Model save
Browse files- README.md +6 -1
- all_results.json +6 -6
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- train_results.json +6 -6
- trainer_state.json +533 -743
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -4,6 +4,7 @@ base_model: EleutherAI/pythia-2.8b-deduped
|
|
| 4 |
tags:
|
| 5 |
- trl
|
| 6 |
- sft
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: pythia-2_8b_sft-gpt-turbo
|
|
@@ -35,9 +36,13 @@ More information needed
|
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
- learning_rate: 2e-05
|
| 38 |
-
- train_batch_size:
|
| 39 |
- eval_batch_size: 4
|
| 40 |
- seed: 42
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 4 |
tags:
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
+
- alignment-handbook
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: pythia-2_8b_sft-gpt-turbo
|
|
|
|
| 36 |
|
| 37 |
The following hyperparameters were used during training:
|
| 38 |
- learning_rate: 2e-05
|
| 39 |
+
- train_batch_size: 16
|
| 40 |
- eval_batch_size: 4
|
| 41 |
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 2
|
| 44 |
+
- total_train_batch_size: 32
|
| 45 |
+
- total_eval_batch_size: 8
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 4.7710598211200614e+17,
|
| 4 |
+
"train_loss": 1.2026635146722562,
|
| 5 |
+
"train_runtime": 2875.3707,
|
| 6 |
+
"train_samples": 41954,
|
| 7 |
+
"train_samples_per_second": 14.591,
|
| 8 |
+
"train_steps_per_second": 0.456
|
| 9 |
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4978209248
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18f01819f93756c7e3055d02f82a12514166e9002f375a6492205f19ba78ce59
|
| 3 |
size 4978209248
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 572254856
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:208094d7e45292a7eed2812c6da4f68a27eb6cca0041fac99d416ae32fac0449
|
| 3 |
size 572254856
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 4.7710598211200614e+17,
|
| 4 |
+
"train_loss": 1.2026635146722562,
|
| 5 |
+
"train_runtime": 2875.3707,
|
| 6 |
+
"train_samples": 41954,
|
| 7 |
+
"train_samples_per_second": 14.591,
|
| 8 |
+
"train_steps_per_second": 0.456
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,1150 +3,940 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
"eval_steps": 100,
|
| 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": 4.
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 5.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 4.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
"grad_norm": 2.921875,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 2.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 2.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 2.
|
| 70 |
-
"learning_rate": 1.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 2.
|
| 77 |
-
"learning_rate": 1.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 2.
|
| 84 |
-
"learning_rate": 1.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 2.
|
| 91 |
-
"learning_rate": 1.
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 2.
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 2.
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 2.
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 2.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 2.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 2.
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 2.
|
| 140 |
-
"learning_rate": 1.
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 2.
|
| 147 |
-
"learning_rate": 1.
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 2.
|
| 154 |
-
"learning_rate": 1.
|
| 155 |
-
"loss": 1.
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 2.
|
| 161 |
-
"learning_rate": 1.
|
| 162 |
-
"loss": 1.
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 2.
|
| 168 |
-
"learning_rate": 1.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 2.
|
| 175 |
-
"learning_rate": 1.
|
| 176 |
-
"loss": 1.
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 2.
|
| 182 |
-
"learning_rate": 1.
|
| 183 |
-
"loss": 1.
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 2.
|
| 189 |
-
"learning_rate": 1.
|
| 190 |
-
"loss":
|
| 191 |
"step": 260
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 2.
|
| 196 |
-
"learning_rate": 1.
|
| 197 |
-
"loss": 1.
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 2.
|
| 203 |
-
"learning_rate": 1.
|
| 204 |
-
"loss": 1.
|
| 205 |
"step": 280
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm": 2.
|
| 210 |
-
"learning_rate": 1.
|
| 211 |
-
"loss":
|
| 212 |
"step": 290
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 2.
|
| 217 |
-
"learning_rate": 1.
|
| 218 |
-
"loss": 1.
|
| 219 |
"step": 300
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 2.
|
| 224 |
-
"learning_rate": 1.
|
| 225 |
-
"loss": 1.
|
| 226 |
"step": 310
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 2.
|
| 231 |
-
"learning_rate": 1.
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 320
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 2.
|
| 238 |
-
"learning_rate": 1.
|
| 239 |
-
"loss": 1.
|
| 240 |
"step": 330
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm": 2.
|
| 245 |
-
"learning_rate": 1.
|
| 246 |
-
"loss": 1.
|
| 247 |
"step": 340
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 2.
|
| 252 |
-
"learning_rate": 1.
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 350
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 2.
|
| 259 |
-
"learning_rate": 1.
|
| 260 |
-
"loss":
|
| 261 |
"step": 360
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm": 2.
|
| 266 |
-
"learning_rate": 1.
|
| 267 |
-
"loss":
|
| 268 |
"step": 370
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm": 2.
|
| 273 |
-
"learning_rate": 1.
|
| 274 |
-
"loss":
|
| 275 |
"step": 380
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm": 2.
|
| 280 |
-
"learning_rate": 1.
|
| 281 |
-
"loss": 1.
|
| 282 |
"step": 390
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
-
"learning_rate": 1.
|
| 288 |
-
"loss":
|
| 289 |
"step": 400
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm": 2.
|
| 294 |
-
"learning_rate": 1.
|
| 295 |
-
"loss": 1.
|
| 296 |
"step": 410
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm": 2.
|
| 301 |
-
"learning_rate": 1.
|
| 302 |
-
"loss":
|
| 303 |
"step": 420
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm": 2.
|
| 308 |
-
"learning_rate": 1.
|
| 309 |
-
"loss": 1.
|
| 310 |
"step": 430
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm": 2.
|
| 315 |
-
"learning_rate": 1.
|
| 316 |
-
"loss": 1.
|
| 317 |
"step": 440
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm": 2.
|
| 322 |
-
"learning_rate": 1.
|
| 323 |
-
"loss": 1.
|
| 324 |
"step": 450
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm": 2.
|
| 329 |
-
"learning_rate": 1.
|
| 330 |
-
"loss": 1.
|
| 331 |
"step": 460
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm": 2.
|
| 336 |
-
"learning_rate": 1.
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 470
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm": 2.
|
| 343 |
-
"learning_rate": 1.
|
| 344 |
-
"loss":
|
| 345 |
"step": 480
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm": 2.
|
| 350 |
-
"learning_rate": 1.
|
| 351 |
-
"loss": 1.
|
| 352 |
"step": 490
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm": 2.
|
| 357 |
-
"learning_rate": 1.
|
| 358 |
-
"loss": 1.
|
| 359 |
"step": 500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm": 2.
|
| 364 |
-
"learning_rate": 1.
|
| 365 |
-
"loss":
|
| 366 |
"step": 510
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm": 2.
|
| 371 |
-
"learning_rate": 1.
|
| 372 |
-
"loss":
|
| 373 |
"step": 520
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm": 2.
|
| 378 |
-
"learning_rate": 1.
|
| 379 |
-
"loss": 1.
|
| 380 |
"step": 530
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm": 2.
|
| 385 |
-
"learning_rate": 1.
|
| 386 |
-
"loss":
|
| 387 |
"step": 540
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm": 2.
|
| 392 |
-
"learning_rate": 1.
|
| 393 |
-
"loss": 1.
|
| 394 |
"step": 550
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm": 2.
|
| 399 |
-
"learning_rate": 1.
|
| 400 |
-
"loss":
|
| 401 |
"step": 560
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm": 2.
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss": 1.
|
| 408 |
"step": 570
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm": 2.
|
| 413 |
-
"learning_rate": 1.
|
| 414 |
-
"loss":
|
| 415 |
"step": 580
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
-
"learning_rate": 1.
|
| 421 |
-
"loss": 1.
|
| 422 |
"step": 590
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm": 2.
|
| 427 |
-
"learning_rate": 1.
|
| 428 |
-
"loss":
|
| 429 |
"step": 600
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate": 1.
|
| 435 |
-
"loss":
|
| 436 |
"step": 610
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm": 2.
|
| 441 |
-
"learning_rate": 1.
|
| 442 |
-
"loss": 1.
|
| 443 |
"step": 620
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm": 2.
|
| 448 |
-
"learning_rate": 1.
|
| 449 |
-
"loss":
|
| 450 |
"step": 630
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm": 2.
|
| 455 |
-
"learning_rate": 1.
|
| 456 |
-
"loss": 1.
|
| 457 |
"step": 640
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm": 2.
|
| 462 |
-
"learning_rate": 1.
|
| 463 |
-
"loss":
|
| 464 |
"step": 650
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm": 2.
|
| 469 |
-
"learning_rate": 1.
|
| 470 |
-
"loss": 1.
|
| 471 |
"step": 660
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm": 2.
|
| 476 |
-
"learning_rate": 1.
|
| 477 |
-
"loss": 1.
|
| 478 |
"step": 670
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm": 2.
|
| 483 |
-
"learning_rate": 1.
|
| 484 |
-
"loss":
|
| 485 |
"step": 680
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm": 2.
|
| 490 |
-
"learning_rate": 1.
|
| 491 |
-
"loss":
|
| 492 |
"step": 690
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm": 2.
|
| 497 |
-
"learning_rate": 1.
|
| 498 |
-
"loss":
|
| 499 |
"step": 700
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
-
"learning_rate": 1.
|
| 505 |
-
"loss":
|
| 506 |
"step": 710
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm": 2.
|
| 511 |
-
"learning_rate": 1.
|
| 512 |
-
"loss": 1.
|
| 513 |
"step": 720
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm": 2.
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss":
|
| 520 |
"step": 730
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm": 2.
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss":
|
| 527 |
"step": 740
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm": 2.
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss":
|
| 534 |
"step": 750
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm": 2.
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss":
|
| 541 |
"step": 760
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm": 2.
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss":
|
| 548 |
"step": 770
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm": 2.
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss":
|
| 555 |
"step": 780
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm": 2.
|
| 560 |
-
"learning_rate":
|
| 561 |
-
"loss":
|
| 562 |
"step": 790
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm": 2.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss":
|
| 569 |
"step": 800
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm": 2.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 1.
|
| 576 |
"step": 810
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm": 2.
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss":
|
| 583 |
"step": 820
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"grad_norm": 2.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss":
|
| 590 |
"step": 830
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"grad_norm": 2.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss":
|
| 597 |
"step": 840
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"grad_norm": 2.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 1.
|
| 604 |
"step": 850
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 0.
|
| 608 |
-
"grad_norm": 2.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss":
|
| 611 |
"step": 860
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"grad_norm": 2.
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss":
|
| 618 |
"step": 870
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
-
"grad_norm": 2.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss":
|
| 625 |
"step": 880
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"grad_norm": 2.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss":
|
| 632 |
"step": 890
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"grad_norm": 2.
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss":
|
| 639 |
"step": 900
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm": 2.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss":
|
| 646 |
"step": 910
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 0.
|
| 650 |
-
"grad_norm": 2.
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss":
|
| 653 |
"step": 920
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"grad_norm": 2.
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss": 1.
|
| 660 |
"step": 930
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 0.
|
| 664 |
-
"grad_norm":
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss":
|
| 667 |
"step": 940
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0.
|
| 671 |
-
"grad_norm": 2.
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss":
|
| 674 |
"step": 950
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"grad_norm": 2.
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss":
|
| 681 |
"step": 960
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 0.
|
| 685 |
-
"grad_norm": 2.
|
| 686 |
-
"learning_rate":
|
| 687 |
-
"loss":
|
| 688 |
"step": 970
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 0.
|
| 692 |
-
"grad_norm": 2.
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss":
|
| 695 |
"step": 980
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 0.
|
| 699 |
-
"grad_norm": 2.
|
| 700 |
-
"learning_rate":
|
| 701 |
-
"loss":
|
| 702 |
"step": 990
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 0.
|
| 706 |
-
"grad_norm": 2.
|
| 707 |
-
"learning_rate":
|
| 708 |
-
"loss":
|
| 709 |
"step": 1000
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 0.
|
| 713 |
-
"grad_norm": 2.
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss":
|
| 716 |
"step": 1010
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 0.
|
| 720 |
-
"grad_norm": 2.
|
| 721 |
-
"learning_rate":
|
| 722 |
-
"loss":
|
| 723 |
"step": 1020
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 0.
|
| 727 |
-
"grad_norm": 2.
|
| 728 |
-
"learning_rate":
|
| 729 |
-
"loss":
|
| 730 |
"step": 1030
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"epoch": 0.
|
| 734 |
-
"grad_norm": 2.
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss":
|
| 737 |
"step": 1040
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 0.
|
| 741 |
-
"grad_norm":
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss":
|
| 744 |
"step": 1050
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 0.
|
| 748 |
-
"grad_norm": 2.
|
| 749 |
-
"learning_rate":
|
| 750 |
-
"loss":
|
| 751 |
"step": 1060
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 0.
|
| 755 |
-
"grad_norm": 2.
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss":
|
| 758 |
"step": 1070
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch": 0.
|
| 762 |
-
"grad_norm": 2.
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss":
|
| 765 |
"step": 1080
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 0.
|
| 769 |
-
"grad_norm": 2.
|
| 770 |
-
"learning_rate":
|
| 771 |
-
"loss":
|
| 772 |
"step": 1090
|
| 773 |
},
|
| 774 |
{
|
| 775 |
-
"epoch": 0.
|
| 776 |
-
"grad_norm": 2.
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss":
|
| 779 |
"step": 1100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch": 0.
|
| 783 |
-
"grad_norm": 2.
|
| 784 |
-
"learning_rate":
|
| 785 |
-
"loss":
|
| 786 |
"step": 1110
|
| 787 |
},
|
| 788 |
{
|
| 789 |
-
"epoch": 0.
|
| 790 |
-
"grad_norm": 2.
|
| 791 |
-
"learning_rate":
|
| 792 |
-
"loss":
|
| 793 |
"step": 1120
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"epoch": 0.
|
| 797 |
-
"grad_norm":
|
| 798 |
-
"learning_rate":
|
| 799 |
-
"loss":
|
| 800 |
"step": 1130
|
| 801 |
},
|
| 802 |
{
|
| 803 |
-
"epoch": 0.
|
| 804 |
-
"grad_norm": 2.
|
| 805 |
-
"learning_rate":
|
| 806 |
-
"loss":
|
| 807 |
"step": 1140
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch": 0.
|
| 811 |
-
"grad_norm":
|
| 812 |
-
"learning_rate":
|
| 813 |
-
"loss":
|
| 814 |
"step": 1150
|
| 815 |
},
|
| 816 |
{
|
| 817 |
-
"epoch": 0.
|
| 818 |
-
"grad_norm": 2.
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss":
|
| 821 |
"step": 1160
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch": 0.
|
| 825 |
-
"grad_norm": 2.
|
| 826 |
-
"learning_rate":
|
| 827 |
-
"loss":
|
| 828 |
"step": 1170
|
| 829 |
},
|
| 830 |
{
|
| 831 |
-
"epoch": 0.
|
| 832 |
-
"grad_norm": 2.
|
| 833 |
-
"learning_rate":
|
| 834 |
-
"loss":
|
| 835 |
"step": 1180
|
| 836 |
},
|
| 837 |
{
|
| 838 |
-
"epoch": 0.
|
| 839 |
-
"grad_norm": 2.
|
| 840 |
-
"learning_rate":
|
| 841 |
-
"loss": 1.
|
| 842 |
"step": 1190
|
| 843 |
},
|
| 844 |
{
|
| 845 |
-
"epoch": 0.
|
| 846 |
-
"grad_norm": 2.
|
| 847 |
-
"learning_rate":
|
| 848 |
-
"loss":
|
| 849 |
"step": 1200
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 0.
|
| 853 |
-
"grad_norm": 2.
|
| 854 |
-
"learning_rate":
|
| 855 |
-
"loss":
|
| 856 |
"step": 1210
|
| 857 |
},
|
| 858 |
{
|
| 859 |
-
"epoch": 0.
|
| 860 |
-
"grad_norm": 2.
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss":
|
| 863 |
"step": 1220
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch": 0.
|
| 867 |
-
"grad_norm": 2.
|
| 868 |
-
"learning_rate":
|
| 869 |
-
"loss":
|
| 870 |
"step": 1230
|
| 871 |
},
|
| 872 |
{
|
| 873 |
-
"epoch": 0.
|
| 874 |
-
"grad_norm":
|
| 875 |
-
"learning_rate":
|
| 876 |
-
"loss":
|
| 877 |
"step": 1240
|
| 878 |
},
|
| 879 |
{
|
| 880 |
-
"epoch": 0.
|
| 881 |
-
"grad_norm": 2.
|
| 882 |
-
"learning_rate":
|
| 883 |
-
"loss":
|
| 884 |
"step": 1250
|
| 885 |
},
|
| 886 |
{
|
| 887 |
-
"epoch": 0.
|
| 888 |
-
"grad_norm": 2.
|
| 889 |
-
"learning_rate":
|
| 890 |
-
"loss": 1.
|
| 891 |
"step": 1260
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch": 0.
|
| 895 |
-
"grad_norm": 2.
|
| 896 |
-
"learning_rate":
|
| 897 |
-
"loss":
|
| 898 |
"step": 1270
|
| 899 |
},
|
| 900 |
{
|
| 901 |
-
"epoch": 0.
|
| 902 |
-
"grad_norm": 2.
|
| 903 |
-
"learning_rate":
|
| 904 |
-
"loss":
|
| 905 |
"step": 1280
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch": 0.
|
| 909 |
-
"grad_norm": 2.
|
| 910 |
-
"learning_rate":
|
| 911 |
-
"loss":
|
| 912 |
"step": 1290
|
| 913 |
},
|
| 914 |
{
|
| 915 |
-
"epoch": 0.
|
| 916 |
-
"grad_norm": 2.
|
| 917 |
-
"learning_rate":
|
| 918 |
-
"loss":
|
| 919 |
"step": 1300
|
| 920 |
},
|
| 921 |
{
|
| 922 |
-
"epoch": 0.
|
| 923 |
-
"grad_norm": 2.
|
| 924 |
-
"learning_rate":
|
| 925 |
-
"loss":
|
| 926 |
"step": 1310
|
| 927 |
},
|
| 928 |
-
{
|
| 929 |
-
"epoch": 0.8188585607940446,
|
| 930 |
-
"grad_norm": 2.421875,
|
| 931 |
-
"learning_rate": 4.027586206896552e-06,
|
| 932 |
-
"loss": 0.9809,
|
| 933 |
-
"step": 1320
|
| 934 |
-
},
|
| 935 |
-
{
|
| 936 |
-
"epoch": 0.825062034739454,
|
| 937 |
-
"grad_norm": 2.390625,
|
| 938 |
-
"learning_rate": 3.889655172413793e-06,
|
| 939 |
-
"loss": 0.9812,
|
| 940 |
-
"step": 1330
|
| 941 |
-
},
|
| 942 |
-
{
|
| 943 |
-
"epoch": 0.8312655086848635,
|
| 944 |
-
"grad_norm": 2.109375,
|
| 945 |
-
"learning_rate": 3.7517241379310347e-06,
|
| 946 |
-
"loss": 0.9781,
|
| 947 |
-
"step": 1340
|
| 948 |
-
},
|
| 949 |
-
{
|
| 950 |
-
"epoch": 0.8374689826302729,
|
| 951 |
-
"grad_norm": 2.234375,
|
| 952 |
-
"learning_rate": 3.6137931034482758e-06,
|
| 953 |
-
"loss": 0.9754,
|
| 954 |
-
"step": 1350
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 0.8436724565756824,
|
| 958 |
-
"grad_norm": 2.109375,
|
| 959 |
-
"learning_rate": 3.4758620689655177e-06,
|
| 960 |
-
"loss": 0.9641,
|
| 961 |
-
"step": 1360
|
| 962 |
-
},
|
| 963 |
-
{
|
| 964 |
-
"epoch": 0.8498759305210918,
|
| 965 |
-
"grad_norm": 2.671875,
|
| 966 |
-
"learning_rate": 3.3379310344827588e-06,
|
| 967 |
-
"loss": 0.9723,
|
| 968 |
-
"step": 1370
|
| 969 |
-
},
|
| 970 |
-
{
|
| 971 |
-
"epoch": 0.8560794044665012,
|
| 972 |
-
"grad_norm": 2.375,
|
| 973 |
-
"learning_rate": 3.2000000000000003e-06,
|
| 974 |
-
"loss": 0.9938,
|
| 975 |
-
"step": 1380
|
| 976 |
-
},
|
| 977 |
-
{
|
| 978 |
-
"epoch": 0.8622828784119106,
|
| 979 |
-
"grad_norm": 2.359375,
|
| 980 |
-
"learning_rate": 3.0620689655172414e-06,
|
| 981 |
-
"loss": 0.9523,
|
| 982 |
-
"step": 1390
|
| 983 |
-
},
|
| 984 |
-
{
|
| 985 |
-
"epoch": 0.8684863523573201,
|
| 986 |
-
"grad_norm": 1.9609375,
|
| 987 |
-
"learning_rate": 2.924137931034483e-06,
|
| 988 |
-
"loss": 0.9508,
|
| 989 |
-
"step": 1400
|
| 990 |
-
},
|
| 991 |
-
{
|
| 992 |
-
"epoch": 0.8746898263027295,
|
| 993 |
-
"grad_norm": 2.28125,
|
| 994 |
-
"learning_rate": 2.7862068965517244e-06,
|
| 995 |
-
"loss": 0.9512,
|
| 996 |
-
"step": 1410
|
| 997 |
-
},
|
| 998 |
-
{
|
| 999 |
-
"epoch": 0.8808933002481389,
|
| 1000 |
-
"grad_norm": 2.109375,
|
| 1001 |
-
"learning_rate": 2.648275862068966e-06,
|
| 1002 |
-
"loss": 0.9547,
|
| 1003 |
-
"step": 1420
|
| 1004 |
-
},
|
| 1005 |
-
{
|
| 1006 |
-
"epoch": 0.8870967741935484,
|
| 1007 |
-
"grad_norm": 2.28125,
|
| 1008 |
-
"learning_rate": 2.510344827586207e-06,
|
| 1009 |
-
"loss": 0.9539,
|
| 1010 |
-
"step": 1430
|
| 1011 |
-
},
|
| 1012 |
-
{
|
| 1013 |
-
"epoch": 0.8933002481389578,
|
| 1014 |
-
"grad_norm": 2.265625,
|
| 1015 |
-
"learning_rate": 2.3724137931034484e-06,
|
| 1016 |
-
"loss": 0.9879,
|
| 1017 |
-
"step": 1440
|
| 1018 |
-
},
|
| 1019 |
-
{
|
| 1020 |
-
"epoch": 0.8995037220843672,
|
| 1021 |
-
"grad_norm": 2.40625,
|
| 1022 |
-
"learning_rate": 2.23448275862069e-06,
|
| 1023 |
-
"loss": 0.9684,
|
| 1024 |
-
"step": 1450
|
| 1025 |
-
},
|
| 1026 |
-
{
|
| 1027 |
-
"epoch": 0.9057071960297767,
|
| 1028 |
-
"grad_norm": 2.09375,
|
| 1029 |
-
"learning_rate": 2.0965517241379315e-06,
|
| 1030 |
-
"loss": 0.9348,
|
| 1031 |
-
"step": 1460
|
| 1032 |
-
},
|
| 1033 |
-
{
|
| 1034 |
-
"epoch": 0.9119106699751861,
|
| 1035 |
-
"grad_norm": 2.25,
|
| 1036 |
-
"learning_rate": 1.9586206896551725e-06,
|
| 1037 |
-
"loss": 0.9711,
|
| 1038 |
-
"step": 1470
|
| 1039 |
-
},
|
| 1040 |
-
{
|
| 1041 |
-
"epoch": 0.9181141439205955,
|
| 1042 |
-
"grad_norm": 2.328125,
|
| 1043 |
-
"learning_rate": 1.820689655172414e-06,
|
| 1044 |
-
"loss": 0.9602,
|
| 1045 |
-
"step": 1480
|
| 1046 |
-
},
|
| 1047 |
-
{
|
| 1048 |
-
"epoch": 0.924317617866005,
|
| 1049 |
-
"grad_norm": 2.234375,
|
| 1050 |
-
"learning_rate": 1.6827586206896553e-06,
|
| 1051 |
-
"loss": 1.0289,
|
| 1052 |
-
"step": 1490
|
| 1053 |
-
},
|
| 1054 |
-
{
|
| 1055 |
-
"epoch": 0.9305210918114144,
|
| 1056 |
-
"grad_norm": 2.359375,
|
| 1057 |
-
"learning_rate": 1.5448275862068968e-06,
|
| 1058 |
-
"loss": 0.9785,
|
| 1059 |
-
"step": 1500
|
| 1060 |
-
},
|
| 1061 |
-
{
|
| 1062 |
-
"epoch": 0.9367245657568238,
|
| 1063 |
-
"grad_norm": 2.4375,
|
| 1064 |
-
"learning_rate": 1.4068965517241381e-06,
|
| 1065 |
-
"loss": 1.0277,
|
| 1066 |
-
"step": 1510
|
| 1067 |
-
},
|
| 1068 |
-
{
|
| 1069 |
-
"epoch": 0.9429280397022333,
|
| 1070 |
-
"grad_norm": 2.109375,
|
| 1071 |
-
"learning_rate": 1.2689655172413794e-06,
|
| 1072 |
-
"loss": 0.982,
|
| 1073 |
-
"step": 1520
|
| 1074 |
-
},
|
| 1075 |
-
{
|
| 1076 |
-
"epoch": 0.9491315136476427,
|
| 1077 |
-
"grad_norm": 2.265625,
|
| 1078 |
-
"learning_rate": 1.1310344827586207e-06,
|
| 1079 |
-
"loss": 0.9723,
|
| 1080 |
-
"step": 1530
|
| 1081 |
-
},
|
| 1082 |
-
{
|
| 1083 |
-
"epoch": 0.9553349875930521,
|
| 1084 |
-
"grad_norm": 2.125,
|
| 1085 |
-
"learning_rate": 9.931034482758622e-07,
|
| 1086 |
-
"loss": 0.9535,
|
| 1087 |
-
"step": 1540
|
| 1088 |
-
},
|
| 1089 |
-
{
|
| 1090 |
-
"epoch": 0.9615384615384616,
|
| 1091 |
-
"grad_norm": 2.21875,
|
| 1092 |
-
"learning_rate": 8.551724137931035e-07,
|
| 1093 |
-
"loss": 0.9652,
|
| 1094 |
-
"step": 1550
|
| 1095 |
-
},
|
| 1096 |
-
{
|
| 1097 |
-
"epoch": 0.967741935483871,
|
| 1098 |
-
"grad_norm": 2.515625,
|
| 1099 |
-
"learning_rate": 7.172413793103449e-07,
|
| 1100 |
-
"loss": 0.9527,
|
| 1101 |
-
"step": 1560
|
| 1102 |
-
},
|
| 1103 |
-
{
|
| 1104 |
-
"epoch": 0.9739454094292804,
|
| 1105 |
-
"grad_norm": 2.25,
|
| 1106 |
-
"learning_rate": 5.793103448275863e-07,
|
| 1107 |
-
"loss": 0.9809,
|
| 1108 |
-
"step": 1570
|
| 1109 |
-
},
|
| 1110 |
-
{
|
| 1111 |
-
"epoch": 0.9801488833746899,
|
| 1112 |
-
"grad_norm": 2.28125,
|
| 1113 |
-
"learning_rate": 4.413793103448276e-07,
|
| 1114 |
-
"loss": 0.9852,
|
| 1115 |
-
"step": 1580
|
| 1116 |
-
},
|
| 1117 |
-
{
|
| 1118 |
-
"epoch": 0.9863523573200993,
|
| 1119 |
-
"grad_norm": 2.109375,
|
| 1120 |
-
"learning_rate": 3.0344827586206897e-07,
|
| 1121 |
-
"loss": 0.9164,
|
| 1122 |
-
"step": 1590
|
| 1123 |
-
},
|
| 1124 |
-
{
|
| 1125 |
-
"epoch": 0.9925558312655087,
|
| 1126 |
-
"grad_norm": 2.40625,
|
| 1127 |
-
"learning_rate": 1.6551724137931037e-07,
|
| 1128 |
-
"loss": 0.9355,
|
| 1129 |
-
"step": 1600
|
| 1130 |
-
},
|
| 1131 |
-
{
|
| 1132 |
-
"epoch": 0.9987593052109182,
|
| 1133 |
-
"grad_norm": 2.015625,
|
| 1134 |
-
"learning_rate": 2.7586206896551723e-08,
|
| 1135 |
-
"loss": 0.9949,
|
| 1136 |
-
"step": 1610
|
| 1137 |
-
},
|
| 1138 |
{
|
| 1139 |
"epoch": 1.0,
|
| 1140 |
-
"step":
|
| 1141 |
-
"total_flos":
|
| 1142 |
-
"train_loss": 1.
|
| 1143 |
-
"train_runtime":
|
| 1144 |
-
"train_samples_per_second":
|
| 1145 |
-
"train_steps_per_second": 0.
|
| 1146 |
}
|
| 1147 |
],
|
| 1148 |
"logging_steps": 10,
|
| 1149 |
-
"max_steps":
|
| 1150 |
"num_input_tokens_seen": 0,
|
| 1151 |
"num_train_epochs": 1,
|
| 1152 |
"save_steps": 100,
|
|
@@ -1162,8 +952,8 @@
|
|
| 1162 |
"attributes": {}
|
| 1163 |
}
|
| 1164 |
},
|
| 1165 |
-
"total_flos":
|
| 1166 |
-
"train_batch_size":
|
| 1167 |
"trial_name": null,
|
| 1168 |
"trial_params": null
|
| 1169 |
}
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1312,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.007621951219512195,
|
| 13 |
+
"grad_norm": 4.6875,
|
| 14 |
+
"learning_rate": 1.5151515151515152e-06,
|
| 15 |
+
"loss": 1.6918,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.01524390243902439,
|
| 20 |
+
"grad_norm": 5.0625,
|
| 21 |
+
"learning_rate": 3.0303030303030305e-06,
|
| 22 |
+
"loss": 1.6227,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.022865853658536585,
|
| 27 |
+
"grad_norm": 4.34375,
|
| 28 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 29 |
+
"loss": 1.6367,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03048780487804878,
|
| 34 |
+
"grad_norm": 4.21875,
|
| 35 |
+
"learning_rate": 6.060606060606061e-06,
|
| 36 |
+
"loss": 1.493,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.038109756097560975,
|
| 41 |
"grad_norm": 2.921875,
|
| 42 |
+
"learning_rate": 7.5757575757575764e-06,
|
| 43 |
+
"loss": 1.441,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.04573170731707317,
|
| 48 |
+
"grad_norm": 3.078125,
|
| 49 |
+
"learning_rate": 9.090909090909091e-06,
|
| 50 |
+
"loss": 1.3488,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.053353658536585365,
|
| 55 |
+
"grad_norm": 2.828125,
|
| 56 |
+
"learning_rate": 1.0606060606060606e-05,
|
| 57 |
+
"loss": 1.3289,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.06097560975609756,
|
| 62 |
+
"grad_norm": 2.765625,
|
| 63 |
+
"learning_rate": 1.2121212121212122e-05,
|
| 64 |
+
"loss": 1.2742,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.06859756097560976,
|
| 69 |
+
"grad_norm": 2.890625,
|
| 70 |
+
"learning_rate": 1.3636363636363637e-05,
|
| 71 |
+
"loss": 1.2758,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.07621951219512195,
|
| 76 |
+
"grad_norm": 2.859375,
|
| 77 |
+
"learning_rate": 1.5151515151515153e-05,
|
| 78 |
+
"loss": 1.2707,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.08384146341463415,
|
| 83 |
+
"grad_norm": 2.890625,
|
| 84 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 85 |
+
"loss": 1.273,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.09146341463414634,
|
| 90 |
+
"grad_norm": 2.765625,
|
| 91 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 92 |
+
"loss": 1.2414,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.09908536585365854,
|
| 97 |
+
"grad_norm": 2.75,
|
| 98 |
+
"learning_rate": 1.96969696969697e-05,
|
| 99 |
+
"loss": 1.275,
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.10670731707317073,
|
| 104 |
+
"grad_norm": 2.78125,
|
| 105 |
+
"learning_rate": 1.986440677966102e-05,
|
| 106 |
+
"loss": 1.2547,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.11432926829268293,
|
| 111 |
+
"grad_norm": 2.578125,
|
| 112 |
+
"learning_rate": 1.969491525423729e-05,
|
| 113 |
+
"loss": 1.2391,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.12195121951219512,
|
| 118 |
+
"grad_norm": 2.75,
|
| 119 |
+
"learning_rate": 1.9525423728813562e-05,
|
| 120 |
+
"loss": 1.2379,
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.12957317073170732,
|
| 125 |
+
"grad_norm": 2.65625,
|
| 126 |
+
"learning_rate": 1.9355932203389832e-05,
|
| 127 |
+
"loss": 1.2105,
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.13719512195121952,
|
| 132 |
+
"grad_norm": 2.53125,
|
| 133 |
+
"learning_rate": 1.9186440677966102e-05,
|
| 134 |
+
"loss": 1.2359,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.1448170731707317,
|
| 139 |
+
"grad_norm": 2.75,
|
| 140 |
+
"learning_rate": 1.9016949152542375e-05,
|
| 141 |
+
"loss": 1.2867,
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.1524390243902439,
|
| 146 |
+
"grad_norm": 2.8125,
|
| 147 |
+
"learning_rate": 1.8847457627118645e-05,
|
| 148 |
+
"loss": 1.1762,
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.1600609756097561,
|
| 153 |
+
"grad_norm": 2.765625,
|
| 154 |
+
"learning_rate": 1.8677966101694918e-05,
|
| 155 |
+
"loss": 1.1992,
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.1676829268292683,
|
| 160 |
+
"grad_norm": 2.75,
|
| 161 |
+
"learning_rate": 1.8508474576271188e-05,
|
| 162 |
+
"loss": 1.232,
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.17530487804878048,
|
| 167 |
+
"grad_norm": 2.65625,
|
| 168 |
+
"learning_rate": 1.8338983050847458e-05,
|
| 169 |
+
"loss": 1.2172,
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.18292682926829268,
|
| 174 |
+
"grad_norm": 2.734375,
|
| 175 |
+
"learning_rate": 1.816949152542373e-05,
|
| 176 |
+
"loss": 1.182,
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.19054878048780488,
|
| 181 |
+
"grad_norm": 2.875,
|
| 182 |
+
"learning_rate": 1.8e-05,
|
| 183 |
+
"loss": 1.2492,
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.19817073170731708,
|
| 188 |
+
"grad_norm": 2.71875,
|
| 189 |
+
"learning_rate": 1.7830508474576274e-05,
|
| 190 |
+
"loss": 1.2312,
|
| 191 |
"step": 260
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.20579268292682926,
|
| 195 |
+
"grad_norm": 2.65625,
|
| 196 |
+
"learning_rate": 1.7661016949152543e-05,
|
| 197 |
+
"loss": 1.241,
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.21341463414634146,
|
| 202 |
+
"grad_norm": 2.78125,
|
| 203 |
+
"learning_rate": 1.7491525423728813e-05,
|
| 204 |
+
"loss": 1.1879,
|
| 205 |
"step": 280
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.22103658536585366,
|
| 209 |
+
"grad_norm": 2.734375,
|
| 210 |
+
"learning_rate": 1.7322033898305086e-05,
|
| 211 |
+
"loss": 1.2211,
|
| 212 |
"step": 290
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.22865853658536586,
|
| 216 |
+
"grad_norm": 2.734375,
|
| 217 |
+
"learning_rate": 1.715254237288136e-05,
|
| 218 |
+
"loss": 1.2152,
|
| 219 |
"step": 300
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.23628048780487804,
|
| 223 |
+
"grad_norm": 2.8125,
|
| 224 |
+
"learning_rate": 1.698305084745763e-05,
|
| 225 |
+
"loss": 1.1938,
|
| 226 |
"step": 310
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.24390243902439024,
|
| 230 |
+
"grad_norm": 2.71875,
|
| 231 |
+
"learning_rate": 1.68135593220339e-05,
|
| 232 |
+
"loss": 1.2266,
|
| 233 |
"step": 320
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.25152439024390244,
|
| 237 |
+
"grad_norm": 2.84375,
|
| 238 |
+
"learning_rate": 1.6644067796610172e-05,
|
| 239 |
+
"loss": 1.1672,
|
| 240 |
"step": 330
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.25914634146341464,
|
| 244 |
+
"grad_norm": 2.71875,
|
| 245 |
+
"learning_rate": 1.6474576271186442e-05,
|
| 246 |
+
"loss": 1.2104,
|
| 247 |
"step": 340
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.26676829268292684,
|
| 251 |
+
"grad_norm": 2.640625,
|
| 252 |
+
"learning_rate": 1.6305084745762715e-05,
|
| 253 |
+
"loss": 1.2098,
|
| 254 |
"step": 350
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.27439024390243905,
|
| 258 |
+
"grad_norm": 2.703125,
|
| 259 |
+
"learning_rate": 1.6135593220338985e-05,
|
| 260 |
+
"loss": 1.1684,
|
| 261 |
"step": 360
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.2820121951219512,
|
| 265 |
+
"grad_norm": 2.765625,
|
| 266 |
+
"learning_rate": 1.5966101694915255e-05,
|
| 267 |
+
"loss": 1.1734,
|
| 268 |
"step": 370
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.2896341463414634,
|
| 272 |
+
"grad_norm": 2.546875,
|
| 273 |
+
"learning_rate": 1.5796610169491528e-05,
|
| 274 |
+
"loss": 1.2008,
|
| 275 |
"step": 380
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.2972560975609756,
|
| 279 |
+
"grad_norm": 2.78125,
|
| 280 |
+
"learning_rate": 1.5627118644067798e-05,
|
| 281 |
+
"loss": 1.1848,
|
| 282 |
"step": 390
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.3048780487804878,
|
| 286 |
+
"grad_norm": 2.625,
|
| 287 |
+
"learning_rate": 1.545762711864407e-05,
|
| 288 |
+
"loss": 1.2033,
|
| 289 |
"step": 400
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.3125,
|
| 293 |
+
"grad_norm": 2.625,
|
| 294 |
+
"learning_rate": 1.528813559322034e-05,
|
| 295 |
+
"loss": 1.1781,
|
| 296 |
"step": 410
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.3201219512195122,
|
| 300 |
+
"grad_norm": 2.96875,
|
| 301 |
+
"learning_rate": 1.511864406779661e-05,
|
| 302 |
+
"loss": 1.2043,
|
| 303 |
"step": 420
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.3277439024390244,
|
| 307 |
+
"grad_norm": 2.453125,
|
| 308 |
+
"learning_rate": 1.4949152542372882e-05,
|
| 309 |
+
"loss": 1.1699,
|
| 310 |
"step": 430
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.3353658536585366,
|
| 314 |
+
"grad_norm": 2.5625,
|
| 315 |
+
"learning_rate": 1.4779661016949153e-05,
|
| 316 |
+
"loss": 1.1871,
|
| 317 |
"step": 440
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.3429878048780488,
|
| 321 |
+
"grad_norm": 2.578125,
|
| 322 |
+
"learning_rate": 1.4610169491525426e-05,
|
| 323 |
+
"loss": 1.1418,
|
| 324 |
"step": 450
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.35060975609756095,
|
| 328 |
+
"grad_norm": 2.640625,
|
| 329 |
+
"learning_rate": 1.4440677966101698e-05,
|
| 330 |
+
"loss": 1.1824,
|
| 331 |
"step": 460
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.35823170731707316,
|
| 335 |
+
"grad_norm": 2.828125,
|
| 336 |
+
"learning_rate": 1.4271186440677966e-05,
|
| 337 |
+
"loss": 1.176,
|
| 338 |
"step": 470
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.36585365853658536,
|
| 342 |
+
"grad_norm": 2.96875,
|
| 343 |
+
"learning_rate": 1.4101694915254239e-05,
|
| 344 |
+
"loss": 1.1785,
|
| 345 |
"step": 480
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.37347560975609756,
|
| 349 |
+
"grad_norm": 2.796875,
|
| 350 |
+
"learning_rate": 1.393220338983051e-05,
|
| 351 |
+
"loss": 1.1758,
|
| 352 |
"step": 490
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.38109756097560976,
|
| 356 |
+
"grad_norm": 2.75,
|
| 357 |
+
"learning_rate": 1.3762711864406782e-05,
|
| 358 |
+
"loss": 1.1988,
|
| 359 |
"step": 500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.38871951219512196,
|
| 363 |
+
"grad_norm": 2.578125,
|
| 364 |
+
"learning_rate": 1.3593220338983053e-05,
|
| 365 |
+
"loss": 1.1822,
|
| 366 |
"step": 510
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.39634146341463417,
|
| 370 |
+
"grad_norm": 2.953125,
|
| 371 |
+
"learning_rate": 1.3423728813559323e-05,
|
| 372 |
+
"loss": 1.1801,
|
| 373 |
"step": 520
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.40396341463414637,
|
| 377 |
+
"grad_norm": 2.640625,
|
| 378 |
+
"learning_rate": 1.3254237288135595e-05,
|
| 379 |
+
"loss": 1.2043,
|
| 380 |
"step": 530
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.4115853658536585,
|
| 384 |
+
"grad_norm": 2.546875,
|
| 385 |
+
"learning_rate": 1.3084745762711866e-05,
|
| 386 |
+
"loss": 1.2098,
|
| 387 |
"step": 540
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.4192073170731707,
|
| 391 |
+
"grad_norm": 2.625,
|
| 392 |
+
"learning_rate": 1.2915254237288137e-05,
|
| 393 |
+
"loss": 1.1973,
|
| 394 |
"step": 550
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.4268292682926829,
|
| 398 |
+
"grad_norm": 2.546875,
|
| 399 |
+
"learning_rate": 1.2745762711864407e-05,
|
| 400 |
+
"loss": 1.2066,
|
| 401 |
"step": 560
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.4344512195121951,
|
| 405 |
+
"grad_norm": 2.4375,
|
| 406 |
+
"learning_rate": 1.2576271186440679e-05,
|
| 407 |
+
"loss": 1.1543,
|
| 408 |
"step": 570
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.4420731707317073,
|
| 412 |
+
"grad_norm": 2.640625,
|
| 413 |
+
"learning_rate": 1.240677966101695e-05,
|
| 414 |
+
"loss": 1.1703,
|
| 415 |
"step": 580
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.4496951219512195,
|
| 419 |
+
"grad_norm": 3.09375,
|
| 420 |
+
"learning_rate": 1.2237288135593222e-05,
|
| 421 |
+
"loss": 1.1566,
|
| 422 |
"step": 590
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.4573170731707317,
|
| 426 |
+
"grad_norm": 2.578125,
|
| 427 |
+
"learning_rate": 1.2067796610169493e-05,
|
| 428 |
+
"loss": 1.1623,
|
| 429 |
"step": 600
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.4649390243902439,
|
| 433 |
+
"grad_norm": 2.734375,
|
| 434 |
+
"learning_rate": 1.1898305084745763e-05,
|
| 435 |
+
"loss": 1.198,
|
| 436 |
"step": 610
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.4725609756097561,
|
| 440 |
+
"grad_norm": 2.78125,
|
| 441 |
+
"learning_rate": 1.1728813559322034e-05,
|
| 442 |
+
"loss": 1.1971,
|
| 443 |
"step": 620
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.4801829268292683,
|
| 447 |
+
"grad_norm": 2.546875,
|
| 448 |
+
"learning_rate": 1.1559322033898306e-05,
|
| 449 |
+
"loss": 1.159,
|
| 450 |
"step": 630
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.4878048780487805,
|
| 454 |
+
"grad_norm": 2.84375,
|
| 455 |
+
"learning_rate": 1.1389830508474577e-05,
|
| 456 |
+
"loss": 1.184,
|
| 457 |
"step": 640
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.4954268292682927,
|
| 461 |
+
"grad_norm": 2.578125,
|
| 462 |
+
"learning_rate": 1.1220338983050849e-05,
|
| 463 |
+
"loss": 1.1859,
|
| 464 |
"step": 650
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.5030487804878049,
|
| 468 |
+
"grad_norm": 2.609375,
|
| 469 |
+
"learning_rate": 1.1050847457627118e-05,
|
| 470 |
+
"loss": 1.1434,
|
| 471 |
"step": 660
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.510670731707317,
|
| 475 |
+
"grad_norm": 2.671875,
|
| 476 |
+
"learning_rate": 1.088135593220339e-05,
|
| 477 |
+
"loss": 1.1775,
|
| 478 |
"step": 670
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.5182926829268293,
|
| 482 |
+
"grad_norm": 2.515625,
|
| 483 |
+
"learning_rate": 1.0711864406779661e-05,
|
| 484 |
+
"loss": 1.15,
|
| 485 |
"step": 680
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.5259146341463414,
|
| 489 |
+
"grad_norm": 2.609375,
|
| 490 |
+
"learning_rate": 1.0542372881355933e-05,
|
| 491 |
+
"loss": 1.1613,
|
| 492 |
"step": 690
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.5335365853658537,
|
| 496 |
+
"grad_norm": 2.5625,
|
| 497 |
+
"learning_rate": 1.0372881355932204e-05,
|
| 498 |
+
"loss": 1.1602,
|
| 499 |
"step": 700
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.5411585365853658,
|
| 503 |
+
"grad_norm": 2.671875,
|
| 504 |
+
"learning_rate": 1.0203389830508474e-05,
|
| 505 |
+
"loss": 1.1879,
|
| 506 |
"step": 710
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.5487804878048781,
|
| 510 |
+
"grad_norm": 2.625,
|
| 511 |
+
"learning_rate": 1.0033898305084746e-05,
|
| 512 |
+
"loss": 1.1941,
|
| 513 |
"step": 720
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.5564024390243902,
|
| 517 |
+
"grad_norm": 2.8125,
|
| 518 |
+
"learning_rate": 9.864406779661017e-06,
|
| 519 |
+
"loss": 1.1574,
|
| 520 |
"step": 730
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.5640243902439024,
|
| 524 |
+
"grad_norm": 2.75,
|
| 525 |
+
"learning_rate": 9.69491525423729e-06,
|
| 526 |
+
"loss": 1.1793,
|
| 527 |
"step": 740
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.5716463414634146,
|
| 531 |
+
"grad_norm": 2.703125,
|
| 532 |
+
"learning_rate": 9.52542372881356e-06,
|
| 533 |
+
"loss": 1.1711,
|
| 534 |
"step": 750
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.5792682926829268,
|
| 538 |
+
"grad_norm": 2.703125,
|
| 539 |
+
"learning_rate": 9.355932203389831e-06,
|
| 540 |
+
"loss": 1.1875,
|
| 541 |
"step": 760
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.586890243902439,
|
| 545 |
+
"grad_norm": 2.8125,
|
| 546 |
+
"learning_rate": 9.186440677966101e-06,
|
| 547 |
+
"loss": 1.19,
|
| 548 |
"step": 770
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.5945121951219512,
|
| 552 |
+
"grad_norm": 2.9375,
|
| 553 |
+
"learning_rate": 9.016949152542374e-06,
|
| 554 |
+
"loss": 1.141,
|
| 555 |
"step": 780
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.6021341463414634,
|
| 559 |
+
"grad_norm": 2.75,
|
| 560 |
+
"learning_rate": 8.847457627118646e-06,
|
| 561 |
+
"loss": 1.1699,
|
| 562 |
"step": 790
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.6097560975609756,
|
| 566 |
+
"grad_norm": 2.65625,
|
| 567 |
+
"learning_rate": 8.677966101694915e-06,
|
| 568 |
+
"loss": 1.1809,
|
| 569 |
"step": 800
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.6173780487804879,
|
| 573 |
+
"grad_norm": 2.703125,
|
| 574 |
+
"learning_rate": 8.508474576271187e-06,
|
| 575 |
+
"loss": 1.1824,
|
| 576 |
"step": 810
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.625,
|
| 580 |
+
"grad_norm": 2.796875,
|
| 581 |
+
"learning_rate": 8.338983050847458e-06,
|
| 582 |
+
"loss": 1.1729,
|
| 583 |
"step": 820
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.6326219512195121,
|
| 587 |
+
"grad_norm": 2.765625,
|
| 588 |
+
"learning_rate": 8.16949152542373e-06,
|
| 589 |
+
"loss": 1.1113,
|
| 590 |
"step": 830
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.6402439024390244,
|
| 594 |
+
"grad_norm": 2.5,
|
| 595 |
+
"learning_rate": 8.000000000000001e-06,
|
| 596 |
+
"loss": 1.1434,
|
| 597 |
"step": 840
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.6478658536585366,
|
| 601 |
+
"grad_norm": 2.515625,
|
| 602 |
+
"learning_rate": 7.830508474576271e-06,
|
| 603 |
+
"loss": 1.1643,
|
| 604 |
"step": 850
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.6554878048780488,
|
| 608 |
+
"grad_norm": 2.65625,
|
| 609 |
+
"learning_rate": 7.661016949152543e-06,
|
| 610 |
+
"loss": 1.1738,
|
| 611 |
"step": 860
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.663109756097561,
|
| 615 |
+
"grad_norm": 2.609375,
|
| 616 |
+
"learning_rate": 7.491525423728814e-06,
|
| 617 |
+
"loss": 1.125,
|
| 618 |
"step": 870
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.6707317073170732,
|
| 622 |
+
"grad_norm": 2.5625,
|
| 623 |
+
"learning_rate": 7.3220338983050855e-06,
|
| 624 |
+
"loss": 1.1723,
|
| 625 |
"step": 880
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.6783536585365854,
|
| 629 |
+
"grad_norm": 2.546875,
|
| 630 |
+
"learning_rate": 7.152542372881357e-06,
|
| 631 |
+
"loss": 1.1617,
|
| 632 |
"step": 890
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.6859756097560976,
|
| 636 |
+
"grad_norm": 2.6875,
|
| 637 |
+
"learning_rate": 6.9830508474576275e-06,
|
| 638 |
+
"loss": 1.1445,
|
| 639 |
"step": 900
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.6935975609756098,
|
| 643 |
+
"grad_norm": 2.546875,
|
| 644 |
+
"learning_rate": 6.813559322033899e-06,
|
| 645 |
+
"loss": 1.1807,
|
| 646 |
"step": 910
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.7012195121951219,
|
| 650 |
+
"grad_norm": 2.65625,
|
| 651 |
+
"learning_rate": 6.64406779661017e-06,
|
| 652 |
+
"loss": 1.1865,
|
| 653 |
"step": 920
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.7088414634146342,
|
| 657 |
+
"grad_norm": 2.734375,
|
| 658 |
+
"learning_rate": 6.474576271186441e-06,
|
| 659 |
+
"loss": 1.198,
|
| 660 |
"step": 930
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.7164634146341463,
|
| 664 |
+
"grad_norm": 2.765625,
|
| 665 |
+
"learning_rate": 6.3050847457627125e-06,
|
| 666 |
+
"loss": 1.1734,
|
| 667 |
"step": 940
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.7240853658536586,
|
| 671 |
+
"grad_norm": 2.609375,
|
| 672 |
+
"learning_rate": 6.135593220338983e-06,
|
| 673 |
+
"loss": 1.1633,
|
| 674 |
"step": 950
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.7317073170731707,
|
| 678 |
+
"grad_norm": 2.578125,
|
| 679 |
+
"learning_rate": 5.9661016949152555e-06,
|
| 680 |
+
"loss": 1.1953,
|
| 681 |
"step": 960
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.739329268292683,
|
| 685 |
+
"grad_norm": 2.5,
|
| 686 |
+
"learning_rate": 5.796610169491525e-06,
|
| 687 |
+
"loss": 1.1879,
|
| 688 |
"step": 970
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.7469512195121951,
|
| 692 |
+
"grad_norm": 2.609375,
|
| 693 |
+
"learning_rate": 5.6271186440677975e-06,
|
| 694 |
+
"loss": 1.1664,
|
| 695 |
"step": 980
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 0.7545731707317073,
|
| 699 |
+
"grad_norm": 2.765625,
|
| 700 |
+
"learning_rate": 5.457627118644067e-06,
|
| 701 |
+
"loss": 1.1789,
|
| 702 |
"step": 990
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 0.7621951219512195,
|
| 706 |
+
"grad_norm": 2.890625,
|
| 707 |
+
"learning_rate": 5.28813559322034e-06,
|
| 708 |
+
"loss": 1.1738,
|
| 709 |
"step": 1000
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.7698170731707317,
|
| 713 |
+
"grad_norm": 2.6875,
|
| 714 |
+
"learning_rate": 5.118644067796611e-06,
|
| 715 |
+
"loss": 1.1582,
|
| 716 |
"step": 1010
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 0.7774390243902439,
|
| 720 |
+
"grad_norm": 2.9375,
|
| 721 |
+
"learning_rate": 4.949152542372882e-06,
|
| 722 |
+
"loss": 1.1391,
|
| 723 |
"step": 1020
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 0.7850609756097561,
|
| 727 |
+
"grad_norm": 2.578125,
|
| 728 |
+
"learning_rate": 4.779661016949153e-06,
|
| 729 |
+
"loss": 1.1652,
|
| 730 |
"step": 1030
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 0.7926829268292683,
|
| 734 |
+
"grad_norm": 2.703125,
|
| 735 |
+
"learning_rate": 4.610169491525424e-06,
|
| 736 |
+
"loss": 1.1426,
|
| 737 |
"step": 1040
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 0.8003048780487805,
|
| 741 |
+
"grad_norm": 2.71875,
|
| 742 |
+
"learning_rate": 4.440677966101695e-06,
|
| 743 |
+
"loss": 1.1713,
|
| 744 |
"step": 1050
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 0.8079268292682927,
|
| 748 |
+
"grad_norm": 2.515625,
|
| 749 |
+
"learning_rate": 4.271186440677967e-06,
|
| 750 |
+
"loss": 1.1727,
|
| 751 |
"step": 1060
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 0.8155487804878049,
|
| 755 |
+
"grad_norm": 2.59375,
|
| 756 |
+
"learning_rate": 4.101694915254237e-06,
|
| 757 |
+
"loss": 1.1695,
|
| 758 |
"step": 1070
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 0.823170731707317,
|
| 762 |
+
"grad_norm": 2.671875,
|
| 763 |
+
"learning_rate": 3.932203389830509e-06,
|
| 764 |
+
"loss": 1.1672,
|
| 765 |
"step": 1080
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 0.8307926829268293,
|
| 769 |
+
"grad_norm": 2.734375,
|
| 770 |
+
"learning_rate": 3.76271186440678e-06,
|
| 771 |
+
"loss": 1.2016,
|
| 772 |
"step": 1090
|
| 773 |
},
|
| 774 |
{
|
| 775 |
+
"epoch": 0.8384146341463414,
|
| 776 |
+
"grad_norm": 2.90625,
|
| 777 |
+
"learning_rate": 3.5932203389830512e-06,
|
| 778 |
+
"loss": 1.1727,
|
| 779 |
"step": 1100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 0.8460365853658537,
|
| 783 |
+
"grad_norm": 2.84375,
|
| 784 |
+
"learning_rate": 3.4237288135593223e-06,
|
| 785 |
+
"loss": 1.1799,
|
| 786 |
"step": 1110
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 0.8536585365853658,
|
| 790 |
+
"grad_norm": 2.8125,
|
| 791 |
+
"learning_rate": 3.2542372881355933e-06,
|
| 792 |
+
"loss": 1.1555,
|
| 793 |
"step": 1120
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 0.8612804878048781,
|
| 797 |
+
"grad_norm": 2.6875,
|
| 798 |
+
"learning_rate": 3.0847457627118648e-06,
|
| 799 |
+
"loss": 1.1678,
|
| 800 |
"step": 1130
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 0.8689024390243902,
|
| 804 |
+
"grad_norm": 2.59375,
|
| 805 |
+
"learning_rate": 2.915254237288136e-06,
|
| 806 |
+
"loss": 1.1578,
|
| 807 |
"step": 1140
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 0.8765243902439024,
|
| 811 |
+
"grad_norm": 2.375,
|
| 812 |
+
"learning_rate": 2.745762711864407e-06,
|
| 813 |
+
"loss": 1.1535,
|
| 814 |
"step": 1150
|
| 815 |
},
|
| 816 |
{
|
| 817 |
+
"epoch": 0.8841463414634146,
|
| 818 |
+
"grad_norm": 2.703125,
|
| 819 |
+
"learning_rate": 2.576271186440678e-06,
|
| 820 |
+
"loss": 1.1602,
|
| 821 |
"step": 1160
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 0.8917682926829268,
|
| 825 |
+
"grad_norm": 2.78125,
|
| 826 |
+
"learning_rate": 2.4067796610169493e-06,
|
| 827 |
+
"loss": 1.1852,
|
| 828 |
"step": 1170
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 0.899390243902439,
|
| 832 |
+
"grad_norm": 2.59375,
|
| 833 |
+
"learning_rate": 2.2372881355932204e-06,
|
| 834 |
+
"loss": 1.1605,
|
| 835 |
"step": 1180
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 0.9070121951219512,
|
| 839 |
+
"grad_norm": 2.59375,
|
| 840 |
+
"learning_rate": 2.0677966101694914e-06,
|
| 841 |
+
"loss": 1.1451,
|
| 842 |
"step": 1190
|
| 843 |
},
|
| 844 |
{
|
| 845 |
+
"epoch": 0.9146341463414634,
|
| 846 |
+
"grad_norm": 2.5625,
|
| 847 |
+
"learning_rate": 1.8983050847457629e-06,
|
| 848 |
+
"loss": 1.1496,
|
| 849 |
"step": 1200
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 0.9222560975609756,
|
| 853 |
+
"grad_norm": 2.859375,
|
| 854 |
+
"learning_rate": 1.728813559322034e-06,
|
| 855 |
+
"loss": 1.1582,
|
| 856 |
"step": 1210
|
| 857 |
},
|
| 858 |
{
|
| 859 |
+
"epoch": 0.9298780487804879,
|
| 860 |
+
"grad_norm": 2.65625,
|
| 861 |
+
"learning_rate": 1.5593220338983054e-06,
|
| 862 |
+
"loss": 1.227,
|
| 863 |
"step": 1220
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 0.9375,
|
| 867 |
+
"grad_norm": 2.703125,
|
| 868 |
+
"learning_rate": 1.3898305084745764e-06,
|
| 869 |
+
"loss": 1.1977,
|
| 870 |
"step": 1230
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"epoch": 0.9451219512195121,
|
| 874 |
+
"grad_norm": 2.671875,
|
| 875 |
+
"learning_rate": 1.2203389830508477e-06,
|
| 876 |
+
"loss": 1.1885,
|
| 877 |
"step": 1240
|
| 878 |
},
|
| 879 |
{
|
| 880 |
+
"epoch": 0.9527439024390244,
|
| 881 |
+
"grad_norm": 2.6875,
|
| 882 |
+
"learning_rate": 1.0508474576271187e-06,
|
| 883 |
+
"loss": 1.1549,
|
| 884 |
"step": 1250
|
| 885 |
},
|
| 886 |
{
|
| 887 |
+
"epoch": 0.9603658536585366,
|
| 888 |
+
"grad_norm": 2.53125,
|
| 889 |
+
"learning_rate": 8.813559322033899e-07,
|
| 890 |
+
"loss": 1.1781,
|
| 891 |
"step": 1260
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 0.9679878048780488,
|
| 895 |
+
"grad_norm": 2.5625,
|
| 896 |
+
"learning_rate": 7.118644067796611e-07,
|
| 897 |
+
"loss": 1.1957,
|
| 898 |
"step": 1270
|
| 899 |
},
|
| 900 |
{
|
| 901 |
+
"epoch": 0.975609756097561,
|
| 902 |
+
"grad_norm": 2.640625,
|
| 903 |
+
"learning_rate": 5.423728813559322e-07,
|
| 904 |
+
"loss": 1.1898,
|
| 905 |
"step": 1280
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 0.9832317073170732,
|
| 909 |
+
"grad_norm": 2.625,
|
| 910 |
+
"learning_rate": 3.7288135593220347e-07,
|
| 911 |
+
"loss": 1.1531,
|
| 912 |
"step": 1290
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 0.9908536585365854,
|
| 916 |
+
"grad_norm": 2.6875,
|
| 917 |
+
"learning_rate": 2.0338983050847458e-07,
|
| 918 |
+
"loss": 1.1773,
|
| 919 |
"step": 1300
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 0.9984756097560976,
|
| 923 |
+
"grad_norm": 2.640625,
|
| 924 |
+
"learning_rate": 3.3898305084745764e-08,
|
| 925 |
+
"loss": 1.0754,
|
| 926 |
"step": 1310
|
| 927 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 928 |
{
|
| 929 |
"epoch": 1.0,
|
| 930 |
+
"step": 1312,
|
| 931 |
+
"total_flos": 4.7710598211200614e+17,
|
| 932 |
+
"train_loss": 1.2026635146722562,
|
| 933 |
+
"train_runtime": 2875.3707,
|
| 934 |
+
"train_samples_per_second": 14.591,
|
| 935 |
+
"train_steps_per_second": 0.456
|
| 936 |
}
|
| 937 |
],
|
| 938 |
"logging_steps": 10,
|
| 939 |
+
"max_steps": 1312,
|
| 940 |
"num_input_tokens_seen": 0,
|
| 941 |
"num_train_epochs": 1,
|
| 942 |
"save_steps": 100,
|
|
|
|
| 952 |
"attributes": {}
|
| 953 |
}
|
| 954 |
},
|
| 955 |
+
"total_flos": 4.7710598211200614e+17,
|
| 956 |
+
"train_batch_size": 16,
|
| 957 |
"trial_name": null,
|
| 958 |
"trial_params": null
|
| 959 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21fa790fc457d5b6b0d17aa60aa932cb9b96c1c3eeb7b854b96ec867dcb5facf
|
| 3 |
size 5176
|