Model save
Browse files- README.md +6 -7
- all_results.json +6 -11
- train_results.json +6 -6
- trainer_state.json +407 -344
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: llama3
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: meta-llama/Meta-Llama-3-8B
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: downstream_0.1p_seed42_level2_syntax
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# downstream_0.1p_seed42_level2_syntax
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 1.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -56,9 +55,9 @@ The following hyperparameters were used during training:
|
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
-
| Training Loss | Epoch
|
| 60 |
-
|:-------------:|:-----
|
| 61 |
-
| 1.
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
|
|
|
| 2 |
license: llama3
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: meta-llama/Meta-Llama-3-8B
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: downstream_0.1p_seed42_level2_syntax
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# downstream_0.1p_seed42_level2_syntax
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.0367
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 1.094 | 1.0 | 454 | 1.0367 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"total_flos": 1.2948113606049792e+16,
|
| 9 |
-
"train_loss": 1.1371603935372596,
|
| 10 |
-
"train_runtime": 15464.5644,
|
| 11 |
-
"train_samples": 90000,
|
| 12 |
-
"train_samples_per_second": 3.378,
|
| 13 |
"train_steps_per_second": 0.026
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.4408836278386688e+16,
|
| 4 |
+
"train_loss": 1.131237539688396,
|
| 5 |
+
"train_runtime": 17315.3957,
|
| 6 |
+
"train_samples": 100000,
|
| 7 |
+
"train_samples_per_second": 3.356,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
"train_steps_per_second": 0.026
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 3.
|
| 8 |
"train_steps_per_second": 0.026
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.4408836278386688e+16,
|
| 4 |
+
"train_loss": 1.131237539688396,
|
| 5 |
+
"train_runtime": 17315.3957,
|
| 6 |
+
"train_samples": 100000,
|
| 7 |
+
"train_samples_per_second": 3.356,
|
| 8 |
"train_steps_per_second": 0.026
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,607 +1,670 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
-
"learning_rate": 4.
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
-
"learning_rate": 2.
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate": 4.
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate": 0.
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 1.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 0.
|
| 162 |
-
"loss": 1.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 1.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss": 1.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 0.
|
| 190 |
-
"loss": 1.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate": 0.
|
| 197 |
-
"loss": 1.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate": 0.
|
| 204 |
-
"loss": 1.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
-
"learning_rate": 0.
|
| 211 |
-
"loss": 1.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate": 0.
|
| 218 |
-
"loss": 1.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate": 0.
|
| 225 |
-
"loss": 1.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate": 0.
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss": 1.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
-
"learning_rate": 0.
|
| 246 |
-
"loss": 1.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate": 0.
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
-
"learning_rate": 0.
|
| 260 |
-
"loss": 1.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
-
"learning_rate": 0.
|
| 267 |
-
"loss": 1.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate": 0.
|
| 274 |
-
"loss": 1.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate": 0.
|
| 281 |
-
"loss": 1.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate": 0.
|
| 288 |
-
"loss": 1.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate": 0.
|
| 295 |
-
"loss": 1.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
-
"learning_rate": 0.
|
| 302 |
-
"loss": 1.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
-
"learning_rate": 0.
|
| 309 |
-
"loss": 1.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate": 0.
|
| 316 |
-
"loss": 1.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
-
"learning_rate": 0.
|
| 323 |
-
"loss": 1.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 1.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 1.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss": 1.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 1.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss": 1.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss": 1.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss": 1.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss": 1.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss": 1.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 1.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate":
|
| 407 |
-
"loss": 1.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss": 1.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss": 1.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss": 1.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss": 1.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 1.
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 1.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss": 1.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 1.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss": 1.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss": 1.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 1.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 1.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss": 1.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 1.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm": 0.
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss": 1.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss": 1.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss": 1.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 1.
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm": 0.
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss": 1.
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 1.
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
-
"learning_rate": 1.
|
| 554 |
-
"loss": 1.
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm": 0.
|
| 560 |
-
"learning_rate": 1.
|
| 561 |
-
"loss": 1.
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 1.
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 1.
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm": 0.
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 1.
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"
|
| 588 |
-
"
|
| 589 |
-
"
|
| 590 |
-
"
|
| 591 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"step":
|
| 596 |
-
"total_flos": 1.
|
| 597 |
-
"train_loss": 1.
|
| 598 |
-
"train_runtime":
|
| 599 |
-
"train_samples_per_second": 3.
|
| 600 |
"train_steps_per_second": 0.026
|
| 601 |
}
|
| 602 |
],
|
| 603 |
"logging_steps": 5,
|
| 604 |
-
"max_steps":
|
| 605 |
"num_input_tokens_seen": 0,
|
| 606 |
"num_train_epochs": 1,
|
| 607 |
"save_steps": 100,
|
|
@@ -617,7 +680,7 @@
|
|
| 617 |
"attributes": {}
|
| 618 |
}
|
| 619 |
},
|
| 620 |
-
"total_flos": 1.
|
| 621 |
"train_batch_size": 8,
|
| 622 |
"trial_name": null,
|
| 623 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 454,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0022026431718061676,
|
| 13 |
+
"grad_norm": 1.0637864900736664,
|
| 14 |
+
"learning_rate": 4.347826086956522e-06,
|
| 15 |
+
"loss": 1.4366,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.011013215859030838,
|
| 20 |
+
"grad_norm": 1.153707883328667,
|
| 21 |
+
"learning_rate": 2.173913043478261e-05,
|
| 22 |
+
"loss": 1.4048,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.022026431718061675,
|
| 27 |
+
"grad_norm": 0.24033826612761172,
|
| 28 |
+
"learning_rate": 4.347826086956522e-05,
|
| 29 |
+
"loss": 1.3638,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03303964757709251,
|
| 34 |
+
"grad_norm": 0.23098755386890235,
|
| 35 |
+
"learning_rate": 6.521739130434783e-05,
|
| 36 |
+
"loss": 1.3035,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.04405286343612335,
|
| 41 |
+
"grad_norm": 0.18769938367180908,
|
| 42 |
+
"learning_rate": 8.695652173913044e-05,
|
| 43 |
+
"loss": 1.2767,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.05506607929515418,
|
| 48 |
+
"grad_norm": 0.14239545432122608,
|
| 49 |
+
"learning_rate": 0.00010869565217391305,
|
| 50 |
+
"loss": 1.2196,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.06607929515418502,
|
| 55 |
+
"grad_norm": 0.10372099126300886,
|
| 56 |
+
"learning_rate": 0.00013043478260869567,
|
| 57 |
+
"loss": 1.21,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.07709251101321586,
|
| 62 |
+
"grad_norm": 0.10109724419729917,
|
| 63 |
+
"learning_rate": 0.00015217391304347827,
|
| 64 |
+
"loss": 1.2161,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.0881057268722467,
|
| 69 |
+
"grad_norm": 0.08677348076375273,
|
| 70 |
+
"learning_rate": 0.00017391304347826088,
|
| 71 |
+
"loss": 1.2002,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.09911894273127753,
|
| 76 |
+
"grad_norm": 0.07988238174290498,
|
| 77 |
+
"learning_rate": 0.0001956521739130435,
|
| 78 |
+
"loss": 1.1907,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.11013215859030837,
|
| 83 |
+
"grad_norm": 0.08057978560906214,
|
| 84 |
+
"learning_rate": 0.0001999525719713366,
|
| 85 |
+
"loss": 1.1572,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.1211453744493392,
|
| 90 |
+
"grad_norm": 0.08125586602301926,
|
| 91 |
+
"learning_rate": 0.0001997599727063717,
|
| 92 |
+
"loss": 1.1575,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.13215859030837004,
|
| 97 |
+
"grad_norm": 0.08174265113795184,
|
| 98 |
+
"learning_rate": 0.00019941952317728147,
|
| 99 |
+
"loss": 1.1662,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.14317180616740088,
|
| 104 |
+
"grad_norm": 0.09240936715995939,
|
| 105 |
+
"learning_rate": 0.00019893172795069144,
|
| 106 |
+
"loss": 1.1561,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.15418502202643172,
|
| 111 |
+
"grad_norm": 0.07863084458052047,
|
| 112 |
+
"learning_rate": 0.0001982973099683902,
|
| 113 |
+
"loss": 1.137,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.16519823788546256,
|
| 118 |
+
"grad_norm": 0.08035585420548778,
|
| 119 |
+
"learning_rate": 0.00019751720947588602,
|
| 120 |
+
"loss": 1.1417,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.1762114537444934,
|
| 125 |
+
"grad_norm": 0.07156903107546937,
|
| 126 |
+
"learning_rate": 0.00019659258262890683,
|
| 127 |
+
"loss": 1.156,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.18722466960352424,
|
| 132 |
+
"grad_norm": 0.07056162758348125,
|
| 133 |
+
"learning_rate": 0.000195524799779908,
|
| 134 |
+
"loss": 1.135,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.19823788546255505,
|
| 139 |
+
"grad_norm": 0.07468007657011916,
|
| 140 |
+
"learning_rate": 0.00019431544344712776,
|
| 141 |
+
"loss": 1.1285,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.2092511013215859,
|
| 146 |
+
"grad_norm": 0.07250524328449914,
|
| 147 |
+
"learning_rate": 0.00019296630596920023,
|
| 148 |
+
"loss": 1.1335,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.22026431718061673,
|
| 153 |
+
"grad_norm": 0.06955764714773345,
|
| 154 |
+
"learning_rate": 0.0001914793868488021,
|
| 155 |
+
"loss": 1.1512,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.23127753303964757,
|
| 160 |
+
"grad_norm": 0.07221286385490396,
|
| 161 |
+
"learning_rate": 0.0001898568897892697,
|
| 162 |
+
"loss": 1.1306,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.2422907488986784,
|
| 167 |
+
"grad_norm": 0.07486140156540808,
|
| 168 |
+
"learning_rate": 0.00018810121942857845,
|
| 169 |
+
"loss": 1.1342,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.2533039647577093,
|
| 174 |
+
"grad_norm": 0.07131722083122453,
|
| 175 |
+
"learning_rate": 0.00018621497777552507,
|
| 176 |
+
"loss": 1.1307,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.2643171806167401,
|
| 181 |
+
"grad_norm": 0.07348336925535924,
|
| 182 |
+
"learning_rate": 0.00018420096035339452,
|
| 183 |
+
"loss": 1.1222,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.2753303964757709,
|
| 188 |
+
"grad_norm": 0.07501908796003587,
|
| 189 |
+
"learning_rate": 0.00018206215205682683,
|
| 190 |
+
"loss": 1.1116,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.28634361233480177,
|
| 195 |
+
"grad_norm": 0.07330385148735,
|
| 196 |
+
"learning_rate": 0.000179801722728024,
|
| 197 |
+
"loss": 1.1373,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.2973568281938326,
|
| 202 |
+
"grad_norm": 0.073420013451975,
|
| 203 |
+
"learning_rate": 0.00017742302245885383,
|
| 204 |
+
"loss": 1.1053,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.30837004405286345,
|
| 209 |
+
"grad_norm": 0.07954193429333527,
|
| 210 |
+
"learning_rate": 0.00017492957662581295,
|
| 211 |
+
"loss": 1.1232,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.31938325991189426,
|
| 216 |
+
"grad_norm": 0.07217417992157714,
|
| 217 |
+
"learning_rate": 0.00017232508066520702,
|
| 218 |
+
"loss": 1.1237,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.3303964757709251,
|
| 223 |
+
"grad_norm": 0.07802555489353481,
|
| 224 |
+
"learning_rate": 0.0001696133945962927,
|
| 225 |
+
"loss": 1.1208,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.34140969162995594,
|
| 230 |
+
"grad_norm": 0.08970080549205696,
|
| 231 |
+
"learning_rate": 0.00016679853730049743,
|
| 232 |
+
"loss": 1.1404,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.3524229074889868,
|
| 237 |
+
"grad_norm": 0.07726401697227049,
|
| 238 |
+
"learning_rate": 0.00016388468056519612,
|
| 239 |
+
"loss": 1.0981,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.3634361233480176,
|
| 244 |
+
"grad_norm": 0.0738469257596569,
|
| 245 |
+
"learning_rate": 0.00016087614290087208,
|
| 246 |
+
"loss": 1.1245,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.3744493392070485,
|
| 251 |
+
"grad_norm": 0.0760961629898493,
|
| 252 |
+
"learning_rate": 0.00015777738314082514,
|
| 253 |
+
"loss": 1.1282,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.3854625550660793,
|
| 258 |
+
"grad_norm": 0.07461575712201869,
|
| 259 |
+
"learning_rate": 0.00015459299383291345,
|
| 260 |
+
"loss": 1.1206,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.3964757709251101,
|
| 265 |
+
"grad_norm": 0.07597642898156688,
|
| 266 |
+
"learning_rate": 0.00015132769443312207,
|
| 267 |
+
"loss": 1.1151,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.40748898678414097,
|
| 272 |
+
"grad_norm": 0.07220979125832422,
|
| 273 |
+
"learning_rate": 0.00014798632431104592,
|
| 274 |
+
"loss": 1.1313,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.4185022026431718,
|
| 279 |
+
"grad_norm": 0.07235062387347811,
|
| 280 |
+
"learning_rate": 0.00014457383557765386,
|
| 281 |
+
"loss": 1.126,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.42951541850220265,
|
| 286 |
+
"grad_norm": 0.07249541910324554,
|
| 287 |
+
"learning_rate": 0.00014109528574596301,
|
| 288 |
+
"loss": 1.1223,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.44052863436123346,
|
| 293 |
+
"grad_norm": 0.07276295542602365,
|
| 294 |
+
"learning_rate": 0.00013755583023550126,
|
| 295 |
+
"loss": 1.0954,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.45154185022026433,
|
| 300 |
+
"grad_norm": 0.08222792797320272,
|
| 301 |
+
"learning_rate": 0.00013396071473166613,
|
| 302 |
+
"loss": 1.1109,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.46255506607929514,
|
| 307 |
+
"grad_norm": 0.07000329815419548,
|
| 308 |
+
"learning_rate": 0.00013031526741130435,
|
| 309 |
+
"loss": 1.1122,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.473568281938326,
|
| 314 |
+
"grad_norm": 0.07273029465748866,
|
| 315 |
+
"learning_rate": 0.0001266248910460341,
|
| 316 |
+
"loss": 1.1098,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.4845814977973568,
|
| 321 |
+
"grad_norm": 0.07891968539218898,
|
| 322 |
+
"learning_rate": 0.0001228950549950134,
|
| 323 |
+
"loss": 1.1235,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.4955947136563877,
|
| 328 |
+
"grad_norm": 0.07948656131261005,
|
| 329 |
+
"learning_rate": 0.00011913128709902181,
|
| 330 |
+
"loss": 1.119,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.5066079295154186,
|
| 335 |
+
"grad_norm": 0.07426812917091022,
|
| 336 |
+
"learning_rate": 0.00011533916548786857,
|
| 337 |
+
"loss": 1.1153,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.5176211453744494,
|
| 342 |
+
"grad_norm": 0.07320813310432901,
|
| 343 |
+
"learning_rate": 0.00011152431031326978,
|
| 344 |
+
"loss": 1.1189,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.5286343612334802,
|
| 349 |
+
"grad_norm": 0.07027215565752497,
|
| 350 |
+
"learning_rate": 0.0001076923754194464,
|
| 351 |
+
"loss": 1.0921,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.539647577092511,
|
| 356 |
+
"grad_norm": 0.07197803964893518,
|
| 357 |
+
"learning_rate": 0.00010384903996378783,
|
| 358 |
+
"loss": 1.139,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.5506607929515418,
|
| 363 |
+
"grad_norm": 0.07476406384806204,
|
| 364 |
+
"learning_rate": 0.0001,
|
| 365 |
+
"loss": 1.113,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.5616740088105727,
|
| 370 |
+
"grad_norm": 0.07773914011299175,
|
| 371 |
+
"learning_rate": 9.615096003621221e-05,
|
| 372 |
+
"loss": 1.1112,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.5726872246696035,
|
| 377 |
+
"grad_norm": 0.07599058071843097,
|
| 378 |
+
"learning_rate": 9.230762458055363e-05,
|
| 379 |
+
"loss": 1.0823,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.5837004405286343,
|
| 384 |
+
"grad_norm": 0.0791178168791927,
|
| 385 |
+
"learning_rate": 8.847568968673026e-05,
|
| 386 |
+
"loss": 1.1322,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.5947136563876652,
|
| 391 |
+
"grad_norm": 0.08321320900809528,
|
| 392 |
+
"learning_rate": 8.466083451213144e-05,
|
| 393 |
+
"loss": 1.1129,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.6057268722466961,
|
| 398 |
+
"grad_norm": 0.07548913525051125,
|
| 399 |
+
"learning_rate": 8.086871290097821e-05,
|
| 400 |
+
"loss": 1.1062,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.6167400881057269,
|
| 405 |
+
"grad_norm": 0.07309802934315239,
|
| 406 |
+
"learning_rate": 7.710494500498662e-05,
|
| 407 |
+
"loss": 1.1129,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.6277533039647577,
|
| 412 |
+
"grad_norm": 0.069929912224867,
|
| 413 |
+
"learning_rate": 7.337510895396591e-05,
|
| 414 |
+
"loss": 1.1142,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.6387665198237885,
|
| 419 |
+
"grad_norm": 0.06916254050976436,
|
| 420 |
+
"learning_rate": 6.968473258869566e-05,
|
| 421 |
+
"loss": 1.1115,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.6497797356828194,
|
| 426 |
+
"grad_norm": 0.07179141392925117,
|
| 427 |
+
"learning_rate": 6.603928526833387e-05,
|
| 428 |
+
"loss": 1.1114,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.6607929515418502,
|
| 433 |
+
"grad_norm": 0.07393948815408477,
|
| 434 |
+
"learning_rate": 6.244416976449875e-05,
|
| 435 |
+
"loss": 1.0976,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.6718061674008811,
|
| 440 |
+
"grad_norm": 0.06872619691818142,
|
| 441 |
+
"learning_rate": 5.890471425403703e-05,
|
| 442 |
+
"loss": 1.1146,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.6828193832599119,
|
| 447 |
+
"grad_norm": 0.07576414521745392,
|
| 448 |
+
"learning_rate": 5.542616442234618e-05,
|
| 449 |
+
"loss": 1.1147,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.6938325991189427,
|
| 454 |
+
"grad_norm": 0.07330914446070216,
|
| 455 |
+
"learning_rate": 5.201367568895408e-05,
|
| 456 |
+
"loss": 1.0951,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.7048458149779736,
|
| 461 |
+
"grad_norm": 0.07224487325459926,
|
| 462 |
+
"learning_rate": 4.8672305566877964e-05,
|
| 463 |
+
"loss": 1.1086,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.7158590308370044,
|
| 468 |
+
"grad_norm": 0.075306080937695,
|
| 469 |
+
"learning_rate": 4.540700616708658e-05,
|
| 470 |
+
"loss": 1.1092,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.7268722466960352,
|
| 475 |
+
"grad_norm": 0.07437067674720446,
|
| 476 |
+
"learning_rate": 4.222261685917489e-05,
|
| 477 |
+
"loss": 1.0921,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.737885462555066,
|
| 482 |
+
"grad_norm": 0.0711166096176443,
|
| 483 |
+
"learning_rate": 3.9123857099127936e-05,
|
| 484 |
+
"loss": 1.1074,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.748898678414097,
|
| 489 |
+
"grad_norm": 0.07140111450874426,
|
| 490 |
+
"learning_rate": 3.6115319434803894e-05,
|
| 491 |
+
"loss": 1.1162,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.7599118942731278,
|
| 496 |
+
"grad_norm": 0.07495611302185953,
|
| 497 |
+
"learning_rate": 3.32014626995026e-05,
|
| 498 |
+
"loss": 1.1003,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.7709251101321586,
|
| 503 |
+
"grad_norm": 0.07717586378795258,
|
| 504 |
+
"learning_rate": 3.0386605403707346e-05,
|
| 505 |
+
"loss": 1.0927,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.7819383259911894,
|
| 510 |
+
"grad_norm": 0.07229554748902306,
|
| 511 |
+
"learning_rate": 2.7674919334793035e-05,
|
| 512 |
+
"loss": 1.1108,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.7929515418502202,
|
| 517 |
+
"grad_norm": 0.0768434750345143,
|
| 518 |
+
"learning_rate": 2.507042337418707e-05,
|
| 519 |
+
"loss": 1.1058,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.8039647577092511,
|
| 524 |
+
"grad_norm": 0.07347159962460627,
|
| 525 |
+
"learning_rate": 2.2576977541146193e-05,
|
| 526 |
+
"loss": 1.1115,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.8149779735682819,
|
| 531 |
+
"grad_norm": 0.0690702684713988,
|
| 532 |
+
"learning_rate": 2.0198277271976052e-05,
|
| 533 |
+
"loss": 1.1091,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.8259911894273128,
|
| 538 |
+
"grad_norm": 0.06989352121524041,
|
| 539 |
+
"learning_rate": 1.793784794317319e-05,
|
| 540 |
+
"loss": 1.0988,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.8370044052863436,
|
| 545 |
+
"grad_norm": 0.06951408734875156,
|
| 546 |
+
"learning_rate": 1.5799039646605486e-05,
|
| 547 |
+
"loss": 1.1,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.8480176211453745,
|
| 552 |
+
"grad_norm": 0.06836481993836284,
|
| 553 |
+
"learning_rate": 1.3785022224474942e-05,
|
| 554 |
+
"loss": 1.102,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.8590308370044053,
|
| 559 |
+
"grad_norm": 0.06852988441116457,
|
| 560 |
+
"learning_rate": 1.1898780571421552e-05,
|
| 561 |
+
"loss": 1.0987,
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.8700440528634361,
|
| 566 |
+
"grad_norm": 0.07030606192740464,
|
| 567 |
+
"learning_rate": 1.0143110210730312e-05,
|
| 568 |
+
"loss": 1.1009,
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.8810572687224669,
|
| 573 |
+
"grad_norm": 0.06918150786144664,
|
| 574 |
+
"learning_rate": 8.520613151197898e-06,
|
| 575 |
+
"loss": 1.1123,
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.8920704845814978,
|
| 580 |
+
"grad_norm": 0.07235026662376734,
|
| 581 |
+
"learning_rate": 7.033694030799787e-06,
|
| 582 |
+
"loss": 1.098,
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.9030837004405287,
|
| 587 |
+
"grad_norm": 0.06951148470437334,
|
| 588 |
+
"learning_rate": 5.684556552872256e-06,
|
| 589 |
+
"loss": 1.0975,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9140969162995595,
|
| 594 |
+
"grad_norm": 0.06872065759217105,
|
| 595 |
+
"learning_rate": 4.475200220092002e-06,
|
| 596 |
+
"loss": 1.0896,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.9251101321585903,
|
| 601 |
+
"grad_norm": 0.06809525315060157,
|
| 602 |
+
"learning_rate": 3.40741737109318e-06,
|
| 603 |
+
"loss": 1.1158,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.9361233480176211,
|
| 608 |
+
"grad_norm": 0.07195357579119646,
|
| 609 |
+
"learning_rate": 2.482790524113998e-06,
|
| 610 |
+
"loss": 1.1097,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.947136563876652,
|
| 615 |
+
"grad_norm": 0.06817141756269769,
|
| 616 |
+
"learning_rate": 1.7026900316098215e-06,
|
| 617 |
+
"loss": 1.1113,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9581497797356828,
|
| 622 |
+
"grad_norm": 0.07218201005051188,
|
| 623 |
+
"learning_rate": 1.0682720493085607e-06,
|
| 624 |
+
"loss": 1.1049,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9691629955947136,
|
| 629 |
+
"grad_norm": 0.06979182488846636,
|
| 630 |
+
"learning_rate": 5.804768227185565e-07,
|
| 631 |
+
"loss": 1.1125,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9801762114537445,
|
| 636 |
+
"grad_norm": 0.06813879663952505,
|
| 637 |
+
"learning_rate": 2.400272936283088e-07,
|
| 638 |
+
"loss": 1.1002,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9911894273127754,
|
| 643 |
+
"grad_norm": 0.06852453577386937,
|
| 644 |
+
"learning_rate": 4.74280286634099e-08,
|
| 645 |
+
"loss": 1.094,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.0,
|
| 650 |
+
"eval_loss": 1.0366965532302856,
|
| 651 |
+
"eval_runtime": 2.2297,
|
| 652 |
+
"eval_samples_per_second": 3.139,
|
| 653 |
+
"eval_steps_per_second": 0.897,
|
| 654 |
+
"step": 454
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 1.0,
|
| 658 |
+
"step": 454,
|
| 659 |
+
"total_flos": 1.4408836278386688e+16,
|
| 660 |
+
"train_loss": 1.131237539688396,
|
| 661 |
+
"train_runtime": 17315.3957,
|
| 662 |
+
"train_samples_per_second": 3.356,
|
| 663 |
"train_steps_per_second": 0.026
|
| 664 |
}
|
| 665 |
],
|
| 666 |
"logging_steps": 5,
|
| 667 |
+
"max_steps": 454,
|
| 668 |
"num_input_tokens_seen": 0,
|
| 669 |
"num_train_epochs": 1,
|
| 670 |
"save_steps": 100,
|
|
|
|
| 680 |
"attributes": {}
|
| 681 |
}
|
| 682 |
},
|
| 683 |
+
"total_flos": 1.4408836278386688e+16,
|
| 684 |
"train_batch_size": 8,
|
| 685 |
"trial_name": null,
|
| 686 |
"trial_params": null
|