Commit
•
4b40d1e
1
Parent(s):
6caeebe
End of training
Browse files- all_results.json +12 -12
- eval_results.json +7 -7
- train_results.json +6 -6
- trainer_state.json +364 -961
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss": 1.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples":
|
| 12 |
-
"train_samples_per_second": 22.
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.1318359673023224,
|
| 4 |
+
"eval_runtime": 200.408,
|
| 5 |
+
"eval_samples": 5045,
|
| 6 |
+
"eval_samples_per_second": 25.174,
|
| 7 |
+
"eval_steps_per_second": 0.788,
|
| 8 |
+
"eval_wer": 0.11208951988728731,
|
| 9 |
+
"train_loss": 1.3699857131640116,
|
| 10 |
+
"train_runtime": 54887.9396,
|
| 11 |
+
"train_samples": 12307,
|
| 12 |
+
"train_samples_per_second": 22.422,
|
| 13 |
+
"train_steps_per_second": 0.175
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.1318359673023224,
|
| 4 |
+
"eval_runtime": 200.408,
|
| 5 |
+
"eval_samples": 5045,
|
| 6 |
+
"eval_samples_per_second": 25.174,
|
| 7 |
+
"eval_steps_per_second": 0.788,
|
| 8 |
+
"eval_wer": 0.11208951988728731
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 22.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.3699857131640116,
|
| 4 |
+
"train_runtime": 54887.9396,
|
| 5 |
+
"train_samples": 12307,
|
| 6 |
+
"train_samples_per_second": 22.422,
|
| 7 |
+
"train_steps_per_second": 0.175
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1279 +1,682 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 99.
|
| 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":
|
| 13 |
-
"loss":
|
| 14 |
-
"step": 50
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 1.1,
|
| 18 |
-
"learning_rate": 1.076923076923077e-05,
|
| 19 |
-
"loss": 14.3202,
|
| 20 |
"step": 100
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch":
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss":
|
| 26 |
-
"step": 150
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 2.2,
|
| 30 |
-
"learning_rate": 2.175824175824176e-05,
|
| 31 |
-
"loss": 3.9265,
|
| 32 |
"step": 200
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
-
"step": 250
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 3.3,
|
| 42 |
-
"learning_rate": 3.274725274725275e-05,
|
| 43 |
-
"loss": 3.1869,
|
| 44 |
"step": 300
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 3.
|
| 50 |
-
"step": 350
|
| 51 |
-
},
|
| 52 |
-
{
|
| 53 |
-
"epoch": 4.39,
|
| 54 |
-
"learning_rate": 4.3736263736263734e-05,
|
| 55 |
-
"loss": 3.1175,
|
| 56 |
"step": 400
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch":
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 3.
|
| 62 |
-
"step": 450
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 5.49,
|
| 66 |
-
"learning_rate": 5.472527472527473e-05,
|
| 67 |
-
"loss": 3.0418,
|
| 68 |
-
"step": 500
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 5.49,
|
| 72 |
-
"eval_loss": 3.0175631046295166,
|
| 73 |
-
"eval_runtime": 180.0397,
|
| 74 |
-
"eval_samples_per_second": 26.9,
|
| 75 |
-
"eval_steps_per_second": 0.844,
|
| 76 |
-
"eval_wer": 1.0,
|
| 77 |
"step": 500
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch": 6.
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss":
|
| 83 |
-
"step": 550
|
| 84 |
-
},
|
| 85 |
-
{
|
| 86 |
-
"epoch": 6.59,
|
| 87 |
-
"learning_rate": 6.571428571428571e-05,
|
| 88 |
-
"loss": 2.9326,
|
| 89 |
"step": 600
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 7.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss":
|
| 95 |
-
"step": 650
|
| 96 |
-
},
|
| 97 |
-
{
|
| 98 |
-
"epoch": 7.69,
|
| 99 |
-
"learning_rate": 7.67032967032967e-05,
|
| 100 |
-
"loss": 2.5847,
|
| 101 |
"step": 700
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 8.
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss":
|
| 107 |
-
"step": 750
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 8.79,
|
| 111 |
-
"learning_rate": 8.76923076923077e-05,
|
| 112 |
-
"loss": 1.5005,
|
| 113 |
"step": 800
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 9.
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss":
|
| 119 |
-
"step": 850
|
| 120 |
-
},
|
| 121 |
-
{
|
| 122 |
-
"epoch": 9.89,
|
| 123 |
-
"learning_rate": 9.868131868131869e-05,
|
| 124 |
-
"loss": 1.2882,
|
| 125 |
"step": 900
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch": 10.
|
| 129 |
-
"learning_rate":
|
| 130 |
-
"loss":
|
| 131 |
-
"step": 950
|
| 132 |
-
},
|
| 133 |
-
{
|
| 134 |
-
"epoch": 10.98,
|
| 135 |
-
"learning_rate": 0.00010967032967032966,
|
| 136 |
-
"loss": 1.1819,
|
| 137 |
"step": 1000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 10.
|
| 141 |
-
"eval_loss":
|
| 142 |
-
"eval_runtime":
|
| 143 |
-
"eval_samples_per_second":
|
| 144 |
-
"eval_steps_per_second": 0.
|
| 145 |
-
"eval_wer": 0
|
| 146 |
"step": 1000
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 11.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss":
|
| 152 |
-
"step": 1050
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch": 12.09,
|
| 156 |
-
"learning_rate": 0.00012065934065934066,
|
| 157 |
-
"loss": 1.1248,
|
| 158 |
"step": 1100
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 12.
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 1.
|
| 164 |
-
"step": 1150
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 13.19,
|
| 168 |
-
"learning_rate": 0.00013164835164835166,
|
| 169 |
-
"loss": 1.0713,
|
| 170 |
"step": 1200
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 13.
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 1.
|
| 176 |
-
"step": 1250
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 14.28,
|
| 180 |
-
"learning_rate": 0.00014263736263736264,
|
| 181 |
-
"loss": 1.0454,
|
| 182 |
"step": 1300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 14.
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 1.
|
| 188 |
-
"step": 1350
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 15.38,
|
| 192 |
-
"learning_rate": 0.00015362637362637362,
|
| 193 |
-
"loss": 1.0207,
|
| 194 |
"step": 1400
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 15.
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss":
|
| 200 |
-
"step": 1450
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 16.48,
|
| 204 |
-
"learning_rate": 0.0001646153846153846,
|
| 205 |
-
"loss": 1.0032,
|
| 206 |
-
"step": 1500
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 16.48,
|
| 210 |
-
"eval_loss": 0.1746312528848648,
|
| 211 |
-
"eval_runtime": 184.5587,
|
| 212 |
-
"eval_samples_per_second": 26.241,
|
| 213 |
-
"eval_steps_per_second": 0.824,
|
| 214 |
-
"eval_wer": 0.15461254612546124,
|
| 215 |
"step": 1500
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss":
|
| 221 |
-
"step": 1550
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 17.58,
|
| 225 |
-
"learning_rate": 0.00017560439560439562,
|
| 226 |
-
"loss": 0.9528,
|
| 227 |
"step": 1600
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
-
"step": 1650
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 18.68,
|
| 237 |
-
"learning_rate": 0.0001865934065934066,
|
| 238 |
-
"loss": 0.9495,
|
| 239 |
"step": 1700
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch":
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss":
|
| 245 |
-
"step": 1750
|
| 246 |
-
},
|
| 247 |
-
{
|
| 248 |
-
"epoch": 19.78,
|
| 249 |
-
"learning_rate": 0.00019758241758241759,
|
| 250 |
-
"loss": 0.9385,
|
| 251 |
"step": 1800
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss":
|
| 257 |
-
"step": 1850
|
| 258 |
-
},
|
| 259 |
-
{
|
| 260 |
-
"epoch": 20.87,
|
| 261 |
-
"learning_rate": 0.00020857142857142857,
|
| 262 |
-
"loss": 0.9028,
|
| 263 |
"step": 1900
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
-
"step": 1950
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 21.97,
|
| 273 |
-
"learning_rate": 0.00021956043956043955,
|
| 274 |
-
"loss": 0.9077,
|
| 275 |
"step": 2000
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"eval_loss": 0.
|
| 280 |
-
"eval_runtime":
|
| 281 |
-
"eval_samples_per_second":
|
| 282 |
-
"eval_steps_per_second": 0.
|
| 283 |
-
"eval_wer": 0.
|
| 284 |
"step": 2000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss":
|
| 290 |
-
"step": 2050
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 23.08,
|
| 294 |
-
"learning_rate": 0.00023054945054945056,
|
| 295 |
-
"loss": 0.9046,
|
| 296 |
"step": 2100
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss":
|
| 302 |
-
"step": 2150
|
| 303 |
-
},
|
| 304 |
-
{
|
| 305 |
-
"epoch": 24.17,
|
| 306 |
-
"learning_rate": 0.00024153846153846155,
|
| 307 |
-
"loss": 0.8888,
|
| 308 |
"step": 2200
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss":
|
| 314 |
-
"step": 2250
|
| 315 |
-
},
|
| 316 |
-
{
|
| 317 |
-
"epoch": 25.27,
|
| 318 |
-
"learning_rate": 0.00024915750915750914,
|
| 319 |
-
"loss": 0.8819,
|
| 320 |
"step": 2300
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch": 25.
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss":
|
| 326 |
-
"step": 2350
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 26.37,
|
| 330 |
-
"learning_rate": 0.0002454945054945055,
|
| 331 |
-
"loss": 0.8757,
|
| 332 |
"step": 2400
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch": 26.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
-
"step": 2450
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 27.47,
|
| 342 |
-
"learning_rate": 0.00024183150183150186,
|
| 343 |
-
"loss": 0.8687,
|
| 344 |
-
"step": 2500
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 27.47,
|
| 348 |
-
"eval_loss": 0.1647317260503769,
|
| 349 |
-
"eval_runtime": 181.9124,
|
| 350 |
-
"eval_samples_per_second": 26.623,
|
| 351 |
-
"eval_steps_per_second": 0.836,
|
| 352 |
-
"eval_wer": 0.1378370706783991,
|
| 353 |
"step": 2500
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
-
"step": 2550
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 28.57,
|
| 363 |
-
"learning_rate": 0.00023816849816849818,
|
| 364 |
-
"loss": 0.8187,
|
| 365 |
"step": 2600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch":
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
-
"step": 2650
|
| 372 |
-
},
|
| 373 |
-
{
|
| 374 |
-
"epoch": 29.67,
|
| 375 |
-
"learning_rate": 0.00023450549450549453,
|
| 376 |
-
"loss": 0.8455,
|
| 377 |
"step": 2700
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 0.
|
| 383 |
-
"step": 2750
|
| 384 |
-
},
|
| 385 |
-
{
|
| 386 |
-
"epoch": 30.77,
|
| 387 |
-
"learning_rate": 0.00023084249084249085,
|
| 388 |
-
"loss": 0.8253,
|
| 389 |
"step": 2800
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
-
"step": 2850
|
| 396 |
-
},
|
| 397 |
-
{
|
| 398 |
-
"epoch": 31.86,
|
| 399 |
-
"learning_rate": 0.0002271794871794872,
|
| 400 |
-
"loss": 0.808,
|
| 401 |
"step": 2900
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
-
"step": 2950
|
| 408 |
-
},
|
| 409 |
-
{
|
| 410 |
-
"epoch": 32.96,
|
| 411 |
-
"learning_rate": 0.00022351648351648352,
|
| 412 |
-
"loss": 0.8081,
|
| 413 |
"step": 3000
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"eval_loss": 0.
|
| 418 |
-
"eval_runtime":
|
| 419 |
-
"eval_samples_per_second":
|
| 420 |
-
"eval_steps_per_second": 0.
|
| 421 |
-
"eval_wer": 0.
|
| 422 |
"step": 3000
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss": 0.
|
| 428 |
-
"step": 3050
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 34.07,
|
| 432 |
-
"learning_rate": 0.00021985347985347986,
|
| 433 |
-
"loss": 0.8035,
|
| 434 |
"step": 3100
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
| 440 |
-
"step": 3150
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 35.16,
|
| 444 |
-
"learning_rate": 0.00021619047619047619,
|
| 445 |
-
"loss": 0.7835,
|
| 446 |
"step": 3200
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss": 0.
|
| 452 |
-
"step": 3250
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 36.26,
|
| 456 |
-
"learning_rate": 0.00021252747252747253,
|
| 457 |
-
"loss": 0.7845,
|
| 458 |
"step": 3300
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
-
"step": 3350
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 37.36,
|
| 468 |
-
"learning_rate": 0.00020886446886446885,
|
| 469 |
-
"loss": 0.7933,
|
| 470 |
"step": 3400
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch":
|
| 474 |
-
"learning_rate":
|
| 475 |
-
"loss": 0.
|
| 476 |
-
"step": 3450
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 38.46,
|
| 480 |
-
"learning_rate": 0.0002052014652014652,
|
| 481 |
-
"loss": 0.7923,
|
| 482 |
"step": 3500
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"
|
| 487 |
-
"
|
| 488 |
-
"eval_samples_per_second": 26.493,
|
| 489 |
-
"eval_steps_per_second": 0.831,
|
| 490 |
-
"eval_wer": 0.12773204655123474,
|
| 491 |
-
"step": 3500
|
| 492 |
-
},
|
| 493 |
-
{
|
| 494 |
-
"epoch": 39.01,
|
| 495 |
-
"learning_rate": 0.00020336996336996338,
|
| 496 |
-
"loss": 0.7861,
|
| 497 |
-
"step": 3550
|
| 498 |
-
},
|
| 499 |
-
{
|
| 500 |
-
"epoch": 39.56,
|
| 501 |
-
"learning_rate": 0.00020157509157509157,
|
| 502 |
-
"loss": 0.767,
|
| 503 |
"step": 3600
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"learning_rate":
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"step": 3650
|
| 510 |
-
},
|
| 511 |
-
{
|
| 512 |
-
"epoch": 40.66,
|
| 513 |
-
"learning_rate": 0.0001979120879120879,
|
| 514 |
-
"loss": 0.7515,
|
| 515 |
"step": 3700
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 0.
|
| 521 |
-
"step": 3750
|
| 522 |
-
},
|
| 523 |
-
{
|
| 524 |
-
"epoch": 41.75,
|
| 525 |
-
"learning_rate": 0.00019424908424908423,
|
| 526 |
-
"loss": 0.7565,
|
| 527 |
"step": 3800
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch":
|
| 531 |
-
"learning_rate":
|
| 532 |
-
"loss": 0.
|
| 533 |
-
"step": 3850
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 42.85,
|
| 537 |
-
"learning_rate": 0.0001905860805860806,
|
| 538 |
-
"loss": 0.7547,
|
| 539 |
"step": 3900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch":
|
| 543 |
-
"learning_rate":
|
| 544 |
-
"loss": 0.
|
| 545 |
-
"step": 3950
|
| 546 |
-
},
|
| 547 |
-
{
|
| 548 |
-
"epoch": 43.95,
|
| 549 |
-
"learning_rate": 0.00018692307692307693,
|
| 550 |
-
"loss": 0.7349,
|
| 551 |
"step": 4000
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch":
|
| 555 |
-
"eval_loss": 0.
|
| 556 |
-
"eval_runtime":
|
| 557 |
-
"eval_samples_per_second":
|
| 558 |
-
"eval_steps_per_second": 0.
|
| 559 |
-
"eval_wer": 0.
|
| 560 |
"step": 4000
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
-
"step": 4050
|
| 567 |
-
},
|
| 568 |
-
{
|
| 569 |
-
"epoch": 45.05,
|
| 570 |
-
"learning_rate": 0.00018326007326007328,
|
| 571 |
-
"loss": 0.7442,
|
| 572 |
"step": 4100
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
-
"step": 4150
|
| 579 |
-
},
|
| 580 |
-
{
|
| 581 |
-
"epoch": 46.15,
|
| 582 |
-
"learning_rate": 0.0001795970695970696,
|
| 583 |
-
"loss": 0.7432,
|
| 584 |
"step": 4200
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
-
"step": 4250
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 47.25,
|
| 594 |
-
"learning_rate": 0.00017593406593406595,
|
| 595 |
-
"loss": 0.7358,
|
| 596 |
"step": 4300
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch":
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
-
"step": 4350
|
| 603 |
-
},
|
| 604 |
-
{
|
| 605 |
-
"epoch": 48.35,
|
| 606 |
-
"learning_rate": 0.00017227106227106227,
|
| 607 |
-
"loss": 0.7327,
|
| 608 |
"step": 4400
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch":
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
-
"step": 4450
|
| 615 |
-
},
|
| 616 |
-
{
|
| 617 |
-
"epoch": 49.45,
|
| 618 |
-
"learning_rate": 0.00016860805860805861,
|
| 619 |
-
"loss": 0.7199,
|
| 620 |
-
"step": 4500
|
| 621 |
-
},
|
| 622 |
-
{
|
| 623 |
-
"epoch": 49.45,
|
| 624 |
-
"eval_loss": 0.16171683371067047,
|
| 625 |
-
"eval_runtime": 181.6699,
|
| 626 |
-
"eval_samples_per_second": 26.658,
|
| 627 |
-
"eval_steps_per_second": 0.837,
|
| 628 |
-
"eval_wer": 0.12770366165200114,
|
| 629 |
"step": 4500
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"learning_rate":
|
| 634 |
-
"loss": 0.
|
| 635 |
-
"step": 4550
|
| 636 |
-
},
|
| 637 |
-
{
|
| 638 |
-
"epoch": 50.55,
|
| 639 |
-
"learning_rate": 0.00016494505494505496,
|
| 640 |
-
"loss": 0.7265,
|
| 641 |
"step": 4600
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"epoch":
|
| 645 |
-
"learning_rate":
|
| 646 |
-
"loss": 0.
|
| 647 |
-
"step": 4650
|
| 648 |
-
},
|
| 649 |
-
{
|
| 650 |
-
"epoch": 51.64,
|
| 651 |
-
"learning_rate": 0.00016128205128205128,
|
| 652 |
-
"loss": 0.7022,
|
| 653 |
"step": 4700
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch":
|
| 657 |
-
"learning_rate":
|
| 658 |
-
"loss": 0.
|
| 659 |
-
"step": 4750
|
| 660 |
-
},
|
| 661 |
-
{
|
| 662 |
-
"epoch": 52.74,
|
| 663 |
-
"learning_rate": 0.00015761904761904763,
|
| 664 |
-
"loss": 0.7239,
|
| 665 |
"step": 4800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
-
"epoch":
|
| 669 |
-
"learning_rate":
|
| 670 |
-
"loss": 0.
|
| 671 |
-
"step": 4850
|
| 672 |
-
},
|
| 673 |
-
{
|
| 674 |
-
"epoch": 53.84,
|
| 675 |
-
"learning_rate": 0.00015395604395604395,
|
| 676 |
-
"loss": 0.6951,
|
| 677 |
"step": 4900
|
| 678 |
},
|
| 679 |
{
|
| 680 |
-
"epoch":
|
| 681 |
-
"learning_rate":
|
| 682 |
-
"loss": 0.
|
| 683 |
-
"step": 4950
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 54.94,
|
| 687 |
-
"learning_rate": 0.0001502930402930403,
|
| 688 |
-
"loss": 0.7028,
|
| 689 |
"step": 5000
|
| 690 |
},
|
| 691 |
{
|
| 692 |
-
"epoch":
|
| 693 |
-
"eval_loss": 0.
|
| 694 |
-
"eval_runtime":
|
| 695 |
-
"eval_samples_per_second":
|
| 696 |
-
"eval_steps_per_second": 0.
|
| 697 |
-
"eval_wer": 0.
|
| 698 |
"step": 5000
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
-
"step": 5050
|
| 705 |
-
},
|
| 706 |
-
{
|
| 707 |
-
"epoch": 56.04,
|
| 708 |
-
"learning_rate": 0.00014663003663003662,
|
| 709 |
-
"loss": 0.6864,
|
| 710 |
"step": 5100
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
-
"step": 5150
|
| 717 |
-
},
|
| 718 |
-
{
|
| 719 |
-
"epoch": 57.14,
|
| 720 |
-
"learning_rate": 0.00014296703296703297,
|
| 721 |
-
"loss": 0.6759,
|
| 722 |
"step": 5200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
-
"step": 5250
|
| 729 |
-
},
|
| 730 |
-
{
|
| 731 |
-
"epoch": 58.24,
|
| 732 |
-
"learning_rate": 0.0001393040293040293,
|
| 733 |
-
"loss": 0.7114,
|
| 734 |
"step": 5300
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
-
"step": 5350
|
| 741 |
-
},
|
| 742 |
-
{
|
| 743 |
-
"epoch": 59.34,
|
| 744 |
-
"learning_rate": 0.00013564102564102566,
|
| 745 |
-
"loss": 0.6862,
|
| 746 |
"step": 5400
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"learning_rate":
|
| 751 |
-
"loss": 0.
|
| 752 |
-
"step": 5450
|
| 753 |
-
},
|
| 754 |
-
{
|
| 755 |
-
"epoch": 60.44,
|
| 756 |
-
"learning_rate": 0.00013197802197802198,
|
| 757 |
-
"loss": 0.6912,
|
| 758 |
"step": 5500
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch":
|
| 762 |
-
"
|
| 763 |
-
"
|
| 764 |
-
"eval_samples_per_second": 27.349,
|
| 765 |
-
"eval_steps_per_second": 0.858,
|
| 766 |
-
"eval_wer": 0.12489355662787398,
|
| 767 |
-
"step": 5500
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 60.98,
|
| 771 |
-
"learning_rate": 0.00013014652014652016,
|
| 772 |
-
"loss": 0.6743,
|
| 773 |
-
"step": 5550
|
| 774 |
-
},
|
| 775 |
-
{
|
| 776 |
-
"epoch": 61.54,
|
| 777 |
-
"learning_rate": 0.00012831501831501833,
|
| 778 |
-
"loss": 0.6683,
|
| 779 |
"step": 5600
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch":
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"loss": 0.
|
| 785 |
-
"step": 5650
|
| 786 |
-
},
|
| 787 |
-
{
|
| 788 |
-
"epoch": 62.63,
|
| 789 |
-
"learning_rate": 0.00012465201465201465,
|
| 790 |
-
"loss": 0.6583,
|
| 791 |
"step": 5700
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
-
"step": 5750
|
| 798 |
-
},
|
| 799 |
-
{
|
| 800 |
-
"epoch": 63.73,
|
| 801 |
-
"learning_rate": 0.000120989010989011,
|
| 802 |
-
"loss": 0.6566,
|
| 803 |
"step": 5800
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 0.
|
| 809 |
-
"step": 5850
|
| 810 |
-
},
|
| 811 |
-
{
|
| 812 |
-
"epoch": 64.83,
|
| 813 |
-
"learning_rate": 0.00011732600732600734,
|
| 814 |
-
"loss": 0.6524,
|
| 815 |
"step": 5900
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss": 0.
|
| 821 |
-
"step": 5950
|
| 822 |
-
},
|
| 823 |
-
{
|
| 824 |
-
"epoch": 65.93,
|
| 825 |
-
"learning_rate": 0.00011366300366300367,
|
| 826 |
-
"loss": 0.6492,
|
| 827 |
"step": 6000
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"eval_loss": 0.
|
| 832 |
-
"eval_runtime":
|
| 833 |
-
"eval_samples_per_second":
|
| 834 |
-
"eval_steps_per_second": 0.
|
| 835 |
-
"eval_wer": 0.
|
| 836 |
"step": 6000
|
| 837 |
},
|
| 838 |
{
|
| 839 |
-
"epoch":
|
| 840 |
-
"learning_rate":
|
| 841 |
-
"loss": 0.
|
| 842 |
-
"step": 6050
|
| 843 |
-
},
|
| 844 |
-
{
|
| 845 |
-
"epoch": 67.03,
|
| 846 |
-
"learning_rate": 0.00011,
|
| 847 |
-
"loss": 0.6542,
|
| 848 |
"step": 6100
|
| 849 |
},
|
| 850 |
{
|
| 851 |
-
"epoch":
|
| 852 |
-
"learning_rate":
|
| 853 |
-
"loss": 0.
|
| 854 |
-
"step": 6150
|
| 855 |
-
},
|
| 856 |
-
{
|
| 857 |
-
"epoch": 68.13,
|
| 858 |
-
"learning_rate": 0.00010633699633699634,
|
| 859 |
-
"loss": 0.6513,
|
| 860 |
"step": 6200
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"epoch":
|
| 864 |
-
"learning_rate":
|
| 865 |
-
"loss": 0.
|
| 866 |
-
"step": 6250
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 69.23,
|
| 870 |
-
"learning_rate": 0.00010267399267399267,
|
| 871 |
-
"loss": 0.6507,
|
| 872 |
"step": 6300
|
| 873 |
},
|
| 874 |
{
|
| 875 |
-
"epoch":
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss": 0.
|
| 878 |
-
"step": 6350
|
| 879 |
-
},
|
| 880 |
-
{
|
| 881 |
-
"epoch": 70.33,
|
| 882 |
-
"learning_rate": 9.901098901098901e-05,
|
| 883 |
-
"loss": 0.6525,
|
| 884 |
"step": 6400
|
| 885 |
},
|
| 886 |
{
|
| 887 |
-
"epoch":
|
| 888 |
-
"learning_rate":
|
| 889 |
-
"loss": 0.
|
| 890 |
-
"step": 6450
|
| 891 |
-
},
|
| 892 |
-
{
|
| 893 |
-
"epoch": 71.43,
|
| 894 |
-
"learning_rate": 9.534798534798534e-05,
|
| 895 |
-
"loss": 0.6407,
|
| 896 |
"step": 6500
|
| 897 |
},
|
| 898 |
{
|
| 899 |
-
"epoch":
|
| 900 |
-
"
|
| 901 |
-
"
|
| 902 |
-
"eval_samples_per_second": 27.096,
|
| 903 |
-
"eval_steps_per_second": 0.85,
|
| 904 |
-
"eval_wer": 0.12398523985239852,
|
| 905 |
-
"step": 6500
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 71.97,
|
| 909 |
-
"learning_rate": 9.351648351648353e-05,
|
| 910 |
-
"loss": 0.6274,
|
| 911 |
-
"step": 6550
|
| 912 |
-
},
|
| 913 |
-
{
|
| 914 |
-
"epoch": 72.52,
|
| 915 |
-
"learning_rate": 9.168498168498169e-05,
|
| 916 |
-
"loss": 0.6338,
|
| 917 |
"step": 6600
|
| 918 |
},
|
| 919 |
{
|
| 920 |
-
"epoch":
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss": 0.
|
| 923 |
-
"step": 6650
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 73.62,
|
| 927 |
-
"learning_rate": 8.805860805860807e-05,
|
| 928 |
-
"loss": 0.6202,
|
| 929 |
"step": 6700
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch":
|
| 933 |
-
"learning_rate":
|
| 934 |
-
"loss": 0.
|
| 935 |
-
"step": 6750
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 74.72,
|
| 939 |
-
"learning_rate": 8.43956043956044e-05,
|
| 940 |
-
"loss": 0.623,
|
| 941 |
"step": 6800
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch":
|
| 945 |
-
"learning_rate":
|
| 946 |
-
"loss": 0.
|
| 947 |
-
"step": 6850
|
| 948 |
-
},
|
| 949 |
-
{
|
| 950 |
-
"epoch": 75.82,
|
| 951 |
-
"learning_rate": 8.073260073260073e-05,
|
| 952 |
-
"loss": 0.6136,
|
| 953 |
"step": 6900
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch":
|
| 957 |
-
"learning_rate":
|
| 958 |
-
"loss": 0.
|
| 959 |
-
"step": 6950
|
| 960 |
-
},
|
| 961 |
-
{
|
| 962 |
-
"epoch": 76.92,
|
| 963 |
-
"learning_rate": 7.706959706959707e-05,
|
| 964 |
-
"loss": 0.6222,
|
| 965 |
"step": 7000
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"eval_loss": 0.
|
| 970 |
-
"eval_runtime":
|
| 971 |
-
"eval_samples_per_second":
|
| 972 |
-
"eval_steps_per_second": 0.
|
| 973 |
-
"eval_wer": 0.
|
| 974 |
"step": 7000
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"learning_rate":
|
| 979 |
-
"loss": 0.
|
| 980 |
-
"step": 7050
|
| 981 |
-
},
|
| 982 |
-
{
|
| 983 |
-
"epoch": 78.02,
|
| 984 |
-
"learning_rate": 7.34065934065934e-05,
|
| 985 |
-
"loss": 0.6338,
|
| 986 |
"step": 7100
|
| 987 |
},
|
| 988 |
{
|
| 989 |
-
"epoch":
|
| 990 |
-
"learning_rate":
|
| 991 |
-
"loss": 0.
|
| 992 |
-
"step": 7150
|
| 993 |
-
},
|
| 994 |
-
{
|
| 995 |
-
"epoch": 79.12,
|
| 996 |
-
"learning_rate": 6.974358974358974e-05,
|
| 997 |
-
"loss": 0.6245,
|
| 998 |
"step": 7200
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
-
"epoch":
|
| 1002 |
-
"learning_rate":
|
| 1003 |
-
"loss": 0.
|
| 1004 |
-
"step": 7250
|
| 1005 |
-
},
|
| 1006 |
-
{
|
| 1007 |
-
"epoch": 80.22,
|
| 1008 |
-
"learning_rate": 6.608058608058607e-05,
|
| 1009 |
-
"loss": 0.6138,
|
| 1010 |
"step": 7300
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
-
"epoch":
|
| 1014 |
-
"learning_rate":
|
| 1015 |
-
"loss": 0.
|
| 1016 |
-
"step": 7350
|
| 1017 |
-
},
|
| 1018 |
-
{
|
| 1019 |
-
"epoch": 81.32,
|
| 1020 |
-
"learning_rate": 6.241758241758242e-05,
|
| 1021 |
-
"loss": 0.6275,
|
| 1022 |
"step": 7400
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
-
"epoch":
|
| 1026 |
-
"learning_rate":
|
| 1027 |
-
"loss": 0.
|
| 1028 |
-
"step": 7450
|
| 1029 |
-
},
|
| 1030 |
-
{
|
| 1031 |
-
"epoch": 82.42,
|
| 1032 |
-
"learning_rate": 5.8754578754578754e-05,
|
| 1033 |
-
"loss": 0.6039,
|
| 1034 |
"step": 7500
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"
|
| 1039 |
-
"
|
| 1040 |
-
"eval_samples_per_second": 27.094,
|
| 1041 |
-
"eval_steps_per_second": 0.85,
|
| 1042 |
-
"eval_wer": 0.12489355662787398,
|
| 1043 |
-
"step": 7500
|
| 1044 |
-
},
|
| 1045 |
-
{
|
| 1046 |
-
"epoch": 82.96,
|
| 1047 |
-
"learning_rate": 5.692307692307693e-05,
|
| 1048 |
-
"loss": 0.6063,
|
| 1049 |
-
"step": 7550
|
| 1050 |
-
},
|
| 1051 |
-
{
|
| 1052 |
-
"epoch": 83.51,
|
| 1053 |
-
"learning_rate": 5.5091575091575095e-05,
|
| 1054 |
-
"loss": 0.6039,
|
| 1055 |
"step": 7600
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
-
"epoch":
|
| 1059 |
-
"learning_rate":
|
| 1060 |
-
"loss": 0.
|
| 1061 |
-
"step": 7650
|
| 1062 |
-
},
|
| 1063 |
-
{
|
| 1064 |
-
"epoch": 84.61,
|
| 1065 |
-
"learning_rate": 5.142857142857143e-05,
|
| 1066 |
-
"loss": 0.5922,
|
| 1067 |
"step": 7700
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
-
"epoch":
|
| 1071 |
-
"learning_rate":
|
| 1072 |
-
"loss": 0.
|
| 1073 |
-
"step": 7750
|
| 1074 |
-
},
|
| 1075 |
-
{
|
| 1076 |
-
"epoch": 85.71,
|
| 1077 |
-
"learning_rate": 4.776556776556776e-05,
|
| 1078 |
-
"loss": 0.5984,
|
| 1079 |
"step": 7800
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
-
"epoch":
|
| 1083 |
-
"learning_rate":
|
| 1084 |
-
"loss": 0.
|
| 1085 |
-
"step": 7850
|
| 1086 |
-
},
|
| 1087 |
-
{
|
| 1088 |
-
"epoch": 86.81,
|
| 1089 |
-
"learning_rate": 4.41025641025641e-05,
|
| 1090 |
-
"loss": 0.5927,
|
| 1091 |
"step": 7900
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
-
"epoch":
|
| 1095 |
-
"learning_rate":
|
| 1096 |
-
"loss": 0.
|
| 1097 |
-
"step": 7950
|
| 1098 |
-
},
|
| 1099 |
-
{
|
| 1100 |
-
"epoch": 87.91,
|
| 1101 |
-
"learning_rate": 4.0439560439560445e-05,
|
| 1102 |
-
"loss": 0.5928,
|
| 1103 |
"step": 8000
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
-
"epoch":
|
| 1107 |
-
"eval_loss": 0.
|
| 1108 |
-
"eval_runtime":
|
| 1109 |
-
"eval_samples_per_second":
|
| 1110 |
-
"eval_steps_per_second": 0.
|
| 1111 |
-
"eval_wer": 0.
|
| 1112 |
"step": 8000
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"epoch":
|
| 1116 |
-
"learning_rate":
|
| 1117 |
-
"loss": 0.
|
| 1118 |
-
"step": 8050
|
| 1119 |
-
},
|
| 1120 |
-
{
|
| 1121 |
-
"epoch": 89.01,
|
| 1122 |
-
"learning_rate": 3.677655677655678e-05,
|
| 1123 |
-
"loss": 0.5962,
|
| 1124 |
"step": 8100
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
-
"epoch":
|
| 1128 |
-
"learning_rate":
|
| 1129 |
-
"loss": 0.
|
| 1130 |
-
"step": 8150
|
| 1131 |
-
},
|
| 1132 |
-
{
|
| 1133 |
-
"epoch": 90.11,
|
| 1134 |
-
"learning_rate": 3.3113553113553114e-05,
|
| 1135 |
-
"loss": 0.6024,
|
| 1136 |
"step": 8200
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch":
|
| 1140 |
-
"learning_rate":
|
| 1141 |
-
"loss": 0.
|
| 1142 |
-
"step": 8250
|
| 1143 |
-
},
|
| 1144 |
-
{
|
| 1145 |
-
"epoch": 91.21,
|
| 1146 |
-
"learning_rate": 2.945054945054945e-05,
|
| 1147 |
-
"loss": 0.6001,
|
| 1148 |
"step": 8300
|
| 1149 |
},
|
| 1150 |
{
|
| 1151 |
-
"epoch":
|
| 1152 |
-
"learning_rate":
|
| 1153 |
-
"loss": 0.
|
| 1154 |
-
"step": 8350
|
| 1155 |
-
},
|
| 1156 |
-
{
|
| 1157 |
-
"epoch": 92.31,
|
| 1158 |
-
"learning_rate": 2.578754578754579e-05,
|
| 1159 |
-
"loss": 0.6078,
|
| 1160 |
"step": 8400
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
-
"epoch":
|
| 1164 |
-
"learning_rate":
|
| 1165 |
-
"loss": 0.
|
| 1166 |
-
"step": 8450
|
| 1167 |
-
},
|
| 1168 |
-
{
|
| 1169 |
-
"epoch": 93.4,
|
| 1170 |
-
"learning_rate": 2.2124542124542124e-05,
|
| 1171 |
-
"loss": 0.6022,
|
| 1172 |
-
"step": 8500
|
| 1173 |
-
},
|
| 1174 |
-
{
|
| 1175 |
-
"epoch": 93.4,
|
| 1176 |
-
"eval_loss": 0.1596660017967224,
|
| 1177 |
-
"eval_runtime": 178.6421,
|
| 1178 |
-
"eval_samples_per_second": 27.11,
|
| 1179 |
-
"eval_steps_per_second": 0.851,
|
| 1180 |
-
"eval_wer": 0.1213170593244394,
|
| 1181 |
"step": 8500
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
-
"epoch":
|
| 1185 |
-
"learning_rate":
|
| 1186 |
-
"loss": 0.
|
| 1187 |
-
"step": 8550
|
| 1188 |
-
},
|
| 1189 |
-
{
|
| 1190 |
-
"epoch": 94.5,
|
| 1191 |
-
"learning_rate": 1.8461538461538465e-05,
|
| 1192 |
-
"loss": 0.5839,
|
| 1193 |
"step": 8600
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
-
"epoch":
|
| 1197 |
-
"learning_rate":
|
| 1198 |
-
"loss": 0.
|
| 1199 |
-
"step":
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
-
"epoch":
|
| 1203 |
-
"learning_rate":
|
| 1204 |
-
"loss": 0.
|
| 1205 |
-
"step":
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
-
"epoch":
|
| 1209 |
-
"learning_rate":
|
| 1210 |
-
"loss": 0.
|
| 1211 |
-
"step":
|
| 1212 |
},
|
| 1213 |
{
|
| 1214 |
-
"epoch":
|
| 1215 |
-
"learning_rate":
|
| 1216 |
-
"loss": 0.
|
| 1217 |
-
"step":
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
-
"epoch":
|
| 1221 |
-
"
|
| 1222 |
-
"
|
| 1223 |
-
"
|
|
|
|
|
|
|
|
|
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
-
"epoch":
|
| 1227 |
-
"learning_rate":
|
| 1228 |
-
"loss": 0.
|
| 1229 |
-
"step":
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
-
"epoch":
|
| 1233 |
-
"learning_rate":
|
| 1234 |
-
"loss": 0.
|
| 1235 |
-
"step":
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
-
"epoch":
|
| 1239 |
-
"learning_rate":
|
| 1240 |
-
"loss": 0.
|
| 1241 |
-
"step":
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
-
"epoch":
|
| 1245 |
-
"
|
| 1246 |
-
"
|
| 1247 |
-
"
|
| 1248 |
-
"eval_steps_per_second": 0.853,
|
| 1249 |
-
"eval_wer": 0.11989781436275901,
|
| 1250 |
-
"step": 9000
|
| 1251 |
},
|
| 1252 |
{
|
| 1253 |
-
"epoch":
|
| 1254 |
-
"learning_rate":
|
| 1255 |
-
"loss": 0.
|
| 1256 |
-
"step":
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch":
|
| 1260 |
-
"learning_rate":
|
| 1261 |
-
"loss": 0.
|
| 1262 |
-
"step":
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
-
"epoch":
|
| 1266 |
-
"step":
|
| 1267 |
-
"total_flos": 1.
|
| 1268 |
-
"train_loss": 1.
|
| 1269 |
-
"train_runtime":
|
| 1270 |
-
"train_samples_per_second": 22.
|
| 1271 |
-
"train_steps_per_second": 0.
|
| 1272 |
}
|
| 1273 |
],
|
| 1274 |
-
"max_steps":
|
| 1275 |
"num_train_epochs": 100,
|
| 1276 |
-
"total_flos": 1.
|
| 1277 |
"trial_name": null,
|
| 1278 |
"trial_params": null
|
| 1279 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.9974025974026,
|
| 5 |
+
"global_step": 9600,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 1.04,
|
| 12 |
+
"learning_rate": 3.828125e-06,
|
| 13 |
+
"loss": 15.104,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 2.08,
|
| 18 |
+
"learning_rate": 7.6953125e-06,
|
| 19 |
+
"loss": 12.2149,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 3.12,
|
| 24 |
+
"learning_rate": 1.16015625e-05,
|
| 25 |
+
"loss": 4.9885,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 4.17,
|
| 30 |
+
"learning_rate": 1.55078125e-05,
|
| 31 |
+
"loss": 3.3788,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 5.21,
|
| 36 |
+
"learning_rate": 1.94140625e-05,
|
| 37 |
+
"loss": 3.1705,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 6.25,
|
| 42 |
+
"learning_rate": 2.3320312499999995e-05,
|
| 43 |
+
"loss": 3.11,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 600
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 7.29,
|
| 48 |
+
"learning_rate": 2.72265625e-05,
|
| 49 |
+
"loss": 3.0526,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
"step": 700
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 8.33,
|
| 54 |
+
"learning_rate": 3.11328125e-05,
|
| 55 |
+
"loss": 3.0113,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
"step": 800
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 9.37,
|
| 60 |
+
"learning_rate": 3.5039062499999995e-05,
|
| 61 |
+
"loss": 2.9601,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
"step": 900
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 10.42,
|
| 66 |
+
"learning_rate": 3.89453125e-05,
|
| 67 |
+
"loss": 2.9099,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
"step": 1000
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 10.42,
|
| 72 |
+
"eval_loss": 2.8369038105010986,
|
| 73 |
+
"eval_runtime": 198.5866,
|
| 74 |
+
"eval_samples_per_second": 25.405,
|
| 75 |
+
"eval_steps_per_second": 0.796,
|
| 76 |
+
"eval_wer": 1.0,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 11.46,
|
| 81 |
+
"learning_rate": 4.28515625e-05,
|
| 82 |
+
"loss": 2.6653,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
"step": 1100
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 12.5,
|
| 87 |
+
"learning_rate": 4.675781249999999e-05,
|
| 88 |
+
"loss": 1.8958,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
"step": 1200
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 13.54,
|
| 93 |
+
"learning_rate": 5.0664062499999996e-05,
|
| 94 |
+
"loss": 1.4812,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
"step": 1300
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 14.58,
|
| 99 |
+
"learning_rate": 5.45703125e-05,
|
| 100 |
+
"loss": 1.3358,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 101 |
"step": 1400
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 15.62,
|
| 105 |
+
"learning_rate": 5.8476562499999996e-05,
|
| 106 |
+
"loss": 1.2522,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
"step": 1500
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 16.66,
|
| 111 |
+
"learning_rate": 6.238281249999999e-05,
|
| 112 |
+
"loss": 1.1919,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
"step": 1600
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 17.71,
|
| 117 |
+
"learning_rate": 6.62890625e-05,
|
| 118 |
+
"loss": 1.1663,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 119 |
"step": 1700
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 18.75,
|
| 123 |
+
"learning_rate": 7.019531249999999e-05,
|
| 124 |
+
"loss": 1.1289,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
"step": 1800
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 19.79,
|
| 129 |
+
"learning_rate": 7.410156249999999e-05,
|
| 130 |
+
"loss": 1.0902,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 131 |
"step": 1900
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 20.83,
|
| 135 |
+
"learning_rate": 7.4248046875e-05,
|
| 136 |
+
"loss": 1.0745,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 137 |
"step": 2000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 20.83,
|
| 141 |
+
"eval_loss": 0.19572903215885162,
|
| 142 |
+
"eval_runtime": 202.06,
|
| 143 |
+
"eval_samples_per_second": 24.968,
|
| 144 |
+
"eval_steps_per_second": 0.782,
|
| 145 |
+
"eval_wer": 0.16725371193237237,
|
| 146 |
"step": 2000
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 21.87,
|
| 150 |
+
"learning_rate": 7.327148437499999e-05,
|
| 151 |
+
"loss": 1.0485,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
"step": 2100
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 22.91,
|
| 156 |
+
"learning_rate": 7.2294921875e-05,
|
| 157 |
+
"loss": 1.0291,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
"step": 2200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 23.96,
|
| 162 |
+
"learning_rate": 7.1318359375e-05,
|
| 163 |
+
"loss": 1.007,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
"step": 2300
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 25.0,
|
| 168 |
+
"learning_rate": 7.0341796875e-05,
|
| 169 |
+
"loss": 1.0008,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 170 |
"step": 2400
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 26.04,
|
| 174 |
+
"learning_rate": 6.9365234375e-05,
|
| 175 |
+
"loss": 0.988,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
"step": 2500
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 27.08,
|
| 180 |
+
"learning_rate": 6.8388671875e-05,
|
| 181 |
+
"loss": 0.9766,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 182 |
"step": 2600
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 28.12,
|
| 186 |
+
"learning_rate": 6.7412109375e-05,
|
| 187 |
+
"loss": 0.9663,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 188 |
"step": 2700
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 29.17,
|
| 192 |
+
"learning_rate": 6.6435546875e-05,
|
| 193 |
+
"loss": 0.9539,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 194 |
"step": 2800
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 30.21,
|
| 198 |
+
"learning_rate": 6.545898437499999e-05,
|
| 199 |
+
"loss": 0.9479,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
"step": 2900
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 31.25,
|
| 204 |
+
"learning_rate": 6.448242187499999e-05,
|
| 205 |
+
"loss": 0.934,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 206 |
"step": 3000
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 31.25,
|
| 210 |
+
"eval_loss": 0.1579357087612152,
|
| 211 |
+
"eval_runtime": 198.7733,
|
| 212 |
+
"eval_samples_per_second": 25.381,
|
| 213 |
+
"eval_steps_per_second": 0.795,
|
| 214 |
+
"eval_wer": 0.1388858784003468,
|
| 215 |
"step": 3000
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 32.29,
|
| 219 |
+
"learning_rate": 6.3505859375e-05,
|
| 220 |
+
"loss": 0.9285,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 221 |
"step": 3100
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 33.33,
|
| 225 |
+
"learning_rate": 6.252929687499999e-05,
|
| 226 |
+
"loss": 0.9121,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 227 |
"step": 3200
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 34.37,
|
| 231 |
+
"learning_rate": 6.155273437499999e-05,
|
| 232 |
+
"loss": 0.9016,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 233 |
"step": 3300
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 35.42,
|
| 237 |
+
"learning_rate": 6.0576171875e-05,
|
| 238 |
+
"loss": 0.9023,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
"step": 3400
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 36.46,
|
| 243 |
+
"learning_rate": 5.9599609374999994e-05,
|
| 244 |
+
"loss": 0.9004,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 245 |
"step": 3500
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 37.5,
|
| 249 |
+
"learning_rate": 5.862304687499999e-05,
|
| 250 |
+
"loss": 0.8844,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 251 |
"step": 3600
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 38.54,
|
| 255 |
+
"learning_rate": 5.7646484375e-05,
|
| 256 |
+
"loss": 0.8771,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 257 |
"step": 3700
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 39.58,
|
| 261 |
+
"learning_rate": 5.6669921875e-05,
|
| 262 |
+
"loss": 0.876,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 263 |
"step": 3800
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 40.62,
|
| 267 |
+
"learning_rate": 5.569335937499999e-05,
|
| 268 |
+
"loss": 0.8708,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 269 |
"step": 3900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 41.66,
|
| 273 |
+
"learning_rate": 5.4716796874999997e-05,
|
| 274 |
+
"loss": 0.8691,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 275 |
"step": 4000
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 41.66,
|
| 279 |
+
"eval_loss": 0.14571049809455872,
|
| 280 |
+
"eval_runtime": 195.237,
|
| 281 |
+
"eval_samples_per_second": 25.84,
|
| 282 |
+
"eval_steps_per_second": 0.809,
|
| 283 |
+
"eval_wer": 0.12899642353961202,
|
| 284 |
"step": 4000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 42.71,
|
| 288 |
+
"learning_rate": 5.3740234374999996e-05,
|
| 289 |
+
"loss": 0.8624,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 290 |
"step": 4100
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 43.75,
|
| 294 |
+
"learning_rate": 5.2763671874999995e-05,
|
| 295 |
+
"loss": 0.8556,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
"step": 4200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 44.79,
|
| 300 |
+
"learning_rate": 5.1787109375e-05,
|
| 301 |
+
"loss": 0.8607,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 302 |
"step": 4300
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 45.83,
|
| 306 |
+
"learning_rate": 5.0810546875e-05,
|
| 307 |
+
"loss": 0.8536,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 308 |
"step": 4400
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 46.87,
|
| 312 |
+
"learning_rate": 4.983398437499999e-05,
|
| 313 |
+
"loss": 0.8493,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 314 |
"step": 4500
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 47.91,
|
| 318 |
+
"learning_rate": 4.8857421875e-05,
|
| 319 |
+
"loss": 0.8456,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 320 |
"step": 4600
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 48.96,
|
| 324 |
+
"learning_rate": 4.7880859375e-05,
|
| 325 |
+
"loss": 0.8333,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 326 |
"step": 4700
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 50.0,
|
| 330 |
+
"learning_rate": 4.6904296874999996e-05,
|
| 331 |
+
"loss": 0.8346,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 332 |
"step": 4800
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 51.04,
|
| 336 |
+
"learning_rate": 4.5927734375e-05,
|
| 337 |
+
"loss": 0.8403,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 338 |
"step": 4900
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 52.08,
|
| 342 |
+
"learning_rate": 4.4951171874999995e-05,
|
| 343 |
+
"loss": 0.8328,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 344 |
"step": 5000
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 52.08,
|
| 348 |
+
"eval_loss": 0.14348936080932617,
|
| 349 |
+
"eval_runtime": 197.7739,
|
| 350 |
+
"eval_samples_per_second": 25.509,
|
| 351 |
+
"eval_steps_per_second": 0.799,
|
| 352 |
+
"eval_wer": 0.12054297171344966,
|
| 353 |
"step": 5000
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 53.12,
|
| 357 |
+
"learning_rate": 4.3974609374999994e-05,
|
| 358 |
+
"loss": 0.8275,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 359 |
"step": 5100
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 54.17,
|
| 363 |
+
"learning_rate": 4.2998046875e-05,
|
| 364 |
+
"loss": 0.8262,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 365 |
"step": 5200
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 55.21,
|
| 369 |
+
"learning_rate": 4.2021484375e-05,
|
| 370 |
+
"loss": 0.8167,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 371 |
"step": 5300
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 56.25,
|
| 375 |
+
"learning_rate": 4.1044921875e-05,
|
| 376 |
+
"loss": 0.8194,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 377 |
"step": 5400
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 57.29,
|
| 381 |
+
"learning_rate": 4.0068359375e-05,
|
| 382 |
+
"loss": 0.8192,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 383 |
"step": 5500
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 58.33,
|
| 387 |
+
"learning_rate": 3.9091796874999996e-05,
|
| 388 |
+
"loss": 0.8176,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 389 |
"step": 5600
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 59.37,
|
| 393 |
+
"learning_rate": 3.8115234374999995e-05,
|
| 394 |
+
"loss": 0.8115,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 395 |
"step": 5700
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 60.42,
|
| 399 |
+
"learning_rate": 3.7138671874999994e-05,
|
| 400 |
+
"loss": 0.8129,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 401 |
"step": 5800
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 61.46,
|
| 405 |
+
"learning_rate": 3.6162109375e-05,
|
| 406 |
+
"loss": 0.8052,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 407 |
"step": 5900
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 62.5,
|
| 411 |
+
"learning_rate": 3.518554687499999e-05,
|
| 412 |
+
"loss": 0.8068,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 413 |
"step": 6000
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 62.5,
|
| 417 |
+
"eval_loss": 0.13501569628715515,
|
| 418 |
+
"eval_runtime": 207.5786,
|
| 419 |
+
"eval_samples_per_second": 24.304,
|
| 420 |
+
"eval_steps_per_second": 0.761,
|
| 421 |
+
"eval_wer": 0.11913406307575593,
|
| 422 |
"step": 6000
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 63.54,
|
| 426 |
+
"learning_rate": 3.4208984375e-05,
|
| 427 |
+
"loss": 0.8004,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 428 |
"step": 6100
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 64.58,
|
| 432 |
+
"learning_rate": 3.32421875e-05,
|
| 433 |
+
"loss": 0.7904,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 434 |
"step": 6200
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 65.62,
|
| 438 |
+
"learning_rate": 3.2265625e-05,
|
| 439 |
+
"loss": 0.7947,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 440 |
"step": 6300
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 66.66,
|
| 444 |
+
"learning_rate": 3.12890625e-05,
|
| 445 |
+
"loss": 0.7981,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 446 |
"step": 6400
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 67.71,
|
| 450 |
+
"learning_rate": 3.0312499999999998e-05,
|
| 451 |
+
"loss": 0.8018,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 452 |
"step": 6500
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 68.75,
|
| 456 |
+
"learning_rate": 2.93359375e-05,
|
| 457 |
+
"loss": 0.7922,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 458 |
"step": 6600
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 69.79,
|
| 462 |
+
"learning_rate": 2.8359374999999996e-05,
|
| 463 |
+
"loss": 0.796,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 464 |
"step": 6700
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 70.83,
|
| 468 |
+
"learning_rate": 2.73828125e-05,
|
| 469 |
+
"loss": 0.7887,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 470 |
"step": 6800
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 71.87,
|
| 474 |
+
"learning_rate": 2.6406249999999998e-05,
|
| 475 |
+
"loss": 0.7796,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 476 |
"step": 6900
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 72.91,
|
| 480 |
+
"learning_rate": 2.5429687499999997e-05,
|
| 481 |
+
"loss": 0.7822,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 482 |
"step": 7000
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 72.91,
|
| 486 |
+
"eval_loss": 0.1346774846315384,
|
| 487 |
+
"eval_runtime": 195.6804,
|
| 488 |
+
"eval_samples_per_second": 25.782,
|
| 489 |
+
"eval_steps_per_second": 0.807,
|
| 490 |
+
"eval_wer": 0.11550341389400672,
|
| 491 |
"step": 7000
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 73.96,
|
| 495 |
+
"learning_rate": 2.4453125e-05,
|
| 496 |
+
"loss": 0.7882,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 497 |
"step": 7100
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 75.0,
|
| 501 |
+
"learning_rate": 2.34765625e-05,
|
| 502 |
+
"loss": 0.7821,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 503 |
"step": 7200
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 76.04,
|
| 507 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 508 |
+
"loss": 0.7814,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 509 |
"step": 7300
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 77.08,
|
| 513 |
+
"learning_rate": 2.1523437499999997e-05,
|
| 514 |
+
"loss": 0.784,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 515 |
"step": 7400
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 78.12,
|
| 519 |
+
"learning_rate": 2.0546875e-05,
|
| 520 |
+
"loss": 0.7772,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 521 |
"step": 7500
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 79.17,
|
| 525 |
+
"learning_rate": 1.95703125e-05,
|
| 526 |
+
"loss": 0.7785,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 527 |
"step": 7600
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 80.21,
|
| 531 |
+
"learning_rate": 1.8593749999999998e-05,
|
| 532 |
+
"loss": 0.7731,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 533 |
"step": 7700
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 81.25,
|
| 537 |
+
"learning_rate": 1.7617187499999997e-05,
|
| 538 |
+
"loss": 0.7726,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 539 |
"step": 7800
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 82.29,
|
| 543 |
+
"learning_rate": 1.6640624999999996e-05,
|
| 544 |
+
"loss": 0.7765,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 545 |
"step": 7900
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 83.33,
|
| 549 |
+
"learning_rate": 1.56640625e-05,
|
| 550 |
+
"loss": 0.7769,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 551 |
"step": 8000
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 83.33,
|
| 555 |
+
"eval_loss": 0.1320926398038864,
|
| 556 |
+
"eval_runtime": 197.3103,
|
| 557 |
+
"eval_samples_per_second": 25.569,
|
| 558 |
+
"eval_steps_per_second": 0.801,
|
| 559 |
+
"eval_wer": 0.11306491817492142,
|
| 560 |
"step": 8000
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 84.37,
|
| 564 |
+
"learning_rate": 1.4687499999999998e-05,
|
| 565 |
+
"loss": 0.7713,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 566 |
"step": 8100
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 85.42,
|
| 570 |
+
"learning_rate": 1.3710937499999999e-05,
|
| 571 |
+
"loss": 0.773,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 572 |
"step": 8200
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 86.46,
|
| 576 |
+
"learning_rate": 1.2734375e-05,
|
| 577 |
+
"loss": 0.7668,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 578 |
"step": 8300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 87.5,
|
| 582 |
+
"learning_rate": 1.1757812499999999e-05,
|
| 583 |
+
"loss": 0.7766,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 584 |
"step": 8400
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 88.54,
|
| 588 |
+
"learning_rate": 1.0781249999999998e-05,
|
| 589 |
+
"loss": 0.7596,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 590 |
"step": 8500
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 89.58,
|
| 594 |
+
"learning_rate": 9.8046875e-06,
|
| 595 |
+
"loss": 0.7561,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 596 |
"step": 8600
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 90.62,
|
| 600 |
+
"learning_rate": 8.837890624999999e-06,
|
| 601 |
+
"loss": 0.7585,
|
| 602 |
+
"step": 8700
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 91.66,
|
| 606 |
+
"learning_rate": 7.861328124999998e-06,
|
| 607 |
+
"loss": 0.7524,
|
| 608 |
+
"step": 8800
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 92.71,
|
| 612 |
+
"learning_rate": 6.884765624999999e-06,
|
| 613 |
+
"loss": 0.7591,
|
| 614 |
+
"step": 8900
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 93.75,
|
| 618 |
+
"learning_rate": 5.908203125e-06,
|
| 619 |
+
"loss": 0.7678,
|
| 620 |
+
"step": 9000
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 93.75,
|
| 624 |
+
"eval_loss": 0.13205420970916748,
|
| 625 |
+
"eval_runtime": 199.4472,
|
| 626 |
+
"eval_samples_per_second": 25.295,
|
| 627 |
+
"eval_steps_per_second": 0.792,
|
| 628 |
+
"eval_wer": 0.11146634875907663,
|
| 629 |
+
"step": 9000
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 94.79,
|
| 633 |
+
"learning_rate": 4.931640624999999e-06,
|
| 634 |
+
"loss": 0.7609,
|
| 635 |
+
"step": 9100
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 95.83,
|
| 639 |
+
"learning_rate": 3.955078125e-06,
|
| 640 |
+
"loss": 0.7611,
|
| 641 |
+
"step": 9200
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 96.87,
|
| 645 |
+
"learning_rate": 2.9785156249999998e-06,
|
| 646 |
+
"loss": 0.7692,
|
| 647 |
+
"step": 9300
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 97.91,
|
| 651 |
+
"learning_rate": 2.0019531249999997e-06,
|
| 652 |
+
"loss": 0.7576,
|
| 653 |
+
"step": 9400
|
|
|
|
|
|
|
|
|
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 98.96,
|
| 657 |
+
"learning_rate": 1.025390625e-06,
|
| 658 |
+
"loss": 0.7552,
|
| 659 |
+
"step": 9500
|
| 660 |
},
|
| 661 |
{
|
| 662 |
+
"epoch": 100.0,
|
| 663 |
+
"learning_rate": 4.8828124999999996e-08,
|
| 664 |
+
"loss": 0.771,
|
| 665 |
+
"step": 9600
|
| 666 |
},
|
| 667 |
{
|
| 668 |
+
"epoch": 100.0,
|
| 669 |
+
"step": 9600,
|
| 670 |
+
"total_flos": 1.4910750173005185e+20,
|
| 671 |
+
"train_loss": 1.3699857131640116,
|
| 672 |
+
"train_runtime": 54887.9396,
|
| 673 |
+
"train_samples_per_second": 22.422,
|
| 674 |
+
"train_steps_per_second": 0.175
|
| 675 |
}
|
| 676 |
],
|
| 677 |
+
"max_steps": 9600,
|
| 678 |
"num_train_epochs": 100,
|
| 679 |
+
"total_flos": 1.4910750173005185e+20,
|
| 680 |
"trial_name": null,
|
| 681 |
"trial_params": null
|
| 682 |
}
|