End of training
Browse files- all_results.json +10 -10
- eval_results.json +6 -6
- runs/Dec19_09-48-52_nipa2022-65403/events.out.tfevents.1671433867.nipa2022-65403.3427.2 +3 -0
- train_results.json +5 -5
- trainer_state.json +144 -315
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
-
"eval_wer":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.02,
|
| 3 |
+
"eval_loss": 0.25927734375,
|
| 4 |
+
"eval_runtime": 704.255,
|
| 5 |
+
"eval_samples_per_second": 3.149,
|
| 6 |
+
"eval_steps_per_second": 0.099,
|
| 7 |
+
"eval_wer": 19.97153700189753,
|
| 8 |
+
"train_loss": 0.5692690700954861,
|
| 9 |
+
"train_runtime": 21274.2497,
|
| 10 |
+
"train_samples_per_second": 2.707,
|
| 11 |
+
"train_steps_per_second": 0.042
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
-
"eval_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.02,
|
| 3 |
+
"eval_loss": 0.25927734375,
|
| 4 |
+
"eval_runtime": 704.255,
|
| 5 |
+
"eval_samples_per_second": 3.149,
|
| 6 |
+
"eval_steps_per_second": 0.099,
|
| 7 |
+
"eval_wer": 19.97153700189753
|
| 8 |
}
|
runs/Dec19_09-48-52_nipa2022-65403/events.out.tfevents.1671433867.nipa2022-65403.3427.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19d9ca51c3c6769bf3220cdc12c5aa68144d421209f9aeedf7e49bd5959578c8
|
| 3 |
+
size 358
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.02,
|
| 3 |
+
"train_loss": 0.5692690700954861,
|
| 4 |
+
"train_runtime": 21274.2497,
|
| 5 |
+
"train_samples_per_second": 2.707,
|
| 6 |
+
"train_steps_per_second": 0.042
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,448 +1,277 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 1.
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 1.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 1.
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 1.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss":
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 2.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 2.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 2.
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 3.
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 3.
|
| 60 |
-
"eval_loss": 0.
|
| 61 |
-
"eval_runtime":
|
| 62 |
-
"eval_samples_per_second": 3.
|
| 63 |
-
"eval_steps_per_second": 0.
|
| 64 |
-
"eval_wer":
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 3.
|
| 69 |
-
"learning_rate":
|
| 70 |
-
"loss": 0.
|
| 71 |
"step": 225
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch": 4.
|
| 75 |
-
"learning_rate":
|
| 76 |
-
"loss": 0.
|
| 77 |
"step": 250
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch": 4.
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 275
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 0.
|
| 89 |
"step": 300
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 5.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 325
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 5.
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 350
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 6.
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 375
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 6.
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 400
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 6.
|
| 117 |
-
"eval_loss": 0.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second": 3.
|
| 120 |
-
"eval_steps_per_second": 0.
|
| 121 |
-
"eval_wer":
|
| 122 |
"step": 400
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 7.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 7.
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 450
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 475
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 8.
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 8.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 525
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch": 9.
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 550
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 9.
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 575
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 600
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"eval_loss": 0.
|
| 175 |
-
"eval_runtime":
|
| 176 |
-
"eval_samples_per_second":
|
| 177 |
-
"eval_steps_per_second": 0.
|
| 178 |
-
"eval_wer":
|
| 179 |
"step": 600
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch": 10.
|
| 183 |
-
"learning_rate": 2.
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 625
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch":
|
| 189 |
-
"learning_rate": 2.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 650
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 11.
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 0.
|
| 197 |
"step": 675
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch": 11.
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 0.
|
| 203 |
"step": 700
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch": 12.
|
| 207 |
-
"learning_rate": 1.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 725
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch": 12.
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss": 0.
|
| 215 |
"step": 750
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 775
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch": 13.
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 800
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 13.
|
| 231 |
-
"eval_loss": 0.
|
| 232 |
-
"eval_runtime":
|
| 233 |
-
"eval_samples_per_second": 3.
|
| 234 |
-
"eval_steps_per_second": 0.
|
| 235 |
-
"eval_wer":
|
| 236 |
"step": 800
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch": 13.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 825
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 14.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 850
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 14.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 875
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 900
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 15.02,
|
| 264 |
-
"
|
| 265 |
-
"
|
| 266 |
-
"
|
| 267 |
-
|
| 268 |
-
|
| 269 |
-
"
|
| 270 |
-
"learning_rate": 1.1480698313070749e-07,
|
| 271 |
-
"loss": 0.1693,
|
| 272 |
-
"step": 950
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 16.01,
|
| 276 |
-
"learning_rate": 1.0606721677313713e-07,
|
| 277 |
-
"loss": 0.2152,
|
| 278 |
-
"step": 975
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 16.03,
|
| 282 |
-
"learning_rate": 9.748775637719916e-08,
|
| 283 |
-
"loss": 0.2034,
|
| 284 |
-
"step": 1000
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 16.03,
|
| 288 |
-
"eval_loss": 0.31787109375,
|
| 289 |
-
"eval_runtime": 744.5018,
|
| 290 |
-
"eval_samples_per_second": 2.979,
|
| 291 |
-
"eval_steps_per_second": 0.094,
|
| 292 |
-
"eval_wer": 27.47628083491461,
|
| 293 |
-
"step": 1000
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 17.0,
|
| 297 |
-
"learning_rate": 8.976315836886941e-08,
|
| 298 |
-
"loss": 0.1694,
|
| 299 |
-
"step": 1025
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 17.02,
|
| 303 |
-
"learning_rate": 8.157864194541929e-08,
|
| 304 |
-
"loss": 0.227,
|
| 305 |
-
"step": 1050
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 17.04,
|
| 309 |
-
"learning_rate": 7.36437876374443e-08,
|
| 310 |
-
"loss": 0.1607,
|
| 311 |
-
"step": 1075
|
| 312 |
-
},
|
| 313 |
-
{
|
| 314 |
-
"epoch": 18.01,
|
| 315 |
-
"learning_rate": 6.598754886780604e-08,
|
| 316 |
-
"loss": 0.2253,
|
| 317 |
-
"step": 1100
|
| 318 |
-
},
|
| 319 |
-
{
|
| 320 |
-
"epoch": 18.03,
|
| 321 |
-
"learning_rate": 5.863786242148148e-08,
|
| 322 |
-
"loss": 0.1737,
|
| 323 |
-
"step": 1125
|
| 324 |
-
},
|
| 325 |
-
{
|
| 326 |
-
"epoch": 19.01,
|
| 327 |
-
"learning_rate": 5.1621546507264605e-08,
|
| 328 |
-
"loss": 0.1891,
|
| 329 |
-
"step": 1150
|
| 330 |
-
},
|
| 331 |
-
{
|
| 332 |
-
"epoch": 19.02,
|
| 333 |
-
"learning_rate": 4.496420290103045e-08,
|
| 334 |
-
"loss": 0.2066,
|
| 335 |
-
"step": 1175
|
| 336 |
-
},
|
| 337 |
-
{
|
| 338 |
-
"epoch": 19.04,
|
| 339 |
-
"learning_rate": 3.869012352763169e-08,
|
| 340 |
-
"loss": 0.1478,
|
| 341 |
-
"step": 1200
|
| 342 |
-
},
|
| 343 |
-
{
|
| 344 |
-
"epoch": 19.04,
|
| 345 |
-
"eval_loss": 0.3193359375,
|
| 346 |
-
"eval_runtime": 742.6411,
|
| 347 |
-
"eval_samples_per_second": 2.987,
|
| 348 |
-
"eval_steps_per_second": 0.094,
|
| 349 |
-
"eval_wer": 27.523719165085385,
|
| 350 |
-
"step": 1200
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"epoch": 20.02,
|
| 354 |
-
"learning_rate": 3.28222018222991e-08,
|
| 355 |
-
"loss": 0.2362,
|
| 356 |
-
"step": 1225
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 20.03,
|
| 360 |
-
"learning_rate": 2.738184919497899e-08,
|
| 361 |
-
"loss": 0.1564,
|
| 362 |
-
"step": 1250
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"epoch": 21.01,
|
| 366 |
-
"learning_rate": 2.2388916902420573e-08,
|
| 367 |
-
"loss": 0.2014,
|
| 368 |
-
"step": 1275
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 21.03,
|
| 372 |
-
"learning_rate": 1.7861623613092718e-08,
|
| 373 |
-
"loss": 0.1918,
|
| 374 |
-
"step": 1300
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"epoch": 22.0,
|
| 378 |
-
"learning_rate": 1.381648892923838e-08,
|
| 379 |
-
"loss": 0.1614,
|
| 380 |
-
"step": 1325
|
| 381 |
-
},
|
| 382 |
-
{
|
| 383 |
-
"epoch": 22.02,
|
| 384 |
-
"learning_rate": 1.0268273108637776e-08,
|
| 385 |
-
"loss": 0.2187,
|
| 386 |
-
"step": 1350
|
| 387 |
-
},
|
| 388 |
-
{
|
| 389 |
-
"epoch": 22.04,
|
| 390 |
-
"learning_rate": 7.229923206028154e-09,
|
| 391 |
-
"loss": 0.1492,
|
| 392 |
-
"step": 1375
|
| 393 |
-
},
|
| 394 |
-
{
|
| 395 |
-
"epoch": 23.01,
|
| 396 |
-
"learning_rate": 4.712525830705338e-09,
|
| 397 |
-
"loss": 0.2169,
|
| 398 |
-
"step": 1400
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 23.01,
|
| 402 |
-
"eval_loss": 0.31982421875,
|
| 403 |
-
"eval_runtime": 739.4797,
|
| 404 |
-
"eval_samples_per_second": 2.999,
|
| 405 |
-
"eval_steps_per_second": 0.095,
|
| 406 |
-
"eval_wer": 27.504743833017077,
|
| 407 |
-
"step": 1400
|
| 408 |
-
},
|
| 409 |
-
{
|
| 410 |
-
"epoch": 23.03,
|
| 411 |
-
"learning_rate": 2.7252666926886157e-09,
|
| 412 |
-
"loss": 0.1702,
|
| 413 |
-
"step": 1425
|
| 414 |
-
},
|
| 415 |
-
{
|
| 416 |
-
"epoch": 24.01,
|
| 417 |
-
"learning_rate": 1.2753970850610251e-09,
|
| 418 |
-
"loss": 0.1815,
|
| 419 |
-
"step": 1450
|
| 420 |
-
},
|
| 421 |
-
{
|
| 422 |
-
"epoch": 24.02,
|
| 423 |
-
"learning_rate": 3.682074247873257e-10,
|
| 424 |
-
"loss": 0.2022,
|
| 425 |
-
"step": 1475
|
| 426 |
-
},
|
| 427 |
-
{
|
| 428 |
-
"epoch": 24.04,
|
| 429 |
-
"learning_rate": 7.0079485561924665e-12,
|
| 430 |
-
"loss": 0.1448,
|
| 431 |
-
"step": 1500
|
| 432 |
-
},
|
| 433 |
-
{
|
| 434 |
-
"epoch": 24.04,
|
| 435 |
-
"step": 1500,
|
| 436 |
-
"total_flos": 2.0191356490809868e+20,
|
| 437 |
-
"train_loss": 0.31971333821614584,
|
| 438 |
-
"train_runtime": 22038.1709,
|
| 439 |
-
"train_samples_per_second": 4.356,
|
| 440 |
-
"train_steps_per_second": 0.068
|
| 441 |
}
|
| 442 |
],
|
| 443 |
-
"max_steps":
|
| 444 |
"num_train_epochs": 9223372036854775807,
|
| 445 |
-
"total_flos":
|
| 446 |
"trial_name": null,
|
| 447 |
"trial_params": null
|
| 448 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 19.97153700189753,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-800",
|
| 4 |
+
"epoch": 15.016666666666667,
|
| 5 |
+
"global_step": 900,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 2e-08,
|
| 13 |
+
"loss": 1.2498,
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"learning_rate": 4.5e-08,
|
| 19 |
+
"loss": 0.9037,
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 1.02,
|
| 24 |
+
"learning_rate": 6.999999999999999e-08,
|
| 25 |
+
"loss": 1.0907,
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 1.05,
|
| 30 |
+
"learning_rate": 9.499999999999999e-08,
|
| 31 |
+
"loss": 0.927,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 2.01,
|
| 36 |
+
"learning_rate": 9.983007697636658e-08,
|
| 37 |
+
"loss": 0.8001,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 2.04,
|
| 42 |
+
"learning_rate": 9.918643144747679e-08,
|
| 43 |
+
"loss": 0.8719,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 2.06,
|
| 48 |
+
"learning_rate": 9.812276182268236e-08,
|
| 49 |
+
"loss": 0.5706,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 3.03,
|
| 54 |
+
"learning_rate": 9.656074673794016e-08,
|
| 55 |
+
"loss": 0.7919,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 3.03,
|
| 60 |
+
"eval_loss": 0.279296875,
|
| 61 |
+
"eval_runtime": 701.0908,
|
| 62 |
+
"eval_samples_per_second": 3.164,
|
| 63 |
+
"eval_steps_per_second": 0.1,
|
| 64 |
+
"eval_wer": 22.58064516129032,
|
| 65 |
"step": 200
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 3.05,
|
| 69 |
+
"learning_rate": 9.455032620941839e-08,
|
| 70 |
+
"loss": 0.5192,
|
| 71 |
"step": 225
|
| 72 |
},
|
| 73 |
{
|
| 74 |
+
"epoch": 4.02,
|
| 75 |
+
"learning_rate": 9.200467769494708e-08,
|
| 76 |
+
"loss": 0.6054,
|
| 77 |
"step": 250
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 4.04,
|
| 81 |
+
"learning_rate": 8.914398527981508e-08,
|
| 82 |
+
"loss": 0.5903,
|
| 83 |
"step": 275
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 5.01,
|
| 87 |
+
"learning_rate": 8.590631488815944e-08,
|
| 88 |
+
"loss": 0.469,
|
| 89 |
"step": 300
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 5.03,
|
| 93 |
+
"learning_rate": 8.23228470557425e-08,
|
| 94 |
+
"loss": 0.6128,
|
| 95 |
"step": 325
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 5.06,
|
| 99 |
+
"learning_rate": 7.84280925367132e-08,
|
| 100 |
+
"loss": 0.4026,
|
| 101 |
"step": 350
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 6.02,
|
| 105 |
+
"learning_rate": 7.408768370508576e-08,
|
| 106 |
+
"loss": 0.5923,
|
| 107 |
"step": 375
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 6.05,
|
| 111 |
+
"learning_rate": 6.967704118273256e-08,
|
| 112 |
+
"loss": 0.4409,
|
| 113 |
"step": 400
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 6.05,
|
| 117 |
+
"eval_loss": 0.26513671875,
|
| 118 |
+
"eval_runtime": 702.5872,
|
| 119 |
+
"eval_samples_per_second": 3.157,
|
| 120 |
+
"eval_steps_per_second": 0.1,
|
| 121 |
+
"eval_wer": 20.60721062618596,
|
| 122 |
"step": 400
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 7.01,
|
| 126 |
+
"learning_rate": 6.507689799722478e-08,
|
| 127 |
+
"loss": 0.4854,
|
| 128 |
"step": 425
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 7.04,
|
| 132 |
+
"learning_rate": 6.03315560421329e-08,
|
| 133 |
+
"loss": 0.4989,
|
| 134 |
"step": 450
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 8.0,
|
| 138 |
+
"learning_rate": 5.5291509972967234e-08,
|
| 139 |
+
"loss": 0.3982,
|
| 140 |
"step": 475
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 8.03,
|
| 144 |
+
"learning_rate": 5.0392695044435566e-08,
|
| 145 |
+
"loss": 0.5425,
|
| 146 |
"step": 500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 8.06,
|
| 150 |
+
"learning_rate": 4.5490098247957034e-08,
|
| 151 |
+
"loss": 0.3773,
|
| 152 |
"step": 525
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 9.02,
|
| 156 |
+
"learning_rate": 4.0630934270713755e-08,
|
| 157 |
+
"loss": 0.5168,
|
| 158 |
"step": 550
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 9.05,
|
| 162 |
+
"learning_rate": 3.586199951809582e-08,
|
| 163 |
+
"loss": 0.4259,
|
| 164 |
"step": 575
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 10.01,
|
| 168 |
+
"learning_rate": 3.1047378773808274e-08,
|
| 169 |
+
"loss": 0.4393,
|
| 170 |
"step": 600
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 10.01,
|
| 174 |
+
"eval_loss": 0.260009765625,
|
| 175 |
+
"eval_runtime": 701.0779,
|
| 176 |
+
"eval_samples_per_second": 3.164,
|
| 177 |
+
"eval_steps_per_second": 0.1,
|
| 178 |
+
"eval_wer": 20.06641366223909,
|
| 179 |
"step": 600
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 10.04,
|
| 183 |
+
"learning_rate": 2.6603509286971338e-08,
|
| 184 |
+
"loss": 0.497,
|
| 185 |
"step": 625
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 11.0,
|
| 189 |
+
"learning_rate": 2.238496079552367e-08,
|
| 190 |
+
"loss": 0.3649,
|
| 191 |
"step": 650
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 11.03,
|
| 195 |
+
"learning_rate": 1.8432360227531113e-08,
|
| 196 |
+
"loss": 0.5444,
|
| 197 |
"step": 675
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 11.06,
|
| 201 |
+
"learning_rate": 1.4783773287174683e-08,
|
| 202 |
+
"loss": 0.3612,
|
| 203 |
"step": 700
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 12.02,
|
| 207 |
+
"learning_rate": 1.134947733186315e-08,
|
| 208 |
+
"loss": 0.4897,
|
| 209 |
"step": 725
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 12.05,
|
| 213 |
+
"learning_rate": 8.426519384872732e-09,
|
| 214 |
+
"loss": 0.4383,
|
| 215 |
"step": 750
|
| 216 |
},
|
| 217 |
{
|
| 218 |
+
"epoch": 13.01,
|
| 219 |
+
"learning_rate": 5.903936782582253e-09,
|
| 220 |
+
"loss": 0.4205,
|
| 221 |
"step": 775
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 13.04,
|
| 225 |
+
"learning_rate": 3.8060233744356625e-09,
|
| 226 |
+
"loss": 0.4975,
|
| 227 |
"step": 800
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 13.04,
|
| 231 |
+
"eval_loss": 0.25927734375,
|
| 232 |
+
"eval_runtime": 700.0881,
|
| 233 |
+
"eval_samples_per_second": 3.168,
|
| 234 |
+
"eval_steps_per_second": 0.1,
|
| 235 |
+
"eval_wer": 19.97153700189753,
|
| 236 |
"step": 800
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 13.06,
|
| 240 |
+
"learning_rate": 2.152983213389559e-09,
|
| 241 |
+
"loss": 0.3587,
|
| 242 |
"step": 825
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 14.03,
|
| 246 |
+
"learning_rate": 9.228083061983804e-10,
|
| 247 |
+
"loss": 0.5509,
|
| 248 |
"step": 850
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 14.05,
|
| 252 |
+
"learning_rate": 2.2190176984600017e-10,
|
| 253 |
+
"loss": 0.3727,
|
| 254 |
"step": 875
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 15.02,
|
| 258 |
+
"learning_rate": 3.8553092647219954e-13,
|
| 259 |
+
"loss": 0.4754,
|
| 260 |
"step": 900
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 15.02,
|
| 264 |
+
"step": 900,
|
| 265 |
+
"total_flos": 1.2318638198947106e+20,
|
| 266 |
+
"train_loss": 0.5692690700954861,
|
| 267 |
+
"train_runtime": 21274.2497,
|
| 268 |
+
"train_samples_per_second": 2.707,
|
| 269 |
+
"train_steps_per_second": 0.042
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
}
|
| 271 |
],
|
| 272 |
+
"max_steps": 900,
|
| 273 |
"num_train_epochs": 9223372036854775807,
|
| 274 |
+
"total_flos": 1.2318638198947106e+20,
|
| 275 |
"trial_name": null,
|
| 276 |
"trial_params": null
|
| 277 |
}
|