Commit
•
141118f
1
Parent(s):
4dbb38d
End of training
Browse files- all_results.json +11 -11
- eval_results.json +7 -7
- train_results.json +5 -5
- trainer_state.json +149 -989
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 4466,
|
| 7 |
-
"eval_samples_per_second": 14.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 1.
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 10152,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 49.99,
|
| 3 |
+
"eval_cer": 0.16094305976711124,
|
| 4 |
+
"eval_loss": 0.5499923825263977,
|
| 5 |
+
"eval_runtime": 304.1423,
|
| 6 |
"eval_samples": 4466,
|
| 7 |
+
"eval_samples_per_second": 14.684,
|
| 8 |
+
"eval_steps_per_second": 1.838,
|
| 9 |
+
"eval_wer": 1.0132020586260908,
|
| 10 |
+
"train_loss": 2.1743111564539657,
|
| 11 |
+
"train_runtime": 44274.9554,
|
| 12 |
"train_samples": 10152,
|
| 13 |
+
"train_samples_per_second": 11.465,
|
| 14 |
+
"train_steps_per_second": 0.089
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 4466,
|
| 7 |
-
"eval_samples_per_second": 14.
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 1.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 49.99,
|
| 3 |
+
"eval_cer": 0.16094305976711124,
|
| 4 |
+
"eval_loss": 0.5499923825263977,
|
| 5 |
+
"eval_runtime": 304.1423,
|
| 6 |
"eval_samples": 4466,
|
| 7 |
+
"eval_samples_per_second": 14.684,
|
| 8 |
+
"eval_steps_per_second": 1.838,
|
| 9 |
+
"eval_wer": 1.0132020586260908
|
| 10 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 10152,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 49.99,
|
| 3 |
+
"train_loss": 2.1743111564539657,
|
| 4 |
+
"train_runtime": 44274.9554,
|
| 5 |
"train_samples": 10152,
|
| 6 |
+
"train_samples_per_second": 11.465,
|
| 7 |
+
"train_steps_per_second": 0.089
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1129 +1,289 @@
|
|
| 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.
|
| 13 |
-
"loss":
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch":
|
| 18 |
-
"learning_rate": 9.
|
| 19 |
-
"loss": 4.
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch":
|
| 24 |
-
"learning_rate": 1.
|
| 25 |
-
"loss":
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch":
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss":
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"learning_rate": 2.
|
| 37 |
-
"loss": 2.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss":
|
| 44 |
"step": 600
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"learning_rate": 3.
|
| 49 |
-
"loss":
|
| 50 |
"step": 700
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch":
|
| 54 |
-
"learning_rate": 3.
|
| 55 |
-
"loss": 1.
|
| 56 |
"step": 800
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch":
|
| 60 |
-
"learning_rate": 4.
|
| 61 |
-
"loss": 1.
|
| 62 |
"step": 900
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch":
|
| 66 |
-
"learning_rate": 4.
|
| 67 |
-
"loss": 1.
|
| 68 |
"step": 1000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"eval_cer": 0.
|
| 73 |
-
"eval_loss": 1.
|
| 74 |
-
"eval_runtime":
|
| 75 |
-
"eval_samples_per_second": 14.
|
| 76 |
-
"eval_steps_per_second": 1.
|
| 77 |
-
"eval_wer": 0.
|
| 78 |
"step": 1000
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch":
|
| 82 |
-
"learning_rate": 5.
|
| 83 |
-
"loss": 1.
|
| 84 |
"step": 1100
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch":
|
| 88 |
-
"learning_rate": 5.
|
| 89 |
-
"loss": 1.
|
| 90 |
"step": 1200
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"learning_rate": 6.
|
| 95 |
-
"loss": 1.
|
| 96 |
"step": 1300
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch":
|
| 100 |
-
"learning_rate": 6.
|
| 101 |
-
"loss": 1.
|
| 102 |
"step": 1400
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"learning_rate": 7.
|
| 107 |
-
"loss": 1.
|
| 108 |
"step": 1500
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch":
|
| 112 |
-
"learning_rate": 7.
|
| 113 |
-
"loss":
|
| 114 |
"step": 1600
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss":
|
| 120 |
"step": 1700
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"learning_rate":
|
| 125 |
-
"loss":
|
| 126 |
"step": 1800
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"learning_rate":
|
| 131 |
-
"loss":
|
| 132 |
"step": 1900
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"learning_rate":
|
| 137 |
-
"loss":
|
| 138 |
"step": 2000
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"eval_cer": 0.
|
| 143 |
-
"eval_loss":
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second": 14.
|
| 146 |
-
"eval_steps_per_second": 1.
|
| 147 |
-
"eval_wer":
|
| 148 |
"step": 2000
|
| 149 |
},
|
| 150 |
{
|
| 151 |
-
"epoch":
|
| 152 |
-
"learning_rate":
|
| 153 |
-
"loss":
|
| 154 |
"step": 2100
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"epoch":
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss":
|
| 160 |
"step": 2200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"epoch":
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"loss":
|
| 166 |
"step": 2300
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch":
|
| 170 |
-
"learning_rate":
|
| 171 |
-
"loss":
|
| 172 |
"step": 2400
|
| 173 |
},
|
| 174 |
{
|
| 175 |
-
"epoch":
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss":
|
| 178 |
"step": 2500
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch":
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss":
|
| 184 |
"step": 2600
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"learning_rate":
|
| 189 |
-
"loss":
|
| 190 |
"step": 2700
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"epoch":
|
| 194 |
-
"learning_rate":
|
| 195 |
-
"loss":
|
| 196 |
"step": 2800
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"epoch":
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss":
|
| 202 |
"step": 2900
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch":
|
| 206 |
-
"learning_rate":
|
| 207 |
-
"loss":
|
| 208 |
"step": 3000
|
| 209 |
},
|
| 210 |
{
|
| 211 |
-
"epoch":
|
| 212 |
-
"eval_cer": 0.
|
| 213 |
-
"eval_loss":
|
| 214 |
-
"eval_runtime":
|
| 215 |
-
"eval_samples_per_second": 14.
|
| 216 |
-
"eval_steps_per_second": 1.
|
| 217 |
-
"eval_wer": 1.
|
| 218 |
"step": 3000
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss":
|
| 224 |
"step": 3100
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss":
|
| 230 |
"step": 3200
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 1.
|
| 236 |
"step": 3300
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 1.
|
| 242 |
"step": 3400
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss":
|
| 248 |
"step": 3500
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 3600
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 1.
|
| 260 |
"step": 3700
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"learning_rate":
|
| 265 |
-
"loss": 1.
|
| 266 |
"step": 3800
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch":
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 1.
|
| 272 |
"step": 3900
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch":
|
| 276 |
-
"
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
|
| 280 |
-
|
| 281 |
-
"
|
| 282 |
-
"eval_cer": 0.26565260623392417,
|
| 283 |
-
"eval_loss": 0.9817910194396973,
|
| 284 |
-
"eval_runtime": 306.1977,
|
| 285 |
-
"eval_samples_per_second": 14.585,
|
| 286 |
-
"eval_steps_per_second": 1.826,
|
| 287 |
-
"eval_wer": 1.0089505482210785,
|
| 288 |
-
"step": 4000
|
| 289 |
-
},
|
| 290 |
-
{
|
| 291 |
-
"epoch": 12.89,
|
| 292 |
-
"learning_rate": 8.492805755395684e-05,
|
| 293 |
-
"loss": 1.9532,
|
| 294 |
-
"step": 4100
|
| 295 |
-
},
|
| 296 |
-
{
|
| 297 |
-
"epoch": 13.21,
|
| 298 |
-
"learning_rate": 8.420863309352519e-05,
|
| 299 |
-
"loss": 1.9614,
|
| 300 |
-
"step": 4200
|
| 301 |
-
},
|
| 302 |
-
{
|
| 303 |
-
"epoch": 13.52,
|
| 304 |
-
"learning_rate": 8.348920863309353e-05,
|
| 305 |
-
"loss": 1.9632,
|
| 306 |
-
"step": 4300
|
| 307 |
-
},
|
| 308 |
-
{
|
| 309 |
-
"epoch": 13.84,
|
| 310 |
-
"learning_rate": 8.276978417266188e-05,
|
| 311 |
-
"loss": 1.9463,
|
| 312 |
-
"step": 4400
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 14.15,
|
| 316 |
-
"learning_rate": 8.205035971223022e-05,
|
| 317 |
-
"loss": 1.9153,
|
| 318 |
-
"step": 4500
|
| 319 |
-
},
|
| 320 |
-
{
|
| 321 |
-
"epoch": 14.47,
|
| 322 |
-
"learning_rate": 8.133093525179857e-05,
|
| 323 |
-
"loss": 1.8858,
|
| 324 |
-
"step": 4600
|
| 325 |
-
},
|
| 326 |
-
{
|
| 327 |
-
"epoch": 14.78,
|
| 328 |
-
"learning_rate": 8.06115107913669e-05,
|
| 329 |
-
"loss": 1.873,
|
| 330 |
-
"step": 4700
|
| 331 |
-
},
|
| 332 |
-
{
|
| 333 |
-
"epoch": 15.09,
|
| 334 |
-
"learning_rate": 7.989208633093525e-05,
|
| 335 |
-
"loss": 1.8554,
|
| 336 |
-
"step": 4800
|
| 337 |
-
},
|
| 338 |
-
{
|
| 339 |
-
"epoch": 15.41,
|
| 340 |
-
"learning_rate": 7.91726618705036e-05,
|
| 341 |
-
"loss": 1.8789,
|
| 342 |
-
"step": 4900
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 15.72,
|
| 346 |
-
"learning_rate": 7.845323741007195e-05,
|
| 347 |
-
"loss": 1.8409,
|
| 348 |
-
"step": 5000
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 15.72,
|
| 352 |
-
"eval_cer": 0.3740518863660643,
|
| 353 |
-
"eval_loss": 1.0381526947021484,
|
| 354 |
-
"eval_runtime": 306.1278,
|
| 355 |
-
"eval_samples_per_second": 14.589,
|
| 356 |
-
"eval_steps_per_second": 1.826,
|
| 357 |
-
"eval_wer": 1.6480196912060863,
|
| 358 |
-
"step": 5000
|
| 359 |
-
},
|
| 360 |
-
{
|
| 361 |
-
"epoch": 16.04,
|
| 362 |
-
"learning_rate": 7.77338129496403e-05,
|
| 363 |
-
"loss": 1.8599,
|
| 364 |
-
"step": 5100
|
| 365 |
-
},
|
| 366 |
-
{
|
| 367 |
-
"epoch": 16.35,
|
| 368 |
-
"learning_rate": 7.701438848920864e-05,
|
| 369 |
-
"loss": 1.8393,
|
| 370 |
-
"step": 5200
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 16.67,
|
| 374 |
-
"learning_rate": 7.629496402877698e-05,
|
| 375 |
-
"loss": 1.8435,
|
| 376 |
-
"step": 5300
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 16.98,
|
| 380 |
-
"learning_rate": 7.557553956834533e-05,
|
| 381 |
-
"loss": 1.8443,
|
| 382 |
-
"step": 5400
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 17.3,
|
| 386 |
-
"learning_rate": 7.485611510791367e-05,
|
| 387 |
-
"loss": 1.7856,
|
| 388 |
-
"step": 5500
|
| 389 |
-
},
|
| 390 |
-
{
|
| 391 |
-
"epoch": 17.61,
|
| 392 |
-
"learning_rate": 7.413669064748202e-05,
|
| 393 |
-
"loss": 1.7948,
|
| 394 |
-
"step": 5600
|
| 395 |
-
},
|
| 396 |
-
{
|
| 397 |
-
"epoch": 17.92,
|
| 398 |
-
"learning_rate": 7.341726618705035e-05,
|
| 399 |
-
"loss": 1.8169,
|
| 400 |
-
"step": 5700
|
| 401 |
-
},
|
| 402 |
-
{
|
| 403 |
-
"epoch": 18.24,
|
| 404 |
-
"learning_rate": 7.269784172661872e-05,
|
| 405 |
-
"loss": 1.7683,
|
| 406 |
-
"step": 5800
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 18.55,
|
| 410 |
-
"learning_rate": 7.197841726618706e-05,
|
| 411 |
-
"loss": 1.7461,
|
| 412 |
-
"step": 5900
|
| 413 |
-
},
|
| 414 |
-
{
|
| 415 |
-
"epoch": 18.87,
|
| 416 |
-
"learning_rate": 7.12589928057554e-05,
|
| 417 |
-
"loss": 1.7449,
|
| 418 |
-
"step": 6000
|
| 419 |
-
},
|
| 420 |
-
{
|
| 421 |
-
"epoch": 18.87,
|
| 422 |
-
"eval_cer": 0.34543796070310384,
|
| 423 |
-
"eval_loss": 0.9962468147277832,
|
| 424 |
-
"eval_runtime": 306.247,
|
| 425 |
-
"eval_samples_per_second": 14.583,
|
| 426 |
-
"eval_steps_per_second": 1.825,
|
| 427 |
-
"eval_wer": 1.6267621391810247,
|
| 428 |
-
"step": 6000
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 19.18,
|
| 432 |
-
"learning_rate": 7.053956834532375e-05,
|
| 433 |
-
"loss": 1.7904,
|
| 434 |
-
"step": 6100
|
| 435 |
-
},
|
| 436 |
-
{
|
| 437 |
-
"epoch": 19.5,
|
| 438 |
-
"learning_rate": 6.982014388489208e-05,
|
| 439 |
-
"loss": 1.7315,
|
| 440 |
-
"step": 6200
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 19.81,
|
| 444 |
-
"learning_rate": 6.910071942446043e-05,
|
| 445 |
-
"loss": 1.7591,
|
| 446 |
-
"step": 6300
|
| 447 |
-
},
|
| 448 |
-
{
|
| 449 |
-
"epoch": 20.13,
|
| 450 |
-
"learning_rate": 6.83884892086331e-05,
|
| 451 |
-
"loss": 1.7338,
|
| 452 |
-
"step": 6400
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 20.44,
|
| 456 |
-
"learning_rate": 6.766906474820143e-05,
|
| 457 |
-
"loss": 1.7211,
|
| 458 |
-
"step": 6500
|
| 459 |
-
},
|
| 460 |
-
{
|
| 461 |
-
"epoch": 20.75,
|
| 462 |
-
"learning_rate": 6.694964028776978e-05,
|
| 463 |
-
"loss": 1.7398,
|
| 464 |
-
"step": 6600
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 21.07,
|
| 468 |
-
"learning_rate": 6.623021582733813e-05,
|
| 469 |
-
"loss": 1.7244,
|
| 470 |
-
"step": 6700
|
| 471 |
-
},
|
| 472 |
-
{
|
| 473 |
-
"epoch": 21.38,
|
| 474 |
-
"learning_rate": 6.551079136690648e-05,
|
| 475 |
-
"loss": 1.7823,
|
| 476 |
-
"step": 6800
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 21.7,
|
| 480 |
-
"learning_rate": 6.479136690647483e-05,
|
| 481 |
-
"loss": 1.7602,
|
| 482 |
-
"step": 6900
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 22.01,
|
| 486 |
-
"learning_rate": 6.407194244604317e-05,
|
| 487 |
-
"loss": 1.7349,
|
| 488 |
-
"step": 7000
|
| 489 |
-
},
|
| 490 |
-
{
|
| 491 |
-
"epoch": 22.01,
|
| 492 |
-
"eval_cer": 0.25974410176762075,
|
| 493 |
-
"eval_loss": 0.9559761881828308,
|
| 494 |
-
"eval_runtime": 307.762,
|
| 495 |
-
"eval_samples_per_second": 14.511,
|
| 496 |
-
"eval_steps_per_second": 1.816,
|
| 497 |
-
"eval_wer": 0.9850078317296934,
|
| 498 |
-
"step": 7000
|
| 499 |
-
},
|
| 500 |
-
{
|
| 501 |
-
"epoch": 22.33,
|
| 502 |
-
"learning_rate": 6.335251798561152e-05,
|
| 503 |
-
"loss": 1.7176,
|
| 504 |
-
"step": 7100
|
| 505 |
-
},
|
| 506 |
-
{
|
| 507 |
-
"epoch": 22.64,
|
| 508 |
-
"learning_rate": 6.263309352517986e-05,
|
| 509 |
-
"loss": 1.7154,
|
| 510 |
-
"step": 7200
|
| 511 |
-
},
|
| 512 |
-
{
|
| 513 |
-
"epoch": 22.96,
|
| 514 |
-
"learning_rate": 6.19136690647482e-05,
|
| 515 |
-
"loss": 1.7136,
|
| 516 |
-
"step": 7300
|
| 517 |
-
},
|
| 518 |
-
{
|
| 519 |
-
"epoch": 23.27,
|
| 520 |
-
"learning_rate": 6.119424460431655e-05,
|
| 521 |
-
"loss": 1.6846,
|
| 522 |
-
"step": 7400
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 23.58,
|
| 526 |
-
"learning_rate": 6.047482014388489e-05,
|
| 527 |
-
"loss": 1.6878,
|
| 528 |
-
"step": 7500
|
| 529 |
-
},
|
| 530 |
-
{
|
| 531 |
-
"epoch": 23.9,
|
| 532 |
-
"learning_rate": 5.975539568345324e-05,
|
| 533 |
-
"loss": 1.7066,
|
| 534 |
-
"step": 7600
|
| 535 |
-
},
|
| 536 |
-
{
|
| 537 |
-
"epoch": 24.21,
|
| 538 |
-
"learning_rate": 5.903597122302159e-05,
|
| 539 |
-
"loss": 1.6664,
|
| 540 |
-
"step": 7700
|
| 541 |
-
},
|
| 542 |
-
{
|
| 543 |
-
"epoch": 24.53,
|
| 544 |
-
"learning_rate": 5.831654676258993e-05,
|
| 545 |
-
"loss": 1.6676,
|
| 546 |
-
"step": 7800
|
| 547 |
-
},
|
| 548 |
-
{
|
| 549 |
-
"epoch": 24.84,
|
| 550 |
-
"learning_rate": 5.759712230215828e-05,
|
| 551 |
-
"loss": 1.6377,
|
| 552 |
-
"step": 7900
|
| 553 |
-
},
|
| 554 |
-
{
|
| 555 |
-
"epoch": 25.16,
|
| 556 |
-
"learning_rate": 5.6877697841726616e-05,
|
| 557 |
-
"loss": 1.6857,
|
| 558 |
-
"step": 8000
|
| 559 |
-
},
|
| 560 |
-
{
|
| 561 |
-
"epoch": 25.16,
|
| 562 |
-
"eval_cer": 0.23246143858524598,
|
| 563 |
-
"eval_loss": 0.8722429275512695,
|
| 564 |
-
"eval_runtime": 307.5785,
|
| 565 |
-
"eval_samples_per_second": 14.52,
|
| 566 |
-
"eval_steps_per_second": 1.817,
|
| 567 |
-
"eval_wer": 0.9668829715820094,
|
| 568 |
-
"step": 8000
|
| 569 |
-
},
|
| 570 |
-
{
|
| 571 |
-
"epoch": 25.47,
|
| 572 |
-
"learning_rate": 5.615827338129497e-05,
|
| 573 |
-
"loss": 1.6433,
|
| 574 |
-
"step": 8100
|
| 575 |
-
},
|
| 576 |
-
{
|
| 577 |
-
"epoch": 25.79,
|
| 578 |
-
"learning_rate": 5.543884892086331e-05,
|
| 579 |
-
"loss": 1.6355,
|
| 580 |
-
"step": 8200
|
| 581 |
-
},
|
| 582 |
-
{
|
| 583 |
-
"epoch": 26.1,
|
| 584 |
-
"learning_rate": 5.4719424460431656e-05,
|
| 585 |
-
"loss": 1.6345,
|
| 586 |
-
"step": 8300
|
| 587 |
-
},
|
| 588 |
-
{
|
| 589 |
-
"epoch": 26.42,
|
| 590 |
-
"learning_rate": 5.4000000000000005e-05,
|
| 591 |
-
"loss": 1.6166,
|
| 592 |
-
"step": 8400
|
| 593 |
-
},
|
| 594 |
-
{
|
| 595 |
-
"epoch": 26.73,
|
| 596 |
-
"learning_rate": 5.328057553956835e-05,
|
| 597 |
-
"loss": 1.6357,
|
| 598 |
-
"step": 8500
|
| 599 |
-
},
|
| 600 |
-
{
|
| 601 |
-
"epoch": 27.04,
|
| 602 |
-
"learning_rate": 5.2561151079136696e-05,
|
| 603 |
-
"loss": 1.6294,
|
| 604 |
-
"step": 8600
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 27.36,
|
| 608 |
-
"learning_rate": 5.1848920863309355e-05,
|
| 609 |
-
"loss": 1.5973,
|
| 610 |
-
"step": 8700
|
| 611 |
-
},
|
| 612 |
-
{
|
| 613 |
-
"epoch": 27.67,
|
| 614 |
-
"learning_rate": 5.11294964028777e-05,
|
| 615 |
-
"loss": 1.5881,
|
| 616 |
-
"step": 8800
|
| 617 |
-
},
|
| 618 |
-
{
|
| 619 |
-
"epoch": 27.99,
|
| 620 |
-
"learning_rate": 5.0410071942446046e-05,
|
| 621 |
-
"loss": 1.6079,
|
| 622 |
-
"step": 8900
|
| 623 |
-
},
|
| 624 |
-
{
|
| 625 |
-
"epoch": 28.3,
|
| 626 |
-
"learning_rate": 4.969064748201439e-05,
|
| 627 |
-
"loss": 1.5637,
|
| 628 |
-
"step": 9000
|
| 629 |
-
},
|
| 630 |
-
{
|
| 631 |
-
"epoch": 28.3,
|
| 632 |
-
"eval_cer": 0.3422495048854046,
|
| 633 |
-
"eval_loss": 0.7635577321052551,
|
| 634 |
-
"eval_runtime": 305.1268,
|
| 635 |
-
"eval_samples_per_second": 14.637,
|
| 636 |
-
"eval_steps_per_second": 1.832,
|
| 637 |
-
"eval_wer": 1.8071156858357575,
|
| 638 |
-
"step": 9000
|
| 639 |
-
},
|
| 640 |
-
{
|
| 641 |
-
"epoch": 28.62,
|
| 642 |
-
"learning_rate": 4.897122302158274e-05,
|
| 643 |
-
"loss": 1.5611,
|
| 644 |
-
"step": 9100
|
| 645 |
-
},
|
| 646 |
-
{
|
| 647 |
-
"epoch": 28.93,
|
| 648 |
-
"learning_rate": 4.825179856115108e-05,
|
| 649 |
-
"loss": 1.5648,
|
| 650 |
-
"step": 9200
|
| 651 |
-
},
|
| 652 |
-
{
|
| 653 |
-
"epoch": 29.25,
|
| 654 |
-
"learning_rate": 4.753237410071943e-05,
|
| 655 |
-
"loss": 1.5724,
|
| 656 |
-
"step": 9300
|
| 657 |
-
},
|
| 658 |
-
{
|
| 659 |
-
"epoch": 29.56,
|
| 660 |
-
"learning_rate": 4.681294964028777e-05,
|
| 661 |
-
"loss": 1.5375,
|
| 662 |
-
"step": 9400
|
| 663 |
-
},
|
| 664 |
-
{
|
| 665 |
-
"epoch": 29.87,
|
| 666 |
-
"learning_rate": 4.609352517985612e-05,
|
| 667 |
-
"loss": 1.5468,
|
| 668 |
-
"step": 9500
|
| 669 |
-
},
|
| 670 |
-
{
|
| 671 |
-
"epoch": 30.19,
|
| 672 |
-
"learning_rate": 4.537410071942446e-05,
|
| 673 |
-
"loss": 1.5512,
|
| 674 |
-
"step": 9600
|
| 675 |
-
},
|
| 676 |
-
{
|
| 677 |
-
"epoch": 30.5,
|
| 678 |
-
"learning_rate": 4.4654676258992804e-05,
|
| 679 |
-
"loss": 1.5394,
|
| 680 |
-
"step": 9700
|
| 681 |
-
},
|
| 682 |
-
{
|
| 683 |
-
"epoch": 30.82,
|
| 684 |
-
"learning_rate": 4.393525179856115e-05,
|
| 685 |
-
"loss": 1.5147,
|
| 686 |
-
"step": 9800
|
| 687 |
-
},
|
| 688 |
-
{
|
| 689 |
-
"epoch": 31.13,
|
| 690 |
-
"learning_rate": 4.32158273381295e-05,
|
| 691 |
-
"loss": 1.5102,
|
| 692 |
-
"step": 9900
|
| 693 |
-
},
|
| 694 |
-
{
|
| 695 |
-
"epoch": 31.45,
|
| 696 |
-
"learning_rate": 4.2496402877697845e-05,
|
| 697 |
-
"loss": 1.5088,
|
| 698 |
-
"step": 10000
|
| 699 |
-
},
|
| 700 |
-
{
|
| 701 |
-
"epoch": 31.45,
|
| 702 |
-
"eval_cer": 0.20850693160433564,
|
| 703 |
-
"eval_loss": 0.7290384769439697,
|
| 704 |
-
"eval_runtime": 306.7599,
|
| 705 |
-
"eval_samples_per_second": 14.559,
|
| 706 |
-
"eval_steps_per_second": 1.822,
|
| 707 |
-
"eval_wer": 1.0398299395837995,
|
| 708 |
-
"step": 10000
|
| 709 |
-
},
|
| 710 |
-
{
|
| 711 |
-
"epoch": 31.76,
|
| 712 |
-
"learning_rate": 4.177697841726619e-05,
|
| 713 |
-
"loss": 1.503,
|
| 714 |
-
"step": 10100
|
| 715 |
-
},
|
| 716 |
-
{
|
| 717 |
-
"epoch": 32.08,
|
| 718 |
-
"learning_rate": 4.1057553956834536e-05,
|
| 719 |
-
"loss": 1.4922,
|
| 720 |
-
"step": 10200
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 32.39,
|
| 724 |
-
"learning_rate": 4.0338129496402885e-05,
|
| 725 |
-
"loss": 1.5002,
|
| 726 |
-
"step": 10300
|
| 727 |
-
},
|
| 728 |
-
{
|
| 729 |
-
"epoch": 32.7,
|
| 730 |
-
"learning_rate": 3.961870503597123e-05,
|
| 731 |
-
"loss": 1.4877,
|
| 732 |
-
"step": 10400
|
| 733 |
-
},
|
| 734 |
-
{
|
| 735 |
-
"epoch": 33.02,
|
| 736 |
-
"learning_rate": 3.889928057553957e-05,
|
| 737 |
-
"loss": 1.5061,
|
| 738 |
-
"step": 10500
|
| 739 |
-
},
|
| 740 |
-
{
|
| 741 |
-
"epoch": 33.33,
|
| 742 |
-
"learning_rate": 3.817985611510791e-05,
|
| 743 |
-
"loss": 1.4751,
|
| 744 |
-
"step": 10600
|
| 745 |
-
},
|
| 746 |
-
{
|
| 747 |
-
"epoch": 33.65,
|
| 748 |
-
"learning_rate": 3.746043165467626e-05,
|
| 749 |
-
"loss": 1.4671,
|
| 750 |
-
"step": 10700
|
| 751 |
-
},
|
| 752 |
-
{
|
| 753 |
-
"epoch": 33.96,
|
| 754 |
-
"learning_rate": 3.674820143884892e-05,
|
| 755 |
-
"loss": 1.4496,
|
| 756 |
-
"step": 10800
|
| 757 |
-
},
|
| 758 |
-
{
|
| 759 |
-
"epoch": 34.28,
|
| 760 |
-
"learning_rate": 3.602877697841727e-05,
|
| 761 |
-
"loss": 1.4533,
|
| 762 |
-
"step": 10900
|
| 763 |
-
},
|
| 764 |
-
{
|
| 765 |
-
"epoch": 34.59,
|
| 766 |
-
"learning_rate": 3.530935251798561e-05,
|
| 767 |
-
"loss": 1.4298,
|
| 768 |
-
"step": 11000
|
| 769 |
-
},
|
| 770 |
-
{
|
| 771 |
-
"epoch": 34.59,
|
| 772 |
-
"eval_cer": 0.21042164862888182,
|
| 773 |
-
"eval_loss": 0.7575956583023071,
|
| 774 |
-
"eval_runtime": 307.6821,
|
| 775 |
-
"eval_samples_per_second": 14.515,
|
| 776 |
-
"eval_steps_per_second": 1.817,
|
| 777 |
-
"eval_wer": 1.0165585142089952,
|
| 778 |
-
"step": 11000
|
| 779 |
-
},
|
| 780 |
-
{
|
| 781 |
-
"epoch": 34.91,
|
| 782 |
-
"learning_rate": 3.458992805755396e-05,
|
| 783 |
-
"loss": 1.4401,
|
| 784 |
-
"step": 11100
|
| 785 |
-
},
|
| 786 |
-
{
|
| 787 |
-
"epoch": 35.22,
|
| 788 |
-
"learning_rate": 3.38705035971223e-05,
|
| 789 |
-
"loss": 1.4155,
|
| 790 |
-
"step": 11200
|
| 791 |
-
},
|
| 792 |
-
{
|
| 793 |
-
"epoch": 35.53,
|
| 794 |
-
"learning_rate": 3.315107913669065e-05,
|
| 795 |
-
"loss": 1.4119,
|
| 796 |
-
"step": 11300
|
| 797 |
-
},
|
| 798 |
-
{
|
| 799 |
-
"epoch": 35.85,
|
| 800 |
-
"learning_rate": 3.243165467625899e-05,
|
| 801 |
-
"loss": 1.4156,
|
| 802 |
-
"step": 11400
|
| 803 |
-
},
|
| 804 |
-
{
|
| 805 |
-
"epoch": 36.16,
|
| 806 |
-
"learning_rate": 3.1712230215827335e-05,
|
| 807 |
-
"loss": 1.4316,
|
| 808 |
-
"step": 11500
|
| 809 |
-
},
|
| 810 |
-
{
|
| 811 |
-
"epoch": 36.48,
|
| 812 |
-
"learning_rate": 3.0992805755395684e-05,
|
| 813 |
-
"loss": 1.3815,
|
| 814 |
-
"step": 11600
|
| 815 |
-
},
|
| 816 |
-
{
|
| 817 |
-
"epoch": 36.79,
|
| 818 |
-
"learning_rate": 3.027338129496403e-05,
|
| 819 |
-
"loss": 1.4087,
|
| 820 |
-
"step": 11700
|
| 821 |
-
},
|
| 822 |
-
{
|
| 823 |
-
"epoch": 37.11,
|
| 824 |
-
"learning_rate": 2.9553956834532376e-05,
|
| 825 |
-
"loss": 1.3889,
|
| 826 |
-
"step": 11800
|
| 827 |
-
},
|
| 828 |
-
{
|
| 829 |
-
"epoch": 37.42,
|
| 830 |
-
"learning_rate": 2.883453237410072e-05,
|
| 831 |
-
"loss": 1.3722,
|
| 832 |
-
"step": 11900
|
| 833 |
-
},
|
| 834 |
-
{
|
| 835 |
-
"epoch": 37.74,
|
| 836 |
-
"learning_rate": 2.8115107913669063e-05,
|
| 837 |
-
"loss": 1.3716,
|
| 838 |
-
"step": 12000
|
| 839 |
-
},
|
| 840 |
-
{
|
| 841 |
-
"epoch": 37.74,
|
| 842 |
-
"eval_cer": 0.21379089317851244,
|
| 843 |
-
"eval_loss": 0.7045890092849731,
|
| 844 |
-
"eval_runtime": 307.0323,
|
| 845 |
-
"eval_samples_per_second": 14.546,
|
| 846 |
-
"eval_steps_per_second": 1.821,
|
| 847 |
-
"eval_wer": 1.1275453121503691,
|
| 848 |
-
"step": 12000
|
| 849 |
-
},
|
| 850 |
-
{
|
| 851 |
-
"epoch": 38.05,
|
| 852 |
-
"learning_rate": 2.7395683453237412e-05,
|
| 853 |
-
"loss": 1.3782,
|
| 854 |
-
"step": 12100
|
| 855 |
-
},
|
| 856 |
-
{
|
| 857 |
-
"epoch": 38.36,
|
| 858 |
-
"learning_rate": 2.6676258992805758e-05,
|
| 859 |
-
"loss": 1.3496,
|
| 860 |
-
"step": 12200
|
| 861 |
-
},
|
| 862 |
-
{
|
| 863 |
-
"epoch": 38.68,
|
| 864 |
-
"learning_rate": 2.5956834532374104e-05,
|
| 865 |
-
"loss": 1.3645,
|
| 866 |
-
"step": 12300
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 38.99,
|
| 870 |
-
"learning_rate": 2.5237410071942446e-05,
|
| 871 |
-
"loss": 1.4,
|
| 872 |
-
"step": 12400
|
| 873 |
-
},
|
| 874 |
-
{
|
| 875 |
-
"epoch": 39.31,
|
| 876 |
-
"learning_rate": 2.4517985611510795e-05,
|
| 877 |
-
"loss": 1.3441,
|
| 878 |
-
"step": 12500
|
| 879 |
-
},
|
| 880 |
-
{
|
| 881 |
-
"epoch": 39.62,
|
| 882 |
-
"learning_rate": 2.3798561151079137e-05,
|
| 883 |
-
"loss": 1.3512,
|
| 884 |
-
"step": 12600
|
| 885 |
-
},
|
| 886 |
-
{
|
| 887 |
-
"epoch": 39.94,
|
| 888 |
-
"learning_rate": 2.3079136690647483e-05,
|
| 889 |
-
"loss": 1.3429,
|
| 890 |
-
"step": 12700
|
| 891 |
-
},
|
| 892 |
-
{
|
| 893 |
-
"epoch": 40.25,
|
| 894 |
-
"learning_rate": 2.2366906474820145e-05,
|
| 895 |
-
"loss": 1.3285,
|
| 896 |
-
"step": 12800
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 40.57,
|
| 900 |
-
"learning_rate": 2.164748201438849e-05,
|
| 901 |
-
"loss": 1.3222,
|
| 902 |
-
"step": 12900
|
| 903 |
-
},
|
| 904 |
-
{
|
| 905 |
-
"epoch": 40.88,
|
| 906 |
-
"learning_rate": 2.0928057553956836e-05,
|
| 907 |
-
"loss": 1.3185,
|
| 908 |
-
"step": 13000
|
| 909 |
-
},
|
| 910 |
-
{
|
| 911 |
-
"epoch": 40.88,
|
| 912 |
-
"eval_cer": 0.21792438100403488,
|
| 913 |
-
"eval_loss": 0.7010895013809204,
|
| 914 |
-
"eval_runtime": 308.0387,
|
| 915 |
-
"eval_samples_per_second": 14.498,
|
| 916 |
-
"eval_steps_per_second": 1.815,
|
| 917 |
-
"eval_wer": 1.1696128887894384,
|
| 918 |
-
"step": 13000
|
| 919 |
-
},
|
| 920 |
-
{
|
| 921 |
-
"epoch": 41.19,
|
| 922 |
-
"learning_rate": 2.02158273381295e-05,
|
| 923 |
-
"loss": 1.3151,
|
| 924 |
-
"step": 13100
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 41.51,
|
| 928 |
-
"learning_rate": 1.949640287769784e-05,
|
| 929 |
-
"loss": 1.2889,
|
| 930 |
-
"step": 13200
|
| 931 |
-
},
|
| 932 |
-
{
|
| 933 |
-
"epoch": 41.82,
|
| 934 |
-
"learning_rate": 1.877697841726619e-05,
|
| 935 |
-
"loss": 1.3339,
|
| 936 |
-
"step": 13300
|
| 937 |
-
},
|
| 938 |
-
{
|
| 939 |
-
"epoch": 42.14,
|
| 940 |
-
"learning_rate": 1.8057553956834532e-05,
|
| 941 |
-
"loss": 1.3292,
|
| 942 |
-
"step": 13400
|
| 943 |
-
},
|
| 944 |
-
{
|
| 945 |
-
"epoch": 42.45,
|
| 946 |
-
"learning_rate": 1.7338129496402878e-05,
|
| 947 |
-
"loss": 1.2849,
|
| 948 |
-
"step": 13500
|
| 949 |
-
},
|
| 950 |
-
{
|
| 951 |
-
"epoch": 42.77,
|
| 952 |
-
"learning_rate": 1.6618705035971224e-05,
|
| 953 |
-
"loss": 1.3046,
|
| 954 |
-
"step": 13600
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 43.08,
|
| 958 |
-
"learning_rate": 1.589928057553957e-05,
|
| 959 |
-
"loss": 1.2722,
|
| 960 |
-
"step": 13700
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 43.4,
|
| 964 |
-
"learning_rate": 1.5179856115107915e-05,
|
| 965 |
-
"loss": 1.2704,
|
| 966 |
-
"step": 13800
|
| 967 |
-
},
|
| 968 |
-
{
|
| 969 |
-
"epoch": 43.71,
|
| 970 |
-
"learning_rate": 1.446043165467626e-05,
|
| 971 |
-
"loss": 1.279,
|
| 972 |
-
"step": 13900
|
| 973 |
-
},
|
| 974 |
-
{
|
| 975 |
-
"epoch": 44.03,
|
| 976 |
-
"learning_rate": 1.3741007194244604e-05,
|
| 977 |
-
"loss": 1.28,
|
| 978 |
-
"step": 14000
|
| 979 |
-
},
|
| 980 |
-
{
|
| 981 |
-
"epoch": 44.03,
|
| 982 |
-
"eval_cer": 0.2023847677275678,
|
| 983 |
-
"eval_loss": 0.675375759601593,
|
| 984 |
-
"eval_runtime": 307.5046,
|
| 985 |
-
"eval_samples_per_second": 14.523,
|
| 986 |
-
"eval_steps_per_second": 1.818,
|
| 987 |
-
"eval_wer": 1.1315730588498545,
|
| 988 |
-
"step": 14000
|
| 989 |
-
},
|
| 990 |
-
{
|
| 991 |
-
"epoch": 44.34,
|
| 992 |
-
"learning_rate": 1.3021582733812952e-05,
|
| 993 |
-
"loss": 1.2604,
|
| 994 |
-
"step": 14100
|
| 995 |
-
},
|
| 996 |
-
{
|
| 997 |
-
"epoch": 44.65,
|
| 998 |
-
"learning_rate": 1.2302158273381296e-05,
|
| 999 |
-
"loss": 1.2631,
|
| 1000 |
-
"step": 14200
|
| 1001 |
-
},
|
| 1002 |
-
{
|
| 1003 |
-
"epoch": 44.97,
|
| 1004 |
-
"learning_rate": 1.1582733812949641e-05,
|
| 1005 |
-
"loss": 1.2742,
|
| 1006 |
-
"step": 14300
|
| 1007 |
-
},
|
| 1008 |
-
{
|
| 1009 |
-
"epoch": 45.28,
|
| 1010 |
-
"learning_rate": 1.0863309352517987e-05,
|
| 1011 |
-
"loss": 1.2552,
|
| 1012 |
-
"step": 14400
|
| 1013 |
-
},
|
| 1014 |
-
{
|
| 1015 |
-
"epoch": 45.6,
|
| 1016 |
-
"learning_rate": 1.0143884892086332e-05,
|
| 1017 |
-
"loss": 1.2328,
|
| 1018 |
-
"step": 14500
|
| 1019 |
-
},
|
| 1020 |
-
{
|
| 1021 |
-
"epoch": 45.91,
|
| 1022 |
-
"learning_rate": 9.424460431654676e-06,
|
| 1023 |
-
"loss": 1.2562,
|
| 1024 |
-
"step": 14600
|
| 1025 |
-
},
|
| 1026 |
-
{
|
| 1027 |
-
"epoch": 46.23,
|
| 1028 |
-
"learning_rate": 8.705035971223022e-06,
|
| 1029 |
-
"loss": 1.2239,
|
| 1030 |
-
"step": 14700
|
| 1031 |
-
},
|
| 1032 |
-
{
|
| 1033 |
-
"epoch": 46.54,
|
| 1034 |
-
"learning_rate": 7.985611510791367e-06,
|
| 1035 |
-
"loss": 1.2502,
|
| 1036 |
-
"step": 14800
|
| 1037 |
-
},
|
| 1038 |
-
{
|
| 1039 |
-
"epoch": 46.86,
|
| 1040 |
-
"learning_rate": 7.266187050359713e-06,
|
| 1041 |
-
"loss": 1.2265,
|
| 1042 |
-
"step": 14900
|
| 1043 |
-
},
|
| 1044 |
-
{
|
| 1045 |
-
"epoch": 47.17,
|
| 1046 |
-
"learning_rate": 6.546762589928059e-06,
|
| 1047 |
-
"loss": 1.2368,
|
| 1048 |
-
"step": 15000
|
| 1049 |
-
},
|
| 1050 |
-
{
|
| 1051 |
-
"epoch": 47.17,
|
| 1052 |
-
"eval_cer": 0.19226881640904273,
|
| 1053 |
-
"eval_loss": 0.6925497055053711,
|
| 1054 |
-
"eval_runtime": 307.2836,
|
| 1055 |
-
"eval_samples_per_second": 14.534,
|
| 1056 |
-
"eval_steps_per_second": 1.819,
|
| 1057 |
-
"eval_wer": 1.0516894159767285,
|
| 1058 |
-
"step": 15000
|
| 1059 |
-
},
|
| 1060 |
-
{
|
| 1061 |
-
"epoch": 47.48,
|
| 1062 |
-
"learning_rate": 5.8273381294964035e-06,
|
| 1063 |
-
"loss": 1.2265,
|
| 1064 |
-
"step": 15100
|
| 1065 |
-
},
|
| 1066 |
-
{
|
| 1067 |
-
"epoch": 47.8,
|
| 1068 |
-
"learning_rate": 5.107913669064748e-06,
|
| 1069 |
-
"loss": 1.2267,
|
| 1070 |
-
"step": 15200
|
| 1071 |
-
},
|
| 1072 |
-
{
|
| 1073 |
-
"epoch": 48.11,
|
| 1074 |
-
"learning_rate": 4.388489208633094e-06,
|
| 1075 |
-
"loss": 1.2267,
|
| 1076 |
-
"step": 15300
|
| 1077 |
-
},
|
| 1078 |
-
{
|
| 1079 |
-
"epoch": 48.43,
|
| 1080 |
-
"learning_rate": 3.669064748201439e-06,
|
| 1081 |
-
"loss": 1.2095,
|
| 1082 |
-
"step": 15400
|
| 1083 |
-
},
|
| 1084 |
-
{
|
| 1085 |
-
"epoch": 48.74,
|
| 1086 |
-
"learning_rate": 2.9496402877697846e-06,
|
| 1087 |
-
"loss": 1.1985,
|
| 1088 |
-
"step": 15500
|
| 1089 |
-
},
|
| 1090 |
-
{
|
| 1091 |
-
"epoch": 49.06,
|
| 1092 |
-
"learning_rate": 2.23021582733813e-06,
|
| 1093 |
-
"loss": 1.2141,
|
| 1094 |
-
"step": 15600
|
| 1095 |
-
},
|
| 1096 |
-
{
|
| 1097 |
-
"epoch": 49.37,
|
| 1098 |
-
"learning_rate": 1.510791366906475e-06,
|
| 1099 |
-
"loss": 1.2087,
|
| 1100 |
-
"step": 15700
|
| 1101 |
-
},
|
| 1102 |
-
{
|
| 1103 |
-
"epoch": 49.69,
|
| 1104 |
-
"learning_rate": 7.913669064748202e-07,
|
| 1105 |
-
"loss": 1.2109,
|
| 1106 |
-
"step": 15800
|
| 1107 |
-
},
|
| 1108 |
-
{
|
| 1109 |
-
"epoch": 50.0,
|
| 1110 |
-
"learning_rate": 7.913669064748202e-08,
|
| 1111 |
-
"loss": 1.2123,
|
| 1112 |
-
"step": 15900
|
| 1113 |
-
},
|
| 1114 |
-
{
|
| 1115 |
-
"epoch": 50.0,
|
| 1116 |
-
"step": 15900,
|
| 1117 |
-
"total_flos": 2.3890518741171333e+20,
|
| 1118 |
-
"train_loss": 1.801840565999349,
|
| 1119 |
-
"train_runtime": 49502.0749,
|
| 1120 |
-
"train_samples_per_second": 10.254,
|
| 1121 |
-
"train_steps_per_second": 0.321
|
| 1122 |
}
|
| 1123 |
],
|
| 1124 |
-
"max_steps":
|
| 1125 |
"num_train_epochs": 50,
|
| 1126 |
-
"total_flos": 2.
|
| 1127 |
"trial_name": null,
|
| 1128 |
"trial_params": null
|
| 1129 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 49.9937106918239,
|
| 5 |
+
"global_step": 3950,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 1.26,
|
| 12 |
+
"learning_rate": 4.849999999999999e-06,
|
| 13 |
+
"loss": 18.5969,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 2.53,
|
| 18 |
+
"learning_rate": 9.849999999999999e-06,
|
| 19 |
+
"loss": 4.2811,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 3.79,
|
| 24 |
+
"learning_rate": 1.485e-05,
|
| 25 |
+
"loss": 3.9192,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 5.06,
|
| 30 |
+
"learning_rate": 1.9849999999999998e-05,
|
| 31 |
+
"loss": 2.7895,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 6.33,
|
| 36 |
+
"learning_rate": 2.4849999999999998e-05,
|
| 37 |
+
"loss": 2.142,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 7.59,
|
| 42 |
+
"learning_rate": 2.985e-05,
|
| 43 |
+
"loss": 1.9161,
|
| 44 |
"step": 600
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 8.86,
|
| 48 |
+
"learning_rate": 3.485e-05,
|
| 49 |
+
"loss": 1.7888,
|
| 50 |
"step": 700
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 10.13,
|
| 54 |
+
"learning_rate": 3.984999999999999e-05,
|
| 55 |
+
"loss": 1.7494,
|
| 56 |
"step": 800
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 11.39,
|
| 60 |
+
"learning_rate": 4.484999999999999e-05,
|
| 61 |
+
"loss": 1.7129,
|
| 62 |
"step": 900
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 12.65,
|
| 66 |
+
"learning_rate": 4.984999999999999e-05,
|
| 67 |
+
"loss": 1.7019,
|
| 68 |
"step": 1000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 12.65,
|
| 72 |
+
"eval_cer": 0.2589387701435627,
|
| 73 |
+
"eval_loss": 1.051009178161621,
|
| 74 |
+
"eval_runtime": 302.233,
|
| 75 |
+
"eval_samples_per_second": 14.777,
|
| 76 |
+
"eval_steps_per_second": 1.85,
|
| 77 |
+
"eval_wer": 0.9832177220854778,
|
| 78 |
"step": 1000
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 13.92,
|
| 82 |
+
"learning_rate": 5.484999999999999e-05,
|
| 83 |
+
"loss": 1.6797,
|
| 84 |
"step": 1100
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 15.19,
|
| 88 |
+
"learning_rate": 5.985e-05,
|
| 89 |
+
"loss": 1.6992,
|
| 90 |
"step": 1200
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 16.45,
|
| 94 |
+
"learning_rate": 6.484999999999999e-05,
|
| 95 |
+
"loss": 1.703,
|
| 96 |
"step": 1300
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 17.72,
|
| 100 |
+
"learning_rate": 6.984999999999999e-05,
|
| 101 |
+
"loss": 1.7304,
|
| 102 |
"step": 1400
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 18.98,
|
| 106 |
+
"learning_rate": 7.484999999999999e-05,
|
| 107 |
+
"loss": 1.7436,
|
| 108 |
"step": 1500
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 20.25,
|
| 112 |
+
"learning_rate": 7.203061224489795e-05,
|
| 113 |
+
"loss": 1.7564,
|
| 114 |
"step": 1600
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 21.52,
|
| 118 |
+
"learning_rate": 6.896938775510203e-05,
|
| 119 |
+
"loss": 1.6982,
|
| 120 |
"step": 1700
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 22.78,
|
| 124 |
+
"learning_rate": 6.590816326530612e-05,
|
| 125 |
+
"loss": 1.7184,
|
| 126 |
"step": 1800
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 24.05,
|
| 130 |
+
"learning_rate": 6.28469387755102e-05,
|
| 131 |
+
"loss": 1.692,
|
| 132 |
"step": 1900
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 25.31,
|
| 136 |
+
"learning_rate": 5.978571428571428e-05,
|
| 137 |
+
"loss": 1.6385,
|
| 138 |
"step": 2000
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 25.31,
|
| 142 |
+
"eval_cer": 0.18507013781031975,
|
| 143 |
+
"eval_loss": 0.6670215129852295,
|
| 144 |
+
"eval_runtime": 300.7027,
|
| 145 |
+
"eval_samples_per_second": 14.852,
|
| 146 |
+
"eval_steps_per_second": 1.859,
|
| 147 |
+
"eval_wer": 0.9914969791899754,
|
| 148 |
"step": 2000
|
| 149 |
},
|
| 150 |
{
|
| 151 |
+
"epoch": 26.58,
|
| 152 |
+
"learning_rate": 5.6724489795918356e-05,
|
| 153 |
+
"loss": 1.6092,
|
| 154 |
"step": 2100
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 27.84,
|
| 158 |
+
"learning_rate": 5.3663265306122446e-05,
|
| 159 |
+
"loss": 1.6209,
|
| 160 |
"step": 2200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 29.11,
|
| 164 |
+
"learning_rate": 5.060204081632652e-05,
|
| 165 |
+
"loss": 1.5933,
|
| 166 |
"step": 2300
|
| 167 |
},
|
| 168 |
{
|
| 169 |
+
"epoch": 30.38,
|
| 170 |
+
"learning_rate": 4.754081632653061e-05,
|
| 171 |
+
"loss": 1.5673,
|
| 172 |
"step": 2400
|
| 173 |
},
|
| 174 |
{
|
| 175 |
+
"epoch": 31.64,
|
| 176 |
+
"learning_rate": 4.447959183673469e-05,
|
| 177 |
+
"loss": 1.5518,
|
| 178 |
"step": 2500
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 32.91,
|
| 182 |
+
"learning_rate": 4.141836734693877e-05,
|
| 183 |
+
"loss": 1.5398,
|
| 184 |
"step": 2600
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 34.18,
|
| 188 |
+
"learning_rate": 3.8357142857142855e-05,
|
| 189 |
+
"loss": 1.5095,
|
| 190 |
"step": 2700
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"epoch": 35.44,
|
| 194 |
+
"learning_rate": 3.529591836734693e-05,
|
| 195 |
+
"loss": 1.508,
|
| 196 |
"step": 2800
|
| 197 |
},
|
| 198 |
{
|
| 199 |
+
"epoch": 36.7,
|
| 200 |
+
"learning_rate": 3.2234693877551015e-05,
|
| 201 |
+
"loss": 1.4769,
|
| 202 |
"step": 2900
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 37.97,
|
| 206 |
+
"learning_rate": 2.9173469387755098e-05,
|
| 207 |
+
"loss": 1.4344,
|
| 208 |
"step": 3000
|
| 209 |
},
|
| 210 |
{
|
| 211 |
+
"epoch": 37.97,
|
| 212 |
+
"eval_cer": 0.17974508788797672,
|
| 213 |
+
"eval_loss": 0.6182843446731567,
|
| 214 |
+
"eval_runtime": 301.8085,
|
| 215 |
+
"eval_samples_per_second": 14.797,
|
| 216 |
+
"eval_steps_per_second": 1.852,
|
| 217 |
+
"eval_wer": 1.0212575520250615,
|
| 218 |
"step": 3000
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 39.24,
|
| 222 |
+
"learning_rate": 2.611224489795918e-05,
|
| 223 |
+
"loss": 1.4331,
|
| 224 |
"step": 3100
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 40.5,
|
| 228 |
+
"learning_rate": 2.3051020408163264e-05,
|
| 229 |
+
"loss": 1.4042,
|
| 230 |
"step": 3200
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 41.77,
|
| 234 |
+
"learning_rate": 2.0020408163265303e-05,
|
| 235 |
+
"loss": 1.3822,
|
| 236 |
"step": 3300
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 43.04,
|
| 240 |
+
"learning_rate": 1.6959183673469386e-05,
|
| 241 |
+
"loss": 1.3768,
|
| 242 |
"step": 3400
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 44.3,
|
| 246 |
+
"learning_rate": 1.3897959183673468e-05,
|
| 247 |
+
"loss": 1.3447,
|
| 248 |
"step": 3500
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 45.57,
|
| 252 |
+
"learning_rate": 1.0836734693877551e-05,
|
| 253 |
+
"loss": 1.3359,
|
| 254 |
"step": 3600
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 46.83,
|
| 258 |
+
"learning_rate": 7.77551020408163e-06,
|
| 259 |
+
"loss": 1.3113,
|
| 260 |
"step": 3700
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 48.1,
|
| 264 |
+
"learning_rate": 4.714285714285714e-06,
|
| 265 |
+
"loss": 1.3053,
|
| 266 |
"step": 3800
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 49.36,
|
| 270 |
+
"learning_rate": 1.6530612244897958e-06,
|
| 271 |
+
"loss": 1.2802,
|
| 272 |
"step": 3900
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 49.99,
|
| 276 |
+
"step": 3950,
|
| 277 |
+
"total_flos": 2.372869475210465e+20,
|
| 278 |
+
"train_loss": 2.1743111564539657,
|
| 279 |
+
"train_runtime": 44274.9554,
|
| 280 |
+
"train_samples_per_second": 11.465,
|
| 281 |
+
"train_steps_per_second": 0.089
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
}
|
| 283 |
],
|
| 284 |
+
"max_steps": 3950,
|
| 285 |
"num_train_epochs": 50,
|
| 286 |
+
"total_flos": 2.372869475210465e+20,
|
| 287 |
"trial_name": null,
|
| 288 |
"trial_params": null
|
| 289 |
}
|