Model save
Browse files- README.md +1 -1
- all_results.json +5 -5
- model.safetensors +1 -1
- runs/Feb14_06-41-42_yekyung-nah-0/events.out.tfevents.1739482920.yekyung-nah-0.3725912.0 +3 -0
- runs/Feb14_06-44-19_yekyung-nah-0/events.out.tfevents.1739483078.yekyung-nah-0.3735356.0 +3 -0
- train_results.json +5 -5
- trainer_state.json +493 -444
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -27,7 +27,7 @@ print(output["generated_text"])
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yknxxh-seoul-national-university/huggingface/runs/
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
| 33 |
|
|
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yknxxh-seoul-national-university/huggingface/runs/zn4j6jmk)
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
| 33 |
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 10000,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 2.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.7685804472008704e+16,
|
| 4 |
+
"train_loss": 1.0384218392182714,
|
| 5 |
+
"train_runtime": 242.5709,
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 38.179,
|
| 8 |
+
"train_steps_per_second": 2.387
|
| 9 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2471645608
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95a541f3397f3ea70e4797acd153ad3d579b4b029fa3dbbe7c2edb68c03c5457
|
| 3 |
size 2471645608
|
runs/Feb14_06-41-42_yekyung-nah-0/events.out.tfevents.1739482920.yekyung-nah-0.3725912.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:311e6f05eac89fd34f5a99c16556c4a074ea9c1f69ce7a845a9914320f8fb630
|
| 3 |
+
size 30335
|
runs/Feb14_06-44-19_yekyung-nah-0/events.out.tfevents.1739483078.yekyung-nah-0.3735356.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35ab42525d078c3a6d2274c74a43ae0a90a88dfc1bc107d7fa43542431819df7
|
| 3 |
+
size 16266
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 10000,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 2.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.7685804472008704e+16,
|
| 4 |
+
"train_loss": 1.0384218392182714,
|
| 5 |
+
"train_runtime": 242.5709,
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 38.179,
|
| 8 |
+
"train_steps_per_second": 2.387
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,786 +3,835 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 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": 7.
|
| 14 |
-
"learning_rate": 3.
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 7.
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 3.
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 5.
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 6.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
"grad_norm": 3.921875,
|
| 56 |
-
"learning_rate": 1.
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 1.
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 2.
|
| 70 |
-
"learning_rate": 1.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 2.
|
| 77 |
-
"learning_rate": 1.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 1.
|
| 84 |
-
"learning_rate": 1.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 1.
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 1.
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 1.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 1.
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 1.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 1.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 1.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 1.
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 1.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate": 1.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate": 1.
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate": 1.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate": 1.
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate": 1.
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate": 1.
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate": 1.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate": 1.
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate": 1.
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate": 1.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate": 1.
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch":
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate": 1.
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate": 1.
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 1.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate": 1.
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 1.
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate": 1.
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 1.
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"learning_rate": 1.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 1.
|
| 307 |
-
"grad_norm":
|
| 308 |
-
"learning_rate": 1.
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 1.
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate": 1.
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 1.
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate": 1.
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 1.
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate": 1.
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 1.
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate": 1.
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 1.
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate": 1.
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 1.
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate": 1.
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 1.
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate": 1.
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 1.
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
-
"learning_rate": 1.
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 1.
|
| 370 |
-
"grad_norm":
|
| 371 |
-
"learning_rate": 1.
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 1.
|
| 377 |
-
"grad_norm":
|
| 378 |
-
"learning_rate": 1.
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 1.
|
| 384 |
-
"grad_norm":
|
| 385 |
-
"learning_rate": 1.
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 1.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate": 1.
|
| 393 |
-
"loss": 0.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 1.
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
-
"learning_rate": 1.
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 1.
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 1.
|
| 412 |
-
"grad_norm":
|
| 413 |
-
"learning_rate": 1.
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 1.
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate": 1.
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 1.
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate": 1.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 1.
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss": 0.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 1.
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 1.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 1.
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 1.
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 1.
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 1.
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 1.
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 1.
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 1.
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 1.
|
| 503 |
-
"grad_norm":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 1.
|
| 510 |
-
"grad_norm":
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 1.
|
| 517 |
-
"grad_norm":
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss": 0.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch":
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch":
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"grad_norm":
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss": 0.
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch":
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch":
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 2.
|
| 559 |
-
"grad_norm": 0.
|
| 560 |
-
"learning_rate":
|
| 561 |
-
"loss": 0.
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 2.
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 2.
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 2.
|
| 580 |
-
"grad_norm":
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 2.
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 2.
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 2.
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 2.
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 2.
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 2.
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 2.
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 2.
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 2.
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 2.
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 2.
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 2.
|
| 664 |
-
"grad_norm": 0.
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss": 0.
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 2.
|
| 671 |
-
"grad_norm": 0.
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 2.
|
| 678 |
-
"grad_norm": 0.
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 2.
|
| 685 |
-
"grad_norm":
|
| 686 |
-
"learning_rate":
|
| 687 |
-
"loss": 0.
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 2.
|
| 692 |
-
"grad_norm":
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 2.
|
| 699 |
-
"grad_norm":
|
| 700 |
-
"learning_rate":
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 2.
|
| 706 |
-
"grad_norm":
|
| 707 |
-
"learning_rate":
|
| 708 |
-
"loss": 0.
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 2.
|
| 713 |
-
"grad_norm":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 2.
|
| 720 |
-
"grad_norm": 0.
|
| 721 |
-
"learning_rate":
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 2.
|
| 727 |
-
"grad_norm": 0.
|
| 728 |
-
"learning_rate":
|
| 729 |
-
"loss": 0.
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"epoch": 2.
|
| 734 |
-
"grad_norm": 0.
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 2.
|
| 741 |
-
"grad_norm": 0.
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 2.
|
| 748 |
-
"grad_norm": 0.
|
| 749 |
-
"learning_rate":
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 2.
|
| 755 |
-
"grad_norm": 0.
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch": 2.
|
| 762 |
-
"grad_norm": 0.
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 2.
|
| 769 |
-
"grad_norm": 0.
|
| 770 |
-
"learning_rate":
|
| 771 |
-
"loss": 0.
|
| 772 |
"step": 540
|
| 773 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 774 |
{
|
| 775 |
"epoch": 3.0,
|
| 776 |
-
"step":
|
| 777 |
-
"total_flos": 2.
|
| 778 |
-
"train_loss": 1.
|
| 779 |
-
"train_runtime":
|
| 780 |
-
"train_samples_per_second":
|
| 781 |
-
"train_steps_per_second": 2.
|
| 782 |
}
|
| 783 |
],
|
| 784 |
"logging_steps": 5,
|
| 785 |
-
"max_steps":
|
| 786 |
"num_input_tokens_seen": 0,
|
| 787 |
"num_train_epochs": 3,
|
| 788 |
"save_steps": 500,
|
|
@@ -798,7 +847,7 @@
|
|
| 798 |
"attributes": {}
|
| 799 |
}
|
| 800 |
},
|
| 801 |
-
"total_flos": 2.
|
| 802 |
"train_batch_size": 16,
|
| 803 |
"trial_name": null,
|
| 804 |
"trial_params": null
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 579,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0051813471502590676,
|
| 13 |
+
"grad_norm": 7.28125,
|
| 14 |
+
"learning_rate": 3.4482758620689656e-07,
|
| 15 |
+
"loss": 2.3287,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.025906735751295335,
|
| 20 |
+
"grad_norm": 7.125,
|
| 21 |
+
"learning_rate": 1.724137931034483e-06,
|
| 22 |
+
"loss": 2.3778,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.05181347150259067,
|
| 27 |
+
"grad_norm": 7.34375,
|
| 28 |
+
"learning_rate": 3.448275862068966e-06,
|
| 29 |
+
"loss": 2.3462,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.07772020725388601,
|
| 34 |
+
"grad_norm": 7.625,
|
| 35 |
+
"learning_rate": 5.172413793103449e-06,
|
| 36 |
+
"loss": 2.303,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.10362694300518134,
|
| 41 |
+
"grad_norm": 6.0625,
|
| 42 |
+
"learning_rate": 6.896551724137932e-06,
|
| 43 |
+
"loss": 2.2303,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.12953367875647667,
|
| 48 |
+
"grad_norm": 6.03125,
|
| 49 |
+
"learning_rate": 8.620689655172414e-06,
|
| 50 |
+
"loss": 2.1691,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.15544041450777202,
|
| 55 |
"grad_norm": 3.921875,
|
| 56 |
+
"learning_rate": 1.0344827586206898e-05,
|
| 57 |
+
"loss": 1.9911,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.18134715025906736,
|
| 62 |
+
"grad_norm": 3.3125,
|
| 63 |
+
"learning_rate": 1.206896551724138e-05,
|
| 64 |
+
"loss": 1.8287,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.20725388601036268,
|
| 69 |
+
"grad_norm": 2.5,
|
| 70 |
+
"learning_rate": 1.3793103448275863e-05,
|
| 71 |
+
"loss": 1.7025,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.23316062176165803,
|
| 76 |
+
"grad_norm": 2.3125,
|
| 77 |
+
"learning_rate": 1.5517241379310346e-05,
|
| 78 |
+
"loss": 1.5651,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.25906735751295334,
|
| 83 |
+
"grad_norm": 1.875,
|
| 84 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 85 |
+
"loss": 1.4727,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.2849740932642487,
|
| 90 |
+
"grad_norm": 2.03125,
|
| 91 |
+
"learning_rate": 1.896551724137931e-05,
|
| 92 |
+
"loss": 1.3068,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.31088082901554404,
|
| 97 |
+
"grad_norm": 1.2578125,
|
| 98 |
+
"learning_rate": 1.9999272808103276e-05,
|
| 99 |
+
"loss": 1.1946,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.33678756476683935,
|
| 104 |
+
"grad_norm": 1.0078125,
|
| 105 |
+
"learning_rate": 1.9991093113822542e-05,
|
| 106 |
+
"loss": 1.1095,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.3626943005181347,
|
| 111 |
+
"grad_norm": 0.91015625,
|
| 112 |
+
"learning_rate": 1.997383219496441e-05,
|
| 113 |
+
"loss": 1.0579,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.38860103626943004,
|
| 118 |
+
"grad_norm": 0.73046875,
|
| 119 |
+
"learning_rate": 1.9947505740530947e-05,
|
| 120 |
+
"loss": 1.0251,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.41450777202072536,
|
| 125 |
+
"grad_norm": 0.7421875,
|
| 126 |
+
"learning_rate": 1.991213767947991e-05,
|
| 127 |
+
"loss": 1.0027,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.44041450777202074,
|
| 132 |
+
"grad_norm": 0.71875,
|
| 133 |
+
"learning_rate": 1.986776015897494e-05,
|
| 134 |
+
"loss": 1.0183,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.46632124352331605,
|
| 139 |
+
"grad_norm": 0.67578125,
|
| 140 |
+
"learning_rate": 1.9814413515165974e-05,
|
| 141 |
+
"loss": 1.0133,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.49222797927461137,
|
| 146 |
+
"grad_norm": 0.640625,
|
| 147 |
+
"learning_rate": 1.975214623652643e-05,
|
| 148 |
+
"loss": 0.9781,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.5181347150259067,
|
| 153 |
+
"grad_norm": 0.6640625,
|
| 154 |
+
"learning_rate": 1.968101491978049e-05,
|
| 155 |
+
"loss": 0.9952,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.5440414507772021,
|
| 160 |
+
"grad_norm": 0.6875,
|
| 161 |
+
"learning_rate": 1.9601084218460494e-05,
|
| 162 |
+
"loss": 0.9748,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.5699481865284974,
|
| 167 |
+
"grad_norm": 0.6171875,
|
| 168 |
+
"learning_rate": 1.9512426784141307e-05,
|
| 169 |
+
"loss": 0.9606,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.5958549222797928,
|
| 174 |
+
"grad_norm": 0.62890625,
|
| 175 |
+
"learning_rate": 1.9415123200404962e-05,
|
| 176 |
+
"loss": 0.985,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.6217616580310881,
|
| 181 |
+
"grad_norm": 0.671875,
|
| 182 |
+
"learning_rate": 1.930926190959566e-05,
|
| 183 |
+
"loss": 0.9631,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.6476683937823834,
|
| 188 |
+
"grad_norm": 0.6328125,
|
| 189 |
+
"learning_rate": 1.9194939132431682e-05,
|
| 190 |
+
"loss": 0.9482,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.6735751295336787,
|
| 195 |
+
"grad_norm": 0.609375,
|
| 196 |
+
"learning_rate": 1.9072258780547317e-05,
|
| 197 |
+
"loss": 0.9733,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.6994818652849741,
|
| 202 |
+
"grad_norm": 0.63671875,
|
| 203 |
+
"learning_rate": 1.894133236204423e-05,
|
| 204 |
+
"loss": 0.9539,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.7253886010362695,
|
| 209 |
+
"grad_norm": 0.70703125,
|
| 210 |
+
"learning_rate": 1.880227888013818e-05,
|
| 211 |
+
"loss": 0.9892,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.7512953367875648,
|
| 216 |
+
"grad_norm": 0.6171875,
|
| 217 |
+
"learning_rate": 1.8655224724993202e-05,
|
| 218 |
+
"loss": 0.9594,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.7772020725388601,
|
| 223 |
+
"grad_norm": 0.609375,
|
| 224 |
+
"learning_rate": 1.850030355884151e-05,
|
| 225 |
+
"loss": 0.9257,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.8031088082901554,
|
| 230 |
+
"grad_norm": 0.62109375,
|
| 231 |
+
"learning_rate": 1.8337656194493637e-05,
|
| 232 |
+
"loss": 0.9688,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.8290155440414507,
|
| 237 |
+
"grad_norm": 0.65625,
|
| 238 |
+
"learning_rate": 1.8167430467349145e-05,
|
| 239 |
+
"loss": 0.9853,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.8549222797927462,
|
| 244 |
+
"grad_norm": 0.62109375,
|
| 245 |
+
"learning_rate": 1.7989781101024305e-05,
|
| 246 |
+
"loss": 0.9375,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.8808290155440415,
|
| 251 |
+
"grad_norm": 0.70703125,
|
| 252 |
+
"learning_rate": 1.7804869566718833e-05,
|
| 253 |
+
"loss": 0.9712,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.9067357512953368,
|
| 258 |
+
"grad_norm": 0.6953125,
|
| 259 |
+
"learning_rate": 1.7612863936449568e-05,
|
| 260 |
+
"loss": 0.951,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.9326424870466321,
|
| 265 |
+
"grad_norm": 0.6875,
|
| 266 |
+
"learning_rate": 1.7413938730284404e-05,
|
| 267 |
+
"loss": 0.9801,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.9585492227979274,
|
| 272 |
+
"grad_norm": 0.640625,
|
| 273 |
+
"learning_rate": 1.7208274757715425e-05,
|
| 274 |
+
"loss": 0.9527,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.9844559585492227,
|
| 279 |
+
"grad_norm": 0.61328125,
|
| 280 |
+
"learning_rate": 1.6996058953315372e-05,
|
| 281 |
+
"loss": 0.9285,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.0103626943005182,
|
| 286 |
+
"grad_norm": 0.6171875,
|
| 287 |
+
"learning_rate": 1.6777484206826793e-05,
|
| 288 |
+
"loss": 0.9499,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 1.0362694300518134,
|
| 293 |
+
"grad_norm": 0.69921875,
|
| 294 |
+
"learning_rate": 1.6552749187838425e-05,
|
| 295 |
+
"loss": 0.9556,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 1.0621761658031088,
|
| 300 |
+
"grad_norm": 0.640625,
|
| 301 |
+
"learning_rate": 1.632205816520799e-05,
|
| 302 |
+
"loss": 0.9324,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 1.0880829015544042,
|
| 307 |
+
"grad_norm": 0.7265625,
|
| 308 |
+
"learning_rate": 1.6085620821395722e-05,
|
| 309 |
+
"loss": 0.9479,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 1.1139896373056994,
|
| 314 |
+
"grad_norm": 0.57421875,
|
| 315 |
+
"learning_rate": 1.5843652061877245e-05,
|
| 316 |
+
"loss": 0.9215,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 1.1398963730569949,
|
| 321 |
+
"grad_norm": 0.5859375,
|
| 322 |
+
"learning_rate": 1.5596371819809104e-05,
|
| 323 |
+
"loss": 0.9367,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 1.16580310880829,
|
| 328 |
+
"grad_norm": 0.5703125,
|
| 329 |
+
"learning_rate": 1.534400485612449e-05,
|
| 330 |
+
"loss": 0.91,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 1.1917098445595855,
|
| 335 |
+
"grad_norm": 0.63671875,
|
| 336 |
+
"learning_rate": 1.5086780555240803e-05,
|
| 337 |
+
"loss": 0.9297,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.2176165803108807,
|
| 342 |
+
"grad_norm": 0.56640625,
|
| 343 |
+
"learning_rate": 1.482493271656482e-05,
|
| 344 |
+
"loss": 0.9426,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.2435233160621761,
|
| 349 |
+
"grad_norm": 0.60546875,
|
| 350 |
+
"learning_rate": 1.4558699341984928e-05,
|
| 351 |
+
"loss": 0.9639,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.2694300518134716,
|
| 356 |
+
"grad_norm": 0.59765625,
|
| 357 |
+
"learning_rate": 1.4288322419543576e-05,
|
| 358 |
+
"loss": 0.946,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 1.2953367875647668,
|
| 363 |
+
"grad_norm": 0.65234375,
|
| 364 |
+
"learning_rate": 1.40140477034866e-05,
|
| 365 |
+
"loss": 0.9438,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.3212435233160622,
|
| 370 |
+
"grad_norm": 0.6015625,
|
| 371 |
+
"learning_rate": 1.373612449088931e-05,
|
| 372 |
+
"loss": 0.9652,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 1.3471502590673574,
|
| 377 |
+
"grad_norm": 0.59765625,
|
| 378 |
+
"learning_rate": 1.3454805395062386e-05,
|
| 379 |
+
"loss": 0.9245,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.3730569948186528,
|
| 384 |
+
"grad_norm": 0.66015625,
|
| 385 |
+
"learning_rate": 1.3170346115943575e-05,
|
| 386 |
+
"loss": 0.9195,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 1.3989637305699483,
|
| 391 |
+
"grad_norm": 0.6015625,
|
| 392 |
+
"learning_rate": 1.2883005207683784e-05,
|
| 393 |
+
"loss": 0.9341,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 1.4248704663212435,
|
| 398 |
+
"grad_norm": 0.6015625,
|
| 399 |
+
"learning_rate": 1.2593043843638978e-05,
|
| 400 |
+
"loss": 0.9238,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 1.450777202072539,
|
| 405 |
+
"grad_norm": 0.62109375,
|
| 406 |
+
"learning_rate": 1.2300725578981308e-05,
|
| 407 |
+
"loss": 0.9271,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 1.4766839378238341,
|
| 412 |
+
"grad_norm": 0.58984375,
|
| 413 |
+
"learning_rate": 1.2006316111145401e-05,
|
| 414 |
+
"loss": 0.9348,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 1.5025906735751295,
|
| 419 |
+
"grad_norm": 0.70703125,
|
| 420 |
+
"learning_rate": 1.1710083038327436e-05,
|
| 421 |
+
"loss": 0.9341,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 1.528497409326425,
|
| 426 |
+
"grad_norm": 0.70703125,
|
| 427 |
+
"learning_rate": 1.1412295616256577e-05,
|
| 428 |
+
"loss": 0.946,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 1.5544041450777202,
|
| 433 |
+
"grad_norm": 0.59375,
|
| 434 |
+
"learning_rate": 1.1113224513459819e-05,
|
| 435 |
+
"loss": 0.9134,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 1.5803108808290154,
|
| 440 |
+
"grad_norm": 0.61328125,
|
| 441 |
+
"learning_rate": 1.0813141565242682e-05,
|
| 442 |
+
"loss": 0.9177,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 1.6062176165803108,
|
| 447 |
+
"grad_norm": 0.65625,
|
| 448 |
+
"learning_rate": 1.0512319526609405e-05,
|
| 449 |
+
"loss": 0.937,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 1.6321243523316062,
|
| 454 |
+
"grad_norm": 0.6015625,
|
| 455 |
+
"learning_rate": 1.021103182434718e-05,
|
| 456 |
+
"loss": 0.9398,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 1.6580310880829017,
|
| 461 |
+
"grad_norm": 0.5625,
|
| 462 |
+
"learning_rate": 9.909552308499792e-06,
|
| 463 |
+
"loss": 0.9289,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 1.6839378238341969,
|
| 468 |
+
"grad_norm": 0.71484375,
|
| 469 |
+
"learning_rate": 9.608155003456529e-06,
|
| 470 |
+
"loss": 0.9075,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 1.709844559585492,
|
| 475 |
+
"grad_norm": 0.6328125,
|
| 476 |
+
"learning_rate": 9.307113858882664e-06,
|
| 477 |
+
"loss": 0.9154,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 1.7357512953367875,
|
| 482 |
+
"grad_norm": 0.55859375,
|
| 483 |
+
"learning_rate": 9.006702500717786e-06,
|
| 484 |
+
"loss": 0.9036,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 1.761658031088083,
|
| 489 |
+
"grad_norm": 0.69921875,
|
| 490 |
+
"learning_rate": 8.707193982468456e-06,
|
| 491 |
+
"loss": 0.9421,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 1.7875647668393784,
|
| 496 |
+
"grad_norm": 0.64453125,
|
| 497 |
+
"learning_rate": 8.408860537021127e-06,
|
| 498 |
+
"loss": 0.9359,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 1.8134715025906736,
|
| 503 |
+
"grad_norm": 0.58984375,
|
| 504 |
+
"learning_rate": 8.111973329200909e-06,
|
| 505 |
+
"loss": 0.9473,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 1.8393782383419688,
|
| 510 |
+
"grad_norm": 0.6328125,
|
| 511 |
+
"learning_rate": 7.816802209301241e-06,
|
| 512 |
+
"loss": 0.9257,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 1.8652849740932642,
|
| 517 |
+
"grad_norm": 0.609375,
|
| 518 |
+
"learning_rate": 7.523615467808249e-06,
|
| 519 |
+
"loss": 0.9486,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 1.8911917098445596,
|
| 524 |
+
"grad_norm": 0.640625,
|
| 525 |
+
"learning_rate": 7.232679591542979e-06,
|
| 526 |
+
"loss": 0.9366,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 1.917098445595855,
|
| 531 |
+
"grad_norm": 0.60546875,
|
| 532 |
+
"learning_rate": 6.944259021442967e-06,
|
| 533 |
+
"loss": 0.9292,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 1.9430051813471503,
|
| 538 |
+
"grad_norm": 0.62109375,
|
| 539 |
+
"learning_rate": 6.6586159122033914e-06,
|
| 540 |
+
"loss": 0.9025,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 1.9689119170984455,
|
| 545 |
+
"grad_norm": 0.609375,
|
| 546 |
+
"learning_rate": 6.3760098939962935e-06,
|
| 547 |
+
"loss": 0.9151,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 1.994818652849741,
|
| 552 |
+
"grad_norm": 0.73046875,
|
| 553 |
+
"learning_rate": 6.096697836484382e-06,
|
| 554 |
+
"loss": 0.9283,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 2.0207253886010363,
|
| 559 |
+
"grad_norm": 0.640625,
|
| 560 |
+
"learning_rate": 5.820933615343975e-06,
|
| 561 |
+
"loss": 0.9249,
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.0466321243523318,
|
| 566 |
+
"grad_norm": 0.5390625,
|
| 567 |
+
"learning_rate": 5.548967881509275e-06,
|
| 568 |
+
"loss": 0.9119,
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.0725388601036268,
|
| 573 |
+
"grad_norm": 0.60546875,
|
| 574 |
+
"learning_rate": 5.281047833347676e-06,
|
| 575 |
+
"loss": 0.9236,
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.098445595854922,
|
| 580 |
+
"grad_norm": 0.609375,
|
| 581 |
+
"learning_rate": 5.017416991973281e-06,
|
| 582 |
+
"loss": 0.9169,
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.1243523316062176,
|
| 587 |
+
"grad_norm": 0.62109375,
|
| 588 |
+
"learning_rate": 4.758314979902734e-06,
|
| 589 |
+
"loss": 0.9315,
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.150259067357513,
|
| 594 |
+
"grad_norm": 0.5859375,
|
| 595 |
+
"learning_rate": 4.503977303254673e-06,
|
| 596 |
+
"loss": 0.9111,
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.1761658031088085,
|
| 601 |
+
"grad_norm": 0.62109375,
|
| 602 |
+
"learning_rate": 4.25463513769064e-06,
|
| 603 |
+
"loss": 0.9495,
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.2020725388601035,
|
| 608 |
+
"grad_norm": 0.59375,
|
| 609 |
+
"learning_rate": 4.010515118292127e-06,
|
| 610 |
+
"loss": 0.9338,
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.227979274611399,
|
| 615 |
+
"grad_norm": 0.5625,
|
| 616 |
+
"learning_rate": 3.771839133564704e-06,
|
| 617 |
+
"loss": 0.9028,
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.2538860103626943,
|
| 622 |
+
"grad_norm": 0.625,
|
| 623 |
+
"learning_rate": 3.5388241237564337e-06,
|
| 624 |
+
"loss": 0.9227,
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.2797927461139897,
|
| 629 |
+
"grad_norm": 0.64453125,
|
| 630 |
+
"learning_rate": 3.311681883673937e-06,
|
| 631 |
+
"loss": 0.9462,
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.305699481865285,
|
| 636 |
+
"grad_norm": 0.6015625,
|
| 637 |
+
"learning_rate": 3.0906188701753127e-06,
|
| 638 |
+
"loss": 0.929,
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 2.33160621761658,
|
| 643 |
+
"grad_norm": 0.609375,
|
| 644 |
+
"learning_rate": 2.875836014514867e-06,
|
| 645 |
+
"loss": 0.9178,
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 2.3575129533678756,
|
| 650 |
+
"grad_norm": 0.58203125,
|
| 651 |
+
"learning_rate": 2.6675285397102856e-06,
|
| 652 |
+
"loss": 0.9052,
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 2.383419689119171,
|
| 657 |
+
"grad_norm": 0.59375,
|
| 658 |
+
"learning_rate": 2.465885783098166e-06,
|
| 659 |
+
"loss": 0.9019,
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 2.4093264248704664,
|
| 664 |
+
"grad_norm": 0.6015625,
|
| 665 |
+
"learning_rate": 2.2710910242392468e-06,
|
| 666 |
+
"loss": 0.9393,
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 2.4352331606217614,
|
| 671 |
+
"grad_norm": 0.6328125,
|
| 672 |
+
"learning_rate": 2.0833213183297475e-06,
|
| 673 |
+
"loss": 0.9261,
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 2.461139896373057,
|
| 678 |
+
"grad_norm": 0.72265625,
|
| 679 |
+
"learning_rate": 1.9027473352702208e-06,
|
| 680 |
+
"loss": 0.9021,
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 2.4870466321243523,
|
| 685 |
+
"grad_norm": 0.5625,
|
| 686 |
+
"learning_rate": 1.729533204538224e-06,
|
| 687 |
+
"loss": 0.9289,
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 2.5129533678756477,
|
| 692 |
+
"grad_norm": 0.625,
|
| 693 |
+
"learning_rate": 1.563836366005782e-06,
|
| 694 |
+
"loss": 0.9077,
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 2.538860103626943,
|
| 699 |
+
"grad_norm": 0.625,
|
| 700 |
+
"learning_rate": 1.4058074268372224e-06,
|
| 701 |
+
"loss": 0.9668,
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 2.5647668393782386,
|
| 706 |
+
"grad_norm": 0.59765625,
|
| 707 |
+
"learning_rate": 1.2555900245975262e-06,
|
| 708 |
+
"loss": 0.9356,
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 2.5906735751295336,
|
| 713 |
+
"grad_norm": 0.5390625,
|
| 714 |
+
"learning_rate": 1.1133206966955213e-06,
|
| 715 |
+
"loss": 0.9311,
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 2.616580310880829,
|
| 720 |
+
"grad_norm": 0.5703125,
|
| 721 |
+
"learning_rate": 9.79128756280675e-07,
|
| 722 |
+
"loss": 0.9411,
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 2.6424870466321244,
|
| 727 |
+
"grad_norm": 0.69140625,
|
| 728 |
+
"learning_rate": 8.531361747062272e-07,
|
| 729 |
+
"loss": 0.9778,
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 2.66839378238342,
|
| 734 |
+
"grad_norm": 0.57421875,
|
| 735 |
+
"learning_rate": 7.354574706655038e-07,
|
| 736 |
+
"loss": 0.9099,
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 2.694300518134715,
|
| 741 |
+
"grad_norm": 0.578125,
|
| 742 |
+
"learning_rate": 6.261996061022335e-07,
|
| 743 |
+
"loss": 0.9045,
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 2.7202072538860103,
|
| 748 |
+
"grad_norm": 0.58203125,
|
| 749 |
+
"learning_rate": 5.254618889893858e-07,
|
| 750 |
+
"loss": 0.8916,
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 2.7461139896373057,
|
| 755 |
+
"grad_norm": 0.61328125,
|
| 756 |
+
"learning_rate": 4.3333588306499584e-07,
|
| 757 |
+
"loss": 0.9151,
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 2.772020725388601,
|
| 762 |
+
"grad_norm": 0.5703125,
|
| 763 |
+
"learning_rate": 3.499053246069362e-07,
|
| 764 |
+
"loss": 0.9418,
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 2.7979274611398965,
|
| 769 |
+
"grad_norm": 0.59765625,
|
| 770 |
+
"learning_rate": 2.7524604632233054e-07,
|
| 771 |
+
"loss": 0.9384,
|
| 772 |
"step": 540
|
| 773 |
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.823834196891192,
|
| 776 |
+
"grad_norm": 0.59375,
|
| 777 |
+
"learning_rate": 2.0942590842078503e-07,
|
| 778 |
+
"loss": 0.9117,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.849740932642487,
|
| 783 |
+
"grad_norm": 0.6328125,
|
| 784 |
+
"learning_rate": 1.5250473693406486e-07,
|
| 785 |
+
"loss": 0.9154,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.8756476683937824,
|
| 790 |
+
"grad_norm": 0.6328125,
|
| 791 |
+
"learning_rate": 1.0453426933830002e-07,
|
| 792 |
+
"loss": 0.942,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.901554404145078,
|
| 797 |
+
"grad_norm": 0.609375,
|
| 798 |
+
"learning_rate": 6.555810752813308e-08,
|
| 799 |
+
"loss": 0.9441,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.927461139896373,
|
| 804 |
+
"grad_norm": 0.59765625,
|
| 805 |
+
"learning_rate": 3.5611678185563106e-08,
|
| 806 |
+
"loss": 0.9353,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.9533678756476682,
|
| 811 |
+
"grad_norm": 0.5859375,
|
| 812 |
+
"learning_rate": 1.4722200579497803e-08,
|
| 813 |
+
"loss": 0.9235,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.9792746113989637,
|
| 818 |
+
"grad_norm": 0.59375,
|
| 819 |
+
"learning_rate": 2.9086618252893717e-09,
|
| 820 |
+
"loss": 0.9246,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
{
|
| 824 |
"epoch": 3.0,
|
| 825 |
+
"step": 579,
|
| 826 |
+
"total_flos": 2.7685804472008704e+16,
|
| 827 |
+
"train_loss": 1.0384218392182714,
|
| 828 |
+
"train_runtime": 242.5709,
|
| 829 |
+
"train_samples_per_second": 38.179,
|
| 830 |
+
"train_steps_per_second": 2.387
|
| 831 |
}
|
| 832 |
],
|
| 833 |
"logging_steps": 5,
|
| 834 |
+
"max_steps": 579,
|
| 835 |
"num_input_tokens_seen": 0,
|
| 836 |
"num_train_epochs": 3,
|
| 837 |
"save_steps": 500,
|
|
|
|
| 847 |
"attributes": {}
|
| 848 |
}
|
| 849 |
},
|
| 850 |
+
"total_flos": 2.7685804472008704e+16,
|
| 851 |
"train_batch_size": 16,
|
| 852 |
"trial_name": null,
|
| 853 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5560
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fa81c0842694b8297e4996a6f1dcfc12daedf7a7564e339a703a7f1e1ac1a96
|
| 3 |
size 5560
|