End of training
Browse files
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 100.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 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": 100.0,
|
| 3 |
+
"eval_loss": 0.21627500653266907,
|
| 4 |
+
"eval_runtime": 133.3587,
|
| 5 |
+
"eval_samples": 2742,
|
| 6 |
+
"eval_samples_per_second": 20.561,
|
| 7 |
+
"eval_steps_per_second": 2.572,
|
| 8 |
+
"eval_wer": 0.32487632188081517,
|
| 9 |
+
"train_loss": 1.3660302423273476,
|
| 10 |
+
"train_runtime": 26867.6077,
|
| 11 |
+
"train_samples": 3292,
|
| 12 |
+
"train_samples_per_second": 12.253,
|
| 13 |
+
"train_steps_per_second": 0.383
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 100.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.21627500653266907,
|
| 4 |
+
"eval_runtime": 133.3587,
|
| 5 |
+
"eval_samples": 2742,
|
| 6 |
+
"eval_samples_per_second": 20.561,
|
| 7 |
+
"eval_steps_per_second": 2.572,
|
| 8 |
+
"eval_wer": 0.32487632188081517
|
| 9 |
}
|
runs/Feb03_08-05-51_job-699ba53c-fea9-4eb2-81af-a97f440eaa45/events.out.tfevents.1643902743.job-699ba53c-fea9-4eb2-81af-a97f440eaa45.626095.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ca1948e90401480b87dead8c3b414610af200d57c96040f29802725beeeac19
|
| 3 |
+
size 358
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 100.0,
|
| 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": 100.0,
|
| 3 |
+
"train_loss": 1.3660302423273476,
|
| 4 |
+
"train_runtime": 26867.6077,
|
| 5 |
+
"train_samples": 3292,
|
| 6 |
+
"train_samples_per_second": 12.253,
|
| 7 |
+
"train_steps_per_second": 0.383
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1486 +1,823 @@
|
|
| 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": 4.950000000000001e-06,
|
| 13 |
-
"loss": 13.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 1.
|
| 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":
|
| 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":
|
| 84 |
-
"eval_steps_per_second": 2.
|
| 85 |
-
"eval_wer":
|
| 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.
|
| 103 |
-
"loss": 1.
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"learning_rate": 6.
|
| 109 |
-
"loss": 1.
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch":
|
| 114 |
-
"learning_rate": 7.
|
| 115 |
-
"loss": 1.
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime":
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
-
"eval_wer": 0.
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"learning_rate": 7.
|
| 130 |
-
"loss": 1.
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch":
|
| 135 |
-
"learning_rate": 8.
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"learning_rate": 8.
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
-
"learning_rate": 9.
|
| 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":
|
| 161 |
-
"eval_samples_per_second":
|
| 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":
|
| 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": 9.
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"eval_loss": 0.
|
| 238 |
-
"eval_runtime":
|
| 239 |
-
"eval_samples_per_second":
|
| 240 |
-
"eval_steps_per_second": 2.
|
| 241 |
-
"eval_wer": 0.
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 1.
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 1.
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"learning_rate":
|
| 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":
|
| 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":
|
| 292 |
-
"loss": 1.
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 1.
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss": 1.
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"eval_loss": 0.
|
| 316 |
-
"eval_runtime":
|
| 317 |
-
"eval_samples_per_second":
|
| 318 |
-
"eval_steps_per_second": 2.
|
| 319 |
-
"eval_wer": 0.
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch":
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 1.
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"learning_rate":
|
| 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": 1.
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch":
|
| 348 |
-
"learning_rate":
|
| 349 |
-
"loss": 1.
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch":
|
| 354 |
-
"eval_loss": 0.
|
| 355 |
-
"eval_runtime":
|
| 356 |
-
"eval_samples_per_second":
|
| 357 |
-
"eval_steps_per_second": 2.
|
| 358 |
-
"eval_wer": 0.
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 1.
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch":
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 1.
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 1.
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 1.
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch":
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 1.
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"eval_loss": 0.
|
| 394 |
-
"eval_runtime":
|
| 395 |
-
"eval_samples_per_second":
|
| 396 |
-
"eval_steps_per_second": 2.
|
| 397 |
-
"eval_wer": 0.
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"learning_rate":
|
| 403 |
-
"loss":
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch":
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss":
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss":
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch":
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss":
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss":
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch":
|
| 432 |
-
"eval_loss": 0.
|
| 433 |
-
"eval_runtime":
|
| 434 |
-
"eval_samples_per_second":
|
| 435 |
-
"eval_steps_per_second": 2.
|
| 436 |
-
"eval_wer": 0.
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss":
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss":
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss":
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss":
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss":
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"eval_loss": 0.
|
| 472 |
-
"eval_runtime":
|
| 473 |
-
"eval_samples_per_second":
|
| 474 |
-
"eval_steps_per_second": 2.
|
| 475 |
-
"eval_wer": 0.
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss":
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss":
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"loss":
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss":
|
| 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":
|
| 544 |
-
"loss": 0.
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"eval_loss": 0.
|
| 550 |
-
"eval_runtime":
|
| 551 |
-
"eval_samples_per_second":
|
| 552 |
-
"eval_steps_per_second": 2.
|
| 553 |
-
"eval_wer": 0.
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch":
|
| 558 |
-
"learning_rate":
|
| 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":
|
| 591 |
-
"eval_steps_per_second":
|
| 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":
|
| 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":
|
| 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":
|
| 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":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 9500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"eval_loss": 0.
|
| 745 |
-
"eval_runtime":
|
| 746 |
-
"eval_samples_per_second":
|
| 747 |
-
"eval_steps_per_second": 2.
|
| 748 |
-
"eval_wer": 0.
|
| 749 |
"step": 9500
|
| 750 |
},
|
| 751 |
{
|
| 752 |
-
"epoch":
|
| 753 |
-
"learning_rate":
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 9600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 9700
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 9800
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 0.
|
| 773 |
"step": 9900
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 10000
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch":
|
| 783 |
-
"eval_loss": 0.
|
| 784 |
-
"eval_runtime":
|
| 785 |
-
"eval_samples_per_second":
|
| 786 |
-
"eval_steps_per_second": 2.
|
| 787 |
-
"eval_wer": 0.
|
| 788 |
"step": 10000
|
| 789 |
},
|
| 790 |
{
|
| 791 |
-
"epoch":
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss": 0.
|
| 794 |
"step": 10100
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch":
|
| 798 |
-
"learning_rate":
|
| 799 |
-
"loss": 0.
|
| 800 |
"step": 10200
|
| 801 |
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 54.78,
|
| 804 |
-
"learning_rate": 5.063690476190477e-05,
|
| 805 |
-
"loss": 0.8777,
|
| 806 |
-
"step": 10300
|
| 807 |
-
},
|
| 808 |
-
{
|
| 809 |
-
"epoch": 55.32,
|
| 810 |
-
"learning_rate": 5.0041666666666666e-05,
|
| 811 |
-
"loss": 0.8805,
|
| 812 |
-
"step": 10400
|
| 813 |
-
},
|
| 814 |
-
{
|
| 815 |
-
"epoch": 55.85,
|
| 816 |
-
"learning_rate": 4.944642857142857e-05,
|
| 817 |
-
"loss": 0.8792,
|
| 818 |
-
"step": 10500
|
| 819 |
-
},
|
| 820 |
-
{
|
| 821 |
-
"epoch": 55.85,
|
| 822 |
-
"eval_loss": 0.22211238741874695,
|
| 823 |
-
"eval_runtime": 149.6455,
|
| 824 |
-
"eval_samples_per_second": 18.337,
|
| 825 |
-
"eval_steps_per_second": 2.292,
|
| 826 |
-
"eval_wer": 0.3656125577776761,
|
| 827 |
-
"step": 10500
|
| 828 |
-
},
|
| 829 |
-
{
|
| 830 |
-
"epoch": 56.38,
|
| 831 |
-
"learning_rate": 4.885119047619048e-05,
|
| 832 |
-
"loss": 0.8805,
|
| 833 |
-
"step": 10600
|
| 834 |
-
},
|
| 835 |
-
{
|
| 836 |
-
"epoch": 56.91,
|
| 837 |
-
"learning_rate": 4.8255952380952386e-05,
|
| 838 |
-
"loss": 0.8653,
|
| 839 |
-
"step": 10700
|
| 840 |
-
},
|
| 841 |
-
{
|
| 842 |
-
"epoch": 57.45,
|
| 843 |
-
"learning_rate": 4.7660714285714284e-05,
|
| 844 |
-
"loss": 0.8749,
|
| 845 |
-
"step": 10800
|
| 846 |
-
},
|
| 847 |
-
{
|
| 848 |
-
"epoch": 57.97,
|
| 849 |
-
"learning_rate": 4.7065476190476195e-05,
|
| 850 |
-
"loss": 0.8713,
|
| 851 |
-
"step": 10900
|
| 852 |
-
},
|
| 853 |
-
{
|
| 854 |
-
"epoch": 58.51,
|
| 855 |
-
"learning_rate": 4.64702380952381e-05,
|
| 856 |
-
"loss": 0.8682,
|
| 857 |
-
"step": 11000
|
| 858 |
-
},
|
| 859 |
-
{
|
| 860 |
-
"epoch": 58.51,
|
| 861 |
-
"eval_loss": 0.22277939319610596,
|
| 862 |
-
"eval_runtime": 153.8771,
|
| 863 |
-
"eval_samples_per_second": 17.832,
|
| 864 |
-
"eval_steps_per_second": 2.229,
|
| 865 |
-
"eval_wer": 0.3322960047595076,
|
| 866 |
-
"step": 11000
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 59.04,
|
| 870 |
-
"learning_rate": 4.5875000000000004e-05,
|
| 871 |
-
"loss": 0.8556,
|
| 872 |
-
"step": 11100
|
| 873 |
-
},
|
| 874 |
-
{
|
| 875 |
-
"epoch": 59.57,
|
| 876 |
-
"learning_rate": 4.528571428571429e-05,
|
| 877 |
-
"loss": 0.8508,
|
| 878 |
-
"step": 11200
|
| 879 |
-
},
|
| 880 |
-
{
|
| 881 |
-
"epoch": 60.11,
|
| 882 |
-
"learning_rate": 4.469047619047619e-05,
|
| 883 |
-
"loss": 0.8594,
|
| 884 |
-
"step": 11300
|
| 885 |
-
},
|
| 886 |
-
{
|
| 887 |
-
"epoch": 60.64,
|
| 888 |
-
"learning_rate": 4.4095238095238096e-05,
|
| 889 |
-
"loss": 0.8441,
|
| 890 |
-
"step": 11400
|
| 891 |
-
},
|
| 892 |
-
{
|
| 893 |
-
"epoch": 61.17,
|
| 894 |
-
"learning_rate": 4.35e-05,
|
| 895 |
-
"loss": 0.8492,
|
| 896 |
-
"step": 11500
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 61.17,
|
| 900 |
-
"eval_loss": 0.2166604995727539,
|
| 901 |
-
"eval_runtime": 147.2191,
|
| 902 |
-
"eval_samples_per_second": 18.639,
|
| 903 |
-
"eval_steps_per_second": 2.33,
|
| 904 |
-
"eval_wer": 0.3446066541577045,
|
| 905 |
-
"step": 11500
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 61.7,
|
| 909 |
-
"learning_rate": 4.290476190476191e-05,
|
| 910 |
-
"loss": 0.8353,
|
| 911 |
-
"step": 11600
|
| 912 |
-
},
|
| 913 |
-
{
|
| 914 |
-
"epoch": 62.23,
|
| 915 |
-
"learning_rate": 4.230952380952381e-05,
|
| 916 |
-
"loss": 0.8407,
|
| 917 |
-
"step": 11700
|
| 918 |
-
},
|
| 919 |
-
{
|
| 920 |
-
"epoch": 62.76,
|
| 921 |
-
"learning_rate": 4.1714285714285714e-05,
|
| 922 |
-
"loss": 0.8335,
|
| 923 |
-
"step": 11800
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 63.3,
|
| 927 |
-
"learning_rate": 4.1119047619047625e-05,
|
| 928 |
-
"loss": 0.8596,
|
| 929 |
-
"step": 11900
|
| 930 |
-
},
|
| 931 |
-
{
|
| 932 |
-
"epoch": 63.83,
|
| 933 |
-
"learning_rate": 4.052380952380952e-05,
|
| 934 |
-
"loss": 0.8365,
|
| 935 |
-
"step": 12000
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 63.83,
|
| 939 |
-
"eval_loss": 0.21556589007377625,
|
| 940 |
-
"eval_runtime": 144.6076,
|
| 941 |
-
"eval_samples_per_second": 18.975,
|
| 942 |
-
"eval_steps_per_second": 2.372,
|
| 943 |
-
"eval_wer": 0.33211294677589126,
|
| 944 |
-
"step": 12000
|
| 945 |
-
},
|
| 946 |
-
{
|
| 947 |
-
"epoch": 64.36,
|
| 948 |
-
"learning_rate": 3.9928571428571434e-05,
|
| 949 |
-
"loss": 0.8412,
|
| 950 |
-
"step": 12100
|
| 951 |
-
},
|
| 952 |
-
{
|
| 953 |
-
"epoch": 64.89,
|
| 954 |
-
"learning_rate": 3.933333333333333e-05,
|
| 955 |
-
"loss": 0.8301,
|
| 956 |
-
"step": 12200
|
| 957 |
-
},
|
| 958 |
-
{
|
| 959 |
-
"epoch": 65.42,
|
| 960 |
-
"learning_rate": 3.873809523809524e-05,
|
| 961 |
-
"loss": 0.8283,
|
| 962 |
-
"step": 12300
|
| 963 |
-
},
|
| 964 |
-
{
|
| 965 |
-
"epoch": 65.95,
|
| 966 |
-
"learning_rate": 3.814285714285715e-05,
|
| 967 |
-
"loss": 0.8257,
|
| 968 |
-
"step": 12400
|
| 969 |
-
},
|
| 970 |
-
{
|
| 971 |
-
"epoch": 66.49,
|
| 972 |
-
"learning_rate": 3.7547619047619045e-05,
|
| 973 |
-
"loss": 0.8298,
|
| 974 |
-
"step": 12500
|
| 975 |
-
},
|
| 976 |
-
{
|
| 977 |
-
"epoch": 66.49,
|
| 978 |
-
"eval_loss": 0.21417230367660522,
|
| 979 |
-
"eval_runtime": 151.8253,
|
| 980 |
-
"eval_samples_per_second": 18.073,
|
| 981 |
-
"eval_steps_per_second": 2.259,
|
| 982 |
-
"eval_wer": 0.3400302045672967,
|
| 983 |
-
"step": 12500
|
| 984 |
-
},
|
| 985 |
-
{
|
| 986 |
-
"epoch": 67.02,
|
| 987 |
-
"learning_rate": 3.6952380952380956e-05,
|
| 988 |
-
"loss": 0.834,
|
| 989 |
-
"step": 12600
|
| 990 |
-
},
|
| 991 |
-
{
|
| 992 |
-
"epoch": 67.55,
|
| 993 |
-
"learning_rate": 3.6357142857142854e-05,
|
| 994 |
-
"loss": 0.8155,
|
| 995 |
-
"step": 12700
|
| 996 |
-
},
|
| 997 |
-
{
|
| 998 |
-
"epoch": 68.08,
|
| 999 |
-
"learning_rate": 3.5761904761904765e-05,
|
| 1000 |
-
"loss": 0.8157,
|
| 1001 |
-
"step": 12800
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 68.61,
|
| 1005 |
-
"learning_rate": 3.516666666666667e-05,
|
| 1006 |
-
"loss": 0.8027,
|
| 1007 |
-
"step": 12900
|
| 1008 |
-
},
|
| 1009 |
-
{
|
| 1010 |
-
"epoch": 69.15,
|
| 1011 |
-
"learning_rate": 3.4571428571428574e-05,
|
| 1012 |
-
"loss": 0.808,
|
| 1013 |
-
"step": 13000
|
| 1014 |
-
},
|
| 1015 |
-
{
|
| 1016 |
-
"epoch": 69.15,
|
| 1017 |
-
"eval_loss": 0.20793649554252625,
|
| 1018 |
-
"eval_runtime": 147.4919,
|
| 1019 |
-
"eval_samples_per_second": 18.604,
|
| 1020 |
-
"eval_steps_per_second": 2.326,
|
| 1021 |
-
"eval_wer": 0.31476820282824586,
|
| 1022 |
-
"step": 13000
|
| 1023 |
-
},
|
| 1024 |
-
{
|
| 1025 |
-
"epoch": 69.68,
|
| 1026 |
-
"learning_rate": 3.397619047619048e-05,
|
| 1027 |
-
"loss": 0.807,
|
| 1028 |
-
"step": 13100
|
| 1029 |
-
},
|
| 1030 |
-
{
|
| 1031 |
-
"epoch": 70.21,
|
| 1032 |
-
"learning_rate": 3.338095238095238e-05,
|
| 1033 |
-
"loss": 0.8164,
|
| 1034 |
-
"step": 13200
|
| 1035 |
-
},
|
| 1036 |
-
{
|
| 1037 |
-
"epoch": 70.74,
|
| 1038 |
-
"learning_rate": 3.278571428571429e-05,
|
| 1039 |
-
"loss": 0.7979,
|
| 1040 |
-
"step": 13300
|
| 1041 |
-
},
|
| 1042 |
-
{
|
| 1043 |
-
"epoch": 71.28,
|
| 1044 |
-
"learning_rate": 3.219047619047619e-05,
|
| 1045 |
-
"loss": 0.815,
|
| 1046 |
-
"step": 13400
|
| 1047 |
-
},
|
| 1048 |
-
{
|
| 1049 |
-
"epoch": 71.81,
|
| 1050 |
-
"learning_rate": 3.15952380952381e-05,
|
| 1051 |
-
"loss": 0.7999,
|
| 1052 |
-
"step": 13500
|
| 1053 |
-
},
|
| 1054 |
-
{
|
| 1055 |
-
"epoch": 71.81,
|
| 1056 |
-
"eval_loss": 0.21165262162685394,
|
| 1057 |
-
"eval_runtime": 145.5259,
|
| 1058 |
-
"eval_samples_per_second": 18.856,
|
| 1059 |
-
"eval_steps_per_second": 2.357,
|
| 1060 |
-
"eval_wer": 0.32254816713193907,
|
| 1061 |
-
"step": 13500
|
| 1062 |
-
},
|
| 1063 |
-
{
|
| 1064 |
-
"epoch": 72.34,
|
| 1065 |
-
"learning_rate": 3.1e-05,
|
| 1066 |
-
"loss": 0.8143,
|
| 1067 |
-
"step": 13600
|
| 1068 |
-
},
|
| 1069 |
-
{
|
| 1070 |
-
"epoch": 72.87,
|
| 1071 |
-
"learning_rate": 3.040476190476191e-05,
|
| 1072 |
-
"loss": 0.7952,
|
| 1073 |
-
"step": 13700
|
| 1074 |
-
},
|
| 1075 |
-
{
|
| 1076 |
-
"epoch": 73.4,
|
| 1077 |
-
"learning_rate": 2.980952380952381e-05,
|
| 1078 |
-
"loss": 0.8075,
|
| 1079 |
-
"step": 13800
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 73.93,
|
| 1083 |
-
"learning_rate": 2.9214285714285715e-05,
|
| 1084 |
-
"loss": 0.8021,
|
| 1085 |
-
"step": 13900
|
| 1086 |
-
},
|
| 1087 |
-
{
|
| 1088 |
-
"epoch": 74.47,
|
| 1089 |
-
"learning_rate": 2.8619047619047623e-05,
|
| 1090 |
-
"loss": 0.7871,
|
| 1091 |
-
"step": 14000
|
| 1092 |
-
},
|
| 1093 |
-
{
|
| 1094 |
-
"epoch": 74.47,
|
| 1095 |
-
"eval_loss": 0.2087966501712799,
|
| 1096 |
-
"eval_runtime": 154.8982,
|
| 1097 |
-
"eval_samples_per_second": 17.715,
|
| 1098 |
-
"eval_steps_per_second": 2.214,
|
| 1099 |
-
"eval_wer": 0.31742254359068234,
|
| 1100 |
-
"step": 14000
|
| 1101 |
-
},
|
| 1102 |
-
{
|
| 1103 |
-
"epoch": 75.0,
|
| 1104 |
-
"learning_rate": 2.8023809523809524e-05,
|
| 1105 |
-
"loss": 0.7898,
|
| 1106 |
-
"step": 14100
|
| 1107 |
-
},
|
| 1108 |
-
{
|
| 1109 |
-
"epoch": 75.53,
|
| 1110 |
-
"learning_rate": 2.742857142857143e-05,
|
| 1111 |
-
"loss": 0.7972,
|
| 1112 |
-
"step": 14200
|
| 1113 |
-
},
|
| 1114 |
-
{
|
| 1115 |
-
"epoch": 76.06,
|
| 1116 |
-
"learning_rate": 2.6833333333333333e-05,
|
| 1117 |
-
"loss": 0.801,
|
| 1118 |
-
"step": 14300
|
| 1119 |
-
},
|
| 1120 |
-
{
|
| 1121 |
-
"epoch": 76.59,
|
| 1122 |
-
"learning_rate": 2.623809523809524e-05,
|
| 1123 |
-
"loss": 0.786,
|
| 1124 |
-
"step": 14400
|
| 1125 |
-
},
|
| 1126 |
-
{
|
| 1127 |
-
"epoch": 77.13,
|
| 1128 |
-
"learning_rate": 2.5642857142857145e-05,
|
| 1129 |
-
"loss": 0.7858,
|
| 1130 |
-
"step": 14500
|
| 1131 |
-
},
|
| 1132 |
-
{
|
| 1133 |
-
"epoch": 77.13,
|
| 1134 |
-
"eval_loss": 0.2059505730867386,
|
| 1135 |
-
"eval_runtime": 147.0151,
|
| 1136 |
-
"eval_samples_per_second": 18.665,
|
| 1137 |
-
"eval_steps_per_second": 2.333,
|
| 1138 |
-
"eval_wer": 0.3008100315775022,
|
| 1139 |
-
"step": 14500
|
| 1140 |
-
},
|
| 1141 |
-
{
|
| 1142 |
-
"epoch": 77.66,
|
| 1143 |
-
"learning_rate": 2.5047619047619046e-05,
|
| 1144 |
-
"loss": 0.7753,
|
| 1145 |
-
"step": 14600
|
| 1146 |
-
},
|
| 1147 |
-
{
|
| 1148 |
-
"epoch": 78.19,
|
| 1149 |
-
"learning_rate": 2.4452380952380954e-05,
|
| 1150 |
-
"loss": 0.7794,
|
| 1151 |
-
"step": 14700
|
| 1152 |
-
},
|
| 1153 |
-
{
|
| 1154 |
-
"epoch": 78.72,
|
| 1155 |
-
"learning_rate": 2.385714285714286e-05,
|
| 1156 |
-
"loss": 0.7775,
|
| 1157 |
-
"step": 14800
|
| 1158 |
-
},
|
| 1159 |
-
{
|
| 1160 |
-
"epoch": 79.25,
|
| 1161 |
-
"learning_rate": 2.3261904761904763e-05,
|
| 1162 |
-
"loss": 0.7753,
|
| 1163 |
-
"step": 14900
|
| 1164 |
-
},
|
| 1165 |
-
{
|
| 1166 |
-
"epoch": 79.78,
|
| 1167 |
-
"learning_rate": 2.2666666666666668e-05,
|
| 1168 |
-
"loss": 0.7764,
|
| 1169 |
-
"step": 15000
|
| 1170 |
-
},
|
| 1171 |
-
{
|
| 1172 |
-
"epoch": 79.78,
|
| 1173 |
-
"eval_loss": 0.2128456085920334,
|
| 1174 |
-
"eval_runtime": 150.2974,
|
| 1175 |
-
"eval_samples_per_second": 18.257,
|
| 1176 |
-
"eval_steps_per_second": 2.282,
|
| 1177 |
-
"eval_wer": 0.3145851448446295,
|
| 1178 |
-
"step": 15000
|
| 1179 |
-
},
|
| 1180 |
-
{
|
| 1181 |
-
"epoch": 80.32,
|
| 1182 |
-
"learning_rate": 2.2071428571428572e-05,
|
| 1183 |
-
"loss": 0.7847,
|
| 1184 |
-
"step": 15100
|
| 1185 |
-
},
|
| 1186 |
-
{
|
| 1187 |
-
"epoch": 80.85,
|
| 1188 |
-
"learning_rate": 2.1476190476190477e-05,
|
| 1189 |
-
"loss": 0.774,
|
| 1190 |
-
"step": 15200
|
| 1191 |
-
},
|
| 1192 |
-
{
|
| 1193 |
-
"epoch": 81.38,
|
| 1194 |
-
"learning_rate": 2.0886904761904763e-05,
|
| 1195 |
-
"loss": 0.7739,
|
| 1196 |
-
"step": 15300
|
| 1197 |
-
},
|
| 1198 |
-
{
|
| 1199 |
-
"epoch": 81.91,
|
| 1200 |
-
"learning_rate": 2.0291666666666667e-05,
|
| 1201 |
-
"loss": 0.7579,
|
| 1202 |
-
"step": 15400
|
| 1203 |
-
},
|
| 1204 |
-
{
|
| 1205 |
-
"epoch": 82.45,
|
| 1206 |
-
"learning_rate": 1.9696428571428572e-05,
|
| 1207 |
-
"loss": 0.7684,
|
| 1208 |
-
"step": 15500
|
| 1209 |
-
},
|
| 1210 |
-
{
|
| 1211 |
-
"epoch": 82.45,
|
| 1212 |
-
"eval_loss": 0.20856936275959015,
|
| 1213 |
-
"eval_runtime": 149.2381,
|
| 1214 |
-
"eval_samples_per_second": 18.387,
|
| 1215 |
-
"eval_steps_per_second": 2.298,
|
| 1216 |
-
"eval_wer": 0.31005445975012585,
|
| 1217 |
-
"step": 15500
|
| 1218 |
-
},
|
| 1219 |
-
{
|
| 1220 |
-
"epoch": 82.97,
|
| 1221 |
-
"learning_rate": 1.9101190476190476e-05,
|
| 1222 |
-
"loss": 0.7584,
|
| 1223 |
-
"step": 15600
|
| 1224 |
-
},
|
| 1225 |
-
{
|
| 1226 |
-
"epoch": 83.51,
|
| 1227 |
-
"learning_rate": 1.850595238095238e-05,
|
| 1228 |
-
"loss": 0.7653,
|
| 1229 |
-
"step": 15700
|
| 1230 |
-
},
|
| 1231 |
-
{
|
| 1232 |
-
"epoch": 84.04,
|
| 1233 |
-
"learning_rate": 1.7910714285714285e-05,
|
| 1234 |
-
"loss": 0.7628,
|
| 1235 |
-
"step": 15800
|
| 1236 |
-
},
|
| 1237 |
-
{
|
| 1238 |
-
"epoch": 84.57,
|
| 1239 |
-
"learning_rate": 1.731547619047619e-05,
|
| 1240 |
-
"loss": 0.76,
|
| 1241 |
-
"step": 15900
|
| 1242 |
-
},
|
| 1243 |
-
{
|
| 1244 |
-
"epoch": 85.11,
|
| 1245 |
-
"learning_rate": 1.6720238095238098e-05,
|
| 1246 |
-
"loss": 0.7717,
|
| 1247 |
-
"step": 16000
|
| 1248 |
-
},
|
| 1249 |
-
{
|
| 1250 |
-
"epoch": 85.11,
|
| 1251 |
-
"eval_loss": 0.20475880801677704,
|
| 1252 |
-
"eval_runtime": 151.8607,
|
| 1253 |
-
"eval_samples_per_second": 18.069,
|
| 1254 |
-
"eval_steps_per_second": 2.259,
|
| 1255 |
-
"eval_wer": 0.3068509450368404,
|
| 1256 |
-
"step": 16000
|
| 1257 |
-
},
|
| 1258 |
-
{
|
| 1259 |
-
"epoch": 85.64,
|
| 1260 |
-
"learning_rate": 1.6125000000000002e-05,
|
| 1261 |
-
"loss": 0.7459,
|
| 1262 |
-
"step": 16100
|
| 1263 |
-
},
|
| 1264 |
-
{
|
| 1265 |
-
"epoch": 86.17,
|
| 1266 |
-
"learning_rate": 1.5529761904761907e-05,
|
| 1267 |
-
"loss": 0.7561,
|
| 1268 |
-
"step": 16200
|
| 1269 |
-
},
|
| 1270 |
-
{
|
| 1271 |
-
"epoch": 86.7,
|
| 1272 |
-
"learning_rate": 1.493452380952381e-05,
|
| 1273 |
-
"loss": 0.7457,
|
| 1274 |
-
"step": 16300
|
| 1275 |
-
},
|
| 1276 |
-
{
|
| 1277 |
-
"epoch": 87.23,
|
| 1278 |
-
"learning_rate": 1.4339285714285716e-05,
|
| 1279 |
-
"loss": 0.7599,
|
| 1280 |
-
"step": 16400
|
| 1281 |
-
},
|
| 1282 |
-
{
|
| 1283 |
-
"epoch": 87.76,
|
| 1284 |
-
"learning_rate": 1.374404761904762e-05,
|
| 1285 |
-
"loss": 0.7435,
|
| 1286 |
-
"step": 16500
|
| 1287 |
-
},
|
| 1288 |
-
{
|
| 1289 |
-
"epoch": 87.76,
|
| 1290 |
-
"eval_loss": 0.20274706184864044,
|
| 1291 |
-
"eval_runtime": 146.6164,
|
| 1292 |
-
"eval_samples_per_second": 18.716,
|
| 1293 |
-
"eval_steps_per_second": 2.339,
|
| 1294 |
-
"eval_wer": 0.3054780101597181,
|
| 1295 |
-
"step": 16500
|
| 1296 |
-
},
|
| 1297 |
-
{
|
| 1298 |
-
"epoch": 88.3,
|
| 1299 |
-
"learning_rate": 1.3148809523809525e-05,
|
| 1300 |
-
"loss": 0.7519,
|
| 1301 |
-
"step": 16600
|
| 1302 |
-
},
|
| 1303 |
-
{
|
| 1304 |
-
"epoch": 88.83,
|
| 1305 |
-
"learning_rate": 1.255357142857143e-05,
|
| 1306 |
-
"loss": 0.7483,
|
| 1307 |
-
"step": 16700
|
| 1308 |
-
},
|
| 1309 |
-
{
|
| 1310 |
-
"epoch": 89.36,
|
| 1311 |
-
"learning_rate": 1.1958333333333334e-05,
|
| 1312 |
-
"loss": 0.7483,
|
| 1313 |
-
"step": 16800
|
| 1314 |
-
},
|
| 1315 |
-
{
|
| 1316 |
-
"epoch": 89.89,
|
| 1317 |
-
"learning_rate": 1.1363095238095238e-05,
|
| 1318 |
-
"loss": 0.7353,
|
| 1319 |
-
"step": 16900
|
| 1320 |
-
},
|
| 1321 |
-
{
|
| 1322 |
-
"epoch": 90.42,
|
| 1323 |
-
"learning_rate": 1.0767857142857143e-05,
|
| 1324 |
-
"loss": 0.7378,
|
| 1325 |
-
"step": 17000
|
| 1326 |
-
},
|
| 1327 |
-
{
|
| 1328 |
-
"epoch": 90.42,
|
| 1329 |
-
"eval_loss": 0.20591045916080475,
|
| 1330 |
-
"eval_runtime": 153.244,
|
| 1331 |
-
"eval_samples_per_second": 17.906,
|
| 1332 |
-
"eval_steps_per_second": 2.238,
|
| 1333 |
-
"eval_wer": 0.29925403871676354,
|
| 1334 |
-
"step": 17000
|
| 1335 |
-
},
|
| 1336 |
-
{
|
| 1337 |
-
"epoch": 90.95,
|
| 1338 |
-
"learning_rate": 1.0172619047619047e-05,
|
| 1339 |
-
"loss": 0.7383,
|
| 1340 |
-
"step": 17100
|
| 1341 |
-
},
|
| 1342 |
-
{
|
| 1343 |
-
"epoch": 91.49,
|
| 1344 |
-
"learning_rate": 9.577380952380953e-06,
|
| 1345 |
-
"loss": 0.7466,
|
| 1346 |
-
"step": 17200
|
| 1347 |
-
},
|
| 1348 |
-
{
|
| 1349 |
-
"epoch": 92.02,
|
| 1350 |
-
"learning_rate": 8.982142857142856e-06,
|
| 1351 |
-
"loss": 0.7377,
|
| 1352 |
-
"step": 17300
|
| 1353 |
-
},
|
| 1354 |
-
{
|
| 1355 |
-
"epoch": 92.55,
|
| 1356 |
-
"learning_rate": 8.386904761904762e-06,
|
| 1357 |
-
"loss": 0.7369,
|
| 1358 |
-
"step": 17400
|
| 1359 |
-
},
|
| 1360 |
-
{
|
| 1361 |
-
"epoch": 93.08,
|
| 1362 |
-
"learning_rate": 7.791666666666667e-06,
|
| 1363 |
-
"loss": 0.7406,
|
| 1364 |
-
"step": 17500
|
| 1365 |
-
},
|
| 1366 |
-
{
|
| 1367 |
-
"epoch": 93.08,
|
| 1368 |
-
"eval_loss": 0.2040216028690338,
|
| 1369 |
-
"eval_runtime": 148.029,
|
| 1370 |
-
"eval_samples_per_second": 18.537,
|
| 1371 |
-
"eval_steps_per_second": 2.317,
|
| 1372 |
-
"eval_wer": 0.2966454624502311,
|
| 1373 |
-
"step": 17500
|
| 1374 |
-
},
|
| 1375 |
-
{
|
| 1376 |
-
"epoch": 93.61,
|
| 1377 |
-
"learning_rate": 7.2023809523809524e-06,
|
| 1378 |
-
"loss": 0.7348,
|
| 1379 |
-
"step": 17600
|
| 1380 |
-
},
|
| 1381 |
-
{
|
| 1382 |
-
"epoch": 94.15,
|
| 1383 |
-
"learning_rate": 6.613095238095239e-06,
|
| 1384 |
-
"loss": 0.7356,
|
| 1385 |
-
"step": 17700
|
| 1386 |
-
},
|
| 1387 |
-
{
|
| 1388 |
-
"epoch": 94.68,
|
| 1389 |
-
"learning_rate": 6.017857142857143e-06,
|
| 1390 |
-
"loss": 0.7305,
|
| 1391 |
-
"step": 17800
|
| 1392 |
-
},
|
| 1393 |
-
{
|
| 1394 |
-
"epoch": 95.21,
|
| 1395 |
-
"learning_rate": 5.422619047619048e-06,
|
| 1396 |
-
"loss": 0.741,
|
| 1397 |
-
"step": 17900
|
| 1398 |
-
},
|
| 1399 |
-
{
|
| 1400 |
-
"epoch": 95.74,
|
| 1401 |
-
"learning_rate": 4.827380952380952e-06,
|
| 1402 |
-
"loss": 0.7361,
|
| 1403 |
-
"step": 18000
|
| 1404 |
-
},
|
| 1405 |
-
{
|
| 1406 |
-
"epoch": 95.74,
|
| 1407 |
-
"eval_loss": 0.20563913881778717,
|
| 1408 |
-
"eval_runtime": 145.4837,
|
| 1409 |
-
"eval_samples_per_second": 18.861,
|
| 1410 |
-
"eval_steps_per_second": 2.358,
|
| 1411 |
-
"eval_wer": 0.30003203514713284,
|
| 1412 |
-
"step": 18000
|
| 1413 |
-
},
|
| 1414 |
-
{
|
| 1415 |
-
"epoch": 96.28,
|
| 1416 |
-
"learning_rate": 4.232142857142858e-06,
|
| 1417 |
-
"loss": 0.7359,
|
| 1418 |
-
"step": 18100
|
| 1419 |
-
},
|
| 1420 |
-
{
|
| 1421 |
-
"epoch": 96.81,
|
| 1422 |
-
"learning_rate": 3.636904761904762e-06,
|
| 1423 |
-
"loss": 0.7246,
|
| 1424 |
-
"step": 18200
|
| 1425 |
-
},
|
| 1426 |
-
{
|
| 1427 |
-
"epoch": 97.34,
|
| 1428 |
-
"learning_rate": 3.041666666666667e-06,
|
| 1429 |
-
"loss": 0.7353,
|
| 1430 |
-
"step": 18300
|
| 1431 |
-
},
|
| 1432 |
-
{
|
| 1433 |
-
"epoch": 97.87,
|
| 1434 |
-
"learning_rate": 2.4464285714285715e-06,
|
| 1435 |
-
"loss": 0.7305,
|
| 1436 |
-
"step": 18400
|
| 1437 |
-
},
|
| 1438 |
-
{
|
| 1439 |
-
"epoch": 98.4,
|
| 1440 |
-
"learning_rate": 1.8511904761904762e-06,
|
| 1441 |
-
"loss": 0.7379,
|
| 1442 |
-
"step": 18500
|
| 1443 |
-
},
|
| 1444 |
-
{
|
| 1445 |
-
"epoch": 98.4,
|
| 1446 |
-
"eval_loss": 0.20308499038219452,
|
| 1447 |
-
"eval_runtime": 146.2002,
|
| 1448 |
-
"eval_samples_per_second": 18.769,
|
| 1449 |
-
"eval_steps_per_second": 2.346,
|
| 1450 |
-
"eval_wer": 0.29756075236831264,
|
| 1451 |
-
"step": 18500
|
| 1452 |
-
},
|
| 1453 |
-
{
|
| 1454 |
-
"epoch": 98.93,
|
| 1455 |
-
"learning_rate": 1.2559523809523812e-06,
|
| 1456 |
-
"loss": 0.724,
|
| 1457 |
-
"step": 18600
|
| 1458 |
-
},
|
| 1459 |
-
{
|
| 1460 |
-
"epoch": 99.47,
|
| 1461 |
-
"learning_rate": 6.607142857142858e-07,
|
| 1462 |
-
"loss": 0.7339,
|
| 1463 |
-
"step": 18700
|
| 1464 |
-
},
|
| 1465 |
{
|
| 1466 |
"epoch": 100.0,
|
| 1467 |
-
"learning_rate":
|
| 1468 |
-
"loss": 0.
|
| 1469 |
-
"step":
|
| 1470 |
},
|
| 1471 |
{
|
| 1472 |
"epoch": 100.0,
|
| 1473 |
-
"step":
|
| 1474 |
-
"total_flos":
|
| 1475 |
-
"train_loss": 1.
|
| 1476 |
-
"train_runtime":
|
| 1477 |
-
"train_samples_per_second":
|
| 1478 |
-
"train_steps_per_second": 0.
|
| 1479 |
}
|
| 1480 |
],
|
| 1481 |
-
"max_steps":
|
| 1482 |
"num_train_epochs": 100,
|
| 1483 |
-
"total_flos":
|
| 1484 |
"trial_name": null,
|
| 1485 |
"trial_params": null
|
| 1486 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 10300,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.97,
|
| 12 |
"learning_rate": 4.950000000000001e-06,
|
| 13 |
+
"loss": 13.4586,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 1.94,
|
| 18 |
"learning_rate": 9.950000000000001e-06,
|
| 19 |
+
"loss": 5.8722,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 2.91,
|
| 24 |
"learning_rate": 1.4950000000000001e-05,
|
| 25 |
+
"loss": 4.0954,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 3.88,
|
| 30 |
"learning_rate": 1.995e-05,
|
| 31 |
+
"loss": 3.572,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 4.85,
|
| 36 |
"learning_rate": 2.495e-05,
|
| 37 |
+
"loss": 3.2914,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 4.85,
|
| 42 |
+
"eval_loss": 3.2282841205596924,
|
| 43 |
+
"eval_runtime": 135.7815,
|
| 44 |
+
"eval_samples_per_second": 20.194,
|
| 45 |
+
"eval_steps_per_second": 2.526,
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
+
"epoch": 5.83,
|
| 51 |
"learning_rate": 2.995e-05,
|
| 52 |
+
"loss": 3.2087,
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 6.8,
|
| 57 |
"learning_rate": 3.495e-05,
|
| 58 |
+
"loss": 3.1802,
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 7.77,
|
| 63 |
"learning_rate": 3.995e-05,
|
| 64 |
+
"loss": 3.1307,
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 8.74,
|
| 69 |
"learning_rate": 4.495e-05,
|
| 70 |
+
"loss": 3.0922,
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
+
"epoch": 9.71,
|
| 75 |
"learning_rate": 4.995e-05,
|
| 76 |
+
"loss": 3.0068,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 9.71,
|
| 81 |
+
"eval_loss": 2.7939300537109375,
|
| 82 |
+
"eval_runtime": 134.6432,
|
| 83 |
+
"eval_samples_per_second": 20.365,
|
| 84 |
+
"eval_steps_per_second": 2.547,
|
| 85 |
+
"eval_wer": 0.997957609040984,
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 10.68,
|
| 90 |
"learning_rate": 5.495e-05,
|
| 91 |
+
"loss": 2.5073,
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 11.65,
|
| 96 |
"learning_rate": 5.995000000000001e-05,
|
| 97 |
+
"loss": 1.8326,
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 12.62,
|
| 102 |
+
"learning_rate": 6.494999999999999e-05,
|
| 103 |
+
"loss": 1.598,
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 13.59,
|
| 108 |
+
"learning_rate": 6.995e-05,
|
| 109 |
+
"loss": 1.5016,
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 14.56,
|
| 114 |
+
"learning_rate": 7.495e-05,
|
| 115 |
+
"loss": 1.4306,
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 14.56,
|
| 120 |
+
"eval_loss": 0.48574715852737427,
|
| 121 |
+
"eval_runtime": 133.5239,
|
| 122 |
+
"eval_samples_per_second": 20.536,
|
| 123 |
+
"eval_steps_per_second": 2.569,
|
| 124 |
+
"eval_wer": 0.6313711251304861,
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 15.53,
|
| 129 |
+
"learning_rate": 7.995e-05,
|
| 130 |
+
"loss": 1.3756,
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 16.5,
|
| 135 |
+
"learning_rate": 8.495e-05,
|
| 136 |
+
"loss": 1.3583,
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 17.48,
|
| 141 |
+
"learning_rate": 8.995e-05,
|
| 142 |
+
"loss": 1.3058,
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 18.45,
|
| 147 |
+
"learning_rate": 9.495e-05,
|
| 148 |
+
"loss": 1.2949,
|
| 149 |
"step": 1900
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 19.42,
|
| 153 |
+
"learning_rate": 9.995e-05,
|
| 154 |
+
"loss": 1.2831,
|
| 155 |
"step": 2000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 19.42,
|
| 159 |
+
"eval_loss": 0.3678707182407379,
|
| 160 |
+
"eval_runtime": 134.1491,
|
| 161 |
+
"eval_samples_per_second": 20.44,
|
| 162 |
+
"eval_steps_per_second": 2.557,
|
| 163 |
+
"eval_wer": 0.6065901148277584,
|
| 164 |
"step": 2000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 20.39,
|
| 168 |
+
"learning_rate": 9.880722891566265e-05,
|
| 169 |
+
"loss": 1.2725,
|
| 170 |
"step": 2100
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 21.36,
|
| 174 |
+
"learning_rate": 9.76144578313253e-05,
|
| 175 |
+
"loss": 1.2436,
|
| 176 |
"step": 2200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 22.33,
|
| 180 |
+
"learning_rate": 9.640963855421687e-05,
|
| 181 |
+
"loss": 1.2363,
|
| 182 |
"step": 2300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 23.3,
|
| 186 |
+
"learning_rate": 9.521686746987952e-05,
|
| 187 |
+
"loss": 1.2243,
|
| 188 |
"step": 2400
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 24.27,
|
| 192 |
+
"learning_rate": 9.402409638554217e-05,
|
| 193 |
+
"loss": 1.2065,
|
| 194 |
"step": 2500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 24.27,
|
| 198 |
+
"eval_loss": 0.33028003573417664,
|
| 199 |
+
"eval_runtime": 134.2277,
|
| 200 |
+
"eval_samples_per_second": 20.428,
|
| 201 |
+
"eval_steps_per_second": 2.555,
|
| 202 |
+
"eval_wer": 0.5559842055099169,
|
| 203 |
"step": 2500
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 25.24,
|
| 207 |
+
"learning_rate": 9.281927710843374e-05,
|
| 208 |
+
"loss": 1.192,
|
| 209 |
"step": 2600
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 26.21,
|
| 213 |
+
"learning_rate": 9.161445783132531e-05,
|
| 214 |
+
"loss": 1.1816,
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 27.18,
|
| 219 |
+
"learning_rate": 9.040963855421686e-05,
|
| 220 |
+
"loss": 1.1869,
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 28.16,
|
| 225 |
+
"learning_rate": 8.920481927710844e-05,
|
| 226 |
+
"loss": 1.1728,
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 29.13,
|
| 231 |
+
"learning_rate": 8.800000000000001e-05,
|
| 232 |
+
"loss": 1.1449,
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 29.13,
|
| 237 |
+
"eval_loss": 0.3007894456386566,
|
| 238 |
+
"eval_runtime": 133.5503,
|
| 239 |
+
"eval_samples_per_second": 20.532,
|
| 240 |
+
"eval_steps_per_second": 2.568,
|
| 241 |
+
"eval_wer": 0.46902373712159035,
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 30.1,
|
| 246 |
+
"learning_rate": 8.679518072289157e-05,
|
| 247 |
+
"loss": 1.1408,
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 31.07,
|
| 252 |
+
"learning_rate": 8.559036144578315e-05,
|
| 253 |
+
"loss": 1.1319,
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 32.04,
|
| 258 |
+
"learning_rate": 8.43855421686747e-05,
|
| 259 |
+
"loss": 1.1178,
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 33.01,
|
| 264 |
+
"learning_rate": 8.318072289156627e-05,
|
| 265 |
+
"loss": 1.1122,
|
| 266 |
"step": 3400
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 33.98,
|
| 270 |
+
"learning_rate": 8.197590361445784e-05,
|
| 271 |
+
"loss": 1.0926,
|
| 272 |
"step": 3500
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 33.98,
|
| 276 |
+
"eval_loss": 0.28173714876174927,
|
| 277 |
+
"eval_runtime": 132.429,
|
| 278 |
+
"eval_samples_per_second": 20.705,
|
| 279 |
+
"eval_steps_per_second": 2.59,
|
| 280 |
+
"eval_wer": 0.4618980619979122,
|
| 281 |
"step": 3500
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 34.95,
|
| 285 |
+
"learning_rate": 8.07710843373494e-05,
|
| 286 |
+
"loss": 1.0935,
|
| 287 |
"step": 3600
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 35.92,
|
| 291 |
+
"learning_rate": 7.956626506024096e-05,
|
| 292 |
+
"loss": 1.0815,
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 36.89,
|
| 297 |
+
"learning_rate": 7.836144578313254e-05,
|
| 298 |
+
"loss": 1.0856,
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 37.86,
|
| 303 |
+
"learning_rate": 7.71566265060241e-05,
|
| 304 |
+
"loss": 1.0732,
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 38.83,
|
| 309 |
+
"learning_rate": 7.595180722891566e-05,
|
| 310 |
+
"loss": 1.0635,
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 38.83,
|
| 315 |
+
"eval_loss": 0.2665168046951294,
|
| 316 |
+
"eval_runtime": 133.7977,
|
| 317 |
+
"eval_samples_per_second": 20.494,
|
| 318 |
+
"eval_steps_per_second": 2.564,
|
| 319 |
+
"eval_wer": 0.4391140561884446,
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 39.81,
|
| 324 |
+
"learning_rate": 7.474698795180723e-05,
|
| 325 |
+
"loss": 1.0614,
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 40.78,
|
| 330 |
+
"learning_rate": 7.35421686746988e-05,
|
| 331 |
+
"loss": 1.0457,
|
| 332 |
"step": 4200
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 41.75,
|
| 336 |
+
"learning_rate": 7.233734939759036e-05,
|
| 337 |
+
"loss": 1.039,
|
| 338 |
"step": 4300
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 42.72,
|
| 342 |
+
"learning_rate": 7.113253012048193e-05,
|
| 343 |
+
"loss": 1.0151,
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 43.69,
|
| 348 |
+
"learning_rate": 6.99277108433735e-05,
|
| 349 |
+
"loss": 1.029,
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 43.69,
|
| 354 |
+
"eval_loss": 0.26156488060951233,
|
| 355 |
+
"eval_runtime": 133.7699,
|
| 356 |
+
"eval_samples_per_second": 20.498,
|
| 357 |
+
"eval_steps_per_second": 2.564,
|
| 358 |
+
"eval_wer": 0.4175100984886307,
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 44.66,
|
| 363 |
+
"learning_rate": 6.873493975903614e-05,
|
| 364 |
+
"loss": 1.0254,
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 45.63,
|
| 369 |
+
"learning_rate": 6.753012048192771e-05,
|
| 370 |
+
"loss": 1.0328,
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 46.6,
|
| 375 |
+
"learning_rate": 6.632530120481928e-05,
|
| 376 |
+
"loss": 1.022,
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 47.57,
|
| 381 |
+
"learning_rate": 6.512048192771085e-05,
|
| 382 |
+
"loss": 1.0021,
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 48.54,
|
| 387 |
+
"learning_rate": 6.391566265060241e-05,
|
| 388 |
+
"loss": 1.0064,
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 48.54,
|
| 393 |
+
"eval_loss": 0.24684669077396393,
|
| 394 |
+
"eval_runtime": 133.5,
|
| 395 |
+
"eval_samples_per_second": 20.539,
|
| 396 |
+
"eval_steps_per_second": 2.569,
|
| 397 |
+
"eval_wer": 0.4051195933372668,
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 49.51,
|
| 402 |
+
"learning_rate": 6.271084337349398e-05,
|
| 403 |
+
"loss": 0.9791,
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 50.49,
|
| 408 |
+
"learning_rate": 6.150602409638555e-05,
|
| 409 |
+
"loss": 0.9722,
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 51.46,
|
| 414 |
+
"learning_rate": 6.030120481927711e-05,
|
| 415 |
+
"loss": 0.9815,
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 52.43,
|
| 420 |
+
"learning_rate": 5.909638554216868e-05,
|
| 421 |
+
"loss": 0.9633,
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 53.4,
|
| 426 |
+
"learning_rate": 5.789156626506025e-05,
|
| 427 |
+
"loss": 0.9659,
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 53.4,
|
| 432 |
+
"eval_loss": 0.2394031286239624,
|
| 433 |
+
"eval_runtime": 133.1725,
|
| 434 |
+
"eval_samples_per_second": 20.59,
|
| 435 |
+
"eval_steps_per_second": 2.576,
|
| 436 |
+
"eval_wer": 0.38596650478827216,
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 54.37,
|
| 441 |
+
"learning_rate": 5.668674698795181e-05,
|
| 442 |
+
"loss": 0.9544,
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 55.34,
|
| 447 |
+
"learning_rate": 5.5481927710843374e-05,
|
| 448 |
+
"loss": 0.9581,
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 56.31,
|
| 453 |
+
"learning_rate": 5.427710843373495e-05,
|
| 454 |
+
"loss": 0.9437,
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 57.28,
|
| 459 |
+
"learning_rate": 5.307228915662651e-05,
|
| 460 |
+
"loss": 0.9378,
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 58.25,
|
| 465 |
+
"learning_rate": 5.186746987951807e-05,
|
| 466 |
+
"loss": 0.9254,
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 58.25,
|
| 471 |
+
"eval_loss": 0.2373155653476715,
|
| 472 |
+
"eval_runtime": 133.3175,
|
| 473 |
+
"eval_samples_per_second": 20.567,
|
| 474 |
+
"eval_steps_per_second": 2.573,
|
| 475 |
+
"eval_wer": 0.3688558071982935,
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 59.22,
|
| 480 |
+
"learning_rate": 5.0662650602409644e-05,
|
| 481 |
+
"loss": 0.9321,
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 60.19,
|
| 486 |
+
"learning_rate": 4.9457831325301205e-05,
|
| 487 |
+
"loss": 0.9122,
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 61.17,
|
| 492 |
+
"learning_rate": 4.825301204819277e-05,
|
| 493 |
+
"loss": 0.9148,
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 62.14,
|
| 498 |
+
"learning_rate": 4.704819277108434e-05,
|
| 499 |
+
"loss": 0.9177,
|
| 500 |
"step": 6400
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 63.11,
|
| 504 |
+
"learning_rate": 4.584337349397591e-05,
|
| 505 |
+
"loss": 0.9209,
|
| 506 |
"step": 6500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 63.11,
|
| 510 |
+
"eval_loss": 0.23466718196868896,
|
| 511 |
+
"eval_runtime": 134.3014,
|
| 512 |
+
"eval_samples_per_second": 20.417,
|
| 513 |
+
"eval_steps_per_second": 2.554,
|
| 514 |
+
"eval_wer": 0.367040348568057,
|
| 515 |
"step": 6500
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 64.08,
|
| 519 |
+
"learning_rate": 4.4638554216867476e-05,
|
| 520 |
+
"loss": 0.8981,
|
| 521 |
"step": 6600
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 65.05,
|
| 525 |
+
"learning_rate": 4.344578313253012e-05,
|
| 526 |
+
"loss": 0.8927,
|
| 527 |
"step": 6700
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 66.02,
|
| 531 |
+
"learning_rate": 4.224096385542169e-05,
|
| 532 |
+
"loss": 0.8986,
|
| 533 |
"step": 6800
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 66.99,
|
| 537 |
+
"learning_rate": 4.1036144578313255e-05,
|
| 538 |
+
"loss": 0.8867,
|
| 539 |
"step": 6900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 67.96,
|
| 543 |
+
"learning_rate": 3.983132530120482e-05,
|
| 544 |
+
"loss": 0.889,
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 67.96,
|
| 549 |
+
"eval_loss": 0.22911565005779266,
|
| 550 |
+
"eval_runtime": 133.5899,
|
| 551 |
+
"eval_samples_per_second": 20.526,
|
| 552 |
+
"eval_steps_per_second": 2.568,
|
| 553 |
+
"eval_wer": 0.36871964780102573,
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 68.93,
|
| 558 |
+
"learning_rate": 3.862650602409639e-05,
|
| 559 |
+
"loss": 0.885,
|
| 560 |
"step": 7100
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 69.9,
|
| 564 |
+
"learning_rate": 3.742168674698796e-05,
|
| 565 |
+
"loss": 0.8772,
|
| 566 |
"step": 7200
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 70.87,
|
| 570 |
+
"learning_rate": 3.62289156626506e-05,
|
| 571 |
+
"loss": 0.8798,
|
| 572 |
"step": 7300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 71.84,
|
| 576 |
+
"learning_rate": 3.502409638554217e-05,
|
| 577 |
+
"loss": 0.8808,
|
| 578 |
"step": 7400
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 72.82,
|
| 582 |
+
"learning_rate": 3.3819277108433736e-05,
|
| 583 |
+
"loss": 0.8859,
|
| 584 |
"step": 7500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 72.82,
|
| 588 |
+
"eval_loss": 0.22717151045799255,
|
| 589 |
+
"eval_runtime": 134.7148,
|
| 590 |
+
"eval_samples_per_second": 20.354,
|
| 591 |
+
"eval_steps_per_second": 2.546,
|
| 592 |
+
"eval_wer": 0.3615939726773476,
|
| 593 |
"step": 7500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 73.79,
|
| 597 |
+
"learning_rate": 3.2614457831325304e-05,
|
| 598 |
+
"loss": 0.8713,
|
| 599 |
"step": 7600
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 74.76,
|
| 603 |
+
"learning_rate": 3.140963855421687e-05,
|
| 604 |
+
"loss": 0.8734,
|
| 605 |
"step": 7700
|
| 606 |
},
|
| 607 |
{
|
| 608 |
+
"epoch": 75.73,
|
| 609 |
+
"learning_rate": 3.0204819277108436e-05,
|
| 610 |
+
"loss": 0.8565,
|
| 611 |
"step": 7800
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 76.7,
|
| 615 |
+
"learning_rate": 2.9e-05,
|
| 616 |
+
"loss": 0.8492,
|
| 617 |
"step": 7900
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 77.67,
|
| 621 |
+
"learning_rate": 2.7795180722891568e-05,
|
| 622 |
+
"loss": 0.8441,
|
| 623 |
"step": 8000
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 77.67,
|
| 627 |
+
"eval_loss": 0.22322185337543488,
|
| 628 |
+
"eval_runtime": 134.4634,
|
| 629 |
+
"eval_samples_per_second": 20.392,
|
| 630 |
+
"eval_steps_per_second": 2.551,
|
| 631 |
+
"eval_wer": 0.35383288703308674,
|
| 632 |
"step": 8000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 78.64,
|
| 636 |
+
"learning_rate": 2.6590361445783136e-05,
|
| 637 |
+
"loss": 0.8516,
|
| 638 |
"step": 8100
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 79.61,
|
| 642 |
+
"learning_rate": 2.5385542168674696e-05,
|
| 643 |
+
"loss": 0.8451,
|
| 644 |
"step": 8200
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 80.58,
|
| 648 |
+
"learning_rate": 2.4180722891566264e-05,
|
| 649 |
+
"loss": 0.8346,
|
| 650 |
"step": 8300
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 81.55,
|
| 654 |
+
"learning_rate": 2.2975903614457832e-05,
|
| 655 |
+
"loss": 0.8378,
|
| 656 |
"step": 8400
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 82.52,
|
| 660 |
+
"learning_rate": 2.17710843373494e-05,
|
| 661 |
+
"loss": 0.8284,
|
| 662 |
"step": 8500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 82.52,
|
| 666 |
+
"eval_loss": 0.22235004603862762,
|
| 667 |
+
"eval_runtime": 133.9778,
|
| 668 |
+
"eval_samples_per_second": 20.466,
|
| 669 |
+
"eval_steps_per_second": 2.56,
|
| 670 |
+
"eval_wer": 0.33817455634729726,
|
| 671 |
"step": 8500
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 83.5,
|
| 675 |
+
"learning_rate": 2.0566265060240967e-05,
|
| 676 |
+
"loss": 0.8269,
|
| 677 |
"step": 8600
|
| 678 |
},
|
| 679 |
{
|
| 680 |
+
"epoch": 84.47,
|
| 681 |
+
"learning_rate": 1.936144578313253e-05,
|
| 682 |
+
"loss": 0.8186,
|
| 683 |
"step": 8700
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 85.44,
|
| 687 |
+
"learning_rate": 1.8156626506024096e-05,
|
| 688 |
+
"loss": 0.8243,
|
| 689 |
"step": 8800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
+
"epoch": 86.41,
|
| 693 |
+
"learning_rate": 1.6951807228915663e-05,
|
| 694 |
+
"loss": 0.8279,
|
| 695 |
"step": 8900
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 87.38,
|
| 699 |
+
"learning_rate": 1.574698795180723e-05,
|
| 700 |
+
"loss": 0.8142,
|
| 701 |
"step": 9000
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"epoch": 87.38,
|
| 705 |
+
"eval_loss": 0.2192818820476532,
|
| 706 |
+
"eval_runtime": 132.2621,
|
| 707 |
+
"eval_samples_per_second": 20.732,
|
| 708 |
+
"eval_steps_per_second": 2.593,
|
| 709 |
+
"eval_wer": 0.33104888122361914,
|
| 710 |
"step": 9000
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 88.35,
|
| 714 |
+
"learning_rate": 1.4542168674698795e-05,
|
| 715 |
+
"loss": 0.8071,
|
| 716 |
"step": 9100
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 89.32,
|
| 720 |
+
"learning_rate": 1.3337349397590363e-05,
|
| 721 |
+
"loss": 0.8075,
|
| 722 |
"step": 9200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 90.29,
|
| 726 |
+
"learning_rate": 1.2132530120481929e-05,
|
| 727 |
+
"loss": 0.8042,
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 91.26,
|
| 732 |
+
"learning_rate": 1.0927710843373493e-05,
|
| 733 |
+
"loss": 0.7916,
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 92.23,
|
| 738 |
+
"learning_rate": 9.722891566265061e-06,
|
| 739 |
+
"loss": 0.8012,
|
| 740 |
"step": 9500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 92.23,
|
| 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": 10300,
|
| 819 |
"num_train_epochs": 100,
|
| 820 |
+
"total_flos": 5.823193156406256e+19,
|
| 821 |
"trial_name": null,
|
| 822 |
"trial_params": null
|
| 823 |
}
|