End of training
Browse files- README.md +3 -2
- all_results.json +10 -10
- eval_results.json +5 -5
- train_results.json +6 -6
- trainer_state.json +1155 -345
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: all_tasks_combined_8b_sft
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# all_tasks_combined_8b_sft
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: all_tasks_combined_8b_sft
|
|
|
|
| 16 |
|
| 17 |
# all_tasks_combined_8b_sft
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the identity and the data_mc_filtered datasets.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4943
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 2.
|
| 7 |
-
"total_flos":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second": 4.
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.9953574744661093,
|
| 3 |
+
"eval_loss": 0.4942765235900879,
|
| 4 |
+
"eval_runtime": 19.3607,
|
| 5 |
+
"eval_samples_per_second": 18.181,
|
| 6 |
+
"eval_steps_per_second": 2.273,
|
| 7 |
+
"total_flos": 292826854195200.0,
|
| 8 |
+
"train_loss": 0.2790517607955389,
|
| 9 |
+
"train_runtime": 12257.2359,
|
| 10 |
+
"train_samples_per_second": 4.217,
|
| 11 |
+
"train_steps_per_second": 0.132
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 2.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.9953574744661093,
|
| 3 |
+
"eval_loss": 0.4942765235900879,
|
| 4 |
+
"eval_runtime": 19.3607,
|
| 5 |
+
"eval_samples_per_second": 18.181,
|
| 6 |
+
"eval_steps_per_second": 2.273
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 4.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.9953574744661093,
|
| 3 |
+
"total_flos": 292826854195200.0,
|
| 4 |
+
"train_loss": 0.2790517607955389,
|
| 5 |
+
"train_runtime": 12257.2359,
|
| 6 |
+
"train_samples_per_second": 4.217,
|
| 7 |
+
"train_steps_per_second": 0.132
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,600 +1,1410 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 50,
|
| 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": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"eval_loss": 0.
|
| 49 |
-
"eval_runtime":
|
| 50 |
-
"eval_samples_per_second":
|
| 51 |
-
"eval_steps_per_second": 2.
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"epoch": 0.
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch":
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 70
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch":
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss": 0.
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch":
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"grad_norm":
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch":
|
| 91 |
-
"eval_loss": 0.
|
| 92 |
-
"eval_runtime":
|
| 93 |
-
"eval_samples_per_second":
|
| 94 |
-
"eval_steps_per_second": 2.
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 110
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 120
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch":
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 130
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"grad_norm": 6.
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 140
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch":
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 9.
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 150
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch":
|
| 134 |
-
"eval_loss": 0.
|
| 135 |
-
"eval_runtime":
|
| 136 |
-
"eval_samples_per_second":
|
| 137 |
-
"eval_steps_per_second": 2.
|
| 138 |
"step": 150
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"grad_norm":
|
| 143 |
-
"learning_rate": 9.
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 160
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch":
|
| 149 |
-
"grad_norm":
|
| 150 |
-
"learning_rate": 9.
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 170
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"grad_norm":
|
| 157 |
-
"learning_rate": 9.
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 180
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"grad_norm":
|
| 164 |
-
"learning_rate": 9.
|
| 165 |
-
"loss": 0.
|
| 166 |
"step": 190
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch":
|
| 170 |
-
"grad_norm":
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"eval_loss": 0.
|
| 178 |
-
"eval_runtime":
|
| 179 |
-
"eval_samples_per_second":
|
| 180 |
-
"eval_steps_per_second": 2.
|
| 181 |
"step": 200
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch":
|
| 185 |
-
"grad_norm": 6.
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 210
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 220
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 230
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch":
|
| 206 |
-
"grad_norm": 4.
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 240
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 250
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"eval_loss": 0.
|
| 221 |
-
"eval_runtime":
|
| 222 |
-
"eval_samples_per_second":
|
| 223 |
-
"eval_steps_per_second": 2.
|
| 224 |
"step": 250
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"grad_norm":
|
| 229 |
-
"learning_rate":
|
| 230 |
-
"loss": 0.
|
| 231 |
"step": 260
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch":
|
| 235 |
-
"grad_norm": 4.
|
| 236 |
-
"learning_rate":
|
| 237 |
-
"loss": 0.
|
| 238 |
"step": 270
|
| 239 |
},
|
| 240 |
{
|
| 241 |
-
"epoch":
|
| 242 |
-
"grad_norm":
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 280
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"grad_norm":
|
| 250 |
-
"learning_rate":
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 290
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch":
|
| 256 |
-
"grad_norm":
|
| 257 |
-
"learning_rate":
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
-
"epoch":
|
| 263 |
-
"eval_loss": 0.
|
| 264 |
-
"eval_runtime":
|
| 265 |
-
"eval_samples_per_second":
|
| 266 |
-
"eval_steps_per_second": 2.
|
| 267 |
"step": 300
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 310
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch":
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 320
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 330
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch":
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 340
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch":
|
| 299 |
-
"grad_norm": 3.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 350
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
-
"eval_loss": 0.
|
| 307 |
-
"eval_runtime":
|
| 308 |
-
"eval_samples_per_second":
|
| 309 |
-
"eval_steps_per_second": 2.
|
| 310 |
"step": 350
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch":
|
| 314 |
-
"grad_norm":
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 360
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch":
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 370
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch":
|
| 328 |
-
"grad_norm": 3.
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 380
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch":
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 390
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch":
|
| 342 |
-
"grad_norm": 4.
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 400
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"eval_loss": 0.
|
| 350 |
-
"eval_runtime":
|
| 351 |
-
"eval_samples_per_second":
|
| 352 |
-
"eval_steps_per_second": 2.
|
| 353 |
"step": 400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 410
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 420
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 430
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 440
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"eval_loss": 0.
|
| 393 |
-
"eval_runtime":
|
| 394 |
-
"eval_samples_per_second":
|
| 395 |
-
"eval_steps_per_second": 2.
|
| 396 |
"step": 450
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"grad_norm": 4.
|
| 401 |
-
"learning_rate":
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 460
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"grad_norm":
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 470
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"grad_norm":
|
| 415 |
-
"learning_rate":
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 480
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate":
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 490
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch":
|
| 435 |
-
"eval_loss": 0.
|
| 436 |
-
"eval_runtime":
|
| 437 |
-
"eval_samples_per_second":
|
| 438 |
-
"eval_steps_per_second": 2.
|
| 439 |
"step": 500
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch":
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 510
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 520
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 530
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch":
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 540
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 550
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch":
|
| 478 |
-
"eval_loss":
|
| 479 |
-
"eval_runtime":
|
| 480 |
-
"eval_samples_per_second":
|
| 481 |
-
"eval_steps_per_second": 2.
|
| 482 |
"step": 550
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"grad_norm":
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss": 0.
|
| 489 |
"step": 560
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch":
|
| 493 |
-
"grad_norm":
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"loss": 0.
|
| 496 |
"step": 570
|
| 497 |
},
|
| 498 |
{
|
| 499 |
-
"epoch":
|
| 500 |
-
"grad_norm":
|
| 501 |
-
"learning_rate":
|
| 502 |
-
"loss": 0.
|
| 503 |
"step": 580
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"grad_norm":
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 590
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"grad_norm":
|
| 515 |
-
"learning_rate":
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 600
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch":
|
| 521 |
-
"eval_loss":
|
| 522 |
-
"eval_runtime":
|
| 523 |
-
"eval_samples_per_second":
|
| 524 |
-
"eval_steps_per_second": 2.
|
| 525 |
"step": 600
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch":
|
| 529 |
-
"grad_norm":
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 610
|
| 533 |
},
|
| 534 |
{
|
| 535 |
-
"epoch":
|
| 536 |
-
"grad_norm":
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 620
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch":
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 630
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 640
|
| 554 |
},
|
| 555 |
{
|
| 556 |
-
"epoch":
|
| 557 |
-
"grad_norm":
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 650
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"eval_loss":
|
| 565 |
-
"eval_runtime":
|
| 566 |
-
"eval_samples_per_second":
|
| 567 |
-
"eval_steps_per_second": 2.
|
| 568 |
"step": 650
|
| 569 |
},
|
| 570 |
{
|
| 571 |
-
"epoch":
|
| 572 |
-
"grad_norm":
|
| 573 |
-
"learning_rate":
|
| 574 |
-
"loss": 0.
|
| 575 |
"step": 660
|
| 576 |
},
|
| 577 |
{
|
| 578 |
-
"epoch":
|
| 579 |
-
"grad_norm":
|
| 580 |
-
"learning_rate":
|
| 581 |
-
"loss": 0.
|
| 582 |
"step": 670
|
| 583 |
},
|
| 584 |
{
|
| 585 |
-
"epoch":
|
| 586 |
-
"
|
| 587 |
-
"
|
| 588 |
-
"
|
| 589 |
-
"
|
| 590 |
-
|
| 591 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 592 |
}
|
| 593 |
],
|
| 594 |
"logging_steps": 10,
|
| 595 |
-
"max_steps":
|
| 596 |
"num_input_tokens_seen": 0,
|
| 597 |
-
"num_train_epochs":
|
| 598 |
"save_steps": 5000,
|
| 599 |
"stateful_callbacks": {
|
| 600 |
"TrainerControl": {
|
|
@@ -608,7 +1418,7 @@
|
|
| 608 |
"attributes": {}
|
| 609 |
}
|
| 610 |
},
|
| 611 |
-
"total_flos":
|
| 612 |
"train_batch_size": 2,
|
| 613 |
"trial_name": null,
|
| 614 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9953574744661093,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 1614,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.018570102135561744,
|
| 13 |
+
"grad_norm": 33.58969944080453,
|
| 14 |
+
"learning_rate": 6.17283950617284e-07,
|
| 15 |
+
"loss": 0.8947,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.03714020427112349,
|
| 20 |
+
"grad_norm": 16.744611450678853,
|
| 21 |
+
"learning_rate": 1.234567901234568e-06,
|
| 22 |
+
"loss": 0.6994,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.055710306406685235,
|
| 27 |
+
"grad_norm": 13.439601262151244,
|
| 28 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 29 |
+
"loss": 0.624,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.07428040854224698,
|
| 34 |
+
"grad_norm": 13.907357312219297,
|
| 35 |
+
"learning_rate": 2.469135802469136e-06,
|
| 36 |
+
"loss": 0.4977,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.09285051067780872,
|
| 41 |
+
"grad_norm": 8.492973049121323,
|
| 42 |
+
"learning_rate": 3.08641975308642e-06,
|
| 43 |
+
"loss": 0.4639,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.09285051067780872,
|
| 48 |
+
"eval_loss": 0.5398163199424744,
|
| 49 |
+
"eval_runtime": 19.4371,
|
| 50 |
+
"eval_samples_per_second": 18.11,
|
| 51 |
+
"eval_steps_per_second": 2.264,
|
| 52 |
"step": 50
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 0.11142061281337047,
|
| 56 |
+
"grad_norm": 8.651091209611899,
|
| 57 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 58 |
+
"loss": 0.4468,
|
| 59 |
"step": 60
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 0.12999071494893222,
|
| 63 |
+
"grad_norm": 5.085215797862169,
|
| 64 |
+
"learning_rate": 4.3209876543209875e-06,
|
| 65 |
+
"loss": 0.3832,
|
| 66 |
"step": 70
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 0.14856081708449395,
|
| 70 |
+
"grad_norm": 7.069622973657299,
|
| 71 |
+
"learning_rate": 4.938271604938272e-06,
|
| 72 |
+
"loss": 0.4276,
|
| 73 |
"step": 80
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 0.1671309192200557,
|
| 77 |
+
"grad_norm": 99.76457865321862,
|
| 78 |
+
"learning_rate": 5.555555555555557e-06,
|
| 79 |
+
"loss": 0.5276,
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.18570102135561745,
|
| 84 |
+
"grad_norm": 6.1883235099602905,
|
| 85 |
+
"learning_rate": 6.17283950617284e-06,
|
| 86 |
+
"loss": 0.4939,
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.18570102135561745,
|
| 91 |
+
"eval_loss": 0.5121593475341797,
|
| 92 |
+
"eval_runtime": 19.3824,
|
| 93 |
+
"eval_samples_per_second": 18.161,
|
| 94 |
+
"eval_steps_per_second": 2.27,
|
| 95 |
"step": 100
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 0.2042711234911792,
|
| 99 |
+
"grad_norm": 7.381670488617432,
|
| 100 |
+
"learning_rate": 6.790123456790124e-06,
|
| 101 |
+
"loss": 0.4866,
|
| 102 |
"step": 110
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.22284122562674094,
|
| 106 |
+
"grad_norm": 5.554268011959653,
|
| 107 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 108 |
+
"loss": 0.4282,
|
| 109 |
"step": 120
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.2414113277623027,
|
| 113 |
+
"grad_norm": 5.807686582470137,
|
| 114 |
+
"learning_rate": 8.024691358024692e-06,
|
| 115 |
+
"loss": 0.4918,
|
| 116 |
"step": 130
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.25998142989786444,
|
| 120 |
+
"grad_norm": 6.799528459280429,
|
| 121 |
+
"learning_rate": 8.641975308641975e-06,
|
| 122 |
+
"loss": 0.4841,
|
| 123 |
"step": 140
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.2785515320334262,
|
| 127 |
+
"grad_norm": 4.94296183453013,
|
| 128 |
+
"learning_rate": 9.25925925925926e-06,
|
| 129 |
+
"loss": 0.4822,
|
| 130 |
"step": 150
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.2785515320334262,
|
| 134 |
+
"eval_loss": 0.5242471694946289,
|
| 135 |
+
"eval_runtime": 19.4304,
|
| 136 |
+
"eval_samples_per_second": 18.116,
|
| 137 |
+
"eval_steps_per_second": 2.264,
|
| 138 |
"step": 150
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 0.2971216341689879,
|
| 142 |
+
"grad_norm": 5.6102713178528205,
|
| 143 |
+
"learning_rate": 9.876543209876543e-06,
|
| 144 |
+
"loss": 0.4886,
|
| 145 |
"step": 160
|
| 146 |
},
|
| 147 |
{
|
| 148 |
+
"epoch": 0.31569173630454966,
|
| 149 |
+
"grad_norm": 4.845388135628488,
|
| 150 |
+
"learning_rate": 9.99925101063302e-06,
|
| 151 |
+
"loss": 0.4529,
|
| 152 |
"step": 170
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.3342618384401114,
|
| 156 |
+
"grad_norm": 5.39966471118419,
|
| 157 |
+
"learning_rate": 9.99620862590714e-06,
|
| 158 |
+
"loss": 0.4463,
|
| 159 |
"step": 180
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 0.3528319405756732,
|
| 163 |
+
"grad_norm": 5.489503360186311,
|
| 164 |
+
"learning_rate": 9.990827457067342e-06,
|
| 165 |
+
"loss": 0.4885,
|
| 166 |
"step": 190
|
| 167 |
},
|
| 168 |
{
|
| 169 |
+
"epoch": 0.3714020427112349,
|
| 170 |
+
"grad_norm": 4.684590902011352,
|
| 171 |
+
"learning_rate": 9.983110023102148e-06,
|
| 172 |
+
"loss": 0.4701,
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 0.3714020427112349,
|
| 177 |
+
"eval_loss": 0.5521253943443298,
|
| 178 |
+
"eval_runtime": 19.3667,
|
| 179 |
+
"eval_samples_per_second": 18.176,
|
| 180 |
+
"eval_steps_per_second": 2.272,
|
| 181 |
"step": 200
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.38997214484679665,
|
| 185 |
+
"grad_norm": 6.186929243018788,
|
| 186 |
+
"learning_rate": 9.973059936633308e-06,
|
| 187 |
+
"loss": 0.4983,
|
| 188 |
"step": 210
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.4085422469823584,
|
| 192 |
+
"grad_norm": 7.186427473045426,
|
| 193 |
+
"learning_rate": 9.960681902224692e-06,
|
| 194 |
+
"loss": 0.4983,
|
| 195 |
"step": 220
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.4271123491179202,
|
| 199 |
+
"grad_norm": 4.948600368895616,
|
| 200 |
+
"learning_rate": 9.945981714180021e-06,
|
| 201 |
+
"loss": 0.493,
|
| 202 |
"step": 230
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.4456824512534819,
|
| 206 |
+
"grad_norm": 4.131993099439563,
|
| 207 |
+
"learning_rate": 9.928966253830492e-06,
|
| 208 |
+
"loss": 0.4753,
|
| 209 |
"step": 240
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 0.46425255338904364,
|
| 213 |
+
"grad_norm": 4.007998907864266,
|
| 214 |
+
"learning_rate": 9.909643486313533e-06,
|
| 215 |
+
"loss": 0.4216,
|
| 216 |
"step": 250
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 0.46425255338904364,
|
| 220 |
+
"eval_loss": 0.5373674631118774,
|
| 221 |
+
"eval_runtime": 19.3954,
|
| 222 |
+
"eval_samples_per_second": 18.149,
|
| 223 |
+
"eval_steps_per_second": 2.269,
|
| 224 |
"step": 250
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.4828226555246054,
|
| 228 |
+
"grad_norm": 4.355924678117399,
|
| 229 |
+
"learning_rate": 9.888022456844251e-06,
|
| 230 |
+
"loss": 0.4874,
|
| 231 |
"step": 260
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 0.5013927576601671,
|
| 235 |
+
"grad_norm": 4.3475514584684385,
|
| 236 |
+
"learning_rate": 9.864113286481237e-06,
|
| 237 |
+
"loss": 0.4499,
|
| 238 |
"step": 270
|
| 239 |
},
|
| 240 |
{
|
| 241 |
+
"epoch": 0.5199628597957289,
|
| 242 |
+
"grad_norm": 4.925328524759934,
|
| 243 |
+
"learning_rate": 9.837927167388793e-06,
|
| 244 |
+
"loss": 0.4604,
|
| 245 |
"step": 280
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 0.5385329619312906,
|
| 249 |
+
"grad_norm": 4.580602008041337,
|
| 250 |
+
"learning_rate": 9.809476357597738e-06,
|
| 251 |
+
"loss": 0.4697,
|
| 252 |
"step": 290
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 0.5571030640668524,
|
| 256 |
+
"grad_norm": 3.9055769337575335,
|
| 257 |
+
"learning_rate": 9.778774175267294e-06,
|
| 258 |
+
"loss": 0.4159,
|
| 259 |
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
+
"epoch": 0.5571030640668524,
|
| 263 |
+
"eval_loss": 0.5146207213401794,
|
| 264 |
+
"eval_runtime": 19.4216,
|
| 265 |
+
"eval_samples_per_second": 18.124,
|
| 266 |
+
"eval_steps_per_second": 2.266,
|
| 267 |
"step": 300
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.5756731662024142,
|
| 271 |
+
"grad_norm": 4.477623029269123,
|
| 272 |
+
"learning_rate": 9.745834992450688e-06,
|
| 273 |
+
"loss": 0.487,
|
| 274 |
"step": 310
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 0.5942432683379758,
|
| 278 |
+
"grad_norm": 3.729028613174465,
|
| 279 |
+
"learning_rate": 9.710674228367422e-06,
|
| 280 |
+
"loss": 0.4751,
|
| 281 |
"step": 320
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.6128133704735376,
|
| 285 |
+
"grad_norm": 4.298167407385586,
|
| 286 |
+
"learning_rate": 9.673308342185366e-06,
|
| 287 |
+
"loss": 0.4289,
|
| 288 |
"step": 330
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 0.6313834726090993,
|
| 292 |
+
"grad_norm": 4.984754626750181,
|
| 293 |
+
"learning_rate": 9.633754825316015e-06,
|
| 294 |
+
"loss": 0.4783,
|
| 295 |
"step": 340
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 0.6499535747446611,
|
| 299 |
+
"grad_norm": 3.9844077010010444,
|
| 300 |
+
"learning_rate": 9.592032193226564e-06,
|
| 301 |
+
"loss": 0.4502,
|
| 302 |
"step": 350
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 0.6499535747446611,
|
| 306 |
+
"eval_loss": 0.5022189617156982,
|
| 307 |
+
"eval_runtime": 19.4166,
|
| 308 |
+
"eval_samples_per_second": 18.129,
|
| 309 |
+
"eval_steps_per_second": 2.266,
|
| 310 |
"step": 350
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.6685236768802229,
|
| 314 |
+
"grad_norm": 4.474043738957621,
|
| 315 |
+
"learning_rate": 9.548159976772593e-06,
|
| 316 |
+
"loss": 0.4341,
|
| 317 |
"step": 360
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.6870937790157846,
|
| 321 |
+
"grad_norm": 12.385083857047688,
|
| 322 |
+
"learning_rate": 9.502158713055444e-06,
|
| 323 |
+
"loss": 0.4612,
|
| 324 |
"step": 370
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.7056638811513464,
|
| 328 |
+
"grad_norm": 3.8491200153587535,
|
| 329 |
+
"learning_rate": 9.454049935808568e-06,
|
| 330 |
+
"loss": 0.4352,
|
| 331 |
"step": 380
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.724233983286908,
|
| 335 |
+
"grad_norm": 4.149978752910591,
|
| 336 |
+
"learning_rate": 9.403856165317322e-06,
|
| 337 |
+
"loss": 0.3996,
|
| 338 |
"step": 390
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.7428040854224698,
|
| 342 |
+
"grad_norm": 4.740837408740574,
|
| 343 |
+
"learning_rate": 9.351600897876964e-06,
|
| 344 |
+
"loss": 0.4625,
|
| 345 |
"step": 400
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.7428040854224698,
|
| 349 |
+
"eval_loss": 0.49847906827926636,
|
| 350 |
+
"eval_runtime": 19.3879,
|
| 351 |
+
"eval_samples_per_second": 18.156,
|
| 352 |
+
"eval_steps_per_second": 2.269,
|
| 353 |
"step": 400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.7613741875580315,
|
| 357 |
+
"grad_norm": 4.734999881046304,
|
| 358 |
+
"learning_rate": 9.297308594793757e-06,
|
| 359 |
+
"loss": 0.3962,
|
| 360 |
"step": 410
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 0.7799442896935933,
|
| 364 |
+
"grad_norm": 3.769866725347126,
|
| 365 |
+
"learning_rate": 9.241004670934348e-06,
|
| 366 |
+
"loss": 0.458,
|
| 367 |
"step": 420
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 0.7985143918291551,
|
| 371 |
+
"grad_norm": 3.6685753005427832,
|
| 372 |
+
"learning_rate": 9.182715482828764e-06,
|
| 373 |
+
"loss": 0.3995,
|
| 374 |
"step": 430
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.8170844939647168,
|
| 378 |
+
"grad_norm": 3.876272765109621,
|
| 379 |
+
"learning_rate": 9.122468316332611e-06,
|
| 380 |
+
"loss": 0.5064,
|
| 381 |
"step": 440
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.8356545961002786,
|
| 385 |
+
"grad_norm": 4.30859512157268,
|
| 386 |
+
"learning_rate": 9.060291373854252e-06,
|
| 387 |
+
"loss": 0.4313,
|
| 388 |
"step": 450
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 0.8356545961002786,
|
| 392 |
+
"eval_loss": 0.47159674763679504,
|
| 393 |
+
"eval_runtime": 19.4024,
|
| 394 |
+
"eval_samples_per_second": 18.142,
|
| 395 |
+
"eval_steps_per_second": 2.268,
|
| 396 |
"step": 450
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 0.8542246982358404,
|
| 400 |
+
"grad_norm": 4.898314330177169,
|
| 401 |
+
"learning_rate": 8.99621376115291e-06,
|
| 402 |
+
"loss": 0.3925,
|
| 403 |
"step": 460
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 0.872794800371402,
|
| 407 |
+
"grad_norm": 3.1139075537025684,
|
| 408 |
+
"learning_rate": 8.930265473713939e-06,
|
| 409 |
+
"loss": 0.4499,
|
| 410 |
"step": 470
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.8913649025069638,
|
| 414 |
+
"grad_norm": 3.627205546071805,
|
| 415 |
+
"learning_rate": 8.862477382707569e-06,
|
| 416 |
+
"loss": 0.3699,
|
| 417 |
"step": 480
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 0.9099350046425255,
|
| 421 |
+
"grad_norm": 4.789522180281674,
|
| 422 |
+
"learning_rate": 8.792881220537752e-06,
|
| 423 |
+
"loss": 0.4436,
|
| 424 |
"step": 490
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 0.9285051067780873,
|
| 428 |
+
"grad_norm": 2.96048981245909,
|
| 429 |
+
"learning_rate": 8.721509565987858e-06,
|
| 430 |
+
"loss": 0.4472,
|
| 431 |
"step": 500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.9285051067780873,
|
| 435 |
+
"eval_loss": 0.47707709670066833,
|
| 436 |
+
"eval_runtime": 19.4043,
|
| 437 |
+
"eval_samples_per_second": 18.14,
|
| 438 |
+
"eval_steps_per_second": 2.268,
|
| 439 |
"step": 500
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 0.947075208913649,
|
| 443 |
+
"grad_norm": 3.043235000042675,
|
| 444 |
+
"learning_rate": 8.64839582897015e-06,
|
| 445 |
+
"loss": 0.4165,
|
| 446 |
"step": 510
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 0.9656453110492108,
|
| 450 |
+
"grad_norm": 4.594551610015674,
|
| 451 |
+
"learning_rate": 8.573574234886217e-06,
|
| 452 |
+
"loss": 0.4319,
|
| 453 |
"step": 520
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 0.9842154131847726,
|
| 457 |
+
"grad_norm": 3.3821933340348482,
|
| 458 |
+
"learning_rate": 8.497079808605659e-06,
|
| 459 |
+
"loss": 0.4254,
|
| 460 |
"step": 530
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 1.0018570102135562,
|
| 464 |
+
"grad_norm": 6.485913778240125,
|
| 465 |
+
"learning_rate": 8.418948358070535e-06,
|
| 466 |
+
"loss": 0.4035,
|
| 467 |
"step": 540
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 1.020427112349118,
|
| 471 |
+
"grad_norm": 3.6050686984381373,
|
| 472 |
+
"learning_rate": 8.339216457533244e-06,
|
| 473 |
+
"loss": 0.2753,
|
| 474 |
"step": 550
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 1.020427112349118,
|
| 478 |
+
"eval_loss": 0.5026086568832397,
|
| 479 |
+
"eval_runtime": 19.4196,
|
| 480 |
+
"eval_samples_per_second": 18.126,
|
| 481 |
+
"eval_steps_per_second": 2.266,
|
| 482 |
"step": 550
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 1.0389972144846797,
|
| 486 |
+
"grad_norm": 3.2597416531200416,
|
| 487 |
+
"learning_rate": 8.257921430435678e-06,
|
| 488 |
+
"loss": 0.2879,
|
| 489 |
"step": 560
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 1.0575673166202415,
|
| 493 |
+
"grad_norm": 3.347364037440774,
|
| 494 |
+
"learning_rate": 8.175101331937692e-06,
|
| 495 |
+
"loss": 0.2567,
|
| 496 |
"step": 570
|
| 497 |
},
|
| 498 |
{
|
| 499 |
+
"epoch": 1.076137418755803,
|
| 500 |
+
"grad_norm": 2.8678452979996316,
|
| 501 |
+
"learning_rate": 8.090794931103026e-06,
|
| 502 |
+
"loss": 0.2597,
|
| 503 |
"step": 580
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 1.0947075208913648,
|
| 507 |
+
"grad_norm": 3.714937118745614,
|
| 508 |
+
"learning_rate": 8.005041692751055e-06,
|
| 509 |
+
"loss": 0.2831,
|
| 510 |
"step": 590
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 1.1132776230269266,
|
| 514 |
+
"grad_norm": 4.1369787389179224,
|
| 515 |
+
"learning_rate": 7.917881758982838e-06,
|
| 516 |
+
"loss": 0.2877,
|
| 517 |
"step": 600
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 1.1132776230269266,
|
| 521 |
+
"eval_loss": 0.4783521890640259,
|
| 522 |
+
"eval_runtime": 19.3969,
|
| 523 |
+
"eval_samples_per_second": 18.147,
|
| 524 |
+
"eval_steps_per_second": 2.268,
|
| 525 |
"step": 600
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 1.1318477251624883,
|
| 529 |
+
"grad_norm": 3.8629861075408964,
|
| 530 |
+
"learning_rate": 7.829355930390126e-06,
|
| 531 |
+
"loss": 0.2965,
|
| 532 |
"step": 610
|
| 533 |
},
|
| 534 |
{
|
| 535 |
+
"epoch": 1.15041782729805,
|
| 536 |
+
"grad_norm": 3.8801891490338334,
|
| 537 |
+
"learning_rate": 7.739505646956136e-06,
|
| 538 |
+
"loss": 0.2889,
|
| 539 |
"step": 620
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 1.1689879294336118,
|
| 543 |
+
"grad_norm": 3.1120305459072584,
|
| 544 |
+
"learning_rate": 7.648372968656995e-06,
|
| 545 |
+
"loss": 0.2891,
|
| 546 |
"step": 630
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 1.1875580315691736,
|
| 550 |
+
"grad_norm": 2.7200004915782636,
|
| 551 |
+
"learning_rate": 7.5560005557729664e-06,
|
| 552 |
+
"loss": 0.2774,
|
| 553 |
"step": 640
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 1.2061281337047354,
|
| 557 |
+
"grad_norm": 3.7661310879129415,
|
| 558 |
+
"learning_rate": 7.462431648918689e-06,
|
| 559 |
+
"loss": 0.3038,
|
| 560 |
"step": 650
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 1.2061281337047354,
|
| 564 |
+
"eval_loss": 0.4794943630695343,
|
| 565 |
+
"eval_runtime": 19.4005,
|
| 566 |
+
"eval_samples_per_second": 18.144,
|
| 567 |
+
"eval_steps_per_second": 2.268,
|
| 568 |
"step": 650
|
| 569 |
},
|
| 570 |
{
|
| 571 |
+
"epoch": 1.2246982358402971,
|
| 572 |
+
"grad_norm": 3.4542640506421876,
|
| 573 |
+
"learning_rate": 7.367710048801715e-06,
|
| 574 |
+
"loss": 0.292,
|
| 575 |
"step": 660
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 1.243268337975859,
|
| 579 |
+
"grad_norm": 2.949849968659152,
|
| 580 |
+
"learning_rate": 7.271880095718895e-06,
|
| 581 |
+
"loss": 0.2753,
|
| 582 |
"step": 670
|
| 583 |
},
|
| 584 |
{
|
| 585 |
+
"epoch": 1.2618384401114207,
|
| 586 |
+
"grad_norm": 3.7882283436275834,
|
| 587 |
+
"learning_rate": 7.1749866488001604e-06,
|
| 588 |
+
"loss": 0.2802,
|
| 589 |
+
"step": 680
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.2804085422469824,
|
| 593 |
+
"grad_norm": 2.7991103167976403,
|
| 594 |
+
"learning_rate": 7.0770750650094335e-06,
|
| 595 |
+
"loss": 0.284,
|
| 596 |
+
"step": 690
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.2989786443825442,
|
| 600 |
+
"grad_norm": 3.1581810974702322,
|
| 601 |
+
"learning_rate": 6.978191177912499e-06,
|
| 602 |
+
"loss": 0.2944,
|
| 603 |
+
"step": 700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.2989786443825442,
|
| 607 |
+
"eval_loss": 0.4681582748889923,
|
| 608 |
+
"eval_runtime": 19.4083,
|
| 609 |
+
"eval_samples_per_second": 18.137,
|
| 610 |
+
"eval_steps_per_second": 2.267,
|
| 611 |
+
"step": 700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.317548746518106,
|
| 615 |
+
"grad_norm": 3.1190982947461943,
|
| 616 |
+
"learning_rate": 6.878381276221777e-06,
|
| 617 |
+
"loss": 0.2786,
|
| 618 |
+
"step": 710
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.3361188486536677,
|
| 622 |
+
"grad_norm": 3.3739780549267757,
|
| 623 |
+
"learning_rate": 6.777692082128024e-06,
|
| 624 |
+
"loss": 0.2748,
|
| 625 |
+
"step": 720
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.3546889507892295,
|
| 629 |
+
"grad_norm": 3.2357722923246315,
|
| 630 |
+
"learning_rate": 6.676170729429132e-06,
|
| 631 |
+
"loss": 0.2635,
|
| 632 |
+
"step": 730
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.3732590529247912,
|
| 636 |
+
"grad_norm": 3.603099890429247,
|
| 637 |
+
"learning_rate": 6.573864741466236e-06,
|
| 638 |
+
"loss": 0.2978,
|
| 639 |
+
"step": 740
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.3918291550603528,
|
| 643 |
+
"grad_norm": 3.3730203042394873,
|
| 644 |
+
"learning_rate": 6.470822008877482e-06,
|
| 645 |
+
"loss": 0.2722,
|
| 646 |
+
"step": 750
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.3918291550603528,
|
| 650 |
+
"eval_loss": 0.4681358337402344,
|
| 651 |
+
"eval_runtime": 19.3821,
|
| 652 |
+
"eval_samples_per_second": 18.161,
|
| 653 |
+
"eval_steps_per_second": 2.27,
|
| 654 |
+
"step": 750
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.4103992571959145,
|
| 658 |
+
"grad_norm": 3.608817378094592,
|
| 659 |
+
"learning_rate": 6.367090767179855e-06,
|
| 660 |
+
"loss": 0.2691,
|
| 661 |
+
"step": 760
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.4289693593314763,
|
| 665 |
+
"grad_norm": 4.122346770691036,
|
| 666 |
+
"learning_rate": 6.262719574189564e-06,
|
| 667 |
+
"loss": 0.2974,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.447539461467038,
|
| 672 |
+
"grad_norm": 3.795562564382414,
|
| 673 |
+
"learning_rate": 6.157757287291557e-06,
|
| 674 |
+
"loss": 0.272,
|
| 675 |
+
"step": 780
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.4661095636025998,
|
| 679 |
+
"grad_norm": 3.4474808363843903,
|
| 680 |
+
"learning_rate": 6.052253040568804e-06,
|
| 681 |
+
"loss": 0.2924,
|
| 682 |
+
"step": 790
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.4846796657381616,
|
| 686 |
+
"grad_norm": 3.5395096255499885,
|
| 687 |
+
"learning_rate": 5.946256221802052e-06,
|
| 688 |
+
"loss": 0.2734,
|
| 689 |
+
"step": 800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.4846796657381616,
|
| 693 |
+
"eval_loss": 0.4480016827583313,
|
| 694 |
+
"eval_runtime": 19.4053,
|
| 695 |
+
"eval_samples_per_second": 18.139,
|
| 696 |
+
"eval_steps_per_second": 2.267,
|
| 697 |
+
"step": 800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.5032497678737233,
|
| 701 |
+
"grad_norm": 2.890746658579431,
|
| 702 |
+
"learning_rate": 5.839816449350824e-06,
|
| 703 |
+
"loss": 0.2277,
|
| 704 |
+
"step": 810
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.521819870009285,
|
| 708 |
+
"grad_norm": 3.8258223876451884,
|
| 709 |
+
"learning_rate": 5.7329835489264855e-06,
|
| 710 |
+
"loss": 0.296,
|
| 711 |
+
"step": 820
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.5403899721448466,
|
| 715 |
+
"grad_norm": 3.528517987679271,
|
| 716 |
+
"learning_rate": 5.62580753026823e-06,
|
| 717 |
+
"loss": 0.2521,
|
| 718 |
+
"step": 830
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.5589600742804084,
|
| 722 |
+
"grad_norm": 2.677594926231651,
|
| 723 |
+
"learning_rate": 5.518338563732945e-06,
|
| 724 |
+
"loss": 0.2523,
|
| 725 |
+
"step": 840
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.5775301764159702,
|
| 729 |
+
"grad_norm": 3.4048411120623876,
|
| 730 |
+
"learning_rate": 5.410626956809864e-06,
|
| 731 |
+
"loss": 0.2826,
|
| 732 |
+
"step": 850
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.5775301764159702,
|
| 736 |
+
"eval_loss": 0.44835272431373596,
|
| 737 |
+
"eval_runtime": 19.3933,
|
| 738 |
+
"eval_samples_per_second": 18.151,
|
| 739 |
+
"eval_steps_per_second": 2.269,
|
| 740 |
+
"step": 850
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.596100278551532,
|
| 744 |
+
"grad_norm": 2.8957625744349014,
|
| 745 |
+
"learning_rate": 5.30272313057105e-06,
|
| 746 |
+
"loss": 0.2436,
|
| 747 |
+
"step": 860
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.6146703806870937,
|
| 751 |
+
"grad_norm": 2.9749580565630933,
|
| 752 |
+
"learning_rate": 5.194677596068689e-06,
|
| 753 |
+
"loss": 0.2633,
|
| 754 |
+
"step": 870
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.6332404828226554,
|
| 758 |
+
"grad_norm": 3.485048080470531,
|
| 759 |
+
"learning_rate": 5.0865409306902755e-06,
|
| 760 |
+
"loss": 0.2525,
|
| 761 |
+
"step": 880
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.6518105849582172,
|
| 765 |
+
"grad_norm": 3.517296425834879,
|
| 766 |
+
"learning_rate": 4.978363754482741e-06,
|
| 767 |
+
"loss": 0.2776,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.670380687093779,
|
| 772 |
+
"grad_norm": 2.773820634987393,
|
| 773 |
+
"learning_rate": 4.870196706456609e-06,
|
| 774 |
+
"loss": 0.2344,
|
| 775 |
+
"step": 900
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.670380687093779,
|
| 779 |
+
"eval_loss": 0.43884018063545227,
|
| 780 |
+
"eval_runtime": 19.3629,
|
| 781 |
+
"eval_samples_per_second": 18.179,
|
| 782 |
+
"eval_steps_per_second": 2.272,
|
| 783 |
+
"step": 900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.6889507892293407,
|
| 787 |
+
"grad_norm": 4.004058405011273,
|
| 788 |
+
"learning_rate": 4.762090420881289e-06,
|
| 789 |
+
"loss": 0.247,
|
| 790 |
+
"step": 910
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.7075208913649025,
|
| 794 |
+
"grad_norm": 4.098064484974279,
|
| 795 |
+
"learning_rate": 4.654095503582568e-06,
|
| 796 |
+
"loss": 0.2864,
|
| 797 |
+
"step": 920
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.7260909935004642,
|
| 801 |
+
"grad_norm": 3.5973991483602945,
|
| 802 |
+
"learning_rate": 4.546262508253429e-06,
|
| 803 |
+
"loss": 0.248,
|
| 804 |
+
"step": 930
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.744661095636026,
|
| 808 |
+
"grad_norm": 4.143958197553495,
|
| 809 |
+
"learning_rate": 4.438641912789277e-06,
|
| 810 |
+
"loss": 0.265,
|
| 811 |
+
"step": 940
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.7632311977715878,
|
| 815 |
+
"grad_norm": 2.8268071610510126,
|
| 816 |
+
"learning_rate": 4.331284095658637e-06,
|
| 817 |
+
"loss": 0.2437,
|
| 818 |
+
"step": 950
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.7632311977715878,
|
| 822 |
+
"eval_loss": 0.42715513706207275,
|
| 823 |
+
"eval_runtime": 19.3915,
|
| 824 |
+
"eval_samples_per_second": 18.152,
|
| 825 |
+
"eval_steps_per_second": 2.269,
|
| 826 |
+
"step": 950
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.7818012999071495,
|
| 830 |
+
"grad_norm": 3.815603949375226,
|
| 831 |
+
"learning_rate": 4.224239312320399e-06,
|
| 832 |
+
"loss": 0.2493,
|
| 833 |
+
"step": 960
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.8003714020427113,
|
| 837 |
+
"grad_norm": 3.7789513906241248,
|
| 838 |
+
"learning_rate": 4.117557671698648e-06,
|
| 839 |
+
"loss": 0.2371,
|
| 840 |
+
"step": 970
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.818941504178273,
|
| 844 |
+
"grad_norm": 2.8941597604671903,
|
| 845 |
+
"learning_rate": 4.011289112726085e-06,
|
| 846 |
+
"loss": 0.2605,
|
| 847 |
+
"step": 980
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 1.8375116063138348,
|
| 851 |
+
"grad_norm": 3.5121780352826337,
|
| 852 |
+
"learning_rate": 3.905483380967027e-06,
|
| 853 |
+
"loss": 0.2457,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.8560817084493966,
|
| 858 |
+
"grad_norm": 3.129527995601254,
|
| 859 |
+
"learning_rate": 3.800190005330918e-06,
|
| 860 |
+
"loss": 0.2113,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 1.8560817084493966,
|
| 865 |
+
"eval_loss": 0.4232879877090454,
|
| 866 |
+
"eval_runtime": 19.3985,
|
| 867 |
+
"eval_samples_per_second": 18.146,
|
| 868 |
+
"eval_steps_per_second": 2.268,
|
| 869 |
+
"step": 1000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.8746518105849583,
|
| 873 |
+
"grad_norm": 3.7626616094158876,
|
| 874 |
+
"learning_rate": 3.695458274887268e-06,
|
| 875 |
+
"loss": 0.2709,
|
| 876 |
+
"step": 1010
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.89322191272052,
|
| 880 |
+
"grad_norm": 3.5799401116393743,
|
| 881 |
+
"learning_rate": 3.5913372157928515e-06,
|
| 882 |
+
"loss": 0.2678,
|
| 883 |
+
"step": 1020
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.9117920148560819,
|
| 887 |
+
"grad_norm": 3.304004098929529,
|
| 888 |
+
"learning_rate": 3.487875568341995e-06,
|
| 889 |
+
"loss": 0.2439,
|
| 890 |
+
"step": 1030
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.9303621169916436,
|
| 894 |
+
"grad_norm": 2.842908582122805,
|
| 895 |
+
"learning_rate": 3.3851217641506657e-06,
|
| 896 |
+
"loss": 0.2691,
|
| 897 |
+
"step": 1040
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.9489322191272052,
|
| 901 |
+
"grad_norm": 2.4953468111237402,
|
| 902 |
+
"learning_rate": 3.2831239034850593e-06,
|
| 903 |
+
"loss": 0.2548,
|
| 904 |
+
"step": 1050
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.9489322191272052,
|
| 908 |
+
"eval_loss": 0.4117203652858734,
|
| 909 |
+
"eval_runtime": 19.4229,
|
| 910 |
+
"eval_samples_per_second": 18.123,
|
| 911 |
+
"eval_steps_per_second": 2.265,
|
| 912 |
+
"step": 1050
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.967502321262767,
|
| 916 |
+
"grad_norm": 3.2206038323962836,
|
| 917 |
+
"learning_rate": 3.1819297327453045e-06,
|
| 918 |
+
"loss": 0.2381,
|
| 919 |
+
"step": 1060
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.9860724233983287,
|
| 923 |
+
"grad_norm": 3.5882038303694372,
|
| 924 |
+
"learning_rate": 3.081586622114809e-06,
|
| 925 |
+
"loss": 0.219,
|
| 926 |
+
"step": 1070
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.0037140204271124,
|
| 930 |
+
"grad_norm": 2.4519833657637693,
|
| 931 |
+
"learning_rate": 2.9821415433857174e-06,
|
| 932 |
+
"loss": 0.245,
|
| 933 |
+
"step": 1080
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.022284122562674,
|
| 937 |
+
"grad_norm": 2.5708061741401345,
|
| 938 |
+
"learning_rate": 2.8836410479708625e-06,
|
| 939 |
+
"loss": 0.1082,
|
| 940 |
+
"step": 1090
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.040854224698236,
|
| 944 |
+
"grad_norm": 2.758692661666751,
|
| 945 |
+
"learning_rate": 2.786131245112495e-06,
|
| 946 |
+
"loss": 0.1126,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.040854224698236,
|
| 951 |
+
"eval_loss": 0.5031464695930481,
|
| 952 |
+
"eval_runtime": 19.3667,
|
| 953 |
+
"eval_samples_per_second": 18.176,
|
| 954 |
+
"eval_steps_per_second": 2.272,
|
| 955 |
+
"step": 1100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.0594243268337977,
|
| 959 |
+
"grad_norm": 2.9605517752760933,
|
| 960 |
+
"learning_rate": 2.689657780298019e-06,
|
| 961 |
+
"loss": 0.1195,
|
| 962 |
+
"step": 1110
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 2.0779944289693595,
|
| 966 |
+
"grad_norm": 2.55505057652882,
|
| 967 |
+
"learning_rate": 2.5942658138927866e-06,
|
| 968 |
+
"loss": 0.1006,
|
| 969 |
+
"step": 1120
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.0965645311049212,
|
| 973 |
+
"grad_norm": 3.246793514684397,
|
| 974 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 975 |
+
"loss": 0.1069,
|
| 976 |
+
"step": 1130
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.115134633240483,
|
| 980 |
+
"grad_norm": 3.2377032295876105,
|
| 981 |
+
"learning_rate": 2.406904465557614e-06,
|
| 982 |
+
"loss": 0.0936,
|
| 983 |
+
"step": 1140
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.1337047353760448,
|
| 987 |
+
"grad_norm": 2.5395933544539218,
|
| 988 |
+
"learning_rate": 2.3150227896819782e-06,
|
| 989 |
+
"loss": 0.1128,
|
| 990 |
+
"step": 1150
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 2.1337047353760448,
|
| 994 |
+
"eval_loss": 0.48206087946891785,
|
| 995 |
+
"eval_runtime": 19.455,
|
| 996 |
+
"eval_samples_per_second": 18.093,
|
| 997 |
+
"eval_steps_per_second": 2.262,
|
| 998 |
+
"step": 1150
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.152274837511606,
|
| 1002 |
+
"grad_norm": 2.5999204788587007,
|
| 1003 |
+
"learning_rate": 2.2243979832679515e-06,
|
| 1004 |
+
"loss": 0.1172,
|
| 1005 |
+
"step": 1160
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.170844939647168,
|
| 1009 |
+
"grad_norm": 2.9958486705632446,
|
| 1010 |
+
"learning_rate": 2.1350724688549906e-06,
|
| 1011 |
+
"loss": 0.1098,
|
| 1012 |
+
"step": 1170
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 2.1894150417827296,
|
| 1016 |
+
"grad_norm": 2.2061344645528633,
|
| 1017 |
+
"learning_rate": 2.0470880607686605e-06,
|
| 1018 |
+
"loss": 0.0953,
|
| 1019 |
+
"step": 1180
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.2079851439182914,
|
| 1023 |
+
"grad_norm": 1.406675633603301,
|
| 1024 |
+
"learning_rate": 1.9604859455468587e-06,
|
| 1025 |
+
"loss": 0.1081,
|
| 1026 |
+
"step": 1190
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 2.226555246053853,
|
| 1030 |
+
"grad_norm": 2.9325208784441203,
|
| 1031 |
+
"learning_rate": 1.8753066626599086e-06,
|
| 1032 |
+
"loss": 0.0993,
|
| 1033 |
+
"step": 1200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.226555246053853,
|
| 1037 |
+
"eval_loss": 0.49974343180656433,
|
| 1038 |
+
"eval_runtime": 19.4104,
|
| 1039 |
+
"eval_samples_per_second": 18.135,
|
| 1040 |
+
"eval_steps_per_second": 2.267,
|
| 1041 |
+
"step": 1200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 2.245125348189415,
|
| 1045 |
+
"grad_norm": 3.1642149256556356,
|
| 1046 |
+
"learning_rate": 1.7915900855335506e-06,
|
| 1047 |
+
"loss": 0.1187,
|
| 1048 |
+
"step": 1210
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 2.2636954503249767,
|
| 1052 |
+
"grad_norm": 3.0092943445643003,
|
| 1053 |
+
"learning_rate": 1.7093754028837345e-06,
|
| 1054 |
+
"loss": 0.1,
|
| 1055 |
+
"step": 1220
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 2.2822655524605384,
|
| 1059 |
+
"grad_norm": 2.553339861021684,
|
| 1060 |
+
"learning_rate": 1.6287011003719105e-06,
|
| 1061 |
+
"loss": 0.1044,
|
| 1062 |
+
"step": 1230
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 2.3008356545961,
|
| 1066 |
+
"grad_norm": 2.021123673041649,
|
| 1067 |
+
"learning_rate": 1.549604942589441e-06,
|
| 1068 |
+
"loss": 0.1012,
|
| 1069 |
+
"step": 1240
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 2.319405756731662,
|
| 1073 |
+
"grad_norm": 2.1421187938993724,
|
| 1074 |
+
"learning_rate": 1.4721239553795485e-06,
|
| 1075 |
+
"loss": 0.0978,
|
| 1076 |
+
"step": 1250
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 2.319405756731662,
|
| 1080 |
+
"eval_loss": 0.4895870089530945,
|
| 1081 |
+
"eval_runtime": 19.3723,
|
| 1082 |
+
"eval_samples_per_second": 18.17,
|
| 1083 |
+
"eval_steps_per_second": 2.271,
|
| 1084 |
+
"step": 1250
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 2.3379758588672237,
|
| 1088 |
+
"grad_norm": 3.298579283457447,
|
| 1089 |
+
"learning_rate": 1.3962944085050833e-06,
|
| 1090 |
+
"loss": 0.1094,
|
| 1091 |
+
"step": 1260
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 2.3565459610027855,
|
| 1095 |
+
"grad_norm": 3.3116222885011846,
|
| 1096 |
+
"learning_rate": 1.3221517986702249e-06,
|
| 1097 |
+
"loss": 0.1023,
|
| 1098 |
+
"step": 1270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 2.3751160631383472,
|
| 1102 |
+
"grad_norm": 2.3986802183164824,
|
| 1103 |
+
"learning_rate": 1.2497308329040475e-06,
|
| 1104 |
+
"loss": 0.0953,
|
| 1105 |
+
"step": 1280
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 2.393686165273909,
|
| 1109 |
+
"grad_norm": 2.6963628811320035,
|
| 1110 |
+
"learning_rate": 1.1790654123137552e-06,
|
| 1111 |
+
"loss": 0.1014,
|
| 1112 |
+
"step": 1290
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.4122562674094707,
|
| 1116 |
+
"grad_norm": 3.541306025884895,
|
| 1117 |
+
"learning_rate": 1.1101886162151764e-06,
|
| 1118 |
+
"loss": 0.1056,
|
| 1119 |
+
"step": 1300
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.4122562674094707,
|
| 1123 |
+
"eval_loss": 0.4979850947856903,
|
| 1124 |
+
"eval_runtime": 19.4616,
|
| 1125 |
+
"eval_samples_per_second": 18.087,
|
| 1126 |
+
"eval_steps_per_second": 2.261,
|
| 1127 |
+
"step": 1300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.4308263695450325,
|
| 1131 |
+
"grad_norm": 1.8964890409301638,
|
| 1132 |
+
"learning_rate": 1.0431326866479457e-06,
|
| 1133 |
+
"loss": 0.0855,
|
| 1134 |
+
"step": 1310
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.4493964716805943,
|
| 1138 |
+
"grad_norm": 2.08584902180452,
|
| 1139 |
+
"learning_rate": 9.779290132826224e-07,
|
| 1140 |
+
"loss": 0.1018,
|
| 1141 |
+
"step": 1320
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 2.467966573816156,
|
| 1145 |
+
"grad_norm": 2.1509863412780357,
|
| 1146 |
+
"learning_rate": 9.146081187268185e-07,
|
| 1147 |
+
"loss": 0.1057,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.486536675951718,
|
| 1152 |
+
"grad_norm": 2.337018038983776,
|
| 1153 |
+
"learning_rate": 8.531996442372048e-07,
|
| 1154 |
+
"loss": 0.0969,
|
| 1155 |
+
"step": 1340
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.5051067780872796,
|
| 1159 |
+
"grad_norm": 2.1781847876681604,
|
| 1160 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1161 |
+
"loss": 0.0897,
|
| 1162 |
+
"step": 1350
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 2.5051067780872796,
|
| 1166 |
+
"eval_loss": 0.4882669448852539,
|
| 1167 |
+
"eval_runtime": 19.5254,
|
| 1168 |
+
"eval_samples_per_second": 18.028,
|
| 1169 |
+
"eval_steps_per_second": 2.253,
|
| 1170 |
+
"step": 1350
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 2.5236768802228413,
|
| 1174 |
+
"grad_norm": 2.6685806136301684,
|
| 1175 |
+
"learning_rate": 7.362340308950783e-07,
|
| 1176 |
+
"loss": 0.0976,
|
| 1177 |
+
"step": 1360
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 2.542246982358403,
|
| 1181 |
+
"grad_norm": 2.6897365042467047,
|
| 1182 |
+
"learning_rate": 6.807316450240425e-07,
|
| 1183 |
+
"loss": 0.0957,
|
| 1184 |
+
"step": 1370
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.560817084493965,
|
| 1188 |
+
"grad_norm": 2.143646748023798,
|
| 1189 |
+
"learning_rate": 6.2725115955164e-07,
|
| 1190 |
+
"loss": 0.1087,
|
| 1191 |
+
"step": 1380
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 2.5793871866295266,
|
| 1195 |
+
"grad_norm": 3.3390598746500264,
|
| 1196 |
+
"learning_rate": 5.758176093231294e-07,
|
| 1197 |
+
"loss": 0.0952,
|
| 1198 |
+
"step": 1390
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 2.5979572887650884,
|
| 1202 |
+
"grad_norm": 3.150702584087223,
|
| 1203 |
+
"learning_rate": 5.264550709892685e-07,
|
| 1204 |
+
"loss": 0.0872,
|
| 1205 |
+
"step": 1400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 2.5979572887650884,
|
| 1209 |
+
"eval_loss": 0.49406710267066956,
|
| 1210 |
+
"eval_runtime": 19.3841,
|
| 1211 |
+
"eval_samples_per_second": 18.159,
|
| 1212 |
+
"eval_steps_per_second": 2.27,
|
| 1213 |
+
"step": 1400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.61652739090065,
|
| 1217 |
+
"grad_norm": 3.077156808484069,
|
| 1218 |
+
"learning_rate": 4.791866517357491e-07,
|
| 1219 |
+
"loss": 0.1028,
|
| 1220 |
+
"step": 1410
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.635097493036212,
|
| 1224 |
+
"grad_norm": 2.2139585731845703,
|
| 1225 |
+
"learning_rate": 4.3403447846645355e-07,
|
| 1226 |
+
"loss": 0.0929,
|
| 1227 |
+
"step": 1420
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.6536675951717736,
|
| 1231 |
+
"grad_norm": 2.509450394120424,
|
| 1232 |
+
"learning_rate": 3.910196874455896e-07,
|
| 1233 |
+
"loss": 0.0956,
|
| 1234 |
+
"step": 1430
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.6722376973073354,
|
| 1238 |
+
"grad_norm": 4.615247501065769,
|
| 1239 |
+
"learning_rate": 3.501624144035559e-07,
|
| 1240 |
+
"loss": 0.1015,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.690807799442897,
|
| 1245 |
+
"grad_norm": 3.284511844581789,
|
| 1246 |
+
"learning_rate": 3.1148178511116624e-07,
|
| 1247 |
+
"loss": 0.0916,
|
| 1248 |
+
"step": 1450
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.690807799442897,
|
| 1252 |
+
"eval_loss": 0.49390342831611633,
|
| 1253 |
+
"eval_runtime": 19.3724,
|
| 1254 |
+
"eval_samples_per_second": 18.17,
|
| 1255 |
+
"eval_steps_per_second": 2.271,
|
| 1256 |
+
"step": 1450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.709377901578459,
|
| 1260 |
+
"grad_norm": 2.9686013973926695,
|
| 1261 |
+
"learning_rate": 2.7499590642665773e-07,
|
| 1262 |
+
"loss": 0.101,
|
| 1263 |
+
"step": 1460
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.7279480037140207,
|
| 1267 |
+
"grad_norm": 2.6344171640489265,
|
| 1268 |
+
"learning_rate": 2.407218578196524e-07,
|
| 1269 |
+
"loss": 0.0899,
|
| 1270 |
+
"step": 1470
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.7465181058495824,
|
| 1274 |
+
"grad_norm": 6.728672130977422,
|
| 1275 |
+
"learning_rate": 2.0867568337605616e-07,
|
| 1276 |
+
"loss": 0.1063,
|
| 1277 |
+
"step": 1480
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.7650882079851438,
|
| 1281 |
+
"grad_norm": 1.8996993963458386,
|
| 1282 |
+
"learning_rate": 1.7887238428763553e-07,
|
| 1283 |
+
"loss": 0.086,
|
| 1284 |
+
"step": 1490
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.7836583101207055,
|
| 1288 |
+
"grad_norm": 2.3753024288893223,
|
| 1289 |
+
"learning_rate": 1.5132591182978107e-07,
|
| 1290 |
+
"loss": 0.0844,
|
| 1291 |
+
"step": 1500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.7836583101207055,
|
| 1295 |
+
"eval_loss": 0.4944659173488617,
|
| 1296 |
+
"eval_runtime": 19.3894,
|
| 1297 |
+
"eval_samples_per_second": 18.154,
|
| 1298 |
+
"eval_steps_per_second": 2.269,
|
| 1299 |
+
"step": 1500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.8022284122562673,
|
| 1303 |
+
"grad_norm": 2.4868387472806917,
|
| 1304 |
+
"learning_rate": 1.2604916083075236e-07,
|
| 1305 |
+
"loss": 0.0936,
|
| 1306 |
+
"step": 1510
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.820798514391829,
|
| 1310 |
+
"grad_norm": 2.5429472050035447,
|
| 1311 |
+
"learning_rate": 1.0305396363545717e-07,
|
| 1312 |
+
"loss": 0.1114,
|
| 1313 |
+
"step": 1520
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.839368616527391,
|
| 1317 |
+
"grad_norm": 2.6782347948140037,
|
| 1318 |
+
"learning_rate": 8.235108456658814e-08,
|
| 1319 |
+
"loss": 0.0933,
|
| 1320 |
+
"step": 1530
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.8579387186629526,
|
| 1324 |
+
"grad_norm": 3.3595802114049533,
|
| 1325 |
+
"learning_rate": 6.395021488572128e-08,
|
| 1326 |
+
"loss": 0.1059,
|
| 1327 |
+
"step": 1540
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.8765088207985143,
|
| 1331 |
+
"grad_norm": 2.5500224216604344,
|
| 1332 |
+
"learning_rate": 4.7859968256719344e-08,
|
| 1333 |
+
"loss": 0.0959,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.8765088207985143,
|
| 1338 |
+
"eval_loss": 0.4942573308944702,
|
| 1339 |
+
"eval_runtime": 19.3998,
|
| 1340 |
+
"eval_samples_per_second": 18.144,
|
| 1341 |
+
"eval_steps_per_second": 2.268,
|
| 1342 |
+
"step": 1550
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.895078922934076,
|
| 1346 |
+
"grad_norm": 2.472877504615022,
|
| 1347 |
+
"learning_rate": 3.408787671357494e-08,
|
| 1348 |
+
"loss": 0.0945,
|
| 1349 |
+
"step": 1560
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.913649025069638,
|
| 1353 |
+
"grad_norm": 2.6027015654240286,
|
| 1354 |
+
"learning_rate": 2.264038713457706e-08,
|
| 1355 |
+
"loss": 0.088,
|
| 1356 |
+
"step": 1570
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.9322191272051996,
|
| 1360 |
+
"grad_norm": 2.9295049029652995,
|
| 1361 |
+
"learning_rate": 1.3522858224450652e-08,
|
| 1362 |
+
"loss": 0.0992,
|
| 1363 |
+
"step": 1580
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.9507892293407614,
|
| 1367 |
+
"grad_norm": 2.7591280720199776,
|
| 1368 |
+
"learning_rate": 6.739558005884883e-09,
|
| 1369 |
+
"loss": 0.0805,
|
| 1370 |
+
"step": 1590
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.969359331476323,
|
| 1374 |
+
"grad_norm": 2.9108217727851398,
|
| 1375 |
+
"learning_rate": 2.2936618216201635e-09,
|
| 1376 |
+
"loss": 0.094,
|
| 1377 |
+
"step": 1600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.969359331476323,
|
| 1381 |
+
"eval_loss": 0.4940944015979767,
|
| 1382 |
+
"eval_runtime": 19.4091,
|
| 1383 |
+
"eval_samples_per_second": 18.136,
|
| 1384 |
+
"eval_steps_per_second": 2.267,
|
| 1385 |
+
"step": 1600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.987929433611885,
|
| 1389 |
+
"grad_norm": 3.2721186890864007,
|
| 1390 |
+
"learning_rate": 1.872508480332824e-10,
|
| 1391 |
+
"loss": 0.0803,
|
| 1392 |
+
"step": 1610
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 2.9953574744661093,
|
| 1396 |
+
"step": 1614,
|
| 1397 |
+
"total_flos": 292826854195200.0,
|
| 1398 |
+
"train_loss": 0.2790517607955389,
|
| 1399 |
+
"train_runtime": 12257.2359,
|
| 1400 |
+
"train_samples_per_second": 4.217,
|
| 1401 |
+
"train_steps_per_second": 0.132
|
| 1402 |
}
|
| 1403 |
],
|
| 1404 |
"logging_steps": 10,
|
| 1405 |
+
"max_steps": 1614,
|
| 1406 |
"num_input_tokens_seen": 0,
|
| 1407 |
+
"num_train_epochs": 3,
|
| 1408 |
"save_steps": 5000,
|
| 1409 |
"stateful_callbacks": {
|
| 1410 |
"TrainerControl": {
|
|
|
|
| 1418 |
"attributes": {}
|
| 1419 |
}
|
| 1420 |
},
|
| 1421 |
+
"total_flos": 292826854195200.0,
|
| 1422 |
"train_batch_size": 2,
|
| 1423 |
"trial_name": null,
|
| 1424 |
"trial_params": null
|
training_eval_loss.png
CHANGED
|
|
training_loss.png
CHANGED
|
|