Model save
Browse files- README.md +8 -8
- all_results.json +6 -5
- train_results.json +6 -5
- trainer_state.json +1415 -311
README.md
CHANGED
|
@@ -36,14 +36,14 @@ More information needed
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 0.0002
|
| 39 |
-
- train_batch_size:
|
| 40 |
- eval_batch_size: 8
|
| 41 |
- seed: 42
|
| 42 |
- distributed_type: multi-GPU
|
| 43 |
-
- num_devices:
|
| 44 |
- gradient_accumulation_steps: 2
|
| 45 |
- total_train_batch_size: 16
|
| 46 |
-
- total_eval_batch_size:
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: cosine
|
| 49 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -55,8 +55,8 @@ The following hyperparameters were used during training:
|
|
| 55 |
|
| 56 |
### Framework versions
|
| 57 |
|
| 58 |
-
- PEFT 0.
|
| 59 |
-
- Transformers 4.
|
| 60 |
-
- Pytorch 2.
|
| 61 |
-
- Datasets
|
| 62 |
-
- Tokenizers 0.
|
|
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 0.0002
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
- eval_batch_size: 8
|
| 41 |
- seed: 42
|
| 42 |
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 2
|
| 44 |
- gradient_accumulation_steps: 2
|
| 45 |
- total_train_batch_size: 16
|
| 46 |
+
- total_eval_batch_size: 16
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: cosine
|
| 49 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 55 |
|
| 56 |
### Framework versions
|
| 57 |
|
| 58 |
+
- PEFT 0.13.1
|
| 59 |
+
- Transformers 4.45.2
|
| 60 |
+
- Pytorch 2.4.1+cu121
|
| 61 |
+
- Datasets 3.0.1
|
| 62 |
+
- Tokenizers 0.20.0
|
all_results.json
CHANGED
|
@@ -1,8 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.025624031790891e+18,
|
| 4 |
+
"train_loss": 0.004042043934228078,
|
| 5 |
+
"train_runtime": 6129.3476,
|
| 6 |
+
"train_samples": 18448,
|
| 7 |
+
"train_samples_per_second": 3.01,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.025624031790891e+18,
|
| 4 |
+
"train_loss": 0.004042043934228078,
|
| 5 |
+
"train_runtime": 6129.3476,
|
| 6 |
+
"train_samples": 18448,
|
| 7 |
+
"train_samples_per_second": 3.01,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,553 +3,1657 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 0.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 0.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate": 0.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate": 0.
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
-
"learning_rate": 0.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate": 0.
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate": 0.
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate": 0.
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
-
"learning_rate": 0.
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate": 0.
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
-
"learning_rate": 0.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
-
"learning_rate": 0.
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate": 0.
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate": 0.
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate": 0.
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate": 0.
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss": 0.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate":
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss": 0.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm": 0.
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss": 0.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch":
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
-
"learning_rate": 0.
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 370
|
| 535 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 536 |
{
|
| 537 |
"epoch": 1.0,
|
| 538 |
-
"step":
|
| 539 |
-
"total_flos":
|
| 540 |
-
"train_loss": 0.
|
| 541 |
-
"train_runtime":
|
| 542 |
-
"train_samples_per_second":
|
| 543 |
-
"train_steps_per_second": 0.
|
| 544 |
}
|
| 545 |
],
|
| 546 |
"logging_steps": 5,
|
| 547 |
-
"max_steps":
|
| 548 |
"num_input_tokens_seen": 0,
|
| 549 |
"num_train_epochs": 1,
|
| 550 |
-
"save_steps":
|
| 551 |
-
"
|
| 552 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 553 |
"trial_name": null,
|
| 554 |
"trial_params": null
|
| 555 |
}
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1153,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0008673026886383347,
|
| 13 |
+
"grad_norm": 0.824800431728363,
|
| 14 |
+
"learning_rate": 1.724137931034483e-06,
|
| 15 |
+
"loss": 0.2659,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.004336513443191674,
|
| 20 |
+
"grad_norm": 0.697605550289154,
|
| 21 |
+
"learning_rate": 8.620689655172414e-06,
|
| 22 |
+
"loss": 0.3205,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.008673026886383347,
|
| 27 |
+
"grad_norm": 0.5697792172431946,
|
| 28 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 29 |
+
"loss": 0.2515,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.013009540329575022,
|
| 34 |
+
"grad_norm": 0.4874624013900757,
|
| 35 |
+
"learning_rate": 2.5862068965517244e-05,
|
| 36 |
+
"loss": 0.1831,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.017346053772766695,
|
| 41 |
+
"grad_norm": 0.32868602871894836,
|
| 42 |
+
"learning_rate": 3.4482758620689657e-05,
|
| 43 |
+
"loss": 0.0883,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.02168256721595837,
|
| 48 |
+
"grad_norm": 0.17428267002105713,
|
| 49 |
+
"learning_rate": 4.3103448275862066e-05,
|
| 50 |
+
"loss": 0.0309,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.026019080659150044,
|
| 55 |
+
"grad_norm": 0.07211117446422577,
|
| 56 |
+
"learning_rate": 5.172413793103449e-05,
|
| 57 |
+
"loss": 0.013,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.030355594102341718,
|
| 62 |
+
"grad_norm": 0.07596682757139206,
|
| 63 |
+
"learning_rate": 6.03448275862069e-05,
|
| 64 |
+
"loss": 0.0043,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.03469210754553339,
|
| 69 |
+
"grad_norm": 0.06741539388895035,
|
| 70 |
+
"learning_rate": 6.896551724137931e-05,
|
| 71 |
+
"loss": 0.0031,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.03902862098872507,
|
| 76 |
+
"grad_norm": 0.020168501883745193,
|
| 77 |
+
"learning_rate": 7.758620689655173e-05,
|
| 78 |
+
"loss": 0.0015,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.04336513443191674,
|
| 83 |
+
"grad_norm": 0.037137292325496674,
|
| 84 |
+
"learning_rate": 8.620689655172413e-05,
|
| 85 |
+
"loss": 0.0015,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.047701647875108416,
|
| 90 |
+
"grad_norm": 0.13907165825366974,
|
| 91 |
+
"learning_rate": 9.482758620689656e-05,
|
| 92 |
+
"loss": 0.0009,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.05203816131830009,
|
| 97 |
+
"grad_norm": 0.0042100874707102776,
|
| 98 |
+
"learning_rate": 0.00010344827586206898,
|
| 99 |
+
"loss": 0.0013,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.05637467476149176,
|
| 104 |
+
"grad_norm": 0.10066337138414383,
|
| 105 |
+
"learning_rate": 0.00011206896551724138,
|
| 106 |
+
"loss": 0.0011,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.060711188204683436,
|
| 111 |
+
"grad_norm": 0.0019178424263373017,
|
| 112 |
+
"learning_rate": 0.0001206896551724138,
|
| 113 |
+
"loss": 0.0021,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.06504770164787511,
|
| 118 |
+
"grad_norm": 0.09226825088262558,
|
| 119 |
+
"learning_rate": 0.0001293103448275862,
|
| 120 |
+
"loss": 0.001,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.06938421509106678,
|
| 125 |
+
"grad_norm": 0.0025779225397855043,
|
| 126 |
+
"learning_rate": 0.00013793103448275863,
|
| 127 |
+
"loss": 0.0004,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.07372072853425846,
|
| 132 |
+
"grad_norm": 0.0068855443969368935,
|
| 133 |
+
"learning_rate": 0.00014655172413793104,
|
| 134 |
+
"loss": 0.0004,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.07805724197745013,
|
| 139 |
+
"grad_norm": 0.0020994944497942924,
|
| 140 |
+
"learning_rate": 0.00015517241379310346,
|
| 141 |
+
"loss": 0.0005,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.0823937554206418,
|
| 146 |
+
"grad_norm": 0.002034812932834029,
|
| 147 |
+
"learning_rate": 0.00016379310344827587,
|
| 148 |
+
"loss": 0.0003,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.08673026886383348,
|
| 153 |
+
"grad_norm": 0.0018760113744065166,
|
| 154 |
+
"learning_rate": 0.00017241379310344826,
|
| 155 |
+
"loss": 0.0025,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.09106678230702515,
|
| 160 |
+
"grad_norm": 0.0013358393916860223,
|
| 161 |
+
"learning_rate": 0.0001810344827586207,
|
| 162 |
+
"loss": 0.0004,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.09540329575021683,
|
| 167 |
+
"grad_norm": 0.01223821472376585,
|
| 168 |
+
"learning_rate": 0.00018965517241379312,
|
| 169 |
+
"loss": 0.0013,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.0997398091934085,
|
| 174 |
+
"grad_norm": 0.010651814751327038,
|
| 175 |
+
"learning_rate": 0.00019827586206896554,
|
| 176 |
+
"loss": 0.0011,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.10407632263660017,
|
| 181 |
+
"grad_norm": 0.0021788859739899635,
|
| 182 |
+
"learning_rate": 0.0001999926577882564,
|
| 183 |
+
"loss": 0.0003,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.10841283607979185,
|
| 188 |
+
"grad_norm": 0.006896303500980139,
|
| 189 |
+
"learning_rate": 0.00019996283190086388,
|
| 190 |
+
"loss": 0.0003,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.11274934952298352,
|
| 195 |
+
"grad_norm": 0.003941075410693884,
|
| 196 |
+
"learning_rate": 0.00019991007028765122,
|
| 197 |
+
"loss": 0.0004,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.1170858629661752,
|
| 202 |
+
"grad_norm": 0.0025480242911726236,
|
| 203 |
+
"learning_rate": 0.0001998343850543768,
|
| 204 |
+
"loss": 0.0004,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.12142237640936687,
|
| 209 |
+
"grad_norm": 0.0022759223356842995,
|
| 210 |
+
"learning_rate": 0.0001997357935664527,
|
| 211 |
+
"loss": 0.0017,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.12575888985255854,
|
| 216 |
+
"grad_norm": 0.004669133573770523,
|
| 217 |
+
"learning_rate": 0.00019961431844496002,
|
| 218 |
+
"loss": 0.0005,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.13009540329575023,
|
| 223 |
+
"grad_norm": 0.0009343024576082826,
|
| 224 |
+
"learning_rate": 0.0001994699875614589,
|
| 225 |
+
"loss": 0.0003,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.1344319167389419,
|
| 230 |
+
"grad_norm": 0.0016963341040536761,
|
| 231 |
+
"learning_rate": 0.00019930283403159357,
|
| 232 |
+
"loss": 0.0002,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.13876843018213356,
|
| 237 |
+
"grad_norm": 0.003908711485564709,
|
| 238 |
+
"learning_rate": 0.000199112896207494,
|
| 239 |
+
"loss": 0.0002,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.14310494362532525,
|
| 244 |
+
"grad_norm": 0.001237453892827034,
|
| 245 |
+
"learning_rate": 0.00019890021766897663,
|
| 246 |
+
"loss": 0.0003,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.1474414570685169,
|
| 251 |
+
"grad_norm": 0.0009205214446410537,
|
| 252 |
+
"learning_rate": 0.00019866484721354499,
|
| 253 |
+
"loss": 0.0002,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.15177797051170858,
|
| 258 |
+
"grad_norm": 0.0009426427423022687,
|
| 259 |
+
"learning_rate": 0.00019840683884519368,
|
| 260 |
+
"loss": 0.0009,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.15611448395490027,
|
| 265 |
+
"grad_norm": 0.005143130198121071,
|
| 266 |
+
"learning_rate": 0.00019812625176201745,
|
| 267 |
+
"loss": 0.0002,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.16045099739809193,
|
| 272 |
+
"grad_norm": 0.0014524105936288834,
|
| 273 |
+
"learning_rate": 0.0001978231503426287,
|
| 274 |
+
"loss": 0.0002,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.1647875108412836,
|
| 279 |
+
"grad_norm": 0.06288997083902359,
|
| 280 |
+
"learning_rate": 0.00019749760413138626,
|
| 281 |
+
"loss": 0.0007,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.1691240242844753,
|
| 286 |
+
"grad_norm": 0.0016669128090143204,
|
| 287 |
+
"learning_rate": 0.0001971496878224389,
|
| 288 |
+
"loss": 0.0014,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.17346053772766695,
|
| 293 |
+
"grad_norm": 0.001623625634238124,
|
| 294 |
+
"learning_rate": 0.00019677948124258748,
|
| 295 |
+
"loss": 0.0002,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.17779705117085862,
|
| 300 |
+
"grad_norm": 0.000822431524284184,
|
| 301 |
+
"learning_rate": 0.00019638706933296915,
|
| 302 |
+
"loss": 0.0001,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.1821335646140503,
|
| 307 |
+
"grad_norm": 0.004115803632885218,
|
| 308 |
+
"learning_rate": 0.00019597254212956822,
|
| 309 |
+
"loss": 0.0004,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.18647007805724197,
|
| 314 |
+
"grad_norm": 0.0006049461662769318,
|
| 315 |
+
"learning_rate": 0.00019553599474255812,
|
| 316 |
+
"loss": 0.0001,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.19080659150043366,
|
| 321 |
+
"grad_norm": 0.0010940938955172896,
|
| 322 |
+
"learning_rate": 0.0001950775273344792,
|
| 323 |
+
"loss": 0.0002,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.19514310494362533,
|
| 328 |
+
"grad_norm": 0.013846734538674355,
|
| 329 |
+
"learning_rate": 0.00019459724509725694,
|
| 330 |
+
"loss": 0.0002,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.199479618386817,
|
| 335 |
+
"grad_norm": 0.0004897746257483959,
|
| 336 |
+
"learning_rate": 0.00019409525822806662,
|
| 337 |
+
"loss": 0.0001,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.20381613183000868,
|
| 342 |
+
"grad_norm": 0.0004935916513204575,
|
| 343 |
+
"learning_rate": 0.00019357168190404936,
|
| 344 |
+
"loss": 0.0001,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.20815264527320035,
|
| 349 |
+
"grad_norm": 0.00045111175859346986,
|
| 350 |
+
"learning_rate": 0.00019302663625588563,
|
| 351 |
+
"loss": 0.0001,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.212489158716392,
|
| 356 |
+
"grad_norm": 0.0004669725021813065,
|
| 357 |
+
"learning_rate": 0.00019246024634023208,
|
| 358 |
+
"loss": 0.0001,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.2168256721595837,
|
| 363 |
+
"grad_norm": 0.0005562947480939329,
|
| 364 |
+
"learning_rate": 0.0001918726421110282,
|
| 365 |
+
"loss": 0.0001,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.22116218560277537,
|
| 370 |
+
"grad_norm": 0.0005943718133494258,
|
| 371 |
+
"learning_rate": 0.00019126395838967941,
|
| 372 |
+
"loss": 0.0001,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.22549869904596703,
|
| 377 |
+
"grad_norm": 0.000462163268821314,
|
| 378 |
+
"learning_rate": 0.00019063433483412347,
|
| 379 |
+
"loss": 0.0001,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.22983521248915872,
|
| 384 |
+
"grad_norm": 0.000997440074570477,
|
| 385 |
+
"learning_rate": 0.00018998391590678665,
|
| 386 |
+
"loss": 0.0001,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.2341717259323504,
|
| 391 |
+
"grad_norm": 0.00212860107421875,
|
| 392 |
+
"learning_rate": 0.00018931285084143818,
|
| 393 |
+
"loss": 0.0002,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.23850823937554205,
|
| 398 |
+
"grad_norm": 0.0011264905333518982,
|
| 399 |
+
"learning_rate": 0.00018862129360894958,
|
| 400 |
+
"loss": 0.0001,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.24284475281873374,
|
| 405 |
+
"grad_norm": 0.0006431519868783653,
|
| 406 |
+
"learning_rate": 0.00018790940288196715,
|
| 407 |
+
"loss": 0.0001,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.2471812662619254,
|
| 412 |
+
"grad_norm": 0.000361749145668,
|
| 413 |
+
"learning_rate": 0.00018717734199850584,
|
| 414 |
+
"loss": 0.0001,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.2515177797051171,
|
| 419 |
+
"grad_norm": 0.0005252111586742103,
|
| 420 |
+
"learning_rate": 0.00018642527892447243,
|
| 421 |
+
"loss": 0.0001,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.25585429314830876,
|
| 426 |
+
"grad_norm": 0.000348928413586691,
|
| 427 |
+
"learning_rate": 0.0001856533862151271,
|
| 428 |
+
"loss": 0.0001,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.26019080659150046,
|
| 433 |
+
"grad_norm": 0.00145808607339859,
|
| 434 |
+
"learning_rate": 0.00018486184097549186,
|
| 435 |
+
"loss": 0.0001,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.2645273200346921,
|
| 440 |
+
"grad_norm": 0.0007445367518812418,
|
| 441 |
+
"learning_rate": 0.00018405082481971516,
|
| 442 |
+
"loss": 0.0004,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.2688638334778838,
|
| 447 |
+
"grad_norm": 0.00031382605084218085,
|
| 448 |
+
"learning_rate": 0.0001832205238294018,
|
| 449 |
+
"loss": 0.0001,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.2732003469210755,
|
| 454 |
+
"grad_norm": 0.0006055298144929111,
|
| 455 |
+
"learning_rate": 0.00018237112851091802,
|
| 456 |
+
"loss": 0.0001,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.2775368603642671,
|
| 461 |
+
"grad_norm": 0.0002734568843152374,
|
| 462 |
+
"learning_rate": 0.00018150283375168114,
|
| 463 |
+
"loss": 0.0001,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.2818733738074588,
|
| 468 |
+
"grad_norm": 0.000268082192633301,
|
| 469 |
+
"learning_rate": 0.00018061583877544414,
|
| 470 |
+
"loss": 0.0001,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.2862098872506505,
|
| 475 |
+
"grad_norm": 0.0006183464429341257,
|
| 476 |
+
"learning_rate": 0.0001797103470965852,
|
| 477 |
+
"loss": 0.0001,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.29054640069384213,
|
| 482 |
+
"grad_norm": 0.0012671047588810325,
|
| 483 |
+
"learning_rate": 0.000178786566473413,
|
| 484 |
+
"loss": 0.0001,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.2948829141370338,
|
| 489 |
+
"grad_norm": 0.00037384298047982156,
|
| 490 |
+
"learning_rate": 0.00017784470886049783,
|
| 491 |
+
"loss": 0.0001,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.2992194275802255,
|
| 496 |
+
"grad_norm": 0.00038407245301641524,
|
| 497 |
+
"learning_rate": 0.0001768849903600406,
|
| 498 |
+
"loss": 0.0001,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.30355594102341715,
|
| 503 |
+
"grad_norm": 0.0005591203807853162,
|
| 504 |
+
"learning_rate": 0.00017590763117228934,
|
| 505 |
+
"loss": 0.0001,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.30789245446660884,
|
| 510 |
+
"grad_norm": 0.0004962561069987714,
|
| 511 |
+
"learning_rate": 0.00017491285554501636,
|
| 512 |
+
"loss": 0.0001,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.31222896790980054,
|
| 517 |
+
"grad_norm": 0.00041289450018666685,
|
| 518 |
+
"learning_rate": 0.00017390089172206592,
|
| 519 |
+
"loss": 0.0002,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.3165654813529922,
|
| 524 |
+
"grad_norm": 0.0003575518203433603,
|
| 525 |
+
"learning_rate": 0.00017287197189098556,
|
| 526 |
+
"loss": 0.0001,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.32090199479618386,
|
| 531 |
+
"grad_norm": 0.00035653176018968225,
|
| 532 |
+
"learning_rate": 0.0001718263321297523,
|
| 533 |
+
"loss": 0.0,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.32523850823937556,
|
| 538 |
+
"grad_norm": 0.00047867343528196216,
|
| 539 |
+
"learning_rate": 0.00017076421235260648,
|
| 540 |
+
"loss": 0.0001,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3295750216825672,
|
| 545 |
+
"grad_norm": 0.0018968080403283238,
|
| 546 |
+
"learning_rate": 0.00016968585625500498,
|
| 547 |
+
"loss": 0.0003,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3339115351257589,
|
| 552 |
+
"grad_norm": 0.0008598743006587029,
|
| 553 |
+
"learning_rate": 0.00016859151125770737,
|
| 554 |
+
"loss": 0.0002,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.3382480485689506,
|
| 559 |
+
"grad_norm": 0.001001562224701047,
|
| 560 |
+
"learning_rate": 0.0001674814284500068,
|
| 561 |
+
"loss": 0.0001,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3425845620121422,
|
| 566 |
+
"grad_norm": 0.0005547697655856609,
|
| 567 |
+
"learning_rate": 0.00016635586253211962,
|
| 568 |
+
"loss": 0.0001,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3469210754553339,
|
| 573 |
+
"grad_norm": 0.0012839736882597208,
|
| 574 |
+
"learning_rate": 0.00016521507175674643,
|
| 575 |
+
"loss": 0.0,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3512575888985256,
|
| 580 |
+
"grad_norm": 0.00024862270220182836,
|
| 581 |
+
"learning_rate": 0.00016405931786981755,
|
| 582 |
+
"loss": 0.0001,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.35559410234171723,
|
| 587 |
+
"grad_norm": 0.0018100353190675378,
|
| 588 |
+
"learning_rate": 0.00016288886605043764,
|
| 589 |
+
"loss": 0.0002,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3599306157849089,
|
| 594 |
+
"grad_norm": 0.0014100705739110708,
|
| 595 |
+
"learning_rate": 0.0001617039848500424,
|
| 596 |
+
"loss": 0.0001,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3642671292281006,
|
| 601 |
+
"grad_norm": 0.5226932764053345,
|
| 602 |
+
"learning_rate": 0.0001605049461307812,
|
| 603 |
+
"loss": 0.0009,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3686036426712923,
|
| 608 |
+
"grad_norm": 0.0063658421859145164,
|
| 609 |
+
"learning_rate": 0.00015929202500314051,
|
| 610 |
+
"loss": 0.0008,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.37294015611448394,
|
| 615 |
+
"grad_norm": 0.0008500253898091614,
|
| 616 |
+
"learning_rate": 0.00015806549976282182,
|
| 617 |
+
"loss": 0.0006,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.37727666955767564,
|
| 622 |
+
"grad_norm": 0.00786112155765295,
|
| 623 |
+
"learning_rate": 0.0001568256518268887,
|
| 624 |
+
"loss": 0.0002,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.38161318300086733,
|
| 629 |
+
"grad_norm": 0.0010554317850619555,
|
| 630 |
+
"learning_rate": 0.00015557276566919784,
|
| 631 |
+
"loss": 0.0009,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.38594969644405897,
|
| 636 |
+
"grad_norm": 0.08147475868463516,
|
| 637 |
+
"learning_rate": 0.0001543071287551287,
|
| 638 |
+
"loss": 0.0005,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.39028620988725066,
|
| 643 |
+
"grad_norm": 0.00513398926705122,
|
| 644 |
+
"learning_rate": 0.0001530290314756265,
|
| 645 |
+
"loss": 0.0005,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.39462272333044235,
|
| 650 |
+
"grad_norm": 0.012065605260431767,
|
| 651 |
+
"learning_rate": 0.00015173876708057438,
|
| 652 |
+
"loss": 0.0001,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.398959236773634,
|
| 657 |
+
"grad_norm": 0.0004873398575000465,
|
| 658 |
+
"learning_rate": 0.00015043663161150937,
|
| 659 |
+
"loss": 0.0011,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4032957502168257,
|
| 664 |
+
"grad_norm": 0.006046785041689873,
|
| 665 |
+
"learning_rate": 0.00014912292383369787,
|
| 666 |
+
"loss": 0.0001,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.40763226366001737,
|
| 671 |
+
"grad_norm": 0.0011943348217755556,
|
| 672 |
+
"learning_rate": 0.0001477979451675861,
|
| 673 |
+
"loss": 0.0001,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.411968777103209,
|
| 678 |
+
"grad_norm": 0.000647057022433728,
|
| 679 |
+
"learning_rate": 0.0001464619996196415,
|
| 680 |
+
"loss": 0.0004,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.4163052905464007,
|
| 685 |
+
"grad_norm": 0.04003003239631653,
|
| 686 |
+
"learning_rate": 0.00014511539371260074,
|
| 687 |
+
"loss": 0.0005,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4206418039895924,
|
| 692 |
+
"grad_norm": 0.02914806827902794,
|
| 693 |
+
"learning_rate": 0.00014375843641514034,
|
| 694 |
+
"loss": 0.0005,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.424978317432784,
|
| 699 |
+
"grad_norm": 0.0006412832881323993,
|
| 700 |
+
"learning_rate": 0.0001423914390709861,
|
| 701 |
+
"loss": 0.001,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4293148308759757,
|
| 706 |
+
"grad_norm": 0.0005903305718675256,
|
| 707 |
+
"learning_rate": 0.00014101471532747752,
|
| 708 |
+
"loss": 0.0011,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4336513443191674,
|
| 713 |
+
"grad_norm": 0.0005716175073757768,
|
| 714 |
+
"learning_rate": 0.00013962858106360398,
|
| 715 |
+
"loss": 0.0012,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.43798785776235905,
|
| 720 |
+
"grad_norm": 0.031988050788640976,
|
| 721 |
+
"learning_rate": 0.00013823335431752853,
|
| 722 |
+
"loss": 0.001,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.44232437120555074,
|
| 727 |
+
"grad_norm": 0.003730722237378359,
|
| 728 |
+
"learning_rate": 0.00013682935521361627,
|
| 729 |
+
"loss": 0.0001,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.44666088464874243,
|
| 734 |
+
"grad_norm": 0.0007267138571478426,
|
| 735 |
+
"learning_rate": 0.00013541690588898448,
|
| 736 |
+
"loss": 0.0001,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.45099739809193407,
|
| 741 |
+
"grad_norm": 0.07632599025964737,
|
| 742 |
+
"learning_rate": 0.00013399633041959047,
|
| 743 |
+
"loss": 0.0003,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.45533391153512576,
|
| 748 |
+
"grad_norm": 0.0006175984744913876,
|
| 749 |
+
"learning_rate": 0.00013256795474587485,
|
| 750 |
+
"loss": 0.0001,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.45967042497831745,
|
| 755 |
+
"grad_norm": 0.0003872651723213494,
|
| 756 |
+
"learning_rate": 0.00013113210659797687,
|
| 757 |
+
"loss": 0.0001,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4640069384215091,
|
| 762 |
+
"grad_norm": 0.0009627993567846715,
|
| 763 |
+
"learning_rate": 0.00012968911542053923,
|
| 764 |
+
"loss": 0.0001,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4683434518647008,
|
| 769 |
+
"grad_norm": 0.0004405930812936276,
|
| 770 |
+
"learning_rate": 0.00012823931229711944,
|
| 771 |
+
"loss": 0.0001,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.47267996530789247,
|
| 776 |
+
"grad_norm": 0.001180918887257576,
|
| 777 |
+
"learning_rate": 0.00012678302987422532,
|
| 778 |
+
"loss": 0.0001,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4770164787510841,
|
| 783 |
+
"grad_norm": 0.0004968827124685049,
|
| 784 |
+
"learning_rate": 0.00012532060228499136,
|
| 785 |
+
"loss": 0.0001,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4813529921942758,
|
| 790 |
+
"grad_norm": 0.0003979886241722852,
|
| 791 |
+
"learning_rate": 0.00012385236507251476,
|
| 792 |
+
"loss": 0.0001,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4856895056374675,
|
| 797 |
+
"grad_norm": 0.010322037152945995,
|
| 798 |
+
"learning_rate": 0.00012237865511286746,
|
| 799 |
+
"loss": 0.0001,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4900260190806591,
|
| 804 |
+
"grad_norm": 0.0002791626611724496,
|
| 805 |
+
"learning_rate": 0.00012089981053780224,
|
| 806 |
+
"loss": 0.0,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4943625325238508,
|
| 811 |
+
"grad_norm": 0.000490942969918251,
|
| 812 |
+
"learning_rate": 0.00011941617065717124,
|
| 813 |
+
"loss": 0.0001,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4986990459670425,
|
| 818 |
+
"grad_norm": 0.0003438853018451482,
|
| 819 |
+
"learning_rate": 0.00011792807588107357,
|
| 820 |
+
"loss": 0.0001,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5030355594102341,
|
| 825 |
+
"grad_norm": 0.0021341927349567413,
|
| 826 |
+
"learning_rate": 0.00011643586764175092,
|
| 827 |
+
"loss": 0.0001,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5073720728534259,
|
| 832 |
+
"grad_norm": 0.00031035722349770367,
|
| 833 |
+
"learning_rate": 0.00011493988831524834,
|
| 834 |
+
"loss": 0.0,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5117085862966175,
|
| 839 |
+
"grad_norm": 0.00061289023142308,
|
| 840 |
+
"learning_rate": 0.00011344048114285882,
|
| 841 |
+
"loss": 0.0,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5160450997398092,
|
| 846 |
+
"grad_norm": 0.00996777880936861,
|
| 847 |
+
"learning_rate": 0.00011193799015236885,
|
| 848 |
+
"loss": 0.0001,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5203816131830009,
|
| 853 |
+
"grad_norm": 0.0004861672641709447,
|
| 854 |
+
"learning_rate": 0.00011043276007912413,
|
| 855 |
+
"loss": 0.0001,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5247181266261925,
|
| 860 |
+
"grad_norm": 0.0003749791067093611,
|
| 861 |
+
"learning_rate": 0.00010892513628693237,
|
| 862 |
+
"loss": 0.0,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5290546400693842,
|
| 867 |
+
"grad_norm": 0.00017323833890259266,
|
| 868 |
+
"learning_rate": 0.00010741546468882223,
|
| 869 |
+
"loss": 0.0,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5333911535125759,
|
| 874 |
+
"grad_norm": 0.0002214506093878299,
|
| 875 |
+
"learning_rate": 0.00010590409166767614,
|
| 876 |
+
"loss": 0.0,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5377276669557676,
|
| 881 |
+
"grad_norm": 0.00023450802837032825,
|
| 882 |
+
"learning_rate": 0.00010439136399675542,
|
| 883 |
+
"loss": 0.0,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5420641803989592,
|
| 888 |
+
"grad_norm": 0.00017662130994722247,
|
| 889 |
+
"learning_rate": 0.00010287762876013563,
|
| 890 |
+
"loss": 0.0,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.546400693842151,
|
| 895 |
+
"grad_norm": 0.00017046746506821364,
|
| 896 |
+
"learning_rate": 0.00010136323327307075,
|
| 897 |
+
"loss": 0.0,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5507372072853426,
|
| 902 |
+
"grad_norm": 0.00033422402339056134,
|
| 903 |
+
"learning_rate": 9.984852500230432e-05,
|
| 904 |
+
"loss": 0.0,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5550737207285342,
|
| 909 |
+
"grad_norm": 0.0006649333517998457,
|
| 910 |
+
"learning_rate": 9.833385148634574e-05,
|
| 911 |
+
"loss": 0.0001,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.559410234171726,
|
| 916 |
+
"grad_norm": 0.0009602860664017498,
|
| 917 |
+
"learning_rate": 9.681956025573022e-05,
|
| 918 |
+
"loss": 0.0,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5637467476149176,
|
| 923 |
+
"grad_norm": 0.000601974839810282,
|
| 924 |
+
"learning_rate": 9.53059987532804e-05,
|
| 925 |
+
"loss": 0.0,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5680832610581092,
|
| 930 |
+
"grad_norm": 0.0002987553598359227,
|
| 931 |
+
"learning_rate": 9.379351425438826e-05,
|
| 932 |
+
"loss": 0.0,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.572419774501301,
|
| 937 |
+
"grad_norm": 0.00017630436923354864,
|
| 938 |
+
"learning_rate": 9.228245378733537e-05,
|
| 939 |
+
"loss": 0.0,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5767562879444926,
|
| 944 |
+
"grad_norm": 0.00021065869077574462,
|
| 945 |
+
"learning_rate": 9.077316405366981e-05,
|
| 946 |
+
"loss": 0.0,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5810928013876843,
|
| 951 |
+
"grad_norm": 0.0003561481134966016,
|
| 952 |
+
"learning_rate": 8.926599134865808e-05,
|
| 953 |
+
"loss": 0.0,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.585429314830876,
|
| 958 |
+
"grad_norm": 0.000257289269939065,
|
| 959 |
+
"learning_rate": 8.776128148183028e-05,
|
| 960 |
+
"loss": 0.0,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5897658282740676,
|
| 965 |
+
"grad_norm": 0.00019555074686650187,
|
| 966 |
+
"learning_rate": 8.625937969763662e-05,
|
| 967 |
+
"loss": 0.0,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5941023417172593,
|
| 972 |
+
"grad_norm": 0.00053867616225034,
|
| 973 |
+
"learning_rate": 8.476063059623375e-05,
|
| 974 |
+
"loss": 0.0,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.598438855160451,
|
| 979 |
+
"grad_norm": 0.0004365757922641933,
|
| 980 |
+
"learning_rate": 8.326537805441884e-05,
|
| 981 |
+
"loss": 0.0,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6027753686036427,
|
| 986 |
+
"grad_norm": 0.0002754295419435948,
|
| 987 |
+
"learning_rate": 8.177396514672939e-05,
|
| 988 |
+
"loss": 0.0,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6071118820468343,
|
| 993 |
+
"grad_norm": 0.0001531521265860647,
|
| 994 |
+
"learning_rate": 8.028673406672763e-05,
|
| 995 |
+
"loss": 0.0001,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.611448395490026,
|
| 1000 |
+
"grad_norm": 0.0001671235222602263,
|
| 1001 |
+
"learning_rate": 7.880402604848662e-05,
|
| 1002 |
+
"loss": 0.0,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6157849089332177,
|
| 1007 |
+
"grad_norm": 0.0001691756333457306,
|
| 1008 |
+
"learning_rate": 7.732618128829656e-05,
|
| 1009 |
+
"loss": 0.0,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6201214223764093,
|
| 1014 |
+
"grad_norm": 0.000139987314469181,
|
| 1015 |
+
"learning_rate": 7.585353886660954e-05,
|
| 1016 |
+
"loss": 0.0,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6244579358196011,
|
| 1021 |
+
"grad_norm": 0.00023128537577576935,
|
| 1022 |
+
"learning_rate": 7.438643667023979e-05,
|
| 1023 |
+
"loss": 0.0,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6287944492627927,
|
| 1028 |
+
"grad_norm": 0.0001494650059612468,
|
| 1029 |
+
"learning_rate": 7.292521131483821e-05,
|
| 1030 |
+
"loss": 0.0,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6331309627059843,
|
| 1035 |
+
"grad_norm": 0.0011407610727474093,
|
| 1036 |
+
"learning_rate": 7.147019806765836e-05,
|
| 1037 |
+
"loss": 0.0,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6374674761491761,
|
| 1042 |
+
"grad_norm": 0.00017924026178661734,
|
| 1043 |
+
"learning_rate": 7.002173077063197e-05,
|
| 1044 |
+
"loss": 0.0,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6418039895923677,
|
| 1049 |
+
"grad_norm": 0.00017581222346052527,
|
| 1050 |
+
"learning_rate": 6.858014176377139e-05,
|
| 1051 |
+
"loss": 0.0,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6461405030355594,
|
| 1056 |
+
"grad_norm": 0.0001402726920787245,
|
| 1057 |
+
"learning_rate": 6.714576180891654e-05,
|
| 1058 |
+
"loss": 0.0,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.6504770164787511,
|
| 1063 |
+
"grad_norm": 0.00018241746874991804,
|
| 1064 |
+
"learning_rate": 6.57189200138442e-05,
|
| 1065 |
+
"loss": 0.0,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6548135299219427,
|
| 1070 |
+
"grad_norm": 0.0002422356337774545,
|
| 1071 |
+
"learning_rate": 6.429994375675661e-05,
|
| 1072 |
+
"loss": 0.0001,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6591500433651344,
|
| 1077 |
+
"grad_norm": 0.000627288012765348,
|
| 1078 |
+
"learning_rate": 6.288915861116706e-05,
|
| 1079 |
+
"loss": 0.0,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6634865568083261,
|
| 1084 |
+
"grad_norm": 0.0002638279111124575,
|
| 1085 |
+
"learning_rate": 6.148688827119937e-05,
|
| 1086 |
+
"loss": 0.0,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6678230702515178,
|
| 1091 |
+
"grad_norm": 0.001158065744675696,
|
| 1092 |
+
"learning_rate": 6.009345447731886e-05,
|
| 1093 |
+
"loss": 0.0,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6721595836947094,
|
| 1098 |
+
"grad_norm": 0.00033756031189113855,
|
| 1099 |
+
"learning_rate": 5.8709176942511136e-05,
|
| 1100 |
+
"loss": 0.0,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6764960971379012,
|
| 1105 |
+
"grad_norm": 0.000685289385728538,
|
| 1106 |
+
"learning_rate": 5.733437327892661e-05,
|
| 1107 |
+
"loss": 0.0,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6808326105810928,
|
| 1112 |
+
"grad_norm": 0.0002907447633333504,
|
| 1113 |
+
"learning_rate": 5.596935892500663e-05,
|
| 1114 |
+
"loss": 0.0,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6851691240242844,
|
| 1119 |
+
"grad_norm": 0.00017344890511594713,
|
| 1120 |
+
"learning_rate": 5.4614447073108375e-05,
|
| 1121 |
+
"loss": 0.0001,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6895056374674762,
|
| 1126 |
+
"grad_norm": 0.00025545316748321056,
|
| 1127 |
+
"learning_rate": 5.326994859764552e-05,
|
| 1128 |
+
"loss": 0.0,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6938421509106678,
|
| 1133 |
+
"grad_norm": 0.000984342535957694,
|
| 1134 |
+
"learning_rate": 5.193617198376004e-05,
|
| 1135 |
+
"loss": 0.0001,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6981786643538594,
|
| 1140 |
+
"grad_norm": 0.0002541212597861886,
|
| 1141 |
+
"learning_rate": 5.0613423256542904e-05,
|
| 1142 |
+
"loss": 0.0,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.7025151777970512,
|
| 1147 |
+
"grad_norm": 0.005005079321563244,
|
| 1148 |
+
"learning_rate": 4.930200591081865e-05,
|
| 1149 |
+
"loss": 0.0,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7068516912402428,
|
| 1154 |
+
"grad_norm": 0.0036286672111600637,
|
| 1155 |
+
"learning_rate": 4.8002220841511045e-05,
|
| 1156 |
+
"loss": 0.0,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.7111882046834345,
|
| 1161 |
+
"grad_norm": 0.0012679731007665396,
|
| 1162 |
+
"learning_rate": 4.671436627460479e-05,
|
| 1163 |
+
"loss": 0.0,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.7155247181266262,
|
| 1168 |
+
"grad_norm": 0.00015154715219978243,
|
| 1169 |
+
"learning_rate": 4.543873769871978e-05,
|
| 1170 |
+
"loss": 0.0,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.7198612315698178,
|
| 1175 |
+
"grad_norm": 0.00019088915723841637,
|
| 1176 |
+
"learning_rate": 4.417562779731355e-05,
|
| 1177 |
+
"loss": 0.0,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.7241977450130095,
|
| 1182 |
+
"grad_norm": 0.000255732040386647,
|
| 1183 |
+
"learning_rate": 4.292532638152713e-05,
|
| 1184 |
+
"loss": 0.0001,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.7285342584562012,
|
| 1189 |
+
"grad_norm": 0.00032757947337813675,
|
| 1190 |
+
"learning_rate": 4.168812032369026e-05,
|
| 1191 |
+
"loss": 0.0,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7328707718993929,
|
| 1196 |
+
"grad_norm": 0.0005451737088151276,
|
| 1197 |
+
"learning_rate": 4.0464293491500326e-05,
|
| 1198 |
+
"loss": 0.0005,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7372072853425846,
|
| 1203 |
+
"grad_norm": 0.3482416570186615,
|
| 1204 |
+
"learning_rate": 3.9254126682891425e-05,
|
| 1205 |
+
"loss": 0.0004,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7415437987857763,
|
| 1210 |
+
"grad_norm": 0.00018617334717418998,
|
| 1211 |
+
"learning_rate": 3.8057897561607014e-05,
|
| 1212 |
+
"loss": 0.0,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7458803122289679,
|
| 1217 |
+
"grad_norm": 0.0004157496150583029,
|
| 1218 |
+
"learning_rate": 3.68758805934923e-05,
|
| 1219 |
+
"loss": 0.0,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7502168256721596,
|
| 1224 |
+
"grad_norm": 0.0003447196795605123,
|
| 1225 |
+
"learning_rate": 3.5708346983519904e-05,
|
| 1226 |
+
"loss": 0.0,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7545533391153513,
|
| 1231 |
+
"grad_norm": 0.0001821848563849926,
|
| 1232 |
+
"learning_rate": 3.455556461356413e-05,
|
| 1233 |
+
"loss": 0.0,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7588898525585429,
|
| 1238 |
+
"grad_norm": 0.0008997240802273154,
|
| 1239 |
+
"learning_rate": 3.3417797980937305e-05,
|
| 1240 |
+
"loss": 0.0,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7632263660017347,
|
| 1245 |
+
"grad_norm": 0.00041676373803056777,
|
| 1246 |
+
"learning_rate": 3.229530813770281e-05,
|
| 1247 |
+
"loss": 0.0,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.7675628794449263,
|
| 1252 |
+
"grad_norm": 0.00024333845067303628,
|
| 1253 |
+
"learning_rate": 3.118835263077874e-05,
|
| 1254 |
+
"loss": 0.0,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7718993928881179,
|
| 1259 |
+
"grad_norm": 0.00044431493734009564,
|
| 1260 |
+
"learning_rate": 3.0097185442845653e-05,
|
| 1261 |
+
"loss": 0.0,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7762359063313097,
|
| 1266 |
+
"grad_norm": 0.003919024486094713,
|
| 1267 |
+
"learning_rate": 2.9022056934072106e-05,
|
| 1268 |
+
"loss": 0.0,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.7805724197745013,
|
| 1273 |
+
"grad_norm": 0.00021601624030154198,
|
| 1274 |
+
"learning_rate": 2.796321378467146e-05,
|
| 1275 |
+
"loss": 0.0001,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.784908933217693,
|
| 1280 |
+
"grad_norm": 0.0003809454501606524,
|
| 1281 |
+
"learning_rate": 2.6920898938302885e-05,
|
| 1282 |
+
"loss": 0.0,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7892454466608847,
|
| 1287 |
+
"grad_norm": 0.0009574461146257818,
|
| 1288 |
+
"learning_rate": 2.5895351546329717e-05,
|
| 1289 |
+
"loss": 0.0,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7935819601040763,
|
| 1294 |
+
"grad_norm": 0.0008413203759118915,
|
| 1295 |
+
"learning_rate": 2.4886806912948035e-05,
|
| 1296 |
+
"loss": 0.0,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.797918473547268,
|
| 1301 |
+
"grad_norm": 0.00016790963127277792,
|
| 1302 |
+
"learning_rate": 2.3895496441197806e-05,
|
| 1303 |
+
"loss": 0.0,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.8022549869904597,
|
| 1308 |
+
"grad_norm": 0.0002135594404535368,
|
| 1309 |
+
"learning_rate": 2.2921647579869275e-05,
|
| 1310 |
+
"loss": 0.0,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.8065915004336514,
|
| 1315 |
+
"grad_norm": 0.0006223685923032463,
|
| 1316 |
+
"learning_rate": 2.1965483771316498e-05,
|
| 1317 |
+
"loss": 0.0,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.810928013876843,
|
| 1322 |
+
"grad_norm": 0.0001733316748868674,
|
| 1323 |
+
"learning_rate": 2.102722440019006e-05,
|
| 1324 |
+
"loss": 0.0,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.8152645273200347,
|
| 1329 |
+
"grad_norm": 0.0006180764175951481,
|
| 1330 |
+
"learning_rate": 2.0107084743101024e-05,
|
| 1331 |
+
"loss": 0.0,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.8196010407632264,
|
| 1336 |
+
"grad_norm": 0.019121604040265083,
|
| 1337 |
+
"learning_rate": 1.9205275919227282e-05,
|
| 1338 |
+
"loss": 0.0,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.823937554206418,
|
| 1343 |
+
"grad_norm": 0.00031261073308996856,
|
| 1344 |
+
"learning_rate": 1.8322004841873842e-05,
|
| 1345 |
+
"loss": 0.0,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.8282740676496098,
|
| 1350 |
+
"grad_norm": 0.00025373531389050186,
|
| 1351 |
+
"learning_rate": 1.7457474170998113e-05,
|
| 1352 |
+
"loss": 0.0001,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.8326105810928014,
|
| 1357 |
+
"grad_norm": 0.00029701556195504963,
|
| 1358 |
+
"learning_rate": 1.661188226671111e-05,
|
| 1359 |
+
"loss": 0.0,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.836947094535993,
|
| 1364 |
+
"grad_norm": 0.00018009443010669202,
|
| 1365 |
+
"learning_rate": 1.5785423143765143e-05,
|
| 1366 |
+
"loss": 0.0,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.8412836079791848,
|
| 1371 |
+
"grad_norm": 0.00026098068337887526,
|
| 1372 |
+
"learning_rate": 1.4978286427038601e-05,
|
| 1373 |
+
"loss": 0.0,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8456201214223764,
|
| 1378 |
+
"grad_norm": 0.00022298283874988556,
|
| 1379 |
+
"learning_rate": 1.4190657308027989e-05,
|
| 1380 |
+
"loss": 0.0,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.849956634865568,
|
| 1385 |
+
"grad_norm": 0.0003458283899817616,
|
| 1386 |
+
"learning_rate": 1.3422716502357102e-05,
|
| 1387 |
+
"loss": 0.0,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8542931483087598,
|
| 1392 |
+
"grad_norm": 0.00045949083869345486,
|
| 1393 |
+
"learning_rate": 1.2674640208313137e-05,
|
| 1394 |
+
"loss": 0.0,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8586296617519514,
|
| 1399 |
+
"grad_norm": 0.00020235533884260803,
|
| 1400 |
+
"learning_rate": 1.1946600066419345e-05,
|
| 1401 |
+
"loss": 0.0,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.8629661751951431,
|
| 1406 |
+
"grad_norm": 0.00016053752915468067,
|
| 1407 |
+
"learning_rate": 1.1238763120053387e-05,
|
| 1408 |
+
"loss": 0.0,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8673026886383348,
|
| 1413 |
+
"grad_norm": 0.0002067391760647297,
|
| 1414 |
+
"learning_rate": 1.0551291777120464e-05,
|
| 1415 |
+
"loss": 0.0,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8716392020815265,
|
| 1420 |
+
"grad_norm": 0.00024100695736706257,
|
| 1421 |
+
"learning_rate": 9.884343772790005e-06,
|
| 1422 |
+
"loss": 0.0,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8759757155247181,
|
| 1427 |
+
"grad_norm": 0.00032670717337168753,
|
| 1428 |
+
"learning_rate": 9.238072133304653e-06,
|
| 1429 |
+
"loss": 0.0,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8803122289679098,
|
| 1434 |
+
"grad_norm": 0.0010708924382925034,
|
| 1435 |
+
"learning_rate": 8.612625140869324e-06,
|
| 1436 |
+
"loss": 0.0,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8846487424111015,
|
| 1441 |
+
"grad_norm": 0.00016485978267155588,
|
| 1442 |
+
"learning_rate": 8.00814629962916e-06,
|
| 1443 |
+
"loss": 0.0,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8889852558542931,
|
| 1448 |
+
"grad_norm": 0.0012277034111320972,
|
| 1449 |
+
"learning_rate": 7.424774302743409e-06,
|
| 1450 |
+
"loss": 0.0,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8933217692974849,
|
| 1455 |
+
"grad_norm": 0.0001766424102243036,
|
| 1456 |
+
"learning_rate": 6.862643000563407e-06,
|
| 1457 |
+
"loss": 0.0,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8976582827406765,
|
| 1462 |
+
"grad_norm": 0.00012875783431809396,
|
| 1463 |
+
"learning_rate": 6.321881369921656e-06,
|
| 1464 |
+
"loss": 0.0,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.9019947961838681,
|
| 1469 |
+
"grad_norm": 0.0003258756478317082,
|
| 1470 |
+
"learning_rate": 5.802613484538888e-06,
|
| 1471 |
+
"loss": 0.0,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.9063313096270599,
|
| 1476 |
+
"grad_norm": 0.0003074019914492965,
|
| 1477 |
+
"learning_rate": 5.304958486556488e-06,
|
| 1478 |
+
"loss": 0.0,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.9106678230702515,
|
| 1483 |
+
"grad_norm": 0.00044591055484488606,
|
| 1484 |
+
"learning_rate": 4.829030559200032e-06,
|
| 1485 |
+
"loss": 0.0,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.9150043365134432,
|
| 1490 |
+
"grad_norm": 0.00019902190251741558,
|
| 1491 |
+
"learning_rate": 4.374938900580883e-06,
|
| 1492 |
+
"loss": 0.0,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.9193408499566349,
|
| 1497 |
+
"grad_norm": 0.0006033729878254235,
|
| 1498 |
+
"learning_rate": 3.942787698641548e-06,
|
| 1499 |
+
"loss": 0.0,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.9236773633998265,
|
| 1504 |
+
"grad_norm": 0.00017164893506560475,
|
| 1505 |
+
"learning_rate": 3.532676107250421e-06,
|
| 1506 |
+
"loss": 0.0,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.9280138768430182,
|
| 1511 |
+
"grad_norm": 0.0002156089904019609,
|
| 1512 |
+
"learning_rate": 3.1446982234517474e-06,
|
| 1513 |
+
"loss": 0.0,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.9323503902862099,
|
| 1518 |
+
"grad_norm": 0.00040788273327052593,
|
| 1519 |
+
"learning_rate": 2.7789430658757275e-06,
|
| 1520 |
+
"loss": 0.0,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.9366869037294016,
|
| 1525 |
+
"grad_norm": 0.0003488777147140354,
|
| 1526 |
+
"learning_rate": 2.4354945543138775e-06,
|
| 1527 |
+
"loss": 0.0,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.9410234171725932,
|
| 1532 |
+
"grad_norm": 0.0004901742213405669,
|
| 1533 |
+
"learning_rate": 2.1144314904642195e-06,
|
| 1534 |
+
"loss": 0.0,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.9453599306157849,
|
| 1539 |
+
"grad_norm": 0.00025740411365404725,
|
| 1540 |
+
"learning_rate": 1.8158275398508784e-06,
|
| 1541 |
+
"loss": 0.0,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.9496964440589766,
|
| 1546 |
+
"grad_norm": 0.0002657132863532752,
|
| 1547 |
+
"learning_rate": 1.539751214922014e-06,
|
| 1548 |
+
"loss": 0.0,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.9540329575021682,
|
| 1553 |
+
"grad_norm": 0.0013277851976454258,
|
| 1554 |
+
"learning_rate": 1.2862658593302046e-06,
|
| 1555 |
+
"loss": 0.0,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.95836947094536,
|
| 1560 |
+
"grad_norm": 0.00018901302246376872,
|
| 1561 |
+
"learning_rate": 1.0554296333987078e-06,
|
| 1562 |
+
"loss": 0.0,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9627059843885516,
|
| 1567 |
+
"grad_norm": 0.00029082829132676125,
|
| 1568 |
+
"learning_rate": 8.472955007769456e-07,
|
| 1569 |
+
"loss": 0.0,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.9670424978317432,
|
| 1574 |
+
"grad_norm": 0.0001757108693709597,
|
| 1575 |
+
"learning_rate": 6.619112162885022e-07,
|
| 1576 |
+
"loss": 0.0,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.971379011274935,
|
| 1581 |
+
"grad_norm": 0.00018936190463136882,
|
| 1582 |
+
"learning_rate": 4.993193149740338e-07,
|
| 1583 |
+
"loss": 0.0,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9757155247181266,
|
| 1588 |
+
"grad_norm": 0.00027079100254923105,
|
| 1589 |
+
"learning_rate": 3.595571023319755e-07,
|
| 1590 |
+
"loss": 0.0,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.9800520381613183,
|
| 1595 |
+
"grad_norm": 0.0003670993319246918,
|
| 1596 |
+
"learning_rate": 2.426566457590651e-07,
|
| 1597 |
+
"loss": 0.0,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.98438855160451,
|
| 1602 |
+
"grad_norm": 0.00011753277067327872,
|
| 1603 |
+
"learning_rate": 1.4864476719270714e-07,
|
| 1604 |
+
"loss": 0.0,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.9887250650477016,
|
| 1609 |
+
"grad_norm": 0.01616135612130165,
|
| 1610 |
+
"learning_rate": 7.754303695688414e-08,
|
| 1611 |
+
"loss": 0.0002,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9930615784908933,
|
| 1616 |
+
"grad_norm": 0.0007944769458845258,
|
| 1617 |
+
"learning_rate": 2.936776881302672e-08,
|
| 1618 |
+
"loss": 0.0002,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.997398091934085,
|
| 1623 |
+
"grad_norm": 0.00044876235187985003,
|
| 1624 |
+
"learning_rate": 4.130016216896682e-09,
|
| 1625 |
+
"loss": 0.0,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
{
|
| 1629 |
"epoch": 1.0,
|
| 1630 |
+
"step": 1153,
|
| 1631 |
+
"total_flos": 1.025624031790891e+18,
|
| 1632 |
+
"train_loss": 0.004042043934228078,
|
| 1633 |
+
"train_runtime": 6129.3476,
|
| 1634 |
+
"train_samples_per_second": 3.01,
|
| 1635 |
+
"train_steps_per_second": 0.188
|
| 1636 |
}
|
| 1637 |
],
|
| 1638 |
"logging_steps": 5,
|
| 1639 |
+
"max_steps": 1153,
|
| 1640 |
"num_input_tokens_seen": 0,
|
| 1641 |
"num_train_epochs": 1,
|
| 1642 |
+
"save_steps": 100,
|
| 1643 |
+
"stateful_callbacks": {
|
| 1644 |
+
"TrainerControl": {
|
| 1645 |
+
"args": {
|
| 1646 |
+
"should_epoch_stop": false,
|
| 1647 |
+
"should_evaluate": false,
|
| 1648 |
+
"should_log": false,
|
| 1649 |
+
"should_save": true,
|
| 1650 |
+
"should_training_stop": true
|
| 1651 |
+
},
|
| 1652 |
+
"attributes": {}
|
| 1653 |
+
}
|
| 1654 |
+
},
|
| 1655 |
+
"total_flos": 1.025624031790891e+18,
|
| 1656 |
+
"train_batch_size": 4,
|
| 1657 |
"trial_name": null,
|
| 1658 |
"trial_params": null
|
| 1659 |
}
|