Zheng Li
commited on
End of training
Browse files- README.md +1 -0
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/May15_01-46-01_cs-Precision-7960-Tower/events.out.tfevents.1747288305.cs-Precision-7960-Tower.143683.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +188 -583
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-base
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- superb
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-base
|
| 5 |
tags:
|
| 6 |
+
- audio-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- superb
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 5.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.8245072080023537,
|
| 4 |
+
"eval_loss": 0.8822575807571411,
|
| 5 |
+
"eval_runtime": 5.2627,
|
| 6 |
+
"eval_samples_per_second": 1291.738,
|
| 7 |
+
"eval_steps_per_second": 40.474,
|
| 8 |
+
"total_flos": 2.36283432911232e+18,
|
| 9 |
+
"train_loss": 1.447761106491089,
|
| 10 |
+
"train_runtime": 310.8685,
|
| 11 |
+
"train_samples_per_second": 821.794,
|
| 12 |
+
"train_steps_per_second": 1.608
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 5.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.8245072080023537,
|
| 4 |
+
"eval_loss": 0.8822575807571411,
|
| 5 |
+
"eval_runtime": 5.2627,
|
| 6 |
+
"eval_samples_per_second": 1291.738,
|
| 7 |
+
"eval_steps_per_second": 40.474
|
| 8 |
}
|
runs/May15_01-46-01_cs-Precision-7960-Tower/events.out.tfevents.1747288305.cs-Precision-7960-Tower.143683.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:954be4cd8f9e374e882f9fbc250f57458be6f0842daece6827bdd0e8ad1e1b40
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 2.36283432911232e+18,
|
| 4 |
+
"train_loss": 1.447761106491089,
|
| 5 |
+
"train_runtime": 310.8685,
|
| 6 |
+
"train_samples_per_second": 821.794,
|
| 7 |
+
"train_steps_per_second": 1.608
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,817 +1,422 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "wav2vec2-base-ft-keyword-spotting/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 4.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.2,
|
| 20 |
-
"grad_norm": 3.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 4.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 3.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.4,
|
| 34 |
-
"grad_norm": 5.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.6,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 2.
|
| 57 |
-
"loss":
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.8,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 2.
|
| 64 |
-
"loss":
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 2.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.0,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
"eval_accuracy": 0.6209179170344219,
|
| 84 |
-
"eval_loss": 1.
|
| 85 |
-
"eval_runtime": 5.
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second":
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.1,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 2.
|
| 94 |
-
"loss": 1.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.2,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 2.
|
| 101 |
-
"loss": 1.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.3,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 2.
|
| 108 |
-
"loss": 1.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.4,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 2.
|
| 115 |
-
"loss": 1.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.5,
|
| 120 |
-
"grad_norm": 3.
|
| 121 |
-
"learning_rate": 2.
|
| 122 |
-
"loss": 1.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.6,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 2.
|
| 129 |
-
"loss": 1.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 1.7,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate": 2.
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.8,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 2.
|
| 143 |
-
"loss": 1.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.9,
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate": 2.
|
| 150 |
-
"loss": 1.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 2.0,
|
| 155 |
-
"grad_norm": 3.
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 1.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.0,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 1.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second":
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.1,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 1.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.2,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 1.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.3,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 1.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.4,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 1.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 2.5,
|
| 199 |
-
"grad_norm": 4.
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 1.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 2.6,
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 1.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.7,
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 1.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 2.8,
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 1.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 2.9,
|
| 227 |
-
"grad_norm": 2.
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss":
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 3.0,
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss":
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 3.0,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss": 0.
|
| 243 |
-
"eval_runtime":
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second":
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 3.1,
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss":
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 3.2,
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss":
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 3.3,
|
| 264 |
-
"grad_norm":
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss":
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.4,
|
| 271 |
-
"grad_norm": 2.
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss":
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 3.5,
|
| 278 |
-
"grad_norm": 2.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss":
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.6,
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss":
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 3.7,
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.8,
|
| 299 |
-
"grad_norm":
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss":
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 3.9,
|
| 306 |
-
"grad_norm": 2.
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss":
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 4.0,
|
| 313 |
-
"grad_norm": 2.
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 4.0,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime": 4.
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.1,
|
| 329 |
-
"grad_norm": 2.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 4.2,
|
| 336 |
-
"grad_norm": 2.
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 4.3,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 4.4,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 4.5,
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 4.6,
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.7,
|
| 371 |
-
"grad_norm": 2.
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 4.8,
|
| 378 |
-
"grad_norm": 2.
|
| 379 |
-
"learning_rate": 1.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 4.9,
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 5.0,
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.0,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime": 5.
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second":
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch": 5.
|
| 408 |
-
"
|
| 409 |
-
"
|
| 410 |
-
"
|
| 411 |
-
"
|
| 412 |
-
|
| 413 |
-
|
| 414 |
-
"epoch": 5.2,
|
| 415 |
-
"grad_norm": 1.6726042032241821,
|
| 416 |
-
"learning_rate": 1.6e-05,
|
| 417 |
-
"loss": 0.2886,
|
| 418 |
-
"step": 520
|
| 419 |
-
},
|
| 420 |
-
{
|
| 421 |
-
"epoch": 5.3,
|
| 422 |
-
"grad_norm": 2.055961847305298,
|
| 423 |
-
"learning_rate": 1.5666666666666667e-05,
|
| 424 |
-
"loss": 0.2745,
|
| 425 |
-
"step": 530
|
| 426 |
-
},
|
| 427 |
-
{
|
| 428 |
-
"epoch": 5.4,
|
| 429 |
-
"grad_norm": 2.2671661376953125,
|
| 430 |
-
"learning_rate": 1.533333333333333e-05,
|
| 431 |
-
"loss": 0.2943,
|
| 432 |
-
"step": 540
|
| 433 |
-
},
|
| 434 |
-
{
|
| 435 |
-
"epoch": 5.5,
|
| 436 |
-
"grad_norm": 2.010511636734009,
|
| 437 |
-
"learning_rate": 1.5e-05,
|
| 438 |
-
"loss": 0.246,
|
| 439 |
-
"step": 550
|
| 440 |
-
},
|
| 441 |
-
{
|
| 442 |
-
"epoch": 5.6,
|
| 443 |
-
"grad_norm": 2.027716636657715,
|
| 444 |
-
"learning_rate": 1.4666666666666666e-05,
|
| 445 |
-
"loss": 0.2962,
|
| 446 |
-
"step": 560
|
| 447 |
-
},
|
| 448 |
-
{
|
| 449 |
-
"epoch": 5.7,
|
| 450 |
-
"grad_norm": 1.839068055152893,
|
| 451 |
-
"learning_rate": 1.4333333333333334e-05,
|
| 452 |
-
"loss": 0.2817,
|
| 453 |
-
"step": 570
|
| 454 |
-
},
|
| 455 |
-
{
|
| 456 |
-
"epoch": 5.8,
|
| 457 |
-
"grad_norm": 2.8274471759796143,
|
| 458 |
-
"learning_rate": 1.4e-05,
|
| 459 |
-
"loss": 0.2512,
|
| 460 |
-
"step": 580
|
| 461 |
-
},
|
| 462 |
-
{
|
| 463 |
-
"epoch": 5.9,
|
| 464 |
-
"grad_norm": 2.0226399898529053,
|
| 465 |
-
"learning_rate": 1.3666666666666667e-05,
|
| 466 |
-
"loss": 0.2756,
|
| 467 |
-
"step": 590
|
| 468 |
-
},
|
| 469 |
-
{
|
| 470 |
-
"epoch": 6.0,
|
| 471 |
-
"grad_norm": 2.212151050567627,
|
| 472 |
-
"learning_rate": 1.3333333333333333e-05,
|
| 473 |
-
"loss": 0.242,
|
| 474 |
-
"step": 600
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 6.0,
|
| 478 |
-
"eval_accuracy": 0.9792586054721977,
|
| 479 |
-
"eval_loss": 0.1708967536687851,
|
| 480 |
-
"eval_runtime": 4.9642,
|
| 481 |
-
"eval_samples_per_second": 1369.415,
|
| 482 |
-
"eval_steps_per_second": 42.908,
|
| 483 |
-
"step": 600
|
| 484 |
-
},
|
| 485 |
-
{
|
| 486 |
-
"epoch": 6.1,
|
| 487 |
-
"grad_norm": 2.019993305206299,
|
| 488 |
-
"learning_rate": 1.3000000000000001e-05,
|
| 489 |
-
"loss": 0.2557,
|
| 490 |
-
"step": 610
|
| 491 |
-
},
|
| 492 |
-
{
|
| 493 |
-
"epoch": 6.2,
|
| 494 |
-
"grad_norm": 2.0448718070983887,
|
| 495 |
-
"learning_rate": 1.2666666666666667e-05,
|
| 496 |
-
"loss": 0.2455,
|
| 497 |
-
"step": 620
|
| 498 |
-
},
|
| 499 |
-
{
|
| 500 |
-
"epoch": 6.3,
|
| 501 |
-
"grad_norm": 2.4769129753112793,
|
| 502 |
-
"learning_rate": 1.2333333333333333e-05,
|
| 503 |
-
"loss": 0.2435,
|
| 504 |
-
"step": 630
|
| 505 |
-
},
|
| 506 |
-
{
|
| 507 |
-
"epoch": 6.4,
|
| 508 |
-
"grad_norm": 2.217947483062744,
|
| 509 |
-
"learning_rate": 1.2e-05,
|
| 510 |
-
"loss": 0.2392,
|
| 511 |
-
"step": 640
|
| 512 |
-
},
|
| 513 |
-
{
|
| 514 |
-
"epoch": 6.5,
|
| 515 |
-
"grad_norm": 2.198836326599121,
|
| 516 |
-
"learning_rate": 1.1666666666666668e-05,
|
| 517 |
-
"loss": 0.2194,
|
| 518 |
-
"step": 650
|
| 519 |
-
},
|
| 520 |
-
{
|
| 521 |
-
"epoch": 6.6,
|
| 522 |
-
"grad_norm": 2.2629637718200684,
|
| 523 |
-
"learning_rate": 1.1333333333333334e-05,
|
| 524 |
-
"loss": 0.2165,
|
| 525 |
-
"step": 660
|
| 526 |
-
},
|
| 527 |
-
{
|
| 528 |
-
"epoch": 6.7,
|
| 529 |
-
"grad_norm": 1.9289292097091675,
|
| 530 |
-
"learning_rate": 1.1e-05,
|
| 531 |
-
"loss": 0.2264,
|
| 532 |
-
"step": 670
|
| 533 |
-
},
|
| 534 |
-
{
|
| 535 |
-
"epoch": 6.8,
|
| 536 |
-
"grad_norm": 2.585972547531128,
|
| 537 |
-
"learning_rate": 1.0666666666666667e-05,
|
| 538 |
-
"loss": 0.2254,
|
| 539 |
-
"step": 680
|
| 540 |
-
},
|
| 541 |
-
{
|
| 542 |
-
"epoch": 6.9,
|
| 543 |
-
"grad_norm": 2.3876583576202393,
|
| 544 |
-
"learning_rate": 1.0333333333333333e-05,
|
| 545 |
-
"loss": 0.2264,
|
| 546 |
-
"step": 690
|
| 547 |
-
},
|
| 548 |
-
{
|
| 549 |
-
"epoch": 7.0,
|
| 550 |
-
"grad_norm": 2.297879695892334,
|
| 551 |
-
"learning_rate": 9.999999999999999e-06,
|
| 552 |
-
"loss": 0.2144,
|
| 553 |
-
"step": 700
|
| 554 |
-
},
|
| 555 |
-
{
|
| 556 |
-
"epoch": 7.0,
|
| 557 |
-
"eval_accuracy": 0.9805825242718447,
|
| 558 |
-
"eval_loss": 0.14598147571086884,
|
| 559 |
-
"eval_runtime": 4.9644,
|
| 560 |
-
"eval_samples_per_second": 1369.359,
|
| 561 |
-
"eval_steps_per_second": 42.906,
|
| 562 |
-
"step": 700
|
| 563 |
-
},
|
| 564 |
-
{
|
| 565 |
-
"epoch": 7.1,
|
| 566 |
-
"grad_norm": 2.2874860763549805,
|
| 567 |
-
"learning_rate": 9.666666666666667e-06,
|
| 568 |
-
"loss": 0.2217,
|
| 569 |
-
"step": 710
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 7.2,
|
| 573 |
-
"grad_norm": 1.6444002389907837,
|
| 574 |
-
"learning_rate": 9.333333333333334e-06,
|
| 575 |
-
"loss": 0.2048,
|
| 576 |
-
"step": 720
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 7.3,
|
| 580 |
-
"grad_norm": 2.211688280105591,
|
| 581 |
-
"learning_rate": 9e-06,
|
| 582 |
-
"loss": 0.2122,
|
| 583 |
-
"step": 730
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 7.4,
|
| 587 |
-
"grad_norm": 2.334533929824829,
|
| 588 |
-
"learning_rate": 8.666666666666666e-06,
|
| 589 |
-
"loss": 0.2142,
|
| 590 |
-
"step": 740
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 7.5,
|
| 594 |
-
"grad_norm": 1.83983314037323,
|
| 595 |
-
"learning_rate": 8.333333333333334e-06,
|
| 596 |
-
"loss": 0.1873,
|
| 597 |
-
"step": 750
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 7.6,
|
| 601 |
-
"grad_norm": 1.8291276693344116,
|
| 602 |
-
"learning_rate": 8e-06,
|
| 603 |
-
"loss": 0.2003,
|
| 604 |
-
"step": 760
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 7.7,
|
| 608 |
-
"grad_norm": 2.4695987701416016,
|
| 609 |
-
"learning_rate": 7.666666666666666e-06,
|
| 610 |
-
"loss": 0.2075,
|
| 611 |
-
"step": 770
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 7.8,
|
| 615 |
-
"grad_norm": 1.7656598091125488,
|
| 616 |
-
"learning_rate": 7.333333333333333e-06,
|
| 617 |
-
"loss": 0.2089,
|
| 618 |
-
"step": 780
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 7.9,
|
| 622 |
-
"grad_norm": 1.8952357769012451,
|
| 623 |
-
"learning_rate": 7e-06,
|
| 624 |
-
"loss": 0.2013,
|
| 625 |
-
"step": 790
|
| 626 |
-
},
|
| 627 |
-
{
|
| 628 |
-
"epoch": 8.0,
|
| 629 |
-
"grad_norm": 1.6189404726028442,
|
| 630 |
-
"learning_rate": 6.666666666666667e-06,
|
| 631 |
-
"loss": 0.1977,
|
| 632 |
-
"step": 800
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 8.0,
|
| 636 |
-
"eval_accuracy": 0.979994115916446,
|
| 637 |
-
"eval_loss": 0.13441701233386993,
|
| 638 |
-
"eval_runtime": 5.0435,
|
| 639 |
-
"eval_samples_per_second": 1347.881,
|
| 640 |
-
"eval_steps_per_second": 42.233,
|
| 641 |
-
"step": 800
|
| 642 |
-
},
|
| 643 |
-
{
|
| 644 |
-
"epoch": 8.1,
|
| 645 |
-
"grad_norm": 2.2733678817749023,
|
| 646 |
-
"learning_rate": 6.333333333333333e-06,
|
| 647 |
-
"loss": 0.1908,
|
| 648 |
-
"step": 810
|
| 649 |
-
},
|
| 650 |
-
{
|
| 651 |
-
"epoch": 8.2,
|
| 652 |
-
"grad_norm": 1.6874712705612183,
|
| 653 |
-
"learning_rate": 6e-06,
|
| 654 |
-
"loss": 0.1819,
|
| 655 |
-
"step": 820
|
| 656 |
-
},
|
| 657 |
-
{
|
| 658 |
-
"epoch": 8.3,
|
| 659 |
-
"grad_norm": 1.4649100303649902,
|
| 660 |
-
"learning_rate": 5.666666666666667e-06,
|
| 661 |
-
"loss": 0.2037,
|
| 662 |
-
"step": 830
|
| 663 |
-
},
|
| 664 |
-
{
|
| 665 |
-
"epoch": 8.4,
|
| 666 |
-
"grad_norm": 1.7231727838516235,
|
| 667 |
-
"learning_rate": 5.333333333333334e-06,
|
| 668 |
-
"loss": 0.1935,
|
| 669 |
-
"step": 840
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 8.5,
|
| 673 |
-
"grad_norm": 1.6908072233200073,
|
| 674 |
-
"learning_rate": 4.9999999999999996e-06,
|
| 675 |
-
"loss": 0.1957,
|
| 676 |
-
"step": 850
|
| 677 |
-
},
|
| 678 |
-
{
|
| 679 |
-
"epoch": 8.6,
|
| 680 |
-
"grad_norm": 2.1523678302764893,
|
| 681 |
-
"learning_rate": 4.666666666666667e-06,
|
| 682 |
-
"loss": 0.1955,
|
| 683 |
-
"step": 860
|
| 684 |
-
},
|
| 685 |
-
{
|
| 686 |
-
"epoch": 8.7,
|
| 687 |
-
"grad_norm": 2.5662710666656494,
|
| 688 |
-
"learning_rate": 4.333333333333333e-06,
|
| 689 |
-
"loss": 0.2008,
|
| 690 |
-
"step": 870
|
| 691 |
-
},
|
| 692 |
-
{
|
| 693 |
-
"epoch": 8.8,
|
| 694 |
-
"grad_norm": 1.984096884727478,
|
| 695 |
-
"learning_rate": 4e-06,
|
| 696 |
-
"loss": 0.186,
|
| 697 |
-
"step": 880
|
| 698 |
-
},
|
| 699 |
-
{
|
| 700 |
-
"epoch": 8.9,
|
| 701 |
-
"grad_norm": 1.4925553798675537,
|
| 702 |
-
"learning_rate": 3.6666666666666666e-06,
|
| 703 |
-
"loss": 0.1905,
|
| 704 |
-
"step": 890
|
| 705 |
-
},
|
| 706 |
-
{
|
| 707 |
-
"epoch": 9.0,
|
| 708 |
-
"grad_norm": 4.499406814575195,
|
| 709 |
-
"learning_rate": 3.3333333333333333e-06,
|
| 710 |
-
"loss": 0.1742,
|
| 711 |
-
"step": 900
|
| 712 |
-
},
|
| 713 |
-
{
|
| 714 |
-
"epoch": 9.0,
|
| 715 |
-
"eval_accuracy": 0.9796999117387467,
|
| 716 |
-
"eval_loss": 0.12820769846439362,
|
| 717 |
-
"eval_runtime": 5.0266,
|
| 718 |
-
"eval_samples_per_second": 1352.417,
|
| 719 |
-
"eval_steps_per_second": 42.375,
|
| 720 |
-
"step": 900
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 9.1,
|
| 724 |
-
"grad_norm": 1.9934319257736206,
|
| 725 |
-
"learning_rate": 3e-06,
|
| 726 |
-
"loss": 0.1915,
|
| 727 |
-
"step": 910
|
| 728 |
-
},
|
| 729 |
-
{
|
| 730 |
-
"epoch": 9.2,
|
| 731 |
-
"grad_norm": 1.9417580366134644,
|
| 732 |
-
"learning_rate": 2.666666666666667e-06,
|
| 733 |
-
"loss": 0.1759,
|
| 734 |
-
"step": 920
|
| 735 |
-
},
|
| 736 |
-
{
|
| 737 |
-
"epoch": 9.3,
|
| 738 |
-
"grad_norm": 2.1848134994506836,
|
| 739 |
-
"learning_rate": 2.3333333333333336e-06,
|
| 740 |
-
"loss": 0.1811,
|
| 741 |
-
"step": 930
|
| 742 |
-
},
|
| 743 |
-
{
|
| 744 |
-
"epoch": 9.4,
|
| 745 |
-
"grad_norm": 1.6830646991729736,
|
| 746 |
-
"learning_rate": 2e-06,
|
| 747 |
-
"loss": 0.1905,
|
| 748 |
-
"step": 940
|
| 749 |
-
},
|
| 750 |
-
{
|
| 751 |
-
"epoch": 9.5,
|
| 752 |
-
"grad_norm": 2.51246976852417,
|
| 753 |
-
"learning_rate": 1.6666666666666667e-06,
|
| 754 |
-
"loss": 0.1889,
|
| 755 |
-
"step": 950
|
| 756 |
-
},
|
| 757 |
-
{
|
| 758 |
-
"epoch": 9.6,
|
| 759 |
-
"grad_norm": 2.315727949142456,
|
| 760 |
-
"learning_rate": 1.3333333333333334e-06,
|
| 761 |
-
"loss": 0.1755,
|
| 762 |
-
"step": 960
|
| 763 |
-
},
|
| 764 |
-
{
|
| 765 |
-
"epoch": 9.7,
|
| 766 |
-
"grad_norm": 1.6819000244140625,
|
| 767 |
-
"learning_rate": 1e-06,
|
| 768 |
-
"loss": 0.1717,
|
| 769 |
-
"step": 970
|
| 770 |
-
},
|
| 771 |
-
{
|
| 772 |
-
"epoch": 9.8,
|
| 773 |
-
"grad_norm": 1.7445892095565796,
|
| 774 |
-
"learning_rate": 6.666666666666667e-07,
|
| 775 |
-
"loss": 0.1756,
|
| 776 |
-
"step": 980
|
| 777 |
-
},
|
| 778 |
-
{
|
| 779 |
-
"epoch": 9.9,
|
| 780 |
-
"grad_norm": 1.542716145515442,
|
| 781 |
-
"learning_rate": 3.3333333333333335e-07,
|
| 782 |
-
"loss": 0.181,
|
| 783 |
-
"step": 990
|
| 784 |
-
},
|
| 785 |
-
{
|
| 786 |
-
"epoch": 10.0,
|
| 787 |
-
"grad_norm": 2.6007721424102783,
|
| 788 |
-
"learning_rate": 0.0,
|
| 789 |
-
"loss": 0.1867,
|
| 790 |
-
"step": 1000
|
| 791 |
-
},
|
| 792 |
-
{
|
| 793 |
-
"epoch": 10.0,
|
| 794 |
-
"eval_accuracy": 0.9811709326272433,
|
| 795 |
-
"eval_loss": 0.12430207431316376,
|
| 796 |
-
"eval_runtime": 5.3794,
|
| 797 |
-
"eval_samples_per_second": 1263.701,
|
| 798 |
-
"eval_steps_per_second": 39.595,
|
| 799 |
-
"step": 1000
|
| 800 |
-
},
|
| 801 |
-
{
|
| 802 |
-
"epoch": 10.0,
|
| 803 |
-
"step": 1000,
|
| 804 |
-
"total_flos": 4.72566865822464e+18,
|
| 805 |
-
"train_loss": 0.7669839228391647,
|
| 806 |
-
"train_runtime": 621.093,
|
| 807 |
-
"train_samples_per_second": 822.646,
|
| 808 |
-
"train_steps_per_second": 1.61
|
| 809 |
}
|
| 810 |
],
|
| 811 |
"logging_steps": 10,
|
| 812 |
-
"max_steps":
|
| 813 |
"num_input_tokens_seen": 0,
|
| 814 |
-
"num_train_epochs":
|
| 815 |
"save_steps": 500,
|
| 816 |
"stateful_callbacks": {
|
| 817 |
"TrainerControl": {
|
|
@@ -825,7 +430,7 @@
|
|
| 825 |
"attributes": {}
|
| 826 |
}
|
| 827 |
},
|
| 828 |
-
"total_flos":
|
| 829 |
"train_batch_size": 128,
|
| 830 |
"trial_name": null,
|
| 831 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8245072080023537,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-base-ft-keyword-spotting/checkpoint-500",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 3.1108202934265137,
|
| 14 |
+
"learning_rate": 6e-06,
|
| 15 |
+
"loss": 4.1853,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.2,
|
| 20 |
+
"grad_norm": 3.643692970275879,
|
| 21 |
+
"learning_rate": 1.2e-05,
|
| 22 |
+
"loss": 4.0071,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3,
|
| 27 |
+
"grad_norm": 6.08334493637085,
|
| 28 |
+
"learning_rate": 1.8e-05,
|
| 29 |
+
"loss": 3.4541,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.4,
|
| 34 |
+
"grad_norm": 5.621030330657959,
|
| 35 |
+
"learning_rate": 2.4e-05,
|
| 36 |
+
"loss": 2.6872,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5,
|
| 41 |
+
"grad_norm": 4.663188457489014,
|
| 42 |
+
"learning_rate": 3e-05,
|
| 43 |
+
"loss": 2.25,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.6,
|
| 48 |
+
"grad_norm": 2.963801383972168,
|
| 49 |
+
"learning_rate": 2.9333333333333333e-05,
|
| 50 |
+
"loss": 2.0037,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.7,
|
| 55 |
+
"grad_norm": 2.0093071460723877,
|
| 56 |
+
"learning_rate": 2.8666666666666668e-05,
|
| 57 |
+
"loss": 1.8419,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.8,
|
| 62 |
+
"grad_norm": 0.7109726071357727,
|
| 63 |
+
"learning_rate": 2.8e-05,
|
| 64 |
+
"loss": 1.8088,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9,
|
| 69 |
+
"grad_norm": 0.4631745219230652,
|
| 70 |
+
"learning_rate": 2.7333333333333335e-05,
|
| 71 |
+
"loss": 1.7483,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.0,
|
| 76 |
+
"grad_norm": 0.49370667338371277,
|
| 77 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 78 |
+
"loss": 1.7728,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.0,
|
| 83 |
"eval_accuracy": 0.6209179170344219,
|
| 84 |
+
"eval_loss": 1.7270870208740234,
|
| 85 |
+
"eval_runtime": 5.1768,
|
| 86 |
+
"eval_samples_per_second": 1313.173,
|
| 87 |
+
"eval_steps_per_second": 41.145,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.1,
|
| 92 |
+
"grad_norm": 1.3470516204833984,
|
| 93 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 94 |
+
"loss": 1.7404,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.2,
|
| 99 |
+
"grad_norm": 2.36470627784729,
|
| 100 |
+
"learning_rate": 2.5333333333333334e-05,
|
| 101 |
+
"loss": 1.6391,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.3,
|
| 106 |
+
"grad_norm": 3.1936564445495605,
|
| 107 |
+
"learning_rate": 2.4666666666666665e-05,
|
| 108 |
+
"loss": 1.5966,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.4,
|
| 113 |
+
"grad_norm": 6.025561332702637,
|
| 114 |
+
"learning_rate": 2.4e-05,
|
| 115 |
+
"loss": 1.5888,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.5,
|
| 120 |
+
"grad_norm": 3.4104502201080322,
|
| 121 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 122 |
+
"loss": 1.4921,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.6,
|
| 127 |
+
"grad_norm": 4.1762871742248535,
|
| 128 |
+
"learning_rate": 2.2666666666666668e-05,
|
| 129 |
+
"loss": 1.5404,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 1.7,
|
| 134 |
+
"grad_norm": 2.305497646331787,
|
| 135 |
+
"learning_rate": 2.2e-05,
|
| 136 |
+
"loss": 1.4838,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.8,
|
| 141 |
+
"grad_norm": 1.8129342794418335,
|
| 142 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 143 |
+
"loss": 1.4184,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.9,
|
| 148 |
+
"grad_norm": 4.611838340759277,
|
| 149 |
+
"learning_rate": 2.0666666666666666e-05,
|
| 150 |
+
"loss": 1.3657,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 2.0,
|
| 155 |
+
"grad_norm": 3.684983730316162,
|
| 156 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 157 |
+
"loss": 1.3042,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 2.0,
|
| 162 |
+
"eval_accuracy": 0.6493086201824065,
|
| 163 |
+
"eval_loss": 1.2064741849899292,
|
| 164 |
+
"eval_runtime": 4.8721,
|
| 165 |
+
"eval_samples_per_second": 1395.278,
|
| 166 |
+
"eval_steps_per_second": 43.718,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.1,
|
| 171 |
+
"grad_norm": 3.525785446166992,
|
| 172 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 173 |
+
"loss": 1.3191,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.2,
|
| 178 |
+
"grad_norm": 3.760420322418213,
|
| 179 |
+
"learning_rate": 1.866666666666667e-05,
|
| 180 |
+
"loss": 1.2842,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.3,
|
| 185 |
+
"grad_norm": 5.203872203826904,
|
| 186 |
+
"learning_rate": 1.8e-05,
|
| 187 |
+
"loss": 1.2544,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.4,
|
| 192 |
+
"grad_norm": 4.353450298309326,
|
| 193 |
+
"learning_rate": 1.7333333333333332e-05,
|
| 194 |
+
"loss": 1.2532,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 2.5,
|
| 199 |
+
"grad_norm": 4.413931846618652,
|
| 200 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 201 |
+
"loss": 1.1657,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 2.6,
|
| 206 |
+
"grad_norm": 2.3682613372802734,
|
| 207 |
+
"learning_rate": 1.6e-05,
|
| 208 |
+
"loss": 1.1996,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.7,
|
| 213 |
+
"grad_norm": 3.3464930057525635,
|
| 214 |
+
"learning_rate": 1.533333333333333e-05,
|
| 215 |
+
"loss": 1.1394,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 2.8,
|
| 220 |
+
"grad_norm": 3.987032890319824,
|
| 221 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 222 |
+
"loss": 1.12,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 2.9,
|
| 227 |
+
"grad_norm": 2.3071091175079346,
|
| 228 |
+
"learning_rate": 1.4e-05,
|
| 229 |
+
"loss": 1.1157,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 3.0,
|
| 234 |
+
"grad_norm": 7.341511249542236,
|
| 235 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 236 |
+
"loss": 1.1086,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 3.0,
|
| 241 |
+
"eval_accuracy": 0.6935863489261548,
|
| 242 |
+
"eval_loss": 0.9966975450515747,
|
| 243 |
+
"eval_runtime": 4.846,
|
| 244 |
+
"eval_samples_per_second": 1402.813,
|
| 245 |
+
"eval_steps_per_second": 43.954,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 3.1,
|
| 250 |
+
"grad_norm": 3.0323097705841064,
|
| 251 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 252 |
+
"loss": 1.0664,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 3.2,
|
| 257 |
+
"grad_norm": 3.4883832931518555,
|
| 258 |
+
"learning_rate": 1.2e-05,
|
| 259 |
+
"loss": 1.0835,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 3.3,
|
| 264 |
+
"grad_norm": 4.886635780334473,
|
| 265 |
+
"learning_rate": 1.1333333333333334e-05,
|
| 266 |
+
"loss": 1.0608,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 3.4,
|
| 271 |
+
"grad_norm": 2.4810945987701416,
|
| 272 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 273 |
+
"loss": 1.0253,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 3.5,
|
| 278 |
+
"grad_norm": 2.3833069801330566,
|
| 279 |
+
"learning_rate": 9.999999999999999e-06,
|
| 280 |
+
"loss": 1.0187,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.6,
|
| 285 |
+
"grad_norm": 2.031707763671875,
|
| 286 |
+
"learning_rate": 9.333333333333334e-06,
|
| 287 |
+
"loss": 1.002,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 3.7,
|
| 292 |
+
"grad_norm": 3.9627315998077393,
|
| 293 |
+
"learning_rate": 8.666666666666666e-06,
|
| 294 |
+
"loss": 0.9833,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.8,
|
| 299 |
+
"grad_norm": 3.683357000350952,
|
| 300 |
+
"learning_rate": 8e-06,
|
| 301 |
+
"loss": 1.0085,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 3.9,
|
| 306 |
+
"grad_norm": 2.747379779815674,
|
| 307 |
+
"learning_rate": 7.333333333333333e-06,
|
| 308 |
+
"loss": 1.0044,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 4.0,
|
| 313 |
+
"grad_norm": 2.9549131393432617,
|
| 314 |
+
"learning_rate": 6.666666666666667e-06,
|
| 315 |
+
"loss": 0.967,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 4.0,
|
| 320 |
+
"eval_accuracy": 0.793027360988526,
|
| 321 |
+
"eval_loss": 0.9590433239936829,
|
| 322 |
+
"eval_runtime": 4.8849,
|
| 323 |
+
"eval_samples_per_second": 1391.628,
|
| 324 |
+
"eval_steps_per_second": 43.604,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 4.1,
|
| 329 |
+
"grad_norm": 2.6056904792785645,
|
| 330 |
+
"learning_rate": 6e-06,
|
| 331 |
+
"loss": 0.981,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 4.2,
|
| 336 |
+
"grad_norm": 2.720146894454956,
|
| 337 |
+
"learning_rate": 5.333333333333334e-06,
|
| 338 |
+
"loss": 0.9899,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 4.3,
|
| 343 |
+
"grad_norm": 3.6615264415740967,
|
| 344 |
+
"learning_rate": 4.666666666666667e-06,
|
| 345 |
+
"loss": 0.9411,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 4.4,
|
| 350 |
+
"grad_norm": 4.013997554779053,
|
| 351 |
+
"learning_rate": 4e-06,
|
| 352 |
+
"loss": 0.9413,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 4.5,
|
| 357 |
+
"grad_norm": 2.698765754699707,
|
| 358 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 359 |
+
"loss": 0.9145,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 4.6,
|
| 364 |
+
"grad_norm": 2.6457314491271973,
|
| 365 |
+
"learning_rate": 2.666666666666667e-06,
|
| 366 |
+
"loss": 0.9069,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.7,
|
| 371 |
+
"grad_norm": 2.491399049758911,
|
| 372 |
+
"learning_rate": 2e-06,
|
| 373 |
+
"loss": 0.8928,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 4.8,
|
| 378 |
+
"grad_norm": 2.6820359230041504,
|
| 379 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 380 |
+
"loss": 0.9105,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 4.9,
|
| 385 |
+
"grad_norm": 5.143585205078125,
|
| 386 |
+
"learning_rate": 6.666666666666667e-07,
|
| 387 |
+
"loss": 0.9215,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 5.0,
|
| 392 |
+
"grad_norm": 4.706216335296631,
|
| 393 |
+
"learning_rate": 0.0,
|
| 394 |
+
"loss": 0.8801,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 5.0,
|
| 399 |
+
"eval_accuracy": 0.8245072080023537,
|
| 400 |
+
"eval_loss": 0.8822575807571411,
|
| 401 |
+
"eval_runtime": 5.2526,
|
| 402 |
+
"eval_samples_per_second": 1294.22,
|
| 403 |
+
"eval_steps_per_second": 40.551,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 5.0,
|
| 408 |
+
"step": 500,
|
| 409 |
+
"total_flos": 2.36283432911232e+18,
|
| 410 |
+
"train_loss": 1.447761106491089,
|
| 411 |
+
"train_runtime": 310.8685,
|
| 412 |
+
"train_samples_per_second": 821.794,
|
| 413 |
+
"train_steps_per_second": 1.608
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 414 |
}
|
| 415 |
],
|
| 416 |
"logging_steps": 10,
|
| 417 |
+
"max_steps": 500,
|
| 418 |
"num_input_tokens_seen": 0,
|
| 419 |
+
"num_train_epochs": 5,
|
| 420 |
"save_steps": 500,
|
| 421 |
"stateful_callbacks": {
|
| 422 |
"TrainerControl": {
|
|
|
|
| 430 |
"attributes": {}
|
| 431 |
}
|
| 432 |
},
|
| 433 |
+
"total_flos": 2.36283432911232e+18,
|
| 434 |
"train_batch_size": 128,
|
| 435 |
"trial_name": null,
|
| 436 |
"trial_params": null
|