Model save
Browse files- README.md +4 -8
- all_results.json +6 -11
- generation_config.json +3 -3
- train_results.json +6 -6
- trainer_state.json +427 -329
README.md
CHANGED
|
@@ -3,15 +3,11 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
| 6 |
-
- alignment-handbook
|
| 7 |
-
- trl
|
| 8 |
-
- sft
|
| 9 |
-
- generated_from_trainer
|
| 10 |
- trl
|
| 11 |
- sft
|
| 12 |
- generated_from_trainer
|
| 13 |
datasets:
|
| 14 |
-
-
|
| 15 |
model-index:
|
| 16 |
- name: mistral-7b-ift
|
| 17 |
results: []
|
|
@@ -22,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 22 |
|
| 23 |
# mistral-7b-ift
|
| 24 |
|
| 25 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the
|
| 26 |
It achieves the following results on the evaluation set:
|
| 27 |
-
- Loss:
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
@@ -61,7 +57,7 @@ The following hyperparameters were used during training:
|
|
| 61 |
|
| 62 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 63 |
|:-------------:|:------:|:----:|:---------------:|
|
| 64 |
-
|
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: mistral-7b-ift
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# mistral-7b-ift
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.0699
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 0.9717 | 0.9989 | 455 | 1.0699 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 71.965,
|
| 7 |
-
"eval_steps_per_second": 1.268,
|
| 8 |
-
"total_flos": 161536404357120.0,
|
| 9 |
-
"train_loss": 11.514628536342958,
|
| 10 |
-
"train_runtime": 5043.2067,
|
| 11 |
"train_samples": 77486,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9989023051591658,
|
| 3 |
+
"total_flos": 190430796840960.0,
|
| 4 |
+
"train_loss": 1.0385934979051024,
|
| 5 |
+
"train_runtime": 5380.501,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 77486,
|
| 7 |
+
"train_samples_per_second": 21.667,
|
| 8 |
+
"train_steps_per_second": 0.085
|
| 9 |
}
|
generation_config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
-
"bos_token_id":
|
| 4 |
-
"eos_token_id":
|
| 5 |
-
"pad_token_id":
|
| 6 |
"transformers_version": "4.44.1"
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 32768,
|
| 4 |
+
"eos_token_id": 32769,
|
| 5 |
+
"pad_token_id": 32769,
|
| 6 |
"transformers_version": "4.44.1"
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 77486,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9989023051591658,
|
| 3 |
+
"total_flos": 190430796840960.0,
|
| 4 |
+
"train_loss": 1.0385934979051024,
|
| 5 |
+
"train_runtime": 5380.501,
|
| 6 |
"train_samples": 77486,
|
| 7 |
+
"train_samples_per_second": 21.667,
|
| 8 |
+
"train_steps_per_second": 0.085
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,579 +1,677 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 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": 1.
|
| 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": 1.
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 1.
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 3.
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 4.
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 4.
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 4.
|
| 85 |
-
"loss":
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss":
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 4.
|
| 99 |
-
"loss":
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 4.
|
| 106 |
-
"loss":
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 4.
|
| 113 |
-
"loss":
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 4.
|
| 120 |
-
"loss":
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 4.
|
| 127 |
-
"loss":
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss":
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 4.
|
| 141 |
-
"loss":
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 4.
|
| 148 |
-
"loss":
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate": 4.
|
| 162 |
-
"loss":
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 4.
|
| 169 |
-
"loss":
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate": 4.
|
| 176 |
-
"loss":
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate": 4.
|
| 183 |
-
"loss":
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate": 4.
|
| 190 |
-
"loss":
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 4.
|
| 197 |
-
"loss":
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
-
"learning_rate": 4.
|
| 204 |
-
"loss":
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate": 4.
|
| 211 |
-
"loss":
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss":
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss":
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss":
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss":
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
-
"learning_rate": 3.
|
| 253 |
-
"loss":
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate": 3.
|
| 260 |
-
"loss":
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate": 3.
|
| 267 |
-
"loss":
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
-
"learning_rate": 3.
|
| 274 |
-
"loss":
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
-
"learning_rate": 3.
|
| 281 |
-
"loss":
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
-
"learning_rate":
|
| 288 |
-
"loss":
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss":
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss":
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm":
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"loss":
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm":
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss":
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss":
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate": 2.
|
| 330 |
-
"loss":
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate": 2.
|
| 337 |
-
"loss":
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss":
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss":
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss":
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm":
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss":
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm":
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss":
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm":
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss":
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm":
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss":
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm":
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss":
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm":
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss":
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm":
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss":
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm":
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss":
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss":
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm":
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss":
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss":
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss":
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss":
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss":
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss":
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss":
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss":
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss":
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm":
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss":
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm":
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss":
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss":
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm":
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss":
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm":
|
| 518 |
-
"learning_rate": 6.
|
| 519 |
-
"loss":
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm":
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss":
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm":
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss":
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss":
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm":
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss":
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm":
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss":
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"
|
| 560 |
-
"
|
| 561 |
-
"
|
| 562 |
-
"
|
| 563 |
-
"step": 386
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch": 0.
|
| 567 |
-
"
|
| 568 |
-
"
|
| 569 |
-
"
|
| 570 |
-
"
|
| 571 |
-
|
| 572 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 573 |
}
|
| 574 |
],
|
| 575 |
"logging_steps": 5,
|
| 576 |
-
"max_steps":
|
| 577 |
"num_input_tokens_seen": 0,
|
| 578 |
"num_train_epochs": 1,
|
| 579 |
"save_steps": 100,
|
|
@@ -589,7 +687,7 @@
|
|
| 589 |
"attributes": {}
|
| 590 |
}
|
| 591 |
},
|
| 592 |
-
"total_flos":
|
| 593 |
"train_batch_size": 8,
|
| 594 |
"trial_name": null,
|
| 595 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9989023051591658,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 455,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0021953896816684962,
|
| 13 |
+
"grad_norm": 37.23542178718614,
|
| 14 |
+
"learning_rate": 1.0869565217391303e-08,
|
| 15 |
+
"loss": 1.4277,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.010976948408342482,
|
| 20 |
+
"grad_norm": 31.768758474830744,
|
| 21 |
+
"learning_rate": 5.434782608695652e-08,
|
| 22 |
+
"loss": 1.4875,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.021953896816684963,
|
| 27 |
+
"grad_norm": 37.21672974282255,
|
| 28 |
+
"learning_rate": 1.0869565217391303e-07,
|
| 29 |
+
"loss": 1.4405,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03293084522502744,
|
| 34 |
+
"grad_norm": 28.728369641123845,
|
| 35 |
+
"learning_rate": 1.6304347826086955e-07,
|
| 36 |
+
"loss": 1.4495,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.043907793633369926,
|
| 41 |
+
"grad_norm": 24.14103448999392,
|
| 42 |
+
"learning_rate": 2.1739130434782607e-07,
|
| 43 |
+
"loss": 1.348,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.054884742041712405,
|
| 48 |
+
"grad_norm": 17.84361561407524,
|
| 49 |
+
"learning_rate": 2.717391304347826e-07,
|
| 50 |
+
"loss": 1.2803,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.06586169045005488,
|
| 55 |
+
"grad_norm": 8.32310884222812,
|
| 56 |
+
"learning_rate": 3.260869565217391e-07,
|
| 57 |
+
"loss": 1.2081,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.07683863885839737,
|
| 62 |
+
"grad_norm": 4.7849107325475435,
|
| 63 |
+
"learning_rate": 3.8043478260869567e-07,
|
| 64 |
+
"loss": 1.1701,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.08781558726673985,
|
| 69 |
+
"grad_norm": 4.142884139685247,
|
| 70 |
+
"learning_rate": 4.3478260869565214e-07,
|
| 71 |
+
"loss": 1.1529,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.09879253567508232,
|
| 76 |
+
"grad_norm": 4.1545233643304105,
|
| 77 |
+
"learning_rate": 4.891304347826087e-07,
|
| 78 |
+
"loss": 1.1303,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.10976948408342481,
|
| 83 |
+
"grad_norm": 3.10192089815295,
|
| 84 |
+
"learning_rate": 4.998820089787287e-07,
|
| 85 |
+
"loss": 1.1088,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.1207464324917673,
|
| 90 |
+
"grad_norm": 3.1563385745763677,
|
| 91 |
+
"learning_rate": 4.99402861330474e-07,
|
| 92 |
+
"loss": 1.1042,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.13172338090010977,
|
| 97 |
+
"grad_norm": 2.5435814181022827,
|
| 98 |
+
"learning_rate": 4.985558887302488e-07,
|
| 99 |
+
"loss": 1.0956,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.14270032930845225,
|
| 104 |
+
"grad_norm": 2.640027183329368,
|
| 105 |
+
"learning_rate": 4.973423403123028e-07,
|
| 106 |
+
"loss": 1.079,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.15367727771679474,
|
| 111 |
+
"grad_norm": 3.037284920443539,
|
| 112 |
+
"learning_rate": 4.957640058451501e-07,
|
| 113 |
+
"loss": 1.0675,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.16465422612513722,
|
| 118 |
+
"grad_norm": 3.1427488556058223,
|
| 119 |
+
"learning_rate": 4.938232130919785e-07,
|
| 120 |
+
"loss": 1.0734,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.1756311745334797,
|
| 125 |
+
"grad_norm": 2.2891126109812543,
|
| 126 |
+
"learning_rate": 4.915228243776124e-07,
|
| 127 |
+
"loss": 1.0663,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.18660812294182216,
|
| 132 |
+
"grad_norm": 3.1770515308884497,
|
| 133 |
+
"learning_rate": 4.888662323670913e-07,
|
| 134 |
+
"loss": 1.0556,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.19758507135016465,
|
| 139 |
+
"grad_norm": 2.497553823382758,
|
| 140 |
+
"learning_rate": 4.858573550620908e-07,
|
| 141 |
+
"loss": 1.067,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.20856201975850713,
|
| 146 |
+
"grad_norm": 2.2786389579554225,
|
| 147 |
+
"learning_rate": 4.825006300225665e-07,
|
| 148 |
+
"loss": 1.0713,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.21953896816684962,
|
| 153 |
+
"grad_norm": 2.4565400673449314,
|
| 154 |
+
"learning_rate": 4.7880100782214e-07,
|
| 155 |
+
"loss": 1.0594,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.2305159165751921,
|
| 160 |
+
"grad_norm": 2.2057241212988274,
|
| 161 |
+
"learning_rate": 4.747639447468816e-07,
|
| 162 |
+
"loss": 1.0622,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.2414928649835346,
|
| 167 |
+
"grad_norm": 2.699908595698097,
|
| 168 |
+
"learning_rate": 4.7039539474825683e-07,
|
| 169 |
+
"loss": 1.0263,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.2524698133918771,
|
| 174 |
+
"grad_norm": 2.595572602014253,
|
| 175 |
+
"learning_rate": 4.657018006621053e-07,
|
| 176 |
+
"loss": 1.0289,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.26344676180021953,
|
| 181 |
+
"grad_norm": 2.367781794798575,
|
| 182 |
+
"learning_rate": 4.6069008470660057e-07,
|
| 183 |
+
"loss": 1.0671,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.27442371020856204,
|
| 188 |
+
"grad_norm": 2.191786685372748,
|
| 189 |
+
"learning_rate": 4.5536763827320803e-07,
|
| 190 |
+
"loss": 1.0331,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.2854006586169045,
|
| 195 |
+
"grad_norm": 2.2558364905064097,
|
| 196 |
+
"learning_rate": 4.4974231102569355e-07,
|
| 197 |
+
"loss": 1.0247,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.29637760702524696,
|
| 202 |
+
"grad_norm": 2.2396909107202134,
|
| 203 |
+
"learning_rate": 4.438223993232634e-07,
|
| 204 |
+
"loss": 1.0097,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.30735455543358947,
|
| 209 |
+
"grad_norm": 2.2953815342987007,
|
| 210 |
+
"learning_rate": 4.3761663398490634e-07,
|
| 211 |
+
"loss": 1.0162,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.31833150384193193,
|
| 216 |
+
"grad_norm": 2.469492662807899,
|
| 217 |
+
"learning_rate": 4.3113416741298616e-07,
|
| 218 |
+
"loss": 1.0233,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.32930845225027444,
|
| 223 |
+
"grad_norm": 2.3585547924597052,
|
| 224 |
+
"learning_rate": 4.2438456009507195e-07,
|
| 225 |
+
"loss": 1.0095,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.3402854006586169,
|
| 230 |
+
"grad_norm": 2.3421663442140837,
|
| 231 |
+
"learning_rate": 4.1737776650391625e-07,
|
| 232 |
+
"loss": 1.0187,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.3512623490669594,
|
| 237 |
+
"grad_norm": 2.5398466939682733,
|
| 238 |
+
"learning_rate": 4.101241204163748e-07,
|
| 239 |
+
"loss": 0.9977,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.36223929747530187,
|
| 244 |
+
"grad_norm": 2.2242476659757275,
|
| 245 |
+
"learning_rate": 4.0263431967291934e-07,
|
| 246 |
+
"loss": 1.027,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.3732162458836443,
|
| 251 |
+
"grad_norm": 2.1753753105769227,
|
| 252 |
+
"learning_rate": 3.949194104002224e-07,
|
| 253 |
+
"loss": 1.008,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.38419319429198684,
|
| 258 |
+
"grad_norm": 2.300275831812692,
|
| 259 |
+
"learning_rate": 3.8699077072008085e-07,
|
| 260 |
+
"loss": 1.0106,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.3951701427003293,
|
| 265 |
+
"grad_norm": 2.369231981384555,
|
| 266 |
+
"learning_rate": 3.7886009396870564e-07,
|
| 267 |
+
"loss": 1.0034,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.4061470911086718,
|
| 272 |
+
"grad_norm": 3.342159958597614,
|
| 273 |
+
"learning_rate": 3.705393714511268e-07,
|
| 274 |
+
"loss": 1.0008,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.41712403951701427,
|
| 279 |
+
"grad_norm": 2.2742818660752344,
|
| 280 |
+
"learning_rate": 3.6204087475614676e-07,
|
| 281 |
+
"loss": 1.0163,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.4281009879253567,
|
| 286 |
+
"grad_norm": 2.224871070156181,
|
| 287 |
+
"learning_rate": 3.533771376579249e-07,
|
| 288 |
+
"loss": 1.0116,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.43907793633369924,
|
| 293 |
+
"grad_norm": 2.201217019563197,
|
| 294 |
+
"learning_rate": 3.445609376308857e-07,
|
| 295 |
+
"loss": 0.9992,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.4500548847420417,
|
| 300 |
+
"grad_norm": 2.244052004363795,
|
| 301 |
+
"learning_rate": 3.356052770052119e-07,
|
| 302 |
+
"loss": 1.0002,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.4610318331503842,
|
| 307 |
+
"grad_norm": 2.1869532116505437,
|
| 308 |
+
"learning_rate": 3.2652336379071506e-07,
|
| 309 |
+
"loss": 1.003,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.47200878155872666,
|
| 314 |
+
"grad_norm": 2.25200716108438,
|
| 315 |
+
"learning_rate": 3.173285921973657e-07,
|
| 316 |
+
"loss": 0.9967,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.4829857299670692,
|
| 321 |
+
"grad_norm": 2.1977221613580915,
|
| 322 |
+
"learning_rate": 3.0803452288121113e-07,
|
| 323 |
+
"loss": 0.9933,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.49396267837541163,
|
| 328 |
+
"grad_norm": 2.1857074442694064,
|
| 329 |
+
"learning_rate": 2.986548629448146e-07,
|
| 330 |
+
"loss": 1.0118,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.5049396267837541,
|
| 335 |
+
"grad_norm": 2.176224528870358,
|
| 336 |
+
"learning_rate": 2.892034457217119e-07,
|
| 337 |
+
"loss": 0.9803,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.5159165751920965,
|
| 342 |
+
"grad_norm": 2.1323657558564895,
|
| 343 |
+
"learning_rate": 2.7969421037470033e-07,
|
| 344 |
+
"loss": 1.011,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.5268935236004391,
|
| 349 |
+
"grad_norm": 2.210400681860978,
|
| 350 |
+
"learning_rate": 2.70141181338047e-07,
|
| 351 |
+
"loss": 0.9835,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.5378704720087816,
|
| 356 |
+
"grad_norm": 2.294325436997366,
|
| 357 |
+
"learning_rate": 2.60558447633938e-07,
|
| 358 |
+
"loss": 0.9994,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.5488474204171241,
|
| 363 |
+
"grad_norm": 2.269474512577858,
|
| 364 |
+
"learning_rate": 2.509601420936717e-07,
|
| 365 |
+
"loss": 1.0056,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.5598243688254665,
|
| 370 |
+
"grad_norm": 2.350884633568068,
|
| 371 |
+
"learning_rate": 2.413604205142404e-07,
|
| 372 |
+
"loss": 0.9975,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.570801317233809,
|
| 377 |
+
"grad_norm": 2.2618918862021387,
|
| 378 |
+
"learning_rate": 2.3177344078104305e-07,
|
| 379 |
+
"loss": 1.0226,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.5817782656421515,
|
| 384 |
+
"grad_norm": 2.335669763682869,
|
| 385 |
+
"learning_rate": 2.2221334198751717e-07,
|
| 386 |
+
"loss": 0.9996,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.5927552140504939,
|
| 391 |
+
"grad_norm": 2.218697106476552,
|
| 392 |
+
"learning_rate": 2.1269422358248534e-07,
|
| 393 |
+
"loss": 0.9764,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.6037321624588364,
|
| 398 |
+
"grad_norm": 2.294055527983993,
|
| 399 |
+
"learning_rate": 2.0323012457597113e-07,
|
| 400 |
+
"loss": 0.9911,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.6147091108671789,
|
| 405 |
+
"grad_norm": 2.131671305300942,
|
| 406 |
+
"learning_rate": 1.9383500283415127e-07,
|
| 407 |
+
"loss": 1.0084,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.6256860592755215,
|
| 412 |
+
"grad_norm": 2.170441232463747,
|
| 413 |
+
"learning_rate": 1.8452271449398015e-07,
|
| 414 |
+
"loss": 0.9935,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.6366630076838639,
|
| 419 |
+
"grad_norm": 2.2800412432483994,
|
| 420 |
+
"learning_rate": 1.753069935278477e-07,
|
| 421 |
+
"loss": 0.9981,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.6476399560922064,
|
| 426 |
+
"grad_norm": 2.2112002252829233,
|
| 427 |
+
"learning_rate": 1.662014314884074e-07,
|
| 428 |
+
"loss": 0.9821,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.6586169045005489,
|
| 433 |
+
"grad_norm": 2.1523725296369696,
|
| 434 |
+
"learning_rate": 1.5721945746344914e-07,
|
| 435 |
+
"loss": 1.0028,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.6695938529088913,
|
| 440 |
+
"grad_norm": 2.2685448996190476,
|
| 441 |
+
"learning_rate": 1.4837431827037786e-07,
|
| 442 |
+
"loss": 0.9843,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.6805708013172338,
|
| 447 |
+
"grad_norm": 2.222219631378057,
|
| 448 |
+
"learning_rate": 1.3967905891950936e-07,
|
| 449 |
+
"loss": 0.9892,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.6915477497255763,
|
| 454 |
+
"grad_norm": 2.17757738873566,
|
| 455 |
+
"learning_rate": 1.3114650337499578e-07,
|
| 456 |
+
"loss": 0.993,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.7025246981339188,
|
| 461 |
+
"grad_norm": 2.202207960410075,
|
| 462 |
+
"learning_rate": 1.227892356417542e-07,
|
| 463 |
+
"loss": 0.98,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.7135016465422612,
|
| 468 |
+
"grad_norm": 2.2364238502729887,
|
| 469 |
+
"learning_rate": 1.1461958120629345e-07,
|
| 470 |
+
"loss": 0.9747,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.7244785949506037,
|
| 475 |
+
"grad_norm": 2.2007426350936314,
|
| 476 |
+
"learning_rate": 1.0664958885880901e-07,
|
| 477 |
+
"loss": 0.9993,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.7354555433589463,
|
| 482 |
+
"grad_norm": 2.4965113706058433,
|
| 483 |
+
"learning_rate": 9.889101292335625e-08,
|
| 484 |
+
"loss": 0.9881,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.7464324917672887,
|
| 489 |
+
"grad_norm": 2.1808970559228955,
|
| 490 |
+
"learning_rate": 9.13552959223089e-08,
|
| 491 |
+
"loss": 0.9961,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.7574094401756312,
|
| 496 |
+
"grad_norm": 2.1878061482574336,
|
| 497 |
+
"learning_rate": 8.405355170066925e-08,
|
| 498 |
+
"loss": 0.9943,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.7683863885839737,
|
| 503 |
+
"grad_norm": 2.1845807299852034,
|
| 504 |
+
"learning_rate": 7.699654903511971e-08,
|
| 505 |
+
"loss": 0.9866,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.7793633369923162,
|
| 510 |
+
"grad_norm": 2.1414775634398375,
|
| 511 |
+
"learning_rate": 7.01946957519886e-08,
|
| 512 |
+
"loss": 0.9759,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.7903402854006586,
|
| 517 |
+
"grad_norm": 2.122717025599496,
|
| 518 |
+
"learning_rate": 6.365802337755364e-08,
|
| 519 |
+
"loss": 1.0019,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.8013172338090011,
|
| 524 |
+
"grad_norm": 2.14871365458936,
|
| 525 |
+
"learning_rate": 5.739617234332131e-08,
|
| 526 |
+
"loss": 0.9961,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.8122941822173436,
|
| 531 |
+
"grad_norm": 2.1563921789431233,
|
| 532 |
+
"learning_rate": 5.14183777681014e-08,
|
| 533 |
+
"loss": 0.979,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.823271130625686,
|
| 538 |
+
"grad_norm": 2.2409621573734566,
|
| 539 |
+
"learning_rate": 4.5733455837846325e-08,
|
| 540 |
+
"loss": 1.0026,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.8342480790340285,
|
| 545 |
+
"grad_norm": 2.2773154227725296,
|
| 546 |
+
"learning_rate": 4.0349790803341274e-08,
|
| 547 |
+
"loss": 1.0058,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.845225027442371,
|
| 552 |
+
"grad_norm": 2.117720539958819,
|
| 553 |
+
"learning_rate": 3.527532261492272e-08,
|
| 554 |
+
"loss": 0.9875,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.8562019758507134,
|
| 559 |
+
"grad_norm": 2.152957888451534,
|
| 560 |
+
"learning_rate": 3.0517535212460946e-08,
|
| 561 |
+
"loss": 0.9784,
|
| 562 |
+
"step": 390
|
|
|
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.867178924259056,
|
| 566 |
+
"grad_norm": 2.1225713025355692,
|
| 567 |
+
"learning_rate": 2.608344548787722e-08,
|
| 568 |
+
"loss": 0.9713,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8781558726673985,
|
| 573 |
+
"grad_norm": 2.1615502933761643,
|
| 574 |
+
"learning_rate": 2.19795929364735e-08,
|
| 575 |
+
"loss": 0.9895,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.889132821075741,
|
| 580 |
+
"grad_norm": 2.167176721017015,
|
| 581 |
+
"learning_rate": 1.8212030012337704e-08,
|
| 582 |
+
"loss": 0.9982,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.9001097694840834,
|
| 587 |
+
"grad_norm": 2.1328147082553994,
|
| 588 |
+
"learning_rate": 1.4786313202048456e-08,
|
| 589 |
+
"loss": 1.0076,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9110867178924259,
|
| 594 |
+
"grad_norm": 2.249033557726424,
|
| 595 |
+
"learning_rate": 1.1707494829843207e-08,
|
| 596 |
+
"loss": 0.9855,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.9220636663007684,
|
| 601 |
+
"grad_norm": 2.8747995493229372,
|
| 602 |
+
"learning_rate": 8.980115606337046e-09,
|
| 603 |
+
"loss": 0.9957,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.9330406147091108,
|
| 608 |
+
"grad_norm": 2.124614068925926,
|
| 609 |
+
"learning_rate": 6.608197931780496e-09,
|
| 610 |
+
"loss": 0.9769,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9440175631174533,
|
| 615 |
+
"grad_norm": 2.176873597616659,
|
| 616 |
+
"learning_rate": 4.595239963733011e-09,
|
| 617 |
+
"loss": 0.9853,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9549945115257958,
|
| 622 |
+
"grad_norm": 2.1342884606915202,
|
| 623 |
+
"learning_rate": 2.9442104579016356e-09,
|
| 624 |
+
"loss": 0.9917,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9659714599341384,
|
| 629 |
+
"grad_norm": 2.1374930773738487,
|
| 630 |
+
"learning_rate": 1.6575443897531294e-09,
|
| 631 |
+
"loss": 0.9909,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9769484083424808,
|
| 636 |
+
"grad_norm": 2.1416862097657865,
|
| 637 |
+
"learning_rate": 7.371393633574252e-10,
|
| 638 |
+
"loss": 0.9845,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9879253567508233,
|
| 643 |
+
"grad_norm": 2.156534272117916,
|
| 644 |
+
"learning_rate": 1.843528127584981e-10,
|
| 645 |
+
"loss": 0.9959,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9989023051591658,
|
| 650 |
+
"grad_norm": 2.1792993284280864,
|
| 651 |
+
"learning_rate": 0.0,
|
| 652 |
+
"loss": 0.9717,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9989023051591658,
|
| 657 |
+
"eval_loss": 1.0698977708816528,
|
| 658 |
+
"eval_runtime": 3.5449,
|
| 659 |
+
"eval_samples_per_second": 75.601,
|
| 660 |
+
"eval_steps_per_second": 1.41,
|
| 661 |
+
"step": 455
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9989023051591658,
|
| 665 |
+
"step": 455,
|
| 666 |
+
"total_flos": 190430796840960.0,
|
| 667 |
+
"train_loss": 1.0385934979051024,
|
| 668 |
+
"train_runtime": 5380.501,
|
| 669 |
+
"train_samples_per_second": 21.667,
|
| 670 |
+
"train_steps_per_second": 0.085
|
| 671 |
}
|
| 672 |
],
|
| 673 |
"logging_steps": 5,
|
| 674 |
+
"max_steps": 455,
|
| 675 |
"num_input_tokens_seen": 0,
|
| 676 |
"num_train_epochs": 1,
|
| 677 |
"save_steps": 100,
|
|
|
|
| 687 |
"attributes": {}
|
| 688 |
}
|
| 689 |
},
|
| 690 |
+
"total_flos": 190430796840960.0,
|
| 691 |
"train_batch_size": 8,
|
| 692 |
"trial_name": null,
|
| 693 |
"trial_params": null
|