End of training
Browse files
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 20.
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss": 1.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"eval_loss": 0.20258904993534088,
|
| 4 |
+
"eval_runtime": 135.1965,
|
| 5 |
+
"eval_samples": 2744,
|
| 6 |
+
"eval_samples_per_second": 20.296,
|
| 7 |
+
"eval_steps_per_second": 2.537,
|
| 8 |
+
"eval_wer": 0.3247906274312388,
|
| 9 |
+
"train_loss": 1.3946523244330225,
|
| 10 |
+
"train_runtime": 41836.5581,
|
| 11 |
+
"train_samples": 6034,
|
| 12 |
+
"train_samples_per_second": 14.423,
|
| 13 |
+
"train_steps_per_second": 0.225
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 20.
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"eval_loss": 0.20258904993534088,
|
| 4 |
+
"eval_runtime": 135.1965,
|
| 5 |
+
"eval_samples": 2744,
|
| 6 |
+
"eval_samples_per_second": 20.296,
|
| 7 |
+
"eval_steps_per_second": 2.537,
|
| 8 |
+
"eval_wer": 0.3247906274312388
|
| 9 |
}
|
runs/Feb03_17-57-48_job-699ba53c-fea9-4eb2-81af-a97f440eaa45/events.out.tfevents.1643953286.job-699ba53c-fea9-4eb2-81af-a97f440eaa45.861419.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc6d3bd6de783de478dded36396b456f500e13b482db6f63358a0c954a46424d
|
| 3 |
+
size 358
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"train_loss": 1.3946523244330225,
|
| 4 |
+
"train_runtime": 41836.5581,
|
| 5 |
+
"train_samples": 6034,
|
| 6 |
+
"train_samples_per_second": 14.423,
|
| 7 |
+
"train_steps_per_second": 0.225
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,823 +1,751 @@
|
|
| 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":
|
| 12 |
"learning_rate": 4.950000000000001e-06,
|
| 13 |
-
"loss": 13.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch":
|
| 18 |
"learning_rate": 9.950000000000001e-06,
|
| 19 |
-
"loss": 5.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch":
|
| 24 |
"learning_rate": 1.4950000000000001e-05,
|
| 25 |
-
"loss": 4.
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch":
|
| 30 |
"learning_rate": 1.995e-05,
|
| 31 |
-
"loss": 3.
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
"learning_rate": 2.495e-05,
|
| 37 |
-
"loss": 3.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"eval_loss": 3.
|
| 43 |
-
"eval_runtime":
|
| 44 |
-
"eval_samples_per_second": 20.
|
| 45 |
-
"eval_steps_per_second": 2.
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
-
"epoch":
|
| 51 |
"learning_rate": 2.995e-05,
|
| 52 |
-
"loss": 3.
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
"learning_rate": 3.495e-05,
|
| 58 |
-
"loss": 3.
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch":
|
| 63 |
"learning_rate": 3.995e-05,
|
| 64 |
-
"loss": 3.
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch":
|
| 69 |
"learning_rate": 4.495e-05,
|
| 70 |
-
"loss": 3.
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch":
|
| 75 |
"learning_rate": 4.995e-05,
|
| 76 |
-
"loss":
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch":
|
| 81 |
-
"eval_loss": 2.
|
| 82 |
-
"eval_runtime":
|
| 83 |
-
"eval_samples_per_second": 20.
|
| 84 |
-
"eval_steps_per_second": 2.
|
| 85 |
-
"eval_wer": 0.
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
"learning_rate": 5.495e-05,
|
| 91 |
-
"loss": 2.
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch":
|
| 96 |
"learning_rate": 5.995000000000001e-05,
|
| 97 |
-
"loss": 1.
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch":
|
| 102 |
"learning_rate": 6.494999999999999e-05,
|
| 103 |
-
"loss": 1.
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
"learning_rate": 6.995e-05,
|
| 109 |
-
"loss": 1.
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch":
|
| 114 |
"learning_rate": 7.495e-05,
|
| 115 |
-
"loss": 1.
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime":
|
| 122 |
-
"eval_samples_per_second": 20.
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
-
"eval_wer": 0.
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
"learning_rate": 7.995e-05,
|
| 130 |
-
"loss": 1.
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch":
|
| 135 |
"learning_rate": 8.495e-05,
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
"learning_rate": 8.995e-05,
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
"learning_rate": 9.495e-05,
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 1900
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"learning_rate": 9.
|
| 154 |
-
"loss": 1.
|
| 155 |
"step": 2000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"eval_loss": 0.
|
| 160 |
-
"eval_runtime": 134.
|
| 161 |
-
"eval_samples_per_second": 20.
|
| 162 |
-
"eval_steps_per_second": 2.
|
| 163 |
-
"eval_wer": 0.
|
| 164 |
"step": 2000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 2100
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 1.
|
| 176 |
"step": 2200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 1.
|
| 182 |
"step": 2300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 1.
|
| 188 |
"step": 2400
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 1.
|
| 194 |
"step": 2500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"eval_loss": 0.
|
| 199 |
-
"eval_runtime":
|
| 200 |
-
"eval_samples_per_second": 20.
|
| 201 |
-
"eval_steps_per_second": 2.
|
| 202 |
-
"eval_wer": 0.
|
| 203 |
"step": 2500
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch":
|
| 207 |
-
"learning_rate": 9.
|
| 208 |
-
"loss": 1.
|
| 209 |
"step": 2600
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"learning_rate": 9.
|
| 214 |
-
"loss": 1.
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate": 8.
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate": 8.
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"eval_loss": 0.
|
| 238 |
-
"eval_runtime":
|
| 239 |
-
"eval_samples_per_second": 20.
|
| 240 |
-
"eval_steps_per_second": 2.
|
| 241 |
-
"eval_wer": 0.
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate": 8.
|
| 247 |
-
"loss": 1.
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"learning_rate": 8.
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate": 8.
|
| 259 |
-
"loss": 1.
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"learning_rate": 8.
|
| 265 |
-
"loss": 1.
|
| 266 |
"step": 3400
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch":
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 1.
|
| 272 |
"step": 3500
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch":
|
| 276 |
-
"eval_loss": 0.
|
| 277 |
-
"eval_runtime":
|
| 278 |
-
"eval_samples_per_second": 20.
|
| 279 |
-
"eval_steps_per_second": 2.
|
| 280 |
-
"eval_wer": 0.
|
| 281 |
"step": 3500
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 1.
|
| 287 |
"step": 3600
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"learning_rate": 7.
|
| 292 |
-
"loss": 1.
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate": 7.
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"learning_rate": 7.
|
| 304 |
-
"loss": 1.
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"learning_rate": 7.
|
| 310 |
-
"loss": 1.
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"eval_loss": 0.
|
| 316 |
-
"eval_runtime":
|
| 317 |
-
"eval_samples_per_second": 20.
|
| 318 |
-
"eval_steps_per_second": 2.
|
| 319 |
-
"eval_wer": 0.
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch":
|
| 324 |
-
"learning_rate": 7.
|
| 325 |
-
"loss": 1.
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"learning_rate": 7.
|
| 331 |
-
"loss": 1.
|
| 332 |
"step": 4200
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 4300
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch":
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss":
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch":
|
| 348 |
-
"learning_rate": 6.
|
| 349 |
-
"loss":
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch":
|
| 354 |
-
"eval_loss": 0.
|
| 355 |
-
"eval_runtime":
|
| 356 |
-
"eval_samples_per_second": 20.
|
| 357 |
-
"eval_steps_per_second": 2.
|
| 358 |
-
"eval_wer": 0.
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"learning_rate": 6.
|
| 364 |
-
"loss":
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch":
|
| 369 |
-
"learning_rate": 6.
|
| 370 |
-
"loss":
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"learning_rate": 6.
|
| 376 |
-
"loss":
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"learning_rate": 6.
|
| 382 |
-
"loss":
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch":
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss":
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"eval_loss": 0.
|
| 394 |
-
"eval_runtime":
|
| 395 |
-
"eval_samples_per_second": 20.
|
| 396 |
-
"eval_steps_per_second": 2.
|
| 397 |
-
"eval_wer": 0.
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"learning_rate":
|
| 403 |
-
"loss": 0.
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch":
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch":
|
| 420 |
-
"learning_rate": 5.
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"learning_rate": 5.
|
| 427 |
-
"loss": 0.
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch":
|
| 432 |
-
"eval_loss": 0.
|
| 433 |
-
"eval_runtime":
|
| 434 |
-
"eval_samples_per_second": 20.
|
| 435 |
-
"eval_steps_per_second": 2.
|
| 436 |
-
"eval_wer": 0.
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"learning_rate": 5.
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"learning_rate": 5.
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"eval_loss": 0.
|
| 472 |
-
"eval_runtime":
|
| 473 |
-
"eval_samples_per_second": 20.
|
| 474 |
-
"eval_steps_per_second": 2.
|
| 475 |
-
"eval_wer": 0.
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"learning_rate": 4.
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"learning_rate": 4.
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"learning_rate": 4.
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 6400
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 6500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch":
|
| 510 |
-
"eval_loss": 0.
|
| 511 |
-
"eval_runtime":
|
| 512 |
-
"eval_samples_per_second":
|
| 513 |
-
"eval_steps_per_second": 2.
|
| 514 |
-
"eval_wer": 0.
|
| 515 |
"step": 6500
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 0.
|
| 521 |
"step": 6600
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"epoch":
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 6700
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch":
|
| 531 |
-
"learning_rate":
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 6800
|
| 534 |
},
|
| 535 |
{
|
| 536 |
-
"epoch":
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 6900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch":
|
| 543 |
-
"learning_rate": 3.
|
| 544 |
-
"loss": 0.
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"eval_loss": 0.
|
| 550 |
-
"eval_runtime":
|
| 551 |
-
"eval_samples_per_second": 20.
|
| 552 |
-
"eval_steps_per_second": 2.
|
| 553 |
-
"eval_wer": 0.
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch":
|
| 558 |
-
"learning_rate": 3.
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 7100
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 7200
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 7300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 7400
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 7500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"eval_loss": 0.
|
| 589 |
-
"eval_runtime":
|
| 590 |
-
"eval_samples_per_second": 20.
|
| 591 |
-
"eval_steps_per_second": 2.
|
| 592 |
-
"eval_wer": 0.
|
| 593 |
"step": 7500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch":
|
| 597 |
-
"learning_rate":
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 7600
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch":
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss": 0.
|
| 605 |
"step": 7700
|
| 606 |
},
|
| 607 |
{
|
| 608 |
-
"epoch":
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 7800
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch":
|
| 615 |
-
"learning_rate": 2.
|
| 616 |
-
"loss": 0.
|
| 617 |
"step": 7900
|
| 618 |
},
|
| 619 |
{
|
| 620 |
-
"epoch":
|
| 621 |
-
"learning_rate":
|
| 622 |
-
"loss": 0.
|
| 623 |
"step": 8000
|
| 624 |
},
|
| 625 |
{
|
| 626 |
-
"epoch":
|
| 627 |
-
"eval_loss": 0.
|
| 628 |
-
"eval_runtime":
|
| 629 |
-
"eval_samples_per_second":
|
| 630 |
-
"eval_steps_per_second": 2.
|
| 631 |
-
"eval_wer": 0.
|
| 632 |
"step": 8000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 8100
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch":
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 8200
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 8300
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 8400
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 8500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"eval_loss": 0.
|
| 667 |
-
"eval_runtime":
|
| 668 |
-
"eval_samples_per_second": 20.
|
| 669 |
-
"eval_steps_per_second": 2.
|
| 670 |
-
"eval_wer": 0.
|
| 671 |
"step": 8500
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"learning_rate":
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 8600
|
| 678 |
},
|
| 679 |
{
|
| 680 |
-
"epoch":
|
| 681 |
-
"learning_rate":
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 8700
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"epoch":
|
| 687 |
-
"learning_rate":
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 8800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
-
"epoch":
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 8900
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch":
|
| 699 |
-
"learning_rate":
|
| 700 |
-
"loss": 0.
|
| 701 |
"step": 9000
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"epoch":
|
| 705 |
-
"eval_loss": 0.
|
| 706 |
-
"eval_runtime":
|
| 707 |
-
"eval_samples_per_second": 20.
|
| 708 |
-
"eval_steps_per_second": 2.
|
| 709 |
-
"eval_wer": 0.
|
| 710 |
"step": 9000
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 9100
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 9200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate": 1.
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"
|
| 739 |
-
"
|
| 740 |
-
"
|
| 741 |
-
|
| 742 |
-
|
| 743 |
-
"
|
| 744 |
-
"eval_loss": 0.21682003140449524,
|
| 745 |
-
"eval_runtime": 133.9404,
|
| 746 |
-
"eval_samples_per_second": 20.472,
|
| 747 |
-
"eval_steps_per_second": 2.561,
|
| 748 |
-
"eval_wer": 0.3276448962919257,
|
| 749 |
-
"step": 9500
|
| 750 |
-
},
|
| 751 |
-
{
|
| 752 |
-
"epoch": 93.2,
|
| 753 |
-
"learning_rate": 8.518072289156627e-06,
|
| 754 |
-
"loss": 0.8055,
|
| 755 |
-
"step": 9600
|
| 756 |
-
},
|
| 757 |
-
{
|
| 758 |
-
"epoch": 94.17,
|
| 759 |
-
"learning_rate": 7.313253012048194e-06,
|
| 760 |
-
"loss": 0.7955,
|
| 761 |
-
"step": 9700
|
| 762 |
-
},
|
| 763 |
-
{
|
| 764 |
-
"epoch": 95.15,
|
| 765 |
-
"learning_rate": 6.108433734939759e-06,
|
| 766 |
-
"loss": 0.7961,
|
| 767 |
-
"step": 9800
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 96.12,
|
| 771 |
-
"learning_rate": 4.903614457831326e-06,
|
| 772 |
-
"loss": 0.7843,
|
| 773 |
-
"step": 9900
|
| 774 |
-
},
|
| 775 |
-
{
|
| 776 |
-
"epoch": 97.09,
|
| 777 |
-
"learning_rate": 3.6987951807228917e-06,
|
| 778 |
-
"loss": 0.7781,
|
| 779 |
-
"step": 10000
|
| 780 |
-
},
|
| 781 |
-
{
|
| 782 |
-
"epoch": 97.09,
|
| 783 |
-
"eval_loss": 0.21628263592720032,
|
| 784 |
-
"eval_runtime": 133.5255,
|
| 785 |
-
"eval_samples_per_second": 20.535,
|
| 786 |
-
"eval_steps_per_second": 2.569,
|
| 787 |
-
"eval_wer": 0.3240593654972087,
|
| 788 |
-
"step": 10000
|
| 789 |
-
},
|
| 790 |
-
{
|
| 791 |
-
"epoch": 98.06,
|
| 792 |
-
"learning_rate": 2.493975903614458e-06,
|
| 793 |
-
"loss": 0.7842,
|
| 794 |
-
"step": 10100
|
| 795 |
-
},
|
| 796 |
-
{
|
| 797 |
-
"epoch": 99.03,
|
| 798 |
-
"learning_rate": 1.2891566265060241e-06,
|
| 799 |
-
"loss": 0.7821,
|
| 800 |
-
"step": 10200
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 100.0,
|
| 804 |
-
"learning_rate": 9.638554216867469e-08,
|
| 805 |
-
"loss": 0.7779,
|
| 806 |
-
"step": 10300
|
| 807 |
-
},
|
| 808 |
-
{
|
| 809 |
-
"epoch": 100.0,
|
| 810 |
-
"step": 10300,
|
| 811 |
-
"total_flos": 5.823193156406256e+19,
|
| 812 |
-
"train_loss": 1.3660302423273476,
|
| 813 |
-
"train_runtime": 26867.6077,
|
| 814 |
-
"train_samples_per_second": 12.253,
|
| 815 |
-
"train_steps_per_second": 0.383
|
| 816 |
}
|
| 817 |
],
|
| 818 |
-
"max_steps":
|
| 819 |
"num_train_epochs": 100,
|
| 820 |
-
"total_flos":
|
| 821 |
"trial_name": null,
|
| 822 |
"trial_params": null
|
| 823 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.994708994709,
|
| 5 |
+
"global_step": 9400,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 1.06,
|
| 12 |
"learning_rate": 4.950000000000001e-06,
|
| 13 |
+
"loss": 13.5577,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 2.13,
|
| 18 |
"learning_rate": 9.950000000000001e-06,
|
| 19 |
+
"loss": 5.868,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 3.19,
|
| 24 |
"learning_rate": 1.4950000000000001e-05,
|
| 25 |
+
"loss": 4.1355,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 4.25,
|
| 30 |
"learning_rate": 1.995e-05,
|
| 31 |
+
"loss": 3.5988,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 5.32,
|
| 36 |
"learning_rate": 2.495e-05,
|
| 37 |
+
"loss": 3.3036,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 5.32,
|
| 42 |
+
"eval_loss": 3.262770652770996,
|
| 43 |
+
"eval_runtime": 136.0729,
|
| 44 |
+
"eval_samples_per_second": 20.166,
|
| 45 |
+
"eval_steps_per_second": 2.521,
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
+
"epoch": 6.38,
|
| 51 |
"learning_rate": 2.995e-05,
|
| 52 |
+
"loss": 3.2243,
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 7.44,
|
| 57 |
"learning_rate": 3.495e-05,
|
| 58 |
+
"loss": 3.1823,
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 8.51,
|
| 63 |
"learning_rate": 3.995e-05,
|
| 64 |
+
"loss": 3.1371,
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 9.57,
|
| 69 |
"learning_rate": 4.495e-05,
|
| 70 |
+
"loss": 3.0942,
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
+
"epoch": 10.63,
|
| 75 |
"learning_rate": 4.995e-05,
|
| 76 |
+
"loss": 2.9734,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 10.63,
|
| 81 |
+
"eval_loss": 2.5676724910736084,
|
| 82 |
+
"eval_runtime": 135.3086,
|
| 83 |
+
"eval_samples_per_second": 20.28,
|
| 84 |
+
"eval_steps_per_second": 2.535,
|
| 85 |
+
"eval_wer": 0.9980321266761246,
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 11.7,
|
| 90 |
"learning_rate": 5.495e-05,
|
| 91 |
+
"loss": 2.2815,
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 12.76,
|
| 96 |
"learning_rate": 5.995000000000001e-05,
|
| 97 |
+
"loss": 1.6698,
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 13.83,
|
| 102 |
"learning_rate": 6.494999999999999e-05,
|
| 103 |
+
"loss": 1.4895,
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 14.89,
|
| 108 |
"learning_rate": 6.995e-05,
|
| 109 |
+
"loss": 1.3959,
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 15.95,
|
| 114 |
"learning_rate": 7.495e-05,
|
| 115 |
+
"loss": 1.3466,
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 15.95,
|
| 120 |
+
"eval_loss": 0.44553351402282715,
|
| 121 |
+
"eval_runtime": 134.1872,
|
| 122 |
+
"eval_samples_per_second": 20.449,
|
| 123 |
+
"eval_steps_per_second": 2.556,
|
| 124 |
+
"eval_wer": 0.6306347535581895,
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 17.02,
|
| 129 |
"learning_rate": 7.995e-05,
|
| 130 |
+
"loss": 1.3288,
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 18.08,
|
| 135 |
"learning_rate": 8.495e-05,
|
| 136 |
+
"loss": 1.2866,
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 19.15,
|
| 141 |
"learning_rate": 8.995e-05,
|
| 142 |
+
"loss": 1.2619,
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 20.21,
|
| 147 |
"learning_rate": 9.495e-05,
|
| 148 |
+
"loss": 1.2496,
|
| 149 |
"step": 1900
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 21.28,
|
| 153 |
+
"learning_rate": 9.99e-05,
|
| 154 |
+
"loss": 1.2424,
|
| 155 |
"step": 2000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 21.28,
|
| 159 |
+
"eval_loss": 0.3603059649467468,
|
| 160 |
+
"eval_runtime": 134.7251,
|
| 161 |
+
"eval_samples_per_second": 20.367,
|
| 162 |
+
"eval_steps_per_second": 2.546,
|
| 163 |
+
"eval_wer": 0.5301359205528351,
|
| 164 |
"step": 2000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 22.34,
|
| 168 |
+
"learning_rate": 9.867567567567569e-05,
|
| 169 |
+
"loss": 1.2253,
|
| 170 |
"step": 2100
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 23.4,
|
| 174 |
+
"learning_rate": 9.732432432432433e-05,
|
| 175 |
+
"loss": 1.209,
|
| 176 |
"step": 2200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 24.47,
|
| 180 |
+
"learning_rate": 9.597297297297298e-05,
|
| 181 |
+
"loss": 1.1984,
|
| 182 |
"step": 2300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 25.53,
|
| 186 |
+
"learning_rate": 9.462162162162162e-05,
|
| 187 |
+
"loss": 1.1735,
|
| 188 |
"step": 2400
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 26.59,
|
| 192 |
+
"learning_rate": 9.327027027027028e-05,
|
| 193 |
+
"loss": 1.1655,
|
| 194 |
"step": 2500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 26.59,
|
| 198 |
+
"eval_loss": 0.3164927661418915,
|
| 199 |
+
"eval_runtime": 135.1028,
|
| 200 |
+
"eval_samples_per_second": 20.31,
|
| 201 |
+
"eval_steps_per_second": 2.539,
|
| 202 |
+
"eval_wer": 0.4739828840785319,
|
| 203 |
"step": 2500
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 27.66,
|
| 207 |
+
"learning_rate": 9.191891891891893e-05,
|
| 208 |
+
"loss": 1.1529,
|
| 209 |
"step": 2600
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 28.72,
|
| 213 |
+
"learning_rate": 9.056756756756757e-05,
|
| 214 |
+
"loss": 1.1408,
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 29.78,
|
| 219 |
+
"learning_rate": 8.921621621621622e-05,
|
| 220 |
+
"loss": 1.1201,
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 30.85,
|
| 225 |
+
"learning_rate": 8.787837837837838e-05,
|
| 226 |
+
"loss": 1.1108,
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 31.91,
|
| 231 |
+
"learning_rate": 8.652702702702703e-05,
|
| 232 |
+
"loss": 1.1026,
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 31.91,
|
| 237 |
+
"eval_loss": 0.2930183410644531,
|
| 238 |
+
"eval_runtime": 135.7738,
|
| 239 |
+
"eval_samples_per_second": 20.21,
|
| 240 |
+
"eval_steps_per_second": 2.526,
|
| 241 |
+
"eval_wer": 0.4400256281177063,
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 32.97,
|
| 246 |
+
"learning_rate": 8.517567567567568e-05,
|
| 247 |
+
"loss": 1.1035,
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 34.04,
|
| 252 |
+
"learning_rate": 8.382432432432433e-05,
|
| 253 |
+
"loss": 1.0976,
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 35.11,
|
| 258 |
+
"learning_rate": 8.247297297297298e-05,
|
| 259 |
+
"loss": 1.0717,
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 36.17,
|
| 264 |
+
"learning_rate": 8.112162162162162e-05,
|
| 265 |
+
"loss": 1.0778,
|
| 266 |
"step": 3400
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 37.23,
|
| 270 |
+
"learning_rate": 7.977027027027028e-05,
|
| 271 |
+
"loss": 1.0655,
|
| 272 |
"step": 3500
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 37.23,
|
| 276 |
+
"eval_loss": 0.26754099130630493,
|
| 277 |
+
"eval_runtime": 134.1895,
|
| 278 |
+
"eval_samples_per_second": 20.449,
|
| 279 |
+
"eval_steps_per_second": 2.556,
|
| 280 |
+
"eval_wer": 0.41590773877625736,
|
| 281 |
"step": 3500
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 38.3,
|
| 285 |
+
"learning_rate": 7.841891891891892e-05,
|
| 286 |
+
"loss": 1.0588,
|
| 287 |
"step": 3600
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 39.36,
|
| 291 |
+
"learning_rate": 7.706756756756757e-05,
|
| 292 |
+
"loss": 1.0401,
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 40.42,
|
| 297 |
+
"learning_rate": 7.571621621621621e-05,
|
| 298 |
+
"loss": 1.0296,
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 41.49,
|
| 303 |
+
"learning_rate": 7.436486486486487e-05,
|
| 304 |
+
"loss": 1.0218,
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 42.55,
|
| 309 |
+
"learning_rate": 7.301351351351352e-05,
|
| 310 |
+
"loss": 1.0239,
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 42.55,
|
| 315 |
+
"eval_loss": 0.25800037384033203,
|
| 316 |
+
"eval_runtime": 134.1224,
|
| 317 |
+
"eval_samples_per_second": 20.459,
|
| 318 |
+
"eval_steps_per_second": 2.557,
|
| 319 |
+
"eval_wer": 0.3912864399798636,
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 43.61,
|
| 324 |
+
"learning_rate": 7.166216216216216e-05,
|
| 325 |
+
"loss": 1.0226,
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 44.68,
|
| 330 |
+
"learning_rate": 7.031081081081081e-05,
|
| 331 |
+
"loss": 1.0165,
|
| 332 |
"step": 4200
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 45.74,
|
| 336 |
+
"learning_rate": 6.895945945945947e-05,
|
| 337 |
+
"loss": 1.0083,
|
| 338 |
"step": 4300
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 46.8,
|
| 342 |
+
"learning_rate": 6.760810810810811e-05,
|
| 343 |
+
"loss": 0.9981,
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 47.87,
|
| 348 |
+
"learning_rate": 6.625675675675676e-05,
|
| 349 |
+
"loss": 0.9938,
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 47.87,
|
| 354 |
+
"eval_loss": 0.23732751607894897,
|
| 355 |
+
"eval_runtime": 135.9562,
|
| 356 |
+
"eval_samples_per_second": 20.183,
|
| 357 |
+
"eval_steps_per_second": 2.523,
|
| 358 |
+
"eval_wer": 0.36977712690494713,
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 48.93,
|
| 363 |
+
"learning_rate": 6.49054054054054e-05,
|
| 364 |
+
"loss": 0.9717,
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 49.99,
|
| 369 |
+
"learning_rate": 6.355405405405406e-05,
|
| 370 |
+
"loss": 0.9795,
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 51.06,
|
| 375 |
+
"learning_rate": 6.22027027027027e-05,
|
| 376 |
+
"loss": 0.9804,
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 52.13,
|
| 381 |
+
"learning_rate": 6.085135135135135e-05,
|
| 382 |
+
"loss": 0.9655,
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 53.19,
|
| 387 |
+
"learning_rate": 5.95e-05,
|
| 388 |
+
"loss": 0.9655,
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 53.19,
|
| 393 |
+
"eval_loss": 0.2379022240638733,
|
| 394 |
+
"eval_runtime": 134.3498,
|
| 395 |
+
"eval_samples_per_second": 20.424,
|
| 396 |
+
"eval_steps_per_second": 2.553,
|
| 397 |
+
"eval_wer": 0.3674889021097433,
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 54.25,
|
| 402 |
+
"learning_rate": 5.8148648648648655e-05,
|
| 403 |
+
"loss": 0.9616,
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 55.32,
|
| 408 |
+
"learning_rate": 5.67972972972973e-05,
|
| 409 |
+
"loss": 0.9633,
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 56.38,
|
| 414 |
+
"learning_rate": 5.544594594594595e-05,
|
| 415 |
+
"loss": 0.94,
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 57.44,
|
| 420 |
+
"learning_rate": 5.40945945945946e-05,
|
| 421 |
+
"loss": 0.9355,
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 58.51,
|
| 426 |
+
"learning_rate": 5.274324324324325e-05,
|
| 427 |
+
"loss": 0.9374,
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 58.51,
|
| 432 |
+
"eval_loss": 0.24859154224395752,
|
| 433 |
+
"eval_runtime": 135.3597,
|
| 434 |
+
"eval_samples_per_second": 20.272,
|
| 435 |
+
"eval_steps_per_second": 2.534,
|
| 436 |
+
"eval_wer": 0.3794792000366116,
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 59.57,
|
| 441 |
+
"learning_rate": 5.1391891891891894e-05,
|
| 442 |
+
"loss": 0.93,
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 60.63,
|
| 447 |
+
"learning_rate": 5.0040540540540546e-05,
|
| 448 |
+
"loss": 0.9212,
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 61.7,
|
| 453 |
+
"learning_rate": 4.868918918918919e-05,
|
| 454 |
+
"loss": 0.9233,
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 62.76,
|
| 459 |
+
"learning_rate": 4.733783783783784e-05,
|
| 460 |
+
"loss": 0.914,
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 63.83,
|
| 465 |
+
"learning_rate": 4.598648648648649e-05,
|
| 466 |
+
"loss": 0.9065,
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 63.83,
|
| 471 |
+
"eval_loss": 0.22428132593631744,
|
| 472 |
+
"eval_runtime": 136.639,
|
| 473 |
+
"eval_samples_per_second": 20.082,
|
| 474 |
+
"eval_steps_per_second": 2.51,
|
| 475 |
+
"eval_wer": 0.3405336140222415,
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 64.89,
|
| 480 |
+
"learning_rate": 4.463513513513514e-05,
|
| 481 |
+
"loss": 0.9023,
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 65.95,
|
| 486 |
+
"learning_rate": 4.3283783783783785e-05,
|
| 487 |
+
"loss": 0.9105,
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 67.02,
|
| 492 |
+
"learning_rate": 4.193243243243244e-05,
|
| 493 |
+
"loss": 0.907,
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 68.08,
|
| 498 |
+
"learning_rate": 4.058108108108108e-05,
|
| 499 |
+
"loss": 0.8939,
|
| 500 |
"step": 6400
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 69.15,
|
| 504 |
+
"learning_rate": 3.9229729729729734e-05,
|
| 505 |
+
"loss": 0.888,
|
| 506 |
"step": 6500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 69.15,
|
| 510 |
+
"eval_loss": 0.21568605303764343,
|
| 511 |
+
"eval_runtime": 139.7942,
|
| 512 |
+
"eval_samples_per_second": 19.629,
|
| 513 |
+
"eval_steps_per_second": 2.454,
|
| 514 |
+
"eval_wer": 0.3277195551690998,
|
| 515 |
"step": 6500
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 70.21,
|
| 519 |
+
"learning_rate": 3.78918918918919e-05,
|
| 520 |
+
"loss": 0.8866,
|
| 521 |
"step": 6600
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 71.28,
|
| 525 |
+
"learning_rate": 3.654054054054054e-05,
|
| 526 |
+
"loss": 0.8881,
|
| 527 |
"step": 6700
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 72.34,
|
| 531 |
+
"learning_rate": 3.5189189189189195e-05,
|
| 532 |
+
"loss": 0.8849,
|
| 533 |
"step": 6800
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 73.4,
|
| 537 |
+
"learning_rate": 3.383783783783784e-05,
|
| 538 |
+
"loss": 0.8648,
|
| 539 |
"step": 6900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 74.47,
|
| 543 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 544 |
+
"loss": 0.8646,
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 74.47,
|
| 549 |
+
"eval_loss": 0.21030458807945251,
|
| 550 |
+
"eval_runtime": 135.9762,
|
| 551 |
+
"eval_samples_per_second": 20.18,
|
| 552 |
+
"eval_steps_per_second": 2.522,
|
| 553 |
+
"eval_wer": 0.3287721385748936,
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 75.53,
|
| 558 |
+
"learning_rate": 3.114864864864865e-05,
|
| 559 |
+
"loss": 0.8767,
|
| 560 |
"step": 7100
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 76.59,
|
| 564 |
+
"learning_rate": 2.97972972972973e-05,
|
| 565 |
+
"loss": 0.8655,
|
| 566 |
"step": 7200
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 77.66,
|
| 570 |
+
"learning_rate": 2.8445945945945946e-05,
|
| 571 |
+
"loss": 0.8672,
|
| 572 |
"step": 7300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 78.72,
|
| 576 |
+
"learning_rate": 2.7094594594594598e-05,
|
| 577 |
+
"loss": 0.8625,
|
| 578 |
"step": 7400
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 79.78,
|
| 582 |
+
"learning_rate": 2.5743243243243243e-05,
|
| 583 |
+
"loss": 0.8602,
|
| 584 |
"step": 7500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 79.78,
|
| 588 |
+
"eval_loss": 0.20880180597305298,
|
| 589 |
+
"eval_runtime": 135.4639,
|
| 590 |
+
"eval_samples_per_second": 20.256,
|
| 591 |
+
"eval_steps_per_second": 2.532,
|
| 592 |
+
"eval_wer": 0.32378380852134914,
|
| 593 |
"step": 7500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 80.85,
|
| 597 |
+
"learning_rate": 2.4391891891891895e-05,
|
| 598 |
+
"loss": 0.8515,
|
| 599 |
"step": 7600
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 81.91,
|
| 603 |
+
"learning_rate": 2.3040540540540543e-05,
|
| 604 |
+
"loss": 0.8554,
|
| 605 |
"step": 7700
|
| 606 |
},
|
| 607 |
{
|
| 608 |
+
"epoch": 82.97,
|
| 609 |
+
"learning_rate": 2.1689189189189192e-05,
|
| 610 |
+
"loss": 0.8563,
|
| 611 |
"step": 7800
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 84.04,
|
| 615 |
+
"learning_rate": 2.033783783783784e-05,
|
| 616 |
+
"loss": 0.8655,
|
| 617 |
"step": 7900
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 85.11,
|
| 621 |
+
"learning_rate": 1.898648648648649e-05,
|
| 622 |
+
"loss": 0.8442,
|
| 623 |
"step": 8000
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 85.11,
|
| 627 |
+
"eval_loss": 0.2045026570558548,
|
| 628 |
+
"eval_runtime": 138.8412,
|
| 629 |
+
"eval_samples_per_second": 19.764,
|
| 630 |
+
"eval_steps_per_second": 2.47,
|
| 631 |
+
"eval_wer": 0.32657544277149786,
|
| 632 |
"step": 8000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 86.17,
|
| 636 |
+
"learning_rate": 1.7635135135135137e-05,
|
| 637 |
+
"loss": 0.8323,
|
| 638 |
"step": 8100
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 87.23,
|
| 642 |
+
"learning_rate": 1.6283783783783786e-05,
|
| 643 |
+
"loss": 0.8384,
|
| 644 |
"step": 8200
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 88.3,
|
| 648 |
+
"learning_rate": 1.4932432432432433e-05,
|
| 649 |
+
"loss": 0.8391,
|
| 650 |
"step": 8300
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 89.36,
|
| 654 |
+
"learning_rate": 1.3581081081081081e-05,
|
| 655 |
+
"loss": 0.8292,
|
| 656 |
"step": 8400
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 90.42,
|
| 660 |
+
"learning_rate": 1.222972972972973e-05,
|
| 661 |
+
"loss": 0.8335,
|
| 662 |
"step": 8500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 90.42,
|
| 666 |
+
"eval_loss": 0.20376762747764587,
|
| 667 |
+
"eval_runtime": 136.8377,
|
| 668 |
+
"eval_samples_per_second": 20.053,
|
| 669 |
+
"eval_steps_per_second": 2.507,
|
| 670 |
+
"eval_wer": 0.3240583954967736,
|
| 671 |
"step": 8500
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 91.49,
|
| 675 |
+
"learning_rate": 1.0878378378378378e-05,
|
| 676 |
+
"loss": 0.8314,
|
| 677 |
"step": 8600
|
| 678 |
},
|
| 679 |
{
|
| 680 |
+
"epoch": 92.55,
|
| 681 |
+
"learning_rate": 9.527027027027027e-06,
|
| 682 |
+
"loss": 0.8254,
|
| 683 |
"step": 8700
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 93.61,
|
| 687 |
+
"learning_rate": 8.175675675675675e-06,
|
| 688 |
+
"loss": 0.8231,
|
| 689 |
"step": 8800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
+
"epoch": 94.68,
|
| 693 |
+
"learning_rate": 6.8243243243243244e-06,
|
| 694 |
+
"loss": 0.8164,
|
| 695 |
"step": 8900
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 95.74,
|
| 699 |
+
"learning_rate": 5.472972972972974e-06,
|
| 700 |
+
"loss": 0.8288,
|
| 701 |
"step": 9000
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"epoch": 95.74,
|
| 705 |
+
"eval_loss": 0.20242640376091003,
|
| 706 |
+
"eval_runtime": 135.0305,
|
| 707 |
+
"eval_samples_per_second": 20.321,
|
| 708 |
+
"eval_steps_per_second": 2.54,
|
| 709 |
+
"eval_wer": 0.32799414214452427,
|
| 710 |
"step": 9000
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 96.8,
|
| 714 |
+
"learning_rate": 4.121621621621622e-06,
|
| 715 |
+
"loss": 0.816,
|
| 716 |
"step": 9100
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 97.87,
|
| 720 |
+
"learning_rate": 2.7702702702702708e-06,
|
| 721 |
+
"loss": 0.8163,
|
| 722 |
"step": 9200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 98.93,
|
| 726 |
+
"learning_rate": 1.418918918918919e-06,
|
| 727 |
+
"loss": 0.8126,
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 99.99,
|
| 732 |
+
"learning_rate": 6.756756756756757e-08,
|
| 733 |
+
"loss": 0.8084,
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 99.99,
|
| 738 |
+
"step": 9400,
|
| 739 |
+
"total_flos": 1.0839754269306731e+20,
|
| 740 |
+
"train_loss": 1.3946523244330225,
|
| 741 |
+
"train_runtime": 41836.5581,
|
| 742 |
+
"train_samples_per_second": 14.423,
|
| 743 |
+
"train_steps_per_second": 0.225
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 744 |
}
|
| 745 |
],
|
| 746 |
+
"max_steps": 9400,
|
| 747 |
"num_train_epochs": 100,
|
| 748 |
+
"total_flos": 1.0839754269306731e+20,
|
| 749 |
"trial_name": null,
|
| 750 |
"trial_params": null
|
| 751 |
}
|