
| 4 |
-
"eval_HasAns_f1":
|
| 5 |
"eval_HasAns_total": 5928,
|
| 6 |
-
"eval_NoAns_exact": 79.
|
| 7 |
-
"eval_NoAns_f1": 79.
|
| 8 |
"eval_NoAns_total": 5945,
|
| 9 |
-
"eval_best_exact":
|
| 10 |
"eval_best_exact_thresh": 0.0,
|
| 11 |
-
"eval_best_f1":
|
| 12 |
"eval_best_f1_thresh": 0.0,
|
| 13 |
-
"eval_exact":
|
| 14 |
-
"eval_f1":
|
| 15 |
-
"eval_loss": 0.
|
| 16 |
-
"eval_runtime":
|
| 17 |
-
"eval_samples":
|
| 18 |
-
"eval_samples_per_second":
|
| 19 |
-
"eval_steps_per_second":
|
| 20 |
"eval_total": 11873
|
| 21 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_HasAns_exact": 75.97840755735493,
|
| 4 |
+
"eval_HasAns_f1": 81.70934619524567,
|
| 5 |
"eval_HasAns_total": 5928,
|
| 6 |
+
"eval_NoAns_exact": 79.84861227922624,
|
| 7 |
+
"eval_NoAns_f1": 79.84861227922624,
|
| 8 |
"eval_NoAns_total": 5945,
|
| 9 |
+
"eval_best_exact": 77.91628063673882,
|
| 10 |
"eval_best_exact_thresh": 0.0,
|
| 11 |
+
"eval_best_f1": 80.77764711912864,
|
| 12 |
"eval_best_f1_thresh": 0.0,
|
| 13 |
+
"eval_exact": 77.91628063673882,
|
| 14 |
+
"eval_f1": 80.77764711912879,
|
| 15 |
+
"eval_loss": 0.717133104801178,
|
| 16 |
+
"eval_runtime": 484.7669,
|
| 17 |
+
"eval_samples": 12309,
|
| 18 |
+
"eval_samples_per_second": 25.392,
|
| 19 |
+
"eval_steps_per_second": 3.175,
|
| 20 |
"eval_total": 11873
|
| 21 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4181cc201092c086683e482fc182b4558f23a612909249443df7ac63964353d
|
| 3 |
+
size 990408885
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 130319,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.5451086903730735,
|
| 4 |
+
"train_runtime": 8999.8943,
|
| 5 |
"train_samples": 130319,
|
| 6 |
+
"train_samples_per_second": 14.48,
|
| 7 |
+
"train_steps_per_second": 7.24
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,541 +1,805 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 0.
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 0.
|
| 62 |
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 5000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 5500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 6000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 6500
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 0.
|
| 92 |
"step": 7000
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 7500
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 8000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 8500
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 9000
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 9500
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 10000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 10500
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 11000
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch":
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 11500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 12000
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 12500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 13000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 13500
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 14000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 14500
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 15000
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 15500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 16000
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 16500
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 17000
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 17500
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 18000
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 18500
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 19000
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 19500
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 20000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 20500
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 21000
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"learning_rate":
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 21500
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch":
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 0.
|
| 272 |
"step": 22000
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch":
|
| 276 |
-
"learning_rate":
|
| 277 |
-
"loss": 0.
|
| 278 |
"step": 22500
|
| 279 |
},
|
| 280 |
{
|
| 281 |
-
"epoch":
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
"step": 23000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 23500
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 24000
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 24500
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
"step": 25000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss": 0.
|
| 314 |
"step": 25500
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch":
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 26000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch":
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 26500
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 27000
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 27500
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch":
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 28000
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch":
|
| 348 |
-
"learning_rate":
|
| 349 |
-
"loss": 0.
|
| 350 |
"step": 28500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch":
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 29000
|
| 357 |
},
|
| 358 |
{
|
| 359 |
-
"epoch":
|
| 360 |
-
"learning_rate":
|
| 361 |
-
"loss": 0.
|
| 362 |
"step": 29500
|
| 363 |
},
|
| 364 |
{
|
| 365 |
-
"epoch":
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 30000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 30500
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 31000
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch":
|
| 384 |
-
"learning_rate":
|
| 385 |
-
"loss": 0.
|
| 386 |
"step": 31500
|
| 387 |
},
|
| 388 |
{
|
| 389 |
-
"epoch":
|
| 390 |
-
"learning_rate":
|
| 391 |
-
"loss": 0.
|
| 392 |
"step": 32000
|
| 393 |
},
|
| 394 |
{
|
| 395 |
-
"epoch":
|
| 396 |
-
"learning_rate":
|
| 397 |
-
"loss": 0.
|
| 398 |
"step": 32500
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"learning_rate":
|
| 403 |
-
"loss": 0.
|
| 404 |
"step": 33000
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch":
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 33500
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 34000
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch":
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 34500
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss": 0.
|
| 428 |
"step": 35000
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch":
|
| 432 |
-
"learning_rate":
|
| 433 |
-
"loss": 0.
|
| 434 |
"step": 35500
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
| 440 |
"step": 36000
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 36500
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 37000
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch":
|
| 456 |
-
"learning_rate":
|
| 457 |
-
"loss": 0.
|
| 458 |
"step": 37500
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 38000
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch":
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 38500
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch":
|
| 474 |
-
"learning_rate":
|
| 475 |
-
"loss": 0.
|
| 476 |
"step": 39000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"learning_rate": 1.
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 39500
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"learning_rate": 1.
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 40000
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"learning_rate": 1.
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 40500
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"learning_rate": 1.
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 41000
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 41500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch":
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 42000
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 42500
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch":
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 43000
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"
|
| 529 |
-
"
|
| 530 |
-
"
|
| 531 |
-
|
| 532 |
-
|
| 533 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 534 |
}
|
| 535 |
],
|
| 536 |
-
"max_steps":
|
| 537 |
-
"num_train_epochs":
|
| 538 |
-
"total_flos":
|
| 539 |
"trial_name": null,
|
| 540 |
"trial_params": null
|
| 541 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 65160,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.9769797421731124e-05,
|
| 13 |
+
"loss": 0.8207,
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 2.9539594843462247e-05,
|
| 19 |
+
"loss": 0.6605,
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 2.930939226519337e-05,
|
| 25 |
+
"loss": 0.585,
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 2.9079189686924494e-05,
|
| 31 |
+
"loss": 0.6657,
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 2.884898710865562e-05,
|
| 37 |
+
"loss": 0.6631,
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 2.861878453038674e-05,
|
| 43 |
+
"loss": 0.6405,
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"learning_rate": 2.8388581952117863e-05,
|
| 49 |
+
"loss": 0.6778,
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.06,
|
| 54 |
+
"learning_rate": 2.815837937384899e-05,
|
| 55 |
+
"loss": 0.6138,
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.07,
|
| 60 |
+
"learning_rate": 2.7928176795580113e-05,
|
| 61 |
+
"loss": 0.6646,
|
| 62 |
"step": 4500
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"learning_rate": 2.7697974217311233e-05,
|
| 67 |
+
"loss": 0.6009,
|
| 68 |
"step": 5000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.08,
|
| 72 |
+
"learning_rate": 2.746777163904236e-05,
|
| 73 |
+
"loss": 0.6567,
|
| 74 |
"step": 5500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.09,
|
| 78 |
+
"learning_rate": 2.7237569060773482e-05,
|
| 79 |
+
"loss": 0.6283,
|
| 80 |
"step": 6000
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.1,
|
| 84 |
+
"learning_rate": 2.7007366482504606e-05,
|
| 85 |
+
"loss": 0.6209,
|
| 86 |
"step": 6500
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.11,
|
| 90 |
+
"learning_rate": 2.677716390423573e-05,
|
| 91 |
+
"loss": 0.575,
|
| 92 |
"step": 7000
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.12,
|
| 96 |
+
"learning_rate": 2.6546961325966852e-05,
|
| 97 |
+
"loss": 0.5816,
|
| 98 |
"step": 7500
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.12,
|
| 102 |
+
"learning_rate": 2.6316758747697975e-05,
|
| 103 |
+
"loss": 0.5344,
|
| 104 |
"step": 8000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.13,
|
| 108 |
+
"learning_rate": 2.6086556169429095e-05,
|
| 109 |
+
"loss": 0.6448,
|
| 110 |
"step": 8500
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.14,
|
| 114 |
+
"learning_rate": 2.585635359116022e-05,
|
| 115 |
+
"loss": 0.5398,
|
| 116 |
"step": 9000
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.15,
|
| 120 |
+
"learning_rate": 2.5626151012891345e-05,
|
| 121 |
+
"loss": 0.5811,
|
| 122 |
"step": 9500
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.15,
|
| 126 |
+
"learning_rate": 2.5395948434622468e-05,
|
| 127 |
+
"loss": 0.568,
|
| 128 |
"step": 10000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.16,
|
| 132 |
+
"learning_rate": 2.516574585635359e-05,
|
| 133 |
+
"loss": 0.5947,
|
| 134 |
"step": 10500
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.17,
|
| 138 |
+
"learning_rate": 2.4935543278084714e-05,
|
| 139 |
+
"loss": 0.5479,
|
| 140 |
"step": 11000
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.18,
|
| 144 |
+
"learning_rate": 2.4705340699815838e-05,
|
| 145 |
+
"loss": 0.5466,
|
| 146 |
"step": 11500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.18,
|
| 150 |
+
"learning_rate": 2.4475138121546964e-05,
|
| 151 |
+
"loss": 0.5062,
|
| 152 |
"step": 12000
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.19,
|
| 156 |
+
"learning_rate": 2.4244935543278084e-05,
|
| 157 |
+
"loss": 0.5806,
|
| 158 |
"step": 12500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.2,
|
| 162 |
+
"learning_rate": 2.4014732965009207e-05,
|
| 163 |
+
"loss": 0.5795,
|
| 164 |
"step": 13000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.21,
|
| 168 |
+
"learning_rate": 2.3784530386740334e-05,
|
| 169 |
+
"loss": 0.5738,
|
| 170 |
"step": 13500
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.21,
|
| 174 |
+
"learning_rate": 2.3554327808471457e-05,
|
| 175 |
+
"loss": 0.5384,
|
| 176 |
"step": 14000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.22,
|
| 180 |
+
"learning_rate": 2.3324125230202577e-05,
|
| 181 |
+
"loss": 0.5797,
|
| 182 |
"step": 14500
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.23,
|
| 186 |
+
"learning_rate": 2.3093922651933703e-05,
|
| 187 |
+
"loss": 0.581,
|
| 188 |
"step": 15000
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.24,
|
| 192 |
+
"learning_rate": 2.2863720073664826e-05,
|
| 193 |
+
"loss": 0.5418,
|
| 194 |
"step": 15500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.25,
|
| 198 |
+
"learning_rate": 2.2633517495395946e-05,
|
| 199 |
+
"loss": 0.5731,
|
| 200 |
"step": 16000
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.25,
|
| 204 |
+
"learning_rate": 2.2403314917127073e-05,
|
| 205 |
+
"loss": 0.5668,
|
| 206 |
"step": 16500
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.26,
|
| 210 |
+
"learning_rate": 2.2173112338858196e-05,
|
| 211 |
+
"loss": 0.5484,
|
| 212 |
"step": 17000
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.27,
|
| 216 |
+
"learning_rate": 2.194290976058932e-05,
|
| 217 |
+
"loss": 0.5582,
|
| 218 |
"step": 17500
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.28,
|
| 222 |
+
"learning_rate": 2.1712707182320442e-05,
|
| 223 |
+
"loss": 0.5201,
|
| 224 |
"step": 18000
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.28,
|
| 228 |
+
"learning_rate": 2.1482504604051566e-05,
|
| 229 |
+
"loss": 0.605,
|
| 230 |
"step": 18500
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.29,
|
| 234 |
+
"learning_rate": 2.125230202578269e-05,
|
| 235 |
+
"loss": 0.5933,
|
| 236 |
"step": 19000
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.3,
|
| 240 |
+
"learning_rate": 2.1022099447513815e-05,
|
| 241 |
+
"loss": 0.5317,
|
| 242 |
"step": 19500
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.31,
|
| 246 |
+
"learning_rate": 2.0791896869244935e-05,
|
| 247 |
+
"loss": 0.5918,
|
| 248 |
"step": 20000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.31,
|
| 252 |
+
"learning_rate": 2.056169429097606e-05,
|
| 253 |
+
"loss": 0.6269,
|
| 254 |
"step": 20500
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"learning_rate": 2.0331491712707185e-05,
|
| 259 |
+
"loss": 0.5553,
|
| 260 |
"step": 21000
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 0.33,
|
| 264 |
+
"learning_rate": 2.0101289134438308e-05,
|
| 265 |
+
"loss": 0.5287,
|
| 266 |
"step": 21500
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 0.34,
|
| 270 |
+
"learning_rate": 1.9871086556169428e-05,
|
| 271 |
+
"loss": 0.5419,
|
| 272 |
"step": 22000
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 0.35,
|
| 276 |
+
"learning_rate": 1.9640883977900554e-05,
|
| 277 |
+
"loss": 0.6056,
|
| 278 |
"step": 22500
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 0.35,
|
| 282 |
+
"learning_rate": 1.9410681399631678e-05,
|
| 283 |
+
"loss": 0.5206,
|
| 284 |
"step": 23000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 0.36,
|
| 288 |
+
"learning_rate": 1.9180478821362797e-05,
|
| 289 |
+
"loss": 0.5659,
|
| 290 |
"step": 23500
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.37,
|
| 294 |
+
"learning_rate": 1.8950276243093924e-05,
|
| 295 |
+
"loss": 0.5336,
|
| 296 |
"step": 24000
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.38,
|
| 300 |
+
"learning_rate": 1.8720073664825047e-05,
|
| 301 |
+
"loss": 0.6182,
|
| 302 |
"step": 24500
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 0.38,
|
| 306 |
+
"learning_rate": 1.848987108655617e-05,
|
| 307 |
+
"loss": 0.5611,
|
| 308 |
"step": 25000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 0.39,
|
| 312 |
+
"learning_rate": 1.825966850828729e-05,
|
| 313 |
+
"loss": 0.5355,
|
| 314 |
"step": 25500
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 0.4,
|
| 318 |
+
"learning_rate": 1.8029465930018417e-05,
|
| 319 |
+
"loss": 0.4977,
|
| 320 |
"step": 26000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 0.41,
|
| 324 |
+
"learning_rate": 1.779926335174954e-05,
|
| 325 |
+
"loss": 0.5233,
|
| 326 |
"step": 26500
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 0.41,
|
| 330 |
+
"learning_rate": 1.7569060773480663e-05,
|
| 331 |
+
"loss": 0.5405,
|
| 332 |
"step": 27000
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 0.42,
|
| 336 |
+
"learning_rate": 1.7338858195211786e-05,
|
| 337 |
+
"loss": 0.5753,
|
| 338 |
"step": 27500
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.43,
|
| 342 |
+
"learning_rate": 1.710865561694291e-05,
|
| 343 |
+
"loss": 0.5489,
|
| 344 |
"step": 28000
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 0.44,
|
| 348 |
+
"learning_rate": 1.6878453038674033e-05,
|
| 349 |
+
"loss": 0.5646,
|
| 350 |
"step": 28500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 0.45,
|
| 354 |
+
"learning_rate": 1.664825046040516e-05,
|
| 355 |
+
"loss": 0.5525,
|
| 356 |
"step": 29000
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 0.45,
|
| 360 |
+
"learning_rate": 1.641804788213628e-05,
|
| 361 |
+
"loss": 0.5312,
|
| 362 |
"step": 29500
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 0.46,
|
| 366 |
+
"learning_rate": 1.6187845303867402e-05,
|
| 367 |
+
"loss": 0.5748,
|
| 368 |
"step": 30000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 0.47,
|
| 372 |
+
"learning_rate": 1.595764272559853e-05,
|
| 373 |
+
"loss": 0.5524,
|
| 374 |
"step": 30500
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.48,
|
| 378 |
+
"learning_rate": 1.572744014732965e-05,
|
| 379 |
+
"loss": 0.5059,
|
| 380 |
"step": 31000
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.48,
|
| 384 |
+
"learning_rate": 1.5497237569060772e-05,
|
| 385 |
+
"loss": 0.5831,
|
| 386 |
"step": 31500
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 0.49,
|
| 390 |
+
"learning_rate": 1.52670349907919e-05,
|
| 391 |
+
"loss": 0.5455,
|
| 392 |
"step": 32000
|
| 393 |
},
|
| 394 |
{
|
| 395 |
+
"epoch": 0.5,
|
| 396 |
+
"learning_rate": 1.503683241252302e-05,
|
| 397 |
+
"loss": 0.4879,
|
| 398 |
"step": 32500
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 0.51,
|
| 402 |
+
"learning_rate": 1.4806629834254145e-05,
|
| 403 |
+
"loss": 0.5425,
|
| 404 |
"step": 33000
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 0.51,
|
| 408 |
+
"learning_rate": 1.4576427255985266e-05,
|
| 409 |
+
"loss": 0.5369,
|
| 410 |
"step": 33500
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.52,
|
| 414 |
+
"learning_rate": 1.4346224677716391e-05,
|
| 415 |
+
"loss": 0.5857,
|
| 416 |
"step": 34000
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 0.53,
|
| 420 |
+
"learning_rate": 1.4116022099447514e-05,
|
| 421 |
+
"loss": 0.5857,
|
| 422 |
"step": 34500
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.54,
|
| 426 |
+
"learning_rate": 1.3885819521178638e-05,
|
| 427 |
+
"loss": 0.4599,
|
| 428 |
"step": 35000
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 0.54,
|
| 432 |
+
"learning_rate": 1.365561694290976e-05,
|
| 433 |
+
"loss": 0.5572,
|
| 434 |
"step": 35500
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 0.55,
|
| 438 |
+
"learning_rate": 1.3425414364640886e-05,
|
| 439 |
+
"loss": 0.495,
|
| 440 |
"step": 36000
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 0.56,
|
| 444 |
+
"learning_rate": 1.3195211786372007e-05,
|
| 445 |
+
"loss": 0.5072,
|
| 446 |
"step": 36500
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 0.57,
|
| 450 |
+
"learning_rate": 1.2965009208103132e-05,
|
| 451 |
+
"loss": 0.5297,
|
| 452 |
"step": 37000
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.58,
|
| 456 |
+
"learning_rate": 1.2734806629834255e-05,
|
| 457 |
+
"loss": 0.5723,
|
| 458 |
"step": 37500
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 0.58,
|
| 462 |
+
"learning_rate": 1.2504604051565377e-05,
|
| 463 |
+
"loss": 0.5406,
|
| 464 |
"step": 38000
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.59,
|
| 468 |
+
"learning_rate": 1.2274401473296502e-05,
|
| 469 |
+
"loss": 0.4575,
|
| 470 |
"step": 38500
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 0.6,
|
| 474 |
+
"learning_rate": 1.2044198895027623e-05,
|
| 475 |
+
"loss": 0.5566,
|
| 476 |
"step": 39000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 0.61,
|
| 480 |
+
"learning_rate": 1.1813996316758748e-05,
|
| 481 |
+
"loss": 0.5406,
|
| 482 |
"step": 39500
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 0.61,
|
| 486 |
+
"learning_rate": 1.1583793738489871e-05,
|
| 487 |
+
"loss": 0.5272,
|
| 488 |
"step": 40000
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 0.62,
|
| 492 |
+
"learning_rate": 1.1353591160220994e-05,
|
| 493 |
+
"loss": 0.5148,
|
| 494 |
"step": 40500
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 0.63,
|
| 498 |
+
"learning_rate": 1.1123388581952118e-05,
|
| 499 |
+
"loss": 0.495,
|
| 500 |
"step": 41000
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 0.64,
|
| 504 |
+
"learning_rate": 1.0893186003683243e-05,
|
| 505 |
+
"loss": 0.5069,
|
| 506 |
"step": 41500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.64,
|
| 510 |
+
"learning_rate": 1.0662983425414364e-05,
|
| 511 |
+
"loss": 0.5322,
|
| 512 |
"step": 42000
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 0.65,
|
| 516 |
+
"learning_rate": 1.0432780847145489e-05,
|
| 517 |
+
"loss": 0.517,
|
| 518 |
"step": 42500
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 0.66,
|
| 522 |
+
"learning_rate": 1.0202578268876612e-05,
|
| 523 |
+
"loss": 0.5464,
|
| 524 |
"step": 43000
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.67,
|
| 528 |
+
"learning_rate": 9.972375690607735e-06,
|
| 529 |
+
"loss": 0.5402,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.68,
|
| 534 |
+
"learning_rate": 9.742173112338858e-06,
|
| 535 |
+
"loss": 0.5274,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.68,
|
| 540 |
+
"learning_rate": 9.511970534069983e-06,
|
| 541 |
+
"loss": 0.4793,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.69,
|
| 546 |
+
"learning_rate": 9.281767955801105e-06,
|
| 547 |
+
"loss": 0.516,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7,
|
| 552 |
+
"learning_rate": 9.051565377532228e-06,
|
| 553 |
+
"loss": 0.5176,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.71,
|
| 558 |
+
"learning_rate": 8.821362799263353e-06,
|
| 559 |
+
"loss": 0.5499,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.71,
|
| 564 |
+
"learning_rate": 8.591160220994474e-06,
|
| 565 |
+
"loss": 0.5122,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.72,
|
| 570 |
+
"learning_rate": 8.3609576427256e-06,
|
| 571 |
+
"loss": 0.4958,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.73,
|
| 576 |
+
"learning_rate": 8.13075506445672e-06,
|
| 577 |
+
"loss": 0.5044,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.74,
|
| 582 |
+
"learning_rate": 7.900552486187846e-06,
|
| 583 |
+
"loss": 0.4872,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.74,
|
| 588 |
+
"learning_rate": 7.670349907918969e-06,
|
| 589 |
+
"loss": 0.5513,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.75,
|
| 594 |
+
"learning_rate": 7.440147329650092e-06,
|
| 595 |
+
"loss": 0.5299,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.76,
|
| 600 |
+
"learning_rate": 7.209944751381215e-06,
|
| 601 |
+
"loss": 0.5378,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.77,
|
| 606 |
+
"learning_rate": 6.9797421731123384e-06,
|
| 607 |
+
"loss": 0.4696,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.78,
|
| 612 |
+
"learning_rate": 6.7495395948434625e-06,
|
| 613 |
+
"loss": 0.554,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.78,
|
| 618 |
+
"learning_rate": 6.519337016574586e-06,
|
| 619 |
+
"loss": 0.5134,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.79,
|
| 624 |
+
"learning_rate": 6.289134438305709e-06,
|
| 625 |
+
"loss": 0.5365,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8,
|
| 630 |
+
"learning_rate": 6.058931860036833e-06,
|
| 631 |
+
"loss": 0.4699,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.81,
|
| 636 |
+
"learning_rate": 5.828729281767956e-06,
|
| 637 |
+
"loss": 0.5396,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.81,
|
| 642 |
+
"learning_rate": 5.598526703499079e-06,
|
| 643 |
+
"loss": 0.5194,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.82,
|
| 648 |
+
"learning_rate": 5.3683241252302025e-06,
|
| 649 |
+
"loss": 0.5561,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.83,
|
| 654 |
+
"learning_rate": 5.1381215469613265e-06,
|
| 655 |
+
"loss": 0.4814,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.84,
|
| 660 |
+
"learning_rate": 4.90791896869245e-06,
|
| 661 |
+
"loss": 0.5144,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.84,
|
| 666 |
+
"learning_rate": 4.677716390423573e-06,
|
| 667 |
+
"loss": 0.5026,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.85,
|
| 672 |
+
"learning_rate": 4.447513812154697e-06,
|
| 673 |
+
"loss": 0.4791,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.86,
|
| 678 |
+
"learning_rate": 4.21731123388582e-06,
|
| 679 |
+
"loss": 0.4993,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.87,
|
| 684 |
+
"learning_rate": 3.9871086556169424e-06,
|
| 685 |
+
"loss": 0.5426,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.87,
|
| 690 |
+
"learning_rate": 3.756906077348067e-06,
|
| 691 |
+
"loss": 0.5015,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.88,
|
| 696 |
+
"learning_rate": 3.52670349907919e-06,
|
| 697 |
+
"loss": 0.48,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.89,
|
| 702 |
+
"learning_rate": 3.296500920810313e-06,
|
| 703 |
+
"loss": 0.5186,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.9,
|
| 708 |
+
"learning_rate": 3.0662983425414365e-06,
|
| 709 |
+
"loss": 0.4555,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.91,
|
| 714 |
+
"learning_rate": 2.8360957642725597e-06,
|
| 715 |
+
"loss": 0.4436,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.91,
|
| 720 |
+
"learning_rate": 2.6058931860036833e-06,
|
| 721 |
+
"loss": 0.4678,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.92,
|
| 726 |
+
"learning_rate": 2.375690607734807e-06,
|
| 727 |
+
"loss": 0.496,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.93,
|
| 732 |
+
"learning_rate": 2.14548802946593e-06,
|
| 733 |
+
"loss": 0.4916,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.94,
|
| 738 |
+
"learning_rate": 1.9152854511970537e-06,
|
| 739 |
+
"loss": 0.4931,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.94,
|
| 744 |
+
"learning_rate": 1.6850828729281769e-06,
|
| 745 |
+
"loss": 0.5168,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.95,
|
| 750 |
+
"learning_rate": 1.4548802946593003e-06,
|
| 751 |
+
"loss": 0.4746,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.96,
|
| 756 |
+
"learning_rate": 1.2246777163904237e-06,
|
| 757 |
+
"loss": 0.4873,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.97,
|
| 762 |
+
"learning_rate": 9.944751381215469e-07,
|
| 763 |
+
"loss": 0.4708,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.97,
|
| 768 |
+
"learning_rate": 7.642725598526704e-07,
|
| 769 |
+
"loss": 0.503,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.98,
|
| 774 |
+
"learning_rate": 5.340699815837937e-07,
|
| 775 |
+
"loss": 0.5206,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.99,
|
| 780 |
+
"learning_rate": 3.0386740331491715e-07,
|
| 781 |
+
"loss": 0.4864,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.0,
|
| 786 |
+
"learning_rate": 7.366482504604051e-08,
|
| 787 |
+
"loss": 0.4967,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.0,
|
| 792 |
+
"step": 65160,
|
| 793 |
+
"total_flos": 6.692767006025318e+16,
|
| 794 |
+
"train_loss": 0.5451086903730735,
|
| 795 |
+
"train_runtime": 8999.8943,
|
| 796 |
+
"train_samples_per_second": 14.48,
|
| 797 |
+
"train_steps_per_second": 7.24
|
| 798 |
}
|
| 799 |
],
|
| 800 |
+
"max_steps": 65160,
|
| 801 |
+
"num_train_epochs": 1,
|
| 802 |
+
"total_flos": 6.692767006025318e+16,
|
| 803 |
"trial_name": null,
|
| 804 |
"trial_params": null
|
| 805 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9daf810b1587fb1c69c59379e5aaa8588e329be6ef1a6db71bc49efb80f57957
|
| 3 |
+
size 3771
|