End of training
Browse files- all_results.json +8 -8
- eval_results.json +4 -4
- runs/Dec13_21-55-01_dante/events.out.tfevents.1671028475.dante.28467.2 +3 -0
- train_results.json +4 -4
- trainer_state.json +156 -414
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime": 97.
|
| 5 |
-
"eval_samples_per_second": 1.
|
| 6 |
"eval_steps_per_second": 0.062,
|
| 7 |
-
"eval_wer":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second": 6.
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.3284420371055603,
|
| 4 |
+
"eval_runtime": 97.0831,
|
| 5 |
+
"eval_samples_per_second": 1.658,
|
| 6 |
"eval_steps_per_second": 0.062,
|
| 7 |
+
"eval_wer": 15.012180267965894,
|
| 8 |
+
"train_loss": 0.598131457010905,
|
| 9 |
+
"train_runtime": 63177.9957,
|
| 10 |
+
"train_samples_per_second": 6.078,
|
| 11 |
+
"train_steps_per_second": 0.047
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime": 97.
|
| 5 |
-
"eval_samples_per_second": 1.
|
| 6 |
"eval_steps_per_second": 0.062,
|
| 7 |
-
"eval_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 0.3284420371055603,
|
| 4 |
+
"eval_runtime": 97.0831,
|
| 5 |
+
"eval_samples_per_second": 1.658,
|
| 6 |
"eval_steps_per_second": 0.062,
|
| 7 |
+
"eval_wer": 15.012180267965894
|
| 8 |
}
|
runs/Dec13_21-55-01_dante/events.out.tfevents.1671028475.dante.28467.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55b34c5fe5b8a01479790c33e07b4db4a1bb3ebb5c2e14f162fd058ba550cbfe
|
| 3 |
+
size 358
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second": 6.
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.598131457010905,
|
| 4 |
+
"train_runtime": 63177.9957,
|
| 5 |
+
"train_samples_per_second": 6.078,
|
| 6 |
+
"train_steps_per_second": 0.047
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,670 +1,412 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 4 |
"epoch": 1.0,
|
| 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": 0.01,
|
| 12 |
-
"learning_rate": 9.800000000000001e-07,
|
| 13 |
-
"loss": 1.7359,
|
| 14 |
-
"step": 50
|
| 15 |
-
},
|
| 16 |
{
|
| 17 |
"epoch": 0.02,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
-
"step":
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.03,
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss":
|
| 26 |
-
"step":
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 0.04,
|
| 30 |
-
"learning_rate": 3.980000000000001e-06,
|
| 31 |
-
"loss": 1.1693,
|
| 32 |
-
"step": 200
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.05,
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
-
"step":
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.06,
|
| 42 |
-
"learning_rate": 5.98e-06,
|
| 43 |
-
"loss": 1.0805,
|
| 44 |
-
"step": 300
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.07,
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss":
|
| 50 |
-
"step":
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.08,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss":
|
| 56 |
-
"step":
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.09,
|
| 60 |
-
"learning_rate": 8.98e-06,
|
| 61 |
-
"loss": 1.0208,
|
| 62 |
-
"step": 450
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.1,
|
| 66 |
-
"learning_rate": 9.980000000000001e-06,
|
| 67 |
-
"loss": 0.9901,
|
| 68 |
-
"step": 500
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 0.11,
|
| 72 |
"learning_rate": 1e-05,
|
| 73 |
-
"loss": 0.
|
| 74 |
-
"step":
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.12,
|
| 78 |
"learning_rate": 1e-05,
|
| 79 |
-
"loss": 0.
|
| 80 |
-
"step":
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.13,
|
| 84 |
"learning_rate": 1e-05,
|
| 85 |
-
"loss": 0.
|
| 86 |
-
"step":
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.14,
|
| 90 |
-
"learning_rate": 1e-05,
|
| 91 |
-
"loss": 0.9238,
|
| 92 |
-
"step": 700
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.15,
|
| 96 |
"learning_rate": 1e-05,
|
| 97 |
-
"loss": 0.
|
| 98 |
-
"step":
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 0.16,
|
| 102 |
-
"learning_rate": 1e-05,
|
| 103 |
-
"loss": 0.8913,
|
| 104 |
-
"step": 800
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.17,
|
| 108 |
"learning_rate": 1e-05,
|
| 109 |
-
"loss": 0.
|
| 110 |
-
"step":
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.18,
|
| 114 |
"learning_rate": 1e-05,
|
| 115 |
-
"loss": 0.
|
| 116 |
-
"step":
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 0.19,
|
| 120 |
-
"learning_rate": 1e-05,
|
| 121 |
-
"loss": 0.8714,
|
| 122 |
-
"step": 950
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.2,
|
| 126 |
-
"learning_rate": 1e-05,
|
| 127 |
-
"loss": 0.8528,
|
| 128 |
-
"step": 1000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.2,
|
| 132 |
-
"eval_loss": 0.44020676612854004,
|
| 133 |
-
"eval_runtime": 98.0744,
|
| 134 |
-
"eval_samples_per_second": 1.642,
|
| 135 |
-
"eval_steps_per_second": 0.061,
|
| 136 |
-
"eval_wer": 22.56394640682095,
|
| 137 |
-
"step": 1000
|
| 138 |
-
},
|
| 139 |
-
{
|
| 140 |
-
"epoch": 0.21,
|
| 141 |
"learning_rate": 1e-05,
|
| 142 |
-
"loss": 0.
|
| 143 |
-
"step":
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.22,
|
| 147 |
"learning_rate": 1e-05,
|
| 148 |
-
"loss": 0.
|
| 149 |
-
"step":
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.23,
|
| 153 |
"learning_rate": 1e-05,
|
| 154 |
-
"loss": 0.
|
| 155 |
-
"step":
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 0.24,
|
| 159 |
-
"learning_rate": 1e-05,
|
| 160 |
-
"loss": 0.8277,
|
| 161 |
-
"step": 1200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.25,
|
| 165 |
"learning_rate": 1e-05,
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step":
|
| 168 |
-
},
|
| 169 |
-
{
|
| 170 |
-
"epoch": 0.26,
|
| 171 |
-
"learning_rate": 1e-05,
|
| 172 |
-
"loss": 0.8254,
|
| 173 |
-
"step": 1300
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.27,
|
| 177 |
"learning_rate": 1e-05,
|
| 178 |
-
"loss": 0.
|
| 179 |
-
"step":
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 0.28,
|
| 183 |
"learning_rate": 1e-05,
|
| 184 |
-
"loss": 0.
|
| 185 |
-
"step":
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 0.29,
|
| 189 |
-
"learning_rate": 1e-05,
|
| 190 |
-
"loss": 0.8161,
|
| 191 |
-
"step": 1450
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.3,
|
| 195 |
"learning_rate": 1e-05,
|
| 196 |
-
"loss": 0.
|
| 197 |
-
"step":
|
| 198 |
-
},
|
| 199 |
-
{
|
| 200 |
-
"epoch": 0.31,
|
| 201 |
-
"learning_rate": 1e-05,
|
| 202 |
-
"loss": 0.8073,
|
| 203 |
-
"step": 1550
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.32,
|
| 207 |
"learning_rate": 1e-05,
|
| 208 |
-
"loss": 0.
|
| 209 |
-
"step":
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.33,
|
| 213 |
"learning_rate": 1e-05,
|
| 214 |
-
"loss": 0.
|
| 215 |
-
"step":
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch": 0.
|
| 219 |
-
"
|
| 220 |
-
"
|
| 221 |
-
"
|
|
|
|
|
|
|
|
|
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 0.35,
|
| 225 |
"learning_rate": 1e-05,
|
| 226 |
-
"loss": 0.
|
| 227 |
-
"step":
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 0.36,
|
| 231 |
-
"learning_rate": 1e-05,
|
| 232 |
-
"loss": 0.776,
|
| 233 |
-
"step": 1800
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.37,
|
| 237 |
"learning_rate": 1e-05,
|
| 238 |
-
"loss": 0.
|
| 239 |
-
"step":
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.38,
|
| 243 |
"learning_rate": 1e-05,
|
| 244 |
-
"loss": 0.
|
| 245 |
-
"step":
|
| 246 |
-
},
|
| 247 |
-
{
|
| 248 |
-
"epoch": 0.39,
|
| 249 |
-
"learning_rate": 1e-05,
|
| 250 |
-
"loss": 0.7652,
|
| 251 |
-
"step": 1950
|
| 252 |
-
},
|
| 253 |
-
{
|
| 254 |
-
"epoch": 0.4,
|
| 255 |
-
"learning_rate": 1e-05,
|
| 256 |
-
"loss": 0.7558,
|
| 257 |
-
"step": 2000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.4,
|
| 261 |
-
"eval_loss": 0.38613995909690857,
|
| 262 |
-
"eval_runtime": 95.704,
|
| 263 |
-
"eval_samples_per_second": 1.682,
|
| 264 |
-
"eval_steps_per_second": 0.063,
|
| 265 |
-
"eval_wer": 19.701583434835566,
|
| 266 |
-
"step": 2000
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 0.41,
|
| 270 |
"learning_rate": 1e-05,
|
| 271 |
-
"loss": 0.
|
| 272 |
-
"step":
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 0.42,
|
| 276 |
"learning_rate": 1e-05,
|
| 277 |
-
"loss": 0.
|
| 278 |
-
"step":
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 0.43,
|
| 282 |
"learning_rate": 1e-05,
|
| 283 |
-
"loss": 0.
|
| 284 |
-
"step":
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 0.44,
|
| 288 |
-
"learning_rate": 1e-05,
|
| 289 |
-
"loss": 0.7414,
|
| 290 |
-
"step": 2200
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.45,
|
| 294 |
"learning_rate": 1e-05,
|
| 295 |
-
"loss": 0.
|
| 296 |
-
"step":
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 0.46,
|
| 300 |
-
"learning_rate": 1e-05,
|
| 301 |
-
"loss": 0.755,
|
| 302 |
-
"step": 2300
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 0.47,
|
| 306 |
"learning_rate": 1e-05,
|
| 307 |
-
"loss": 0.
|
| 308 |
-
"step":
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 0.48,
|
| 312 |
"learning_rate": 1e-05,
|
| 313 |
-
"loss": 0.
|
| 314 |
-
"step":
|
| 315 |
-
},
|
| 316 |
-
{
|
| 317 |
-
"epoch": 0.49,
|
| 318 |
-
"learning_rate": 1e-05,
|
| 319 |
-
"loss": 0.7345,
|
| 320 |
-
"step": 2450
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 0.5,
|
| 324 |
"learning_rate": 1e-05,
|
| 325 |
-
"loss": 0.
|
| 326 |
-
"step":
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.51,
|
| 330 |
-
"learning_rate": 1e-05,
|
| 331 |
-
"loss": 0.727,
|
| 332 |
-
"step": 2550
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.52,
|
| 336 |
"learning_rate": 1e-05,
|
| 337 |
-
"loss": 0.
|
| 338 |
-
"step":
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.53,
|
| 342 |
"learning_rate": 1e-05,
|
| 343 |
-
"loss": 0.
|
| 344 |
-
"step":
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 0.54,
|
| 348 |
-
"learning_rate": 1e-05,
|
| 349 |
-
"loss": 0.7246,
|
| 350 |
-
"step": 2700
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 0.55,
|
| 354 |
"learning_rate": 1e-05,
|
| 355 |
-
"loss": 0.
|
| 356 |
-
"step":
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 0.56,
|
| 360 |
-
"learning_rate": 1e-05,
|
| 361 |
-
"loss": 0.7184,
|
| 362 |
-
"step": 2800
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 0.57,
|
| 366 |
"learning_rate": 1e-05,
|
| 367 |
-
"loss": 0.
|
| 368 |
-
"step":
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.58,
|
| 372 |
"learning_rate": 1e-05,
|
| 373 |
-
"loss": 0.
|
| 374 |
-
"step":
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"epoch": 0.59,
|
| 378 |
-
"learning_rate": 1e-05,
|
| 379 |
-
"loss": 0.7143,
|
| 380 |
-
"step": 2950
|
| 381 |
-
},
|
| 382 |
-
{
|
| 383 |
-
"epoch": 0.6,
|
| 384 |
-
"learning_rate": 1e-05,
|
| 385 |
-
"loss": 0.721,
|
| 386 |
-
"step": 3000
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 0.6,
|
| 390 |
-
"eval_loss": 0.3545149862766266,
|
| 391 |
-
"eval_runtime": 97.6917,
|
| 392 |
-
"eval_samples_per_second": 1.648,
|
| 393 |
-
"eval_steps_per_second": 0.061,
|
| 394 |
-
"eval_wer": 17.84409257003654,
|
| 395 |
-
"step": 3000
|
| 396 |
-
},
|
| 397 |
-
{
|
| 398 |
-
"epoch": 0.61,
|
| 399 |
"learning_rate": 1e-05,
|
| 400 |
-
"loss": 0.
|
| 401 |
-
"step":
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.62,
|
| 405 |
"learning_rate": 1e-05,
|
| 406 |
-
"loss": 0.
|
| 407 |
-
"step":
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 0.63,
|
| 411 |
"learning_rate": 1e-05,
|
| 412 |
-
"loss": 0.
|
| 413 |
-
"step":
|
| 414 |
-
},
|
| 415 |
-
{
|
| 416 |
-
"epoch": 0.64,
|
| 417 |
-
"learning_rate": 1e-05,
|
| 418 |
-
"loss": 0.7095,
|
| 419 |
-
"step": 3200
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 0.65,
|
| 423 |
"learning_rate": 1e-05,
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"step":
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch": 0.
|
| 429 |
"learning_rate": 1e-05,
|
| 430 |
-
"loss": 0.
|
| 431 |
-
"step":
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 0.67,
|
| 435 |
-
"
|
| 436 |
-
"
|
| 437 |
-
"
|
|
|
|
|
|
|
|
|
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 0.68,
|
| 441 |
"learning_rate": 1e-05,
|
| 442 |
-
"loss": 0.
|
| 443 |
-
"step":
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 0.69,
|
| 447 |
-
"learning_rate": 1e-05,
|
| 448 |
-
"loss": 0.7045,
|
| 449 |
-
"step": 3450
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 0.7,
|
| 453 |
"learning_rate": 1e-05,
|
| 454 |
-
"loss": 0.
|
| 455 |
-
"step":
|
| 456 |
-
},
|
| 457 |
-
{
|
| 458 |
-
"epoch": 0.71,
|
| 459 |
-
"learning_rate": 1e-05,
|
| 460 |
-
"loss": 0.695,
|
| 461 |
-
"step": 3550
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 0.72,
|
| 465 |
"learning_rate": 1e-05,
|
| 466 |
-
"loss": 0.
|
| 467 |
-
"step":
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 0.73,
|
| 471 |
"learning_rate": 1e-05,
|
| 472 |
-
"loss": 0.
|
| 473 |
-
"step":
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 0.74,
|
| 477 |
-
"learning_rate": 1e-05,
|
| 478 |
-
"loss": 0.6955,
|
| 479 |
-
"step": 3700
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 0.75,
|
| 483 |
"learning_rate": 1e-05,
|
| 484 |
-
"loss": 0.
|
| 485 |
-
"step":
|
| 486 |
-
},
|
| 487 |
-
{
|
| 488 |
-
"epoch": 0.76,
|
| 489 |
-
"learning_rate": 1e-05,
|
| 490 |
-
"loss": 0.675,
|
| 491 |
-
"step": 3800
|
| 492 |
},
|
| 493 |
{
|
| 494 |
"epoch": 0.77,
|
| 495 |
"learning_rate": 1e-05,
|
| 496 |
-
"loss": 0.
|
| 497 |
-
"step":
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 0.78,
|
| 501 |
"learning_rate": 1e-05,
|
| 502 |
-
"loss": 0.
|
| 503 |
-
"step":
|
| 504 |
-
},
|
| 505 |
-
{
|
| 506 |
-
"epoch": 0.79,
|
| 507 |
-
"learning_rate": 1e-05,
|
| 508 |
-
"loss": 0.6814,
|
| 509 |
-
"step": 3950
|
| 510 |
-
},
|
| 511 |
-
{
|
| 512 |
-
"epoch": 0.8,
|
| 513 |
-
"learning_rate": 1e-05,
|
| 514 |
-
"loss": 0.6803,
|
| 515 |
-
"step": 4000
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 0.8,
|
| 519 |
-
"eval_loss": 0.33644479513168335,
|
| 520 |
-
"eval_runtime": 96.6414,
|
| 521 |
-
"eval_samples_per_second": 1.666,
|
| 522 |
-
"eval_steps_per_second": 0.062,
|
| 523 |
-
"eval_wer": 16.56516443361754,
|
| 524 |
-
"step": 4000
|
| 525 |
-
},
|
| 526 |
-
{
|
| 527 |
-
"epoch": 0.81,
|
| 528 |
"learning_rate": 1e-05,
|
| 529 |
-
"loss": 0.
|
| 530 |
-
"step":
|
| 531 |
},
|
| 532 |
{
|
| 533 |
"epoch": 0.82,
|
| 534 |
"learning_rate": 1e-05,
|
| 535 |
-
"loss": 0.
|
| 536 |
-
"step":
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 0.83,
|
| 540 |
"learning_rate": 1e-05,
|
| 541 |
-
"loss": 0.
|
| 542 |
-
"step":
|
| 543 |
-
},
|
| 544 |
-
{
|
| 545 |
-
"epoch": 0.84,
|
| 546 |
-
"learning_rate": 1e-05,
|
| 547 |
-
"loss": 0.6774,
|
| 548 |
-
"step": 4200
|
| 549 |
},
|
| 550 |
{
|
| 551 |
"epoch": 0.85,
|
| 552 |
"learning_rate": 1e-05,
|
| 553 |
-
"loss": 0.
|
| 554 |
-
"step":
|
| 555 |
-
},
|
| 556 |
-
{
|
| 557 |
-
"epoch": 0.86,
|
| 558 |
-
"learning_rate": 1e-05,
|
| 559 |
-
"loss": 0.6762,
|
| 560 |
-
"step": 4300
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 0.87,
|
| 564 |
"learning_rate": 1e-05,
|
| 565 |
-
"loss": 0.
|
| 566 |
-
"step":
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 0.88,
|
| 570 |
"learning_rate": 1e-05,
|
| 571 |
-
"loss": 0.
|
| 572 |
-
"step":
|
| 573 |
-
},
|
| 574 |
-
{
|
| 575 |
-
"epoch": 0.89,
|
| 576 |
-
"learning_rate": 1e-05,
|
| 577 |
-
"loss": 0.675,
|
| 578 |
-
"step": 4450
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.9,
|
| 582 |
"learning_rate": 1e-05,
|
| 583 |
-
"loss": 0.
|
| 584 |
-
"step":
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 0.91,
|
| 588 |
-
"learning_rate": 1e-05,
|
| 589 |
-
"loss": 0.6689,
|
| 590 |
-
"step": 4550
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 0.92,
|
| 594 |
"learning_rate": 1e-05,
|
| 595 |
-
"loss": 0.
|
| 596 |
-
"step":
|
| 597 |
},
|
| 598 |
{
|
| 599 |
"epoch": 0.93,
|
| 600 |
"learning_rate": 1e-05,
|
| 601 |
-
"loss": 0.
|
| 602 |
-
"step":
|
| 603 |
-
},
|
| 604 |
-
{
|
| 605 |
-
"epoch": 0.94,
|
| 606 |
-
"learning_rate": 1e-05,
|
| 607 |
-
"loss": 0.6796,
|
| 608 |
-
"step": 4700
|
| 609 |
},
|
| 610 |
{
|
| 611 |
"epoch": 0.95,
|
| 612 |
"learning_rate": 1e-05,
|
| 613 |
-
"loss": 0.
|
| 614 |
-
"step":
|
| 615 |
-
},
|
| 616 |
-
{
|
| 617 |
-
"epoch": 0.96,
|
| 618 |
-
"learning_rate": 1e-05,
|
| 619 |
-
"loss": 0.6562,
|
| 620 |
-
"step": 4800
|
| 621 |
},
|
| 622 |
{
|
| 623 |
"epoch": 0.97,
|
| 624 |
"learning_rate": 1e-05,
|
| 625 |
-
"loss": 0.
|
| 626 |
-
"step":
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 0.98,
|
| 630 |
"learning_rate": 1e-05,
|
| 631 |
-
"loss": 0.
|
| 632 |
-
"step":
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 0.99,
|
| 636 |
-
"learning_rate": 1e-05,
|
| 637 |
-
"loss": 0.6599,
|
| 638 |
-
"step": 4950
|
| 639 |
},
|
| 640 |
{
|
| 641 |
"epoch": 1.0,
|
| 642 |
"learning_rate": 1e-05,
|
| 643 |
-
"loss": 0.
|
| 644 |
-
"step":
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 1.0,
|
| 648 |
-
"eval_loss": 0.
|
| 649 |
-
"eval_runtime":
|
| 650 |
-
"eval_samples_per_second": 1.
|
| 651 |
-
"eval_steps_per_second": 0.
|
| 652 |
-
"eval_wer":
|
| 653 |
-
"step":
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 1.0,
|
| 657 |
-
"step":
|
| 658 |
-
"total_flos": 1.
|
| 659 |
-
"train_loss": 0.
|
| 660 |
-
"train_runtime":
|
| 661 |
-
"train_samples_per_second": 6.
|
| 662 |
-
"train_steps_per_second": 0.
|
| 663 |
}
|
| 664 |
],
|
| 665 |
-
"max_steps":
|
| 666 |
"num_train_epochs": 9223372036854775807,
|
| 667 |
-
"total_flos": 1.
|
| 668 |
"trial_name": null,
|
| 669 |
"trial_params": null
|
| 670 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 15.012180267965894,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-3000",
|
| 4 |
"epoch": 1.0,
|
| 5 |
+
"global_step": 3000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 1e-05,
|
| 13 |
+
"loss": 0.651,
|
| 14 |
+
"step": 50
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 1e-05,
|
| 19 |
+
"loss": 0.6439,
|
| 20 |
+
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 1e-05,
|
| 25 |
+
"loss": 0.6508,
|
| 26 |
+
"step": 150
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.07,
|
| 30 |
+
"learning_rate": 1e-05,
|
| 31 |
+
"loss": 0.6457,
|
| 32 |
+
"step": 200
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.08,
|
| 36 |
+
"learning_rate": 1e-05,
|
| 37 |
+
"loss": 0.633,
|
| 38 |
+
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 0.6258,
|
| 44 |
+
"step": 300
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.12,
|
| 48 |
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 0.6325,
|
| 50 |
+
"step": 350
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.13,
|
| 54 |
"learning_rate": 1e-05,
|
| 55 |
+
"loss": 0.6192,
|
| 56 |
+
"step": 400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.15,
|
| 60 |
"learning_rate": 1e-05,
|
| 61 |
+
"loss": 0.63,
|
| 62 |
+
"step": 450
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.17,
|
| 66 |
"learning_rate": 1e-05,
|
| 67 |
+
"loss": 0.623,
|
| 68 |
+
"step": 500
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.18,
|
| 72 |
"learning_rate": 1e-05,
|
| 73 |
+
"loss": 0.6099,
|
| 74 |
+
"step": 550
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.2,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
"learning_rate": 1e-05,
|
| 79 |
+
"loss": 0.6128,
|
| 80 |
+
"step": 600
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.22,
|
| 84 |
"learning_rate": 1e-05,
|
| 85 |
+
"loss": 0.6155,
|
| 86 |
+
"step": 650
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.23,
|
| 90 |
"learning_rate": 1e-05,
|
| 91 |
+
"loss": 0.6169,
|
| 92 |
+
"step": 700
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.25,
|
| 96 |
"learning_rate": 1e-05,
|
| 97 |
+
"loss": 0.6073,
|
| 98 |
+
"step": 750
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.27,
|
| 102 |
"learning_rate": 1e-05,
|
| 103 |
+
"loss": 0.6076,
|
| 104 |
+
"step": 800
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.28,
|
| 108 |
"learning_rate": 1e-05,
|
| 109 |
+
"loss": 0.6165,
|
| 110 |
+
"step": 850
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.3,
|
| 114 |
"learning_rate": 1e-05,
|
| 115 |
+
"loss": 0.6096,
|
| 116 |
+
"step": 900
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.32,
|
| 120 |
"learning_rate": 1e-05,
|
| 121 |
+
"loss": 0.6059,
|
| 122 |
+
"step": 950
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.33,
|
| 126 |
"learning_rate": 1e-05,
|
| 127 |
+
"loss": 0.5975,
|
| 128 |
+
"step": 1000
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.33,
|
| 132 |
+
"eval_loss": 0.33536991477012634,
|
| 133 |
+
"eval_runtime": 97.6999,
|
| 134 |
+
"eval_samples_per_second": 1.648,
|
| 135 |
+
"eval_steps_per_second": 0.061,
|
| 136 |
+
"eval_wer": 15.77344701583435,
|
| 137 |
+
"step": 1000
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.35,
|
| 141 |
"learning_rate": 1e-05,
|
| 142 |
+
"loss": 0.6067,
|
| 143 |
+
"step": 1050
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.37,
|
| 147 |
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 0.6052,
|
| 149 |
+
"step": 1100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.38,
|
| 153 |
"learning_rate": 1e-05,
|
| 154 |
+
"loss": 0.6001,
|
| 155 |
+
"step": 1150
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 0.4,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 159 |
"learning_rate": 1e-05,
|
| 160 |
+
"loss": 0.5959,
|
| 161 |
+
"step": 1200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 0.42,
|
| 165 |
"learning_rate": 1e-05,
|
| 166 |
+
"loss": 0.5945,
|
| 167 |
+
"step": 1250
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.43,
|
| 171 |
"learning_rate": 1e-05,
|
| 172 |
+
"loss": 0.6036,
|
| 173 |
+
"step": 1300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.45,
|
| 177 |
"learning_rate": 1e-05,
|
| 178 |
+
"loss": 0.6011,
|
| 179 |
+
"step": 1350
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 0.47,
|
| 183 |
"learning_rate": 1e-05,
|
| 184 |
+
"loss": 0.5921,
|
| 185 |
+
"step": 1400
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.48,
|
| 189 |
"learning_rate": 1e-05,
|
| 190 |
+
"loss": 0.6054,
|
| 191 |
+
"step": 1450
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.5,
|
| 195 |
"learning_rate": 1e-05,
|
| 196 |
+
"loss": 0.5883,
|
| 197 |
+
"step": 1500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 0.52,
|
| 201 |
"learning_rate": 1e-05,
|
| 202 |
+
"loss": 0.6017,
|
| 203 |
+
"step": 1550
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.53,
|
| 207 |
"learning_rate": 1e-05,
|
| 208 |
+
"loss": 0.5966,
|
| 209 |
+
"step": 1600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.55,
|
| 213 |
"learning_rate": 1e-05,
|
| 214 |
+
"loss": 0.587,
|
| 215 |
+
"step": 1650
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 0.57,
|
| 219 |
"learning_rate": 1e-05,
|
| 220 |
+
"loss": 0.5909,
|
| 221 |
+
"step": 1700
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 0.58,
|
| 225 |
"learning_rate": 1e-05,
|
| 226 |
+
"loss": 0.5973,
|
| 227 |
+
"step": 1750
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.6,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 231 |
"learning_rate": 1e-05,
|
| 232 |
+
"loss": 0.5885,
|
| 233 |
+
"step": 1800
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.62,
|
| 237 |
"learning_rate": 1e-05,
|
| 238 |
+
"loss": 0.5824,
|
| 239 |
+
"step": 1850
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.63,
|
| 243 |
"learning_rate": 1e-05,
|
| 244 |
+
"loss": 0.5919,
|
| 245 |
+
"step": 1900
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 0.65,
|
| 249 |
"learning_rate": 1e-05,
|
| 250 |
+
"loss": 0.5857,
|
| 251 |
+
"step": 1950
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 0.67,
|
| 255 |
"learning_rate": 1e-05,
|
| 256 |
+
"loss": 0.5783,
|
| 257 |
+
"step": 2000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.67,
|
| 261 |
+
"eval_loss": 0.33269065618515015,
|
| 262 |
+
"eval_runtime": 99.8788,
|
| 263 |
+
"eval_samples_per_second": 1.612,
|
| 264 |
+
"eval_steps_per_second": 0.06,
|
| 265 |
+
"eval_wer": 16.35200974421437,
|
| 266 |
+
"step": 2000
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 0.68,
|
| 270 |
"learning_rate": 1e-05,
|
| 271 |
+
"loss": 0.5788,
|
| 272 |
+
"step": 2050
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 0.7,
|
| 276 |
"learning_rate": 1e-05,
|
| 277 |
+
"loss": 0.5876,
|
| 278 |
+
"step": 2100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 0.72,
|
| 282 |
"learning_rate": 1e-05,
|
| 283 |
+
"loss": 0.5825,
|
| 284 |
+
"step": 2150
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 0.73,
|
| 288 |
"learning_rate": 1e-05,
|
| 289 |
+
"loss": 0.5743,
|
| 290 |
+
"step": 2200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 0.75,
|
| 294 |
"learning_rate": 1e-05,
|
| 295 |
+
"loss": 0.5813,
|
| 296 |
+
"step": 2250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.77,
|
| 300 |
"learning_rate": 1e-05,
|
| 301 |
+
"loss": 0.5882,
|
| 302 |
+
"step": 2300
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 0.78,
|
| 306 |
"learning_rate": 1e-05,
|
| 307 |
+
"loss": 0.5824,
|
| 308 |
+
"step": 2350
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 0.8,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 312 |
"learning_rate": 1e-05,
|
| 313 |
+
"loss": 0.5829,
|
| 314 |
+
"step": 2400
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 0.82,
|
| 318 |
"learning_rate": 1e-05,
|
| 319 |
+
"loss": 0.5779,
|
| 320 |
+
"step": 2450
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 0.83,
|
| 324 |
"learning_rate": 1e-05,
|
| 325 |
+
"loss": 0.5697,
|
| 326 |
+
"step": 2500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 0.85,
|
| 330 |
"learning_rate": 1e-05,
|
| 331 |
+
"loss": 0.5713,
|
| 332 |
+
"step": 2550
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.87,
|
| 336 |
"learning_rate": 1e-05,
|
| 337 |
+
"loss": 0.5695,
|
| 338 |
+
"step": 2600
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.88,
|
| 342 |
"learning_rate": 1e-05,
|
| 343 |
+
"loss": 0.5779,
|
| 344 |
+
"step": 2650
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 0.9,
|
| 348 |
"learning_rate": 1e-05,
|
| 349 |
+
"loss": 0.5734,
|
| 350 |
+
"step": 2700
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 0.92,
|
| 354 |
"learning_rate": 1e-05,
|
| 355 |
+
"loss": 0.5682,
|
| 356 |
+
"step": 2750
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 0.93,
|
| 360 |
"learning_rate": 1e-05,
|
| 361 |
+
"loss": 0.5708,
|
| 362 |
+
"step": 2800
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 0.95,
|
| 366 |
"learning_rate": 1e-05,
|
| 367 |
+
"loss": 0.5762,
|
| 368 |
+
"step": 2850
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 0.97,
|
| 372 |
"learning_rate": 1e-05,
|
| 373 |
+
"loss": 0.5762,
|
| 374 |
+
"step": 2900
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 0.98,
|
| 378 |
"learning_rate": 1e-05,
|
| 379 |
+
"loss": 0.572,
|
| 380 |
+
"step": 2950
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 1.0,
|
| 384 |
"learning_rate": 1e-05,
|
| 385 |
+
"loss": 0.5788,
|
| 386 |
+
"step": 3000
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 1.0,
|
| 390 |
+
"eval_loss": 0.3284420371055603,
|
| 391 |
+
"eval_runtime": 99.6761,
|
| 392 |
+
"eval_samples_per_second": 1.615,
|
| 393 |
+
"eval_steps_per_second": 0.06,
|
| 394 |
+
"eval_wer": 15.012180267965894,
|
| 395 |
+
"step": 3000
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.0,
|
| 399 |
+
"step": 3000,
|
| 400 |
+
"total_flos": 1.1081679372288e+20,
|
| 401 |
+
"train_loss": 0.598131457010905,
|
| 402 |
+
"train_runtime": 63177.9957,
|
| 403 |
+
"train_samples_per_second": 6.078,
|
| 404 |
+
"train_steps_per_second": 0.047
|
| 405 |
}
|
| 406 |
],
|
| 407 |
+
"max_steps": 3000,
|
| 408 |
"num_train_epochs": 9223372036854775807,
|
| 409 |
+
"total_flos": 1.1081679372288e+20,
|
| 410 |
"trial_name": null,
|
| 411 |
"trial_params": null
|
| 412 |
}
|