Model save
Browse files- README.md +3 -2
- all_results.json +6 -6
- generation_config.json +2 -3
- train_results.json +6 -6
- trainer_state.json +1892 -324
README.md
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
model_name: PreThink_MemAgent
|
| 4 |
tags:
|
|
@@ -10,7 +11,7 @@ licence: license
|
|
| 10 |
|
| 11 |
# Model Card for PreThink_MemAgent
|
| 12 |
|
| 13 |
-
This model is a fine-tuned version of [
|
| 14 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 15 |
|
| 16 |
## Quick start
|
|
@@ -26,7 +27,7 @@ print(output["generated_text"])
|
|
| 26 |
|
| 27 |
## Training procedure
|
| 28 |
|
| 29 |
-
[<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/mingzeli/PreThink_MemAgent/runs/
|
| 30 |
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model: Qwen/Qwen3-4B
|
| 3 |
library_name: transformers
|
| 4 |
model_name: PreThink_MemAgent
|
| 5 |
tags:
|
|
|
|
| 11 |
|
| 12 |
# Model Card for PreThink_MemAgent
|
| 13 |
|
| 14 |
+
This model is a fine-tuned version of [Qwen/Qwen3-4B](https://huggingface.co/Qwen/Qwen3-4B).
|
| 15 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
|
| 17 |
## Quick start
|
|
|
|
| 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/mingzeli/PreThink_MemAgent/runs/59jzce83)
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with SFT.
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"total_flos":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 90995475070976.0,
|
| 3 |
+
"train_loss": 1.261348171858117,
|
| 4 |
+
"train_runtime": 1909.4501,
|
| 5 |
+
"train_samples": 4095,
|
| 6 |
+
"train_samples_per_second": 4.289,
|
| 7 |
+
"train_steps_per_second": 0.134
|
| 8 |
}
|
generation_config.json
CHANGED
|
@@ -3,9 +3,8 @@
|
|
| 3 |
"do_sample": true,
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
"pad_token_id": 151643,
|
| 6 |
-
"
|
| 7 |
-
"temperature": 0.7,
|
| 8 |
"top_k": 20,
|
| 9 |
-
"top_p": 0.
|
| 10 |
"transformers_version": "4.52.3"
|
| 11 |
}
|
|
|
|
| 3 |
"do_sample": true,
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
"pad_token_id": 151643,
|
| 6 |
+
"temperature": 0.6,
|
|
|
|
| 7 |
"top_k": 20,
|
| 8 |
+
"top_p": 0.95,
|
| 9 |
"transformers_version": "4.52.3"
|
| 10 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"total_flos":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 90995475070976.0,
|
| 3 |
+
"train_loss": 1.261348171858117,
|
| 4 |
+
"train_runtime": 1909.4501,
|
| 5 |
+
"train_samples": 4095,
|
| 6 |
+
"train_samples_per_second": 4.289,
|
| 7 |
+
"train_steps_per_second": 0.134
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -4,521 +4,2089 @@
|
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
"epoch": 2.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"epoch": 0.
|
| 14 |
-
"grad_norm":
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
-
"loss":
|
| 17 |
-
"num_tokens":
|
| 18 |
"step": 1
|
| 19 |
},
|
| 20 |
{
|
| 21 |
-
"epoch": 0.
|
| 22 |
-
"grad_norm": 19.
|
| 23 |
-
"learning_rate":
|
| 24 |
-
"loss": 2.
|
| 25 |
-
"num_tokens":
|
| 26 |
"step": 2
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"grad_norm":
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
-
"num_tokens":
|
| 34 |
"step": 3
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"epoch": 0.
|
| 38 |
-
"grad_norm":
|
| 39 |
-
"learning_rate":
|
| 40 |
-
"loss":
|
| 41 |
-
"num_tokens":
|
| 42 |
"step": 4
|
| 43 |
},
|
| 44 |
{
|
| 45 |
-
"epoch": 0.
|
| 46 |
-
"grad_norm":
|
| 47 |
-
"learning_rate":
|
| 48 |
-
"loss":
|
| 49 |
-
"num_tokens":
|
| 50 |
"step": 5
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
-
"grad_norm":
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss": 1.
|
| 57 |
-
"num_tokens":
|
| 58 |
"step": 6
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
-
"num_tokens":
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss":
|
| 73 |
-
"num_tokens":
|
| 74 |
"step": 8
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
-
"grad_norm":
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
| 81 |
-
"num_tokens":
|
| 82 |
"step": 9
|
| 83 |
},
|
| 84 |
{
|
| 85 |
-
"epoch": 0.
|
| 86 |
-
"grad_norm":
|
| 87 |
-
"learning_rate": 9.
|
| 88 |
-
"loss":
|
| 89 |
-
"num_tokens":
|
| 90 |
"step": 10
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 0.
|
| 94 |
-
"grad_norm":
|
| 95 |
-
"learning_rate": 9.
|
| 96 |
-
"loss":
|
| 97 |
-
"num_tokens":
|
| 98 |
"step": 11
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
-
"grad_norm": 1.
|
| 103 |
-
"learning_rate": 9.
|
| 104 |
-
"loss":
|
| 105 |
-
"num_tokens":
|
| 106 |
"step": 12
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch": 0.
|
| 110 |
-
"grad_norm": 1.
|
| 111 |
-
"learning_rate": 9.
|
| 112 |
-
"loss":
|
| 113 |
-
"num_tokens":
|
| 114 |
"step": 13
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 1.
|
| 119 |
-
"learning_rate": 9.
|
| 120 |
-
"loss":
|
| 121 |
-
"num_tokens":
|
| 122 |
"step": 14
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate": 9.
|
| 128 |
-
"loss":
|
| 129 |
-
"num_tokens":
|
| 130 |
"step": 15
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch": 0.
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss":
|
| 137 |
-
"num_tokens":
|
| 138 |
"step": 16
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 0.
|
| 142 |
-
"grad_norm":
|
| 143 |
-
"learning_rate":
|
| 144 |
-
"loss":
|
| 145 |
-
"num_tokens":
|
| 146 |
"step": 17
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 150 |
-
"grad_norm": 0.
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss":
|
| 153 |
-
"num_tokens":
|
| 154 |
"step": 18
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"epoch": 0.
|
| 158 |
-
"grad_norm": 0.
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss":
|
| 161 |
-
"num_tokens":
|
| 162 |
"step": 19
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch": 0.
|
| 166 |
-
"grad_norm": 0.
|
| 167 |
-
"learning_rate":
|
| 168 |
-
"loss":
|
| 169 |
-
"num_tokens":
|
| 170 |
"step": 20
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
-
"num_tokens":
|
| 178 |
"step": 21
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss":
|
| 185 |
-
"num_tokens":
|
| 186 |
"step": 22
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch": 0.
|
| 190 |
-
"grad_norm": 0.
|
| 191 |
-
"learning_rate":
|
| 192 |
-
"loss":
|
| 193 |
-
"num_tokens":
|
| 194 |
"step": 23
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"grad_norm": 0.
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss":
|
| 201 |
-
"num_tokens":
|
| 202 |
"step": 24
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch": 0.
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss":
|
| 209 |
-
"num_tokens":
|
| 210 |
"step": 25
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch": 0.
|
| 214 |
-
"grad_norm": 0.
|
| 215 |
-
"learning_rate":
|
| 216 |
-
"loss":
|
| 217 |
-
"num_tokens":
|
| 218 |
"step": 26
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"grad_norm": 0.
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss":
|
| 225 |
-
"num_tokens":
|
| 226 |
"step": 27
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
-
"num_tokens":
|
| 234 |
"step": 28
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss":
|
| 241 |
-
"num_tokens":
|
| 242 |
"step": 29
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"grad_norm":
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss":
|
| 249 |
-
"num_tokens":
|
| 250 |
"step": 30
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch":
|
| 254 |
-
"
|
| 255 |
-
"
|
| 256 |
-
"
|
| 257 |
-
"
|
| 258 |
-
"eval_steps_per_second": 9.575,
|
| 259 |
-
"step": 30
|
| 260 |
-
},
|
| 261 |
-
{
|
| 262 |
-
"epoch": 1.0334728033472804,
|
| 263 |
-
"grad_norm": 0.7928451711132057,
|
| 264 |
-
"learning_rate": 5.74362508863438e-06,
|
| 265 |
-
"loss": 0.5278,
|
| 266 |
-
"num_tokens": 3877387.0,
|
| 267 |
"step": 31
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"grad_norm": 0.
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss":
|
| 274 |
-
"num_tokens":
|
| 275 |
"step": 32
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss":
|
| 282 |
-
"num_tokens":
|
| 283 |
"step": 33
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch":
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss":
|
| 290 |
-
"num_tokens":
|
| 291 |
"step": 34
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch":
|
| 295 |
-
"grad_norm": 0.
|
| 296 |
-
"learning_rate":
|
| 297 |
-
"loss":
|
| 298 |
-
"num_tokens":
|
| 299 |
"step": 35
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
-
"learning_rate":
|
| 305 |
-
"loss":
|
| 306 |
-
"num_tokens":
|
| 307 |
"step": 36
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch":
|
| 311 |
-
"grad_norm": 0.
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss":
|
| 314 |
-
"num_tokens":
|
| 315 |
"step": 37
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch":
|
| 319 |
-
"grad_norm": 0.
|
| 320 |
-
"learning_rate":
|
| 321 |
-
"loss":
|
| 322 |
-
"num_tokens":
|
| 323 |
"step": 38
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"grad_norm": 0.
|
| 328 |
-
"learning_rate":
|
| 329 |
-
"loss":
|
| 330 |
-
"num_tokens":
|
| 331 |
"step": 39
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch":
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss":
|
| 338 |
-
"num_tokens":
|
| 339 |
"step": 40
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss":
|
| 346 |
-
"num_tokens":
|
| 347 |
"step": 41
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"grad_norm":
|
| 352 |
-
"learning_rate":
|
| 353 |
-
"loss":
|
| 354 |
-
"num_tokens":
|
| 355 |
"step": 42
|
| 356 |
},
|
| 357 |
{
|
| 358 |
-
"epoch":
|
| 359 |
-
"grad_norm": 0.
|
| 360 |
-
"learning_rate":
|
| 361 |
-
"loss":
|
| 362 |
-
"num_tokens":
|
| 363 |
"step": 43
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
-
"grad_norm": 0.
|
| 368 |
-
"learning_rate":
|
| 369 |
-
"loss":
|
| 370 |
-
"num_tokens":
|
| 371 |
"step": 44
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
-
"learning_rate":
|
| 377 |
-
"loss":
|
| 378 |
-
"num_tokens":
|
| 379 |
"step": 45
|
| 380 |
},
|
| 381 |
{
|
| 382 |
-
"epoch":
|
| 383 |
-
"grad_norm": 0.
|
| 384 |
-
"learning_rate":
|
| 385 |
-
"loss":
|
| 386 |
-
"num_tokens":
|
| 387 |
"step": 46
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch":
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss":
|
| 394 |
-
"num_tokens":
|
| 395 |
"step": 47
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"grad_norm": 0.
|
| 400 |
-
"learning_rate":
|
| 401 |
-
"loss":
|
| 402 |
-
"num_tokens":
|
| 403 |
"step": 48
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss":
|
| 410 |
-
"num_tokens":
|
| 411 |
"step": 49
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch":
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss":
|
| 418 |
-
"num_tokens":
|
| 419 |
"step": 50
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"grad_norm": 0.
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss":
|
| 426 |
-
"num_tokens":
|
| 427 |
"step": 51
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch":
|
| 431 |
-
"grad_norm": 0.
|
| 432 |
-
"learning_rate":
|
| 433 |
-
"loss":
|
| 434 |
-
"num_tokens":
|
| 435 |
"step": 52
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"grad_norm": 0.
|
| 440 |
-
"learning_rate":
|
| 441 |
-
"loss":
|
| 442 |
-
"num_tokens":
|
| 443 |
"step": 53
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss":
|
| 450 |
-
"num_tokens":
|
| 451 |
"step": 54
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch":
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
-
"learning_rate":
|
| 457 |
-
"loss":
|
| 458 |
-
"num_tokens":
|
| 459 |
"step": 55
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
-
"learning_rate":
|
| 465 |
-
"loss":
|
| 466 |
-
"num_tokens":
|
| 467 |
"step": 56
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss":
|
| 474 |
-
"num_tokens":
|
| 475 |
"step": 57
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch":
|
| 479 |
-
"grad_norm": 0.
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss":
|
| 482 |
-
"num_tokens":
|
| 483 |
"step": 58
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch":
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss":
|
| 490 |
-
"num_tokens":
|
| 491 |
"step": 59
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"grad_norm":
|
| 496 |
-
"learning_rate":
|
| 497 |
-
"loss":
|
| 498 |
-
"num_tokens":
|
| 499 |
"step": 60
|
| 500 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 501 |
{
|
| 502 |
"epoch": 2.0,
|
| 503 |
-
"
|
| 504 |
-
"
|
| 505 |
-
"
|
| 506 |
-
"
|
| 507 |
-
"
|
| 508 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 2.0,
|
| 512 |
-
"step":
|
| 513 |
-
"total_flos":
|
| 514 |
-
"train_loss":
|
| 515 |
-
"train_runtime":
|
| 516 |
-
"train_samples_per_second":
|
| 517 |
-
"train_steps_per_second": 0.
|
| 518 |
}
|
| 519 |
],
|
| 520 |
"logging_steps": 1,
|
| 521 |
-
"max_steps":
|
| 522 |
"num_input_tokens_seen": 0,
|
| 523 |
"num_train_epochs": 2,
|
| 524 |
"save_steps": 500,
|
|
@@ -534,8 +2102,8 @@
|
|
| 534 |
"attributes": {}
|
| 535 |
}
|
| 536 |
},
|
| 537 |
-
"total_flos":
|
| 538 |
-
"train_batch_size":
|
| 539 |
"trial_name": null,
|
| 540 |
"trial_params": null
|
| 541 |
}
|
|
|
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
"epoch": 2.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 256,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"epoch": 0.0078125,
|
| 14 |
+
"grad_norm": 19.84408767851053,
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
+
"loss": 2.138,
|
| 17 |
+
"num_tokens": 117626.0,
|
| 18 |
"step": 1
|
| 19 |
},
|
| 20 |
{
|
| 21 |
+
"epoch": 0.015625,
|
| 22 |
+
"grad_norm": 19.978311286067118,
|
| 23 |
+
"learning_rate": 1.25e-06,
|
| 24 |
+
"loss": 2.1823,
|
| 25 |
+
"num_tokens": 258329.0,
|
| 26 |
"step": 2
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.0234375,
|
| 30 |
+
"grad_norm": 24.54585887790054,
|
| 31 |
+
"learning_rate": 2.5e-06,
|
| 32 |
+
"loss": 2.1749,
|
| 33 |
+
"num_tokens": 358207.0,
|
| 34 |
"step": 3
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"epoch": 0.03125,
|
| 38 |
+
"grad_norm": 20.358913101197274,
|
| 39 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 40 |
+
"loss": 2.1594,
|
| 41 |
+
"num_tokens": 458149.0,
|
| 42 |
"step": 4
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 0.0390625,
|
| 46 |
+
"grad_norm": 13.258634593907978,
|
| 47 |
+
"learning_rate": 5e-06,
|
| 48 |
+
"loss": 2.0771,
|
| 49 |
+
"num_tokens": 628361.0,
|
| 50 |
"step": 5
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.046875,
|
| 54 |
+
"grad_norm": 7.556697764368537,
|
| 55 |
+
"learning_rate": 6.25e-06,
|
| 56 |
+
"loss": 1.9613,
|
| 57 |
+
"num_tokens": 740692.0,
|
| 58 |
"step": 6
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.0546875,
|
| 62 |
+
"grad_norm": 3.95920619952582,
|
| 63 |
+
"learning_rate": 7.500000000000001e-06,
|
| 64 |
+
"loss": 1.8988,
|
| 65 |
+
"num_tokens": 831264.0,
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 0.0625,
|
| 70 |
+
"grad_norm": 2.9437498387368195,
|
| 71 |
+
"learning_rate": 8.750000000000001e-06,
|
| 72 |
+
"loss": 1.8619,
|
| 73 |
+
"num_tokens": 949429.0,
|
| 74 |
"step": 8
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.0703125,
|
| 78 |
+
"grad_norm": 2.2758434326337658,
|
| 79 |
+
"learning_rate": 1e-05,
|
| 80 |
+
"loss": 1.8093,
|
| 81 |
+
"num_tokens": 1064518.0,
|
| 82 |
"step": 9
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"epoch": 0.078125,
|
| 86 |
+
"grad_norm": 1.8335493249829427,
|
| 87 |
+
"learning_rate": 9.99963894522398e-06,
|
| 88 |
+
"loss": 1.7104,
|
| 89 |
+
"num_tokens": 1159190.0,
|
| 90 |
"step": 10
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 0.0859375,
|
| 94 |
+
"grad_norm": 1.4037843921728306,
|
| 95 |
+
"learning_rate": 9.998555838833937e-06,
|
| 96 |
+
"loss": 1.7486,
|
| 97 |
+
"num_tokens": 1294966.0,
|
| 98 |
"step": 11
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.09375,
|
| 102 |
+
"grad_norm": 1.2243446595107301,
|
| 103 |
+
"learning_rate": 9.996750854634648e-06,
|
| 104 |
+
"loss": 1.6837,
|
| 105 |
+
"num_tokens": 1450365.0,
|
| 106 |
"step": 12
|
| 107 |
},
|
| 108 |
{
|
| 109 |
+
"epoch": 0.1015625,
|
| 110 |
+
"grad_norm": 1.300830012625964,
|
| 111 |
+
"learning_rate": 9.994224282269737e-06,
|
| 112 |
+
"loss": 1.6776,
|
| 113 |
+
"num_tokens": 1565582.0,
|
| 114 |
"step": 13
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.109375,
|
| 118 |
+
"grad_norm": 1.1943668176817122,
|
| 119 |
+
"learning_rate": 9.990976527175218e-06,
|
| 120 |
+
"loss": 1.5839,
|
| 121 |
+
"num_tokens": 1665314.0,
|
| 122 |
"step": 14
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.1171875,
|
| 126 |
+
"grad_norm": 0.9227937107276539,
|
| 127 |
+
"learning_rate": 9.987008110514413e-06,
|
| 128 |
+
"loss": 1.5602,
|
| 129 |
+
"num_tokens": 1782228.0,
|
| 130 |
"step": 15
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.125,
|
| 134 |
+
"grad_norm": 0.8398734704934597,
|
| 135 |
+
"learning_rate": 9.982319669094345e-06,
|
| 136 |
+
"loss": 1.4927,
|
| 137 |
+
"num_tokens": 1897118.0,
|
| 138 |
"step": 16
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 0.1328125,
|
| 142 |
+
"grad_norm": 1.1768893009483088,
|
| 143 |
+
"learning_rate": 9.976911955263528e-06,
|
| 144 |
+
"loss": 1.2554,
|
| 145 |
+
"num_tokens": 1960907.0,
|
| 146 |
"step": 17
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.140625,
|
| 150 |
+
"grad_norm": 0.9112944132015599,
|
| 151 |
+
"learning_rate": 9.970785836791259e-06,
|
| 152 |
+
"loss": 1.476,
|
| 153 |
+
"num_tokens": 2042783.0,
|
| 154 |
"step": 18
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 0.1484375,
|
| 158 |
+
"grad_norm": 0.7712029840767504,
|
| 159 |
+
"learning_rate": 9.963942296728345e-06,
|
| 160 |
+
"loss": 1.6416,
|
| 161 |
+
"num_tokens": 2201530.0,
|
| 162 |
"step": 19
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 0.15625,
|
| 166 |
+
"grad_norm": 0.8015840857538936,
|
| 167 |
+
"learning_rate": 9.95638243324938e-06,
|
| 168 |
+
"loss": 1.5349,
|
| 169 |
+
"num_tokens": 2319514.0,
|
| 170 |
"step": 20
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.1640625,
|
| 174 |
+
"grad_norm": 0.7713388134519369,
|
| 175 |
+
"learning_rate": 9.948107459476502e-06,
|
| 176 |
+
"loss": 1.5451,
|
| 177 |
+
"num_tokens": 2432219.0,
|
| 178 |
"step": 21
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 0.171875,
|
| 182 |
+
"grad_norm": 0.8173027499660734,
|
| 183 |
+
"learning_rate": 9.939118703284738e-06,
|
| 184 |
+
"loss": 1.4754,
|
| 185 |
+
"num_tokens": 2542893.0,
|
| 186 |
"step": 22
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 0.1796875,
|
| 190 |
+
"grad_norm": 0.9978504392316467,
|
| 191 |
+
"learning_rate": 9.929417607088905e-06,
|
| 192 |
+
"loss": 1.3757,
|
| 193 |
+
"num_tokens": 2614787.0,
|
| 194 |
"step": 23
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.1875,
|
| 198 |
+
"grad_norm": 0.8355911569239793,
|
| 199 |
+
"learning_rate": 9.91900572761217e-06,
|
| 200 |
+
"loss": 1.5055,
|
| 201 |
+
"num_tokens": 2712982.0,
|
| 202 |
"step": 24
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.1953125,
|
| 206 |
+
"grad_norm": 0.7569239678224411,
|
| 207 |
+
"learning_rate": 9.907884735636226e-06,
|
| 208 |
+
"loss": 1.5263,
|
| 209 |
+
"num_tokens": 2830889.0,
|
| 210 |
"step": 25
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 0.203125,
|
| 214 |
+
"grad_norm": 0.7984856259432604,
|
| 215 |
+
"learning_rate": 9.896056415733193e-06,
|
| 216 |
+
"loss": 1.4859,
|
| 217 |
+
"num_tokens": 2929523.0,
|
| 218 |
"step": 26
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.2109375,
|
| 222 |
+
"grad_norm": 0.8963535651369063,
|
| 223 |
+
"learning_rate": 9.883522665979243e-06,
|
| 224 |
+
"loss": 1.1895,
|
| 225 |
+
"num_tokens": 3012934.0,
|
| 226 |
"step": 27
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.21875,
|
| 230 |
+
"grad_norm": 0.8176677500878967,
|
| 231 |
+
"learning_rate": 9.870285497650027e-06,
|
| 232 |
+
"loss": 1.2961,
|
| 233 |
+
"num_tokens": 3121547.0,
|
| 234 |
"step": 28
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.2265625,
|
| 238 |
+
"grad_norm": 0.6886775851676376,
|
| 239 |
+
"learning_rate": 9.85634703489792e-06,
|
| 240 |
+
"loss": 1.5701,
|
| 241 |
+
"num_tokens": 3261439.0,
|
| 242 |
"step": 29
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.234375,
|
| 246 |
+
"grad_norm": 1.054056421985855,
|
| 247 |
+
"learning_rate": 9.841709514411166e-06,
|
| 248 |
+
"loss": 1.1395,
|
| 249 |
+
"num_tokens": 3317189.0,
|
| 250 |
"step": 30
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"epoch": 0.2421875,
|
| 254 |
+
"grad_norm": 0.7585315349302701,
|
| 255 |
+
"learning_rate": 9.826375285054957e-06,
|
| 256 |
+
"loss": 1.5146,
|
| 257 |
+
"num_tokens": 3419969.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 258 |
"step": 31
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 0.25,
|
| 262 |
+
"grad_norm": 0.9305699804034667,
|
| 263 |
+
"learning_rate": 9.810346807494512e-06,
|
| 264 |
+
"loss": 1.2012,
|
| 265 |
+
"num_tokens": 3511442.0,
|
| 266 |
"step": 32
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 0.2578125,
|
| 270 |
+
"grad_norm": 0.7156092213048836,
|
| 271 |
+
"learning_rate": 9.79362665380022e-06,
|
| 272 |
+
"loss": 1.4797,
|
| 273 |
+
"num_tokens": 3619865.0,
|
| 274 |
"step": 33
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 0.265625,
|
| 278 |
+
"grad_norm": 0.841576109361547,
|
| 279 |
+
"learning_rate": 9.776217507034903e-06,
|
| 280 |
+
"loss": 1.466,
|
| 281 |
+
"num_tokens": 3709774.0,
|
| 282 |
"step": 34
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.2734375,
|
| 286 |
+
"grad_norm": 0.7424654589083722,
|
| 287 |
+
"learning_rate": 9.758122160823267e-06,
|
| 288 |
+
"loss": 1.4378,
|
| 289 |
+
"num_tokens": 3819957.0,
|
| 290 |
"step": 35
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.28125,
|
| 294 |
+
"grad_norm": 0.8039177435008359,
|
| 295 |
+
"learning_rate": 9.739343518903608e-06,
|
| 296 |
+
"loss": 1.2134,
|
| 297 |
+
"num_tokens": 3917385.0,
|
| 298 |
"step": 36
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 0.2890625,
|
| 302 |
+
"grad_norm": 0.8053568262913987,
|
| 303 |
+
"learning_rate": 9.719884594661864e-06,
|
| 304 |
+
"loss": 1.3512,
|
| 305 |
+
"num_tokens": 4008929.0,
|
| 306 |
"step": 37
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 0.296875,
|
| 310 |
+
"grad_norm": 0.6980096362807309,
|
| 311 |
+
"learning_rate": 9.699748510648047e-06,
|
| 312 |
+
"loss": 1.4841,
|
| 313 |
+
"num_tokens": 4133134.0,
|
| 314 |
"step": 38
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 0.3046875,
|
| 318 |
+
"grad_norm": 0.6917558387439214,
|
| 319 |
+
"learning_rate": 9.67893849807518e-06,
|
| 320 |
+
"loss": 1.4458,
|
| 321 |
+
"num_tokens": 4246552.0,
|
| 322 |
"step": 39
|
| 323 |
},
|
| 324 |
{
|
| 325 |
+
"epoch": 0.3125,
|
| 326 |
+
"grad_norm": 0.6394840525038438,
|
| 327 |
+
"learning_rate": 9.65745789630079e-06,
|
| 328 |
+
"loss": 1.5404,
|
| 329 |
+
"num_tokens": 4373626.0,
|
| 330 |
"step": 40
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 0.3203125,
|
| 334 |
+
"grad_norm": 0.6887272352271856,
|
| 335 |
+
"learning_rate": 9.635310152291038e-06,
|
| 336 |
+
"loss": 1.4129,
|
| 337 |
+
"num_tokens": 4486683.0,
|
| 338 |
"step": 41
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.328125,
|
| 342 |
+
"grad_norm": 1.077660016010402,
|
| 343 |
+
"learning_rate": 9.61249882006759e-06,
|
| 344 |
+
"loss": 1.4161,
|
| 345 |
+
"num_tokens": 4576770.0,
|
| 346 |
"step": 42
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 0.3359375,
|
| 350 |
+
"grad_norm": 0.7070509210444217,
|
| 351 |
+
"learning_rate": 9.589027560137313e-06,
|
| 352 |
+
"loss": 1.4538,
|
| 353 |
+
"num_tokens": 4677751.0,
|
| 354 |
"step": 43
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 0.34375,
|
| 358 |
+
"grad_norm": 0.6701041110613191,
|
| 359 |
+
"learning_rate": 9.564900138904871e-06,
|
| 360 |
+
"loss": 1.5049,
|
| 361 |
+
"num_tokens": 4799919.0,
|
| 362 |
"step": 44
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 0.3515625,
|
| 366 |
+
"grad_norm": 0.7359656562432312,
|
| 367 |
+
"learning_rate": 9.540120428068337e-06,
|
| 368 |
+
"loss": 1.2227,
|
| 369 |
+
"num_tokens": 4926312.0,
|
| 370 |
"step": 45
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"epoch": 0.359375,
|
| 374 |
+
"grad_norm": 0.7152824867853115,
|
| 375 |
+
"learning_rate": 9.514692403997914e-06,
|
| 376 |
+
"loss": 1.3409,
|
| 377 |
+
"num_tokens": 5027266.0,
|
| 378 |
"step": 46
|
| 379 |
},
|
| 380 |
{
|
| 381 |
+
"epoch": 0.3671875,
|
| 382 |
+
"grad_norm": 0.7040751604750374,
|
| 383 |
+
"learning_rate": 9.488620147097838e-06,
|
| 384 |
+
"loss": 1.4359,
|
| 385 |
+
"num_tokens": 5141862.0,
|
| 386 |
"step": 47
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 0.375,
|
| 390 |
+
"grad_norm": 0.651532634041541,
|
| 391 |
+
"learning_rate": 9.461907841151607e-06,
|
| 392 |
+
"loss": 1.4755,
|
| 393 |
+
"num_tokens": 5267659.0,
|
| 394 |
"step": 48
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.3828125,
|
| 398 |
+
"grad_norm": 0.5616342800863031,
|
| 399 |
+
"learning_rate": 9.43455977265062e-06,
|
| 400 |
+
"loss": 1.5181,
|
| 401 |
+
"num_tokens": 5438666.0,
|
| 402 |
"step": 49
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 0.390625,
|
| 406 |
+
"grad_norm": 0.6700860717197045,
|
| 407 |
+
"learning_rate": 9.406580330106318e-06,
|
| 408 |
+
"loss": 1.4133,
|
| 409 |
+
"num_tokens": 5554222.0,
|
| 410 |
"step": 50
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.3984375,
|
| 414 |
+
"grad_norm": 0.6159208895448742,
|
| 415 |
+
"learning_rate": 9.377974003345974e-06,
|
| 416 |
+
"loss": 1.4504,
|
| 417 |
+
"num_tokens": 5717355.0,
|
| 418 |
"step": 51
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 0.40625,
|
| 422 |
+
"grad_norm": 0.6680323696906962,
|
| 423 |
+
"learning_rate": 9.34874538279221e-06,
|
| 424 |
+
"loss": 1.3925,
|
| 425 |
+
"num_tokens": 5841564.0,
|
| 426 |
"step": 52
|
| 427 |
},
|
| 428 |
{
|
| 429 |
+
"epoch": 0.4140625,
|
| 430 |
+
"grad_norm": 0.7263410130443334,
|
| 431 |
+
"learning_rate": 9.31889915872638e-06,
|
| 432 |
+
"loss": 1.4245,
|
| 433 |
+
"num_tokens": 5937886.0,
|
| 434 |
"step": 53
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 0.421875,
|
| 438 |
+
"grad_norm": 0.7007572818882284,
|
| 439 |
+
"learning_rate": 9.288440120535926e-06,
|
| 440 |
+
"loss": 1.4077,
|
| 441 |
+
"num_tokens": 6041841.0,
|
| 442 |
"step": 54
|
| 443 |
},
|
| 444 |
{
|
| 445 |
+
"epoch": 0.4296875,
|
| 446 |
+
"grad_norm": 0.6601371708453048,
|
| 447 |
+
"learning_rate": 9.257373155945835e-06,
|
| 448 |
+
"loss": 1.4731,
|
| 449 |
+
"num_tokens": 6156101.0,
|
| 450 |
"step": 55
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.4375,
|
| 454 |
+
"grad_norm": 0.6939996039565332,
|
| 455 |
+
"learning_rate": 9.225703250234299e-06,
|
| 456 |
+
"loss": 1.3667,
|
| 457 |
+
"num_tokens": 6271687.0,
|
| 458 |
"step": 56
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 0.4453125,
|
| 462 |
+
"grad_norm": 0.75123890327989,
|
| 463 |
+
"learning_rate": 9.193435485432744e-06,
|
| 464 |
+
"loss": 1.1228,
|
| 465 |
+
"num_tokens": 6354155.0,
|
| 466 |
"step": 57
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 0.453125,
|
| 470 |
+
"grad_norm": 0.7317912835914634,
|
| 471 |
+
"learning_rate": 9.160575039510328e-06,
|
| 472 |
+
"loss": 1.5066,
|
| 473 |
+
"num_tokens": 6478538.0,
|
| 474 |
"step": 58
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 0.4609375,
|
| 478 |
+
"grad_norm": 0.6861465030588054,
|
| 479 |
+
"learning_rate": 9.127127185543024e-06,
|
| 480 |
+
"loss": 1.4914,
|
| 481 |
+
"num_tokens": 6603772.0,
|
| 482 |
"step": 59
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 0.46875,
|
| 486 |
+
"grad_norm": 1.0773974899134051,
|
| 487 |
+
"learning_rate": 9.093097290867461e-06,
|
| 488 |
+
"loss": 1.4173,
|
| 489 |
+
"num_tokens": 6715588.0,
|
| 490 |
"step": 60
|
| 491 |
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.4765625,
|
| 494 |
+
"grad_norm": 0.7321867863627476,
|
| 495 |
+
"learning_rate": 9.058490816219644e-06,
|
| 496 |
+
"loss": 1.3357,
|
| 497 |
+
"num_tokens": 6808195.0,
|
| 498 |
+
"step": 61
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.484375,
|
| 502 |
+
"grad_norm": 0.6212863326614315,
|
| 503 |
+
"learning_rate": 9.02331331485866e-06,
|
| 504 |
+
"loss": 1.4426,
|
| 505 |
+
"num_tokens": 6939460.0,
|
| 506 |
+
"step": 62
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4921875,
|
| 510 |
+
"grad_norm": 0.7277630583423017,
|
| 511 |
+
"learning_rate": 8.987570431675566e-06,
|
| 512 |
+
"loss": 1.1815,
|
| 513 |
+
"num_tokens": 7041581.0,
|
| 514 |
+
"step": 63
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5,
|
| 518 |
+
"grad_norm": 0.6741034535246058,
|
| 519 |
+
"learning_rate": 8.95126790228755e-06,
|
| 520 |
+
"loss": 1.3185,
|
| 521 |
+
"num_tokens": 7143508.0,
|
| 522 |
+
"step": 64
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.5078125,
|
| 526 |
+
"grad_norm": 0.6928802103062359,
|
| 527 |
+
"learning_rate": 8.91441155211756e-06,
|
| 528 |
+
"loss": 1.3314,
|
| 529 |
+
"num_tokens": 7241533.0,
|
| 530 |
+
"step": 65
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.515625,
|
| 534 |
+
"grad_norm": 0.7217592903894582,
|
| 535 |
+
"learning_rate": 8.877007295459482e-06,
|
| 536 |
+
"loss": 1.3246,
|
| 537 |
+
"num_tokens": 7361609.0,
|
| 538 |
+
"step": 66
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.5234375,
|
| 542 |
+
"grad_norm": 0.8143940532493384,
|
| 543 |
+
"learning_rate": 8.8390611345291e-06,
|
| 544 |
+
"loss": 1.207,
|
| 545 |
+
"num_tokens": 7444097.0,
|
| 546 |
+
"step": 67
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.53125,
|
| 550 |
+
"grad_norm": 0.7772204912891099,
|
| 551 |
+
"learning_rate": 8.800579158500918e-06,
|
| 552 |
+
"loss": 1.0387,
|
| 553 |
+
"num_tokens": 7510299.0,
|
| 554 |
+
"step": 68
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.5390625,
|
| 558 |
+
"grad_norm": 0.7276830703321551,
|
| 559 |
+
"learning_rate": 8.76156754253104e-06,
|
| 560 |
+
"loss": 1.1605,
|
| 561 |
+
"num_tokens": 7607438.0,
|
| 562 |
+
"step": 69
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.546875,
|
| 566 |
+
"grad_norm": 0.8260950847252159,
|
| 567 |
+
"learning_rate": 8.722032546766254e-06,
|
| 568 |
+
"loss": 1.0935,
|
| 569 |
+
"num_tokens": 7682088.0,
|
| 570 |
+
"step": 70
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.5546875,
|
| 574 |
+
"grad_norm": 0.7473650466374854,
|
| 575 |
+
"learning_rate": 8.681980515339464e-06,
|
| 576 |
+
"loss": 1.3942,
|
| 577 |
+
"num_tokens": 7794866.0,
|
| 578 |
+
"step": 71
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.5625,
|
| 582 |
+
"grad_norm": 0.7733771683964346,
|
| 583 |
+
"learning_rate": 8.641417875351666e-06,
|
| 584 |
+
"loss": 1.1644,
|
| 585 |
+
"num_tokens": 7874130.0,
|
| 586 |
+
"step": 72
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.5703125,
|
| 590 |
+
"grad_norm": 0.7082200673345814,
|
| 591 |
+
"learning_rate": 8.60035113584059e-06,
|
| 592 |
+
"loss": 1.3781,
|
| 593 |
+
"num_tokens": 7981885.0,
|
| 594 |
+
"step": 73
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.578125,
|
| 598 |
+
"grad_norm": 0.6687638416856552,
|
| 599 |
+
"learning_rate": 8.558786886736209e-06,
|
| 600 |
+
"loss": 1.4017,
|
| 601 |
+
"num_tokens": 8108112.0,
|
| 602 |
+
"step": 74
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.5859375,
|
| 606 |
+
"grad_norm": 0.6865370309152183,
|
| 607 |
+
"learning_rate": 8.516731797803254e-06,
|
| 608 |
+
"loss": 1.3747,
|
| 609 |
+
"num_tokens": 8230249.0,
|
| 610 |
+
"step": 75
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.59375,
|
| 614 |
+
"grad_norm": 0.675772595290931,
|
| 615 |
+
"learning_rate": 8.474192617570935e-06,
|
| 616 |
+
"loss": 1.3315,
|
| 617 |
+
"num_tokens": 8334831.0,
|
| 618 |
+
"step": 76
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.6015625,
|
| 622 |
+
"grad_norm": 0.7678913060807983,
|
| 623 |
+
"learning_rate": 8.431176172250002e-06,
|
| 624 |
+
"loss": 1.2808,
|
| 625 |
+
"num_tokens": 8417045.0,
|
| 626 |
+
"step": 77
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.609375,
|
| 630 |
+
"grad_norm": 0.7462288083458992,
|
| 631 |
+
"learning_rate": 8.387689364637354e-06,
|
| 632 |
+
"loss": 1.3512,
|
| 633 |
+
"num_tokens": 8509320.0,
|
| 634 |
+
"step": 78
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.6171875,
|
| 638 |
+
"grad_norm": 1.3926589967476652,
|
| 639 |
+
"learning_rate": 8.343739173008368e-06,
|
| 640 |
+
"loss": 1.0666,
|
| 641 |
+
"num_tokens": 8596281.0,
|
| 642 |
+
"step": 79
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.625,
|
| 646 |
+
"grad_norm": 0.746543141672666,
|
| 647 |
+
"learning_rate": 8.299332649997081e-06,
|
| 648 |
+
"loss": 1.1313,
|
| 649 |
+
"num_tokens": 8697879.0,
|
| 650 |
+
"step": 80
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.6328125,
|
| 654 |
+
"grad_norm": 0.8937416904764481,
|
| 655 |
+
"learning_rate": 8.254476921464483e-06,
|
| 656 |
+
"loss": 0.7537,
|
| 657 |
+
"num_tokens": 8743432.0,
|
| 658 |
+
"step": 81
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.640625,
|
| 662 |
+
"grad_norm": 0.7775086176515215,
|
| 663 |
+
"learning_rate": 8.209179185355024e-06,
|
| 664 |
+
"loss": 1.3185,
|
| 665 |
+
"num_tokens": 8847090.0,
|
| 666 |
+
"step": 82
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.6484375,
|
| 670 |
+
"grad_norm": 0.7037452500578896,
|
| 671 |
+
"learning_rate": 8.163446710541578e-06,
|
| 672 |
+
"loss": 1.4299,
|
| 673 |
+
"num_tokens": 8961830.0,
|
| 674 |
+
"step": 83
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.65625,
|
| 678 |
+
"grad_norm": 0.7342849003195578,
|
| 679 |
+
"learning_rate": 8.117286835659014e-06,
|
| 680 |
+
"loss": 1.3388,
|
| 681 |
+
"num_tokens": 9057492.0,
|
| 682 |
+
"step": 84
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.6640625,
|
| 686 |
+
"grad_norm": 0.697534362210573,
|
| 687 |
+
"learning_rate": 8.070706967926565e-06,
|
| 688 |
+
"loss": 1.1589,
|
| 689 |
+
"num_tokens": 9179203.0,
|
| 690 |
+
"step": 85
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.671875,
|
| 694 |
+
"grad_norm": 0.7308248561338963,
|
| 695 |
+
"learning_rate": 8.023714581959222e-06,
|
| 696 |
+
"loss": 1.2598,
|
| 697 |
+
"num_tokens": 9273409.0,
|
| 698 |
+
"step": 86
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.6796875,
|
| 702 |
+
"grad_norm": 0.6751218137706414,
|
| 703 |
+
"learning_rate": 7.976317218568261e-06,
|
| 704 |
+
"loss": 1.4331,
|
| 705 |
+
"num_tokens": 9413531.0,
|
| 706 |
+
"step": 87
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.6875,
|
| 710 |
+
"grad_norm": 0.8001036099910734,
|
| 711 |
+
"learning_rate": 7.928522483551212e-06,
|
| 712 |
+
"loss": 1.2232,
|
| 713 |
+
"num_tokens": 9501317.0,
|
| 714 |
+
"step": 88
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.6953125,
|
| 718 |
+
"grad_norm": 0.6791770324998796,
|
| 719 |
+
"learning_rate": 7.880338046471331e-06,
|
| 720 |
+
"loss": 1.3473,
|
| 721 |
+
"num_tokens": 9623024.0,
|
| 722 |
+
"step": 89
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.703125,
|
| 726 |
+
"grad_norm": 0.6053250285438644,
|
| 727 |
+
"learning_rate": 7.831771639426909e-06,
|
| 728 |
+
"loss": 1.3879,
|
| 729 |
+
"num_tokens": 9769816.0,
|
| 730 |
+
"step": 90
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.7109375,
|
| 734 |
+
"grad_norm": 0.6825722474422947,
|
| 735 |
+
"learning_rate": 7.782831055810483e-06,
|
| 736 |
+
"loss": 1.3736,
|
| 737 |
+
"num_tokens": 9895110.0,
|
| 738 |
+
"step": 91
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.71875,
|
| 742 |
+
"grad_norm": 0.6983636023906173,
|
| 743 |
+
"learning_rate": 7.733524149058258e-06,
|
| 744 |
+
"loss": 1.3785,
|
| 745 |
+
"num_tokens": 10008605.0,
|
| 746 |
+
"step": 92
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.7265625,
|
| 750 |
+
"grad_norm": 0.9972849515830047,
|
| 751 |
+
"learning_rate": 7.683858831389866e-06,
|
| 752 |
+
"loss": 1.0233,
|
| 753 |
+
"num_tokens": 10083447.0,
|
| 754 |
+
"step": 93
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.734375,
|
| 758 |
+
"grad_norm": 0.7049819856389679,
|
| 759 |
+
"learning_rate": 7.633843072538703e-06,
|
| 760 |
+
"loss": 1.4001,
|
| 761 |
+
"num_tokens": 10191867.0,
|
| 762 |
+
"step": 94
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.7421875,
|
| 766 |
+
"grad_norm": 0.71997277540597,
|
| 767 |
+
"learning_rate": 7.583484898473044e-06,
|
| 768 |
+
"loss": 1.3869,
|
| 769 |
+
"num_tokens": 10306190.0,
|
| 770 |
+
"step": 95
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.75,
|
| 774 |
+
"grad_norm": 0.6121122302000951,
|
| 775 |
+
"learning_rate": 7.5327923901081126e-06,
|
| 776 |
+
"loss": 1.3928,
|
| 777 |
+
"num_tokens": 10458490.0,
|
| 778 |
+
"step": 96
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.7578125,
|
| 782 |
+
"grad_norm": 0.6704503866289738,
|
| 783 |
+
"learning_rate": 7.481773682009356e-06,
|
| 784 |
+
"loss": 1.387,
|
| 785 |
+
"num_tokens": 10571586.0,
|
| 786 |
+
"step": 97
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.765625,
|
| 790 |
+
"grad_norm": 0.6718516593876417,
|
| 791 |
+
"learning_rate": 7.430436961087094e-06,
|
| 792 |
+
"loss": 1.3909,
|
| 793 |
+
"num_tokens": 10694451.0,
|
| 794 |
+
"step": 98
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.7734375,
|
| 798 |
+
"grad_norm": 0.7316148335388524,
|
| 799 |
+
"learning_rate": 7.3787904652827845e-06,
|
| 800 |
+
"loss": 1.3022,
|
| 801 |
+
"num_tokens": 10803515.0,
|
| 802 |
+
"step": 99
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.78125,
|
| 806 |
+
"grad_norm": 0.7542306844884376,
|
| 807 |
+
"learning_rate": 7.326842482247082e-06,
|
| 808 |
+
"loss": 1.2066,
|
| 809 |
+
"num_tokens": 10888678.0,
|
| 810 |
+
"step": 100
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.7890625,
|
| 814 |
+
"grad_norm": 0.7486963885583306,
|
| 815 |
+
"learning_rate": 7.274601348009934e-06,
|
| 816 |
+
"loss": 1.2883,
|
| 817 |
+
"num_tokens": 10990858.0,
|
| 818 |
+
"step": 101
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.796875,
|
| 822 |
+
"grad_norm": 0.7645043338977128,
|
| 823 |
+
"learning_rate": 7.222075445642904e-06,
|
| 824 |
+
"loss": 1.1832,
|
| 825 |
+
"num_tokens": 11103396.0,
|
| 826 |
+
"step": 102
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.8046875,
|
| 830 |
+
"grad_norm": 0.9445118218620935,
|
| 831 |
+
"learning_rate": 7.169273203913956e-06,
|
| 832 |
+
"loss": 1.304,
|
| 833 |
+
"num_tokens": 11215478.0,
|
| 834 |
+
"step": 103
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.8125,
|
| 838 |
+
"grad_norm": 0.6840886879876881,
|
| 839 |
+
"learning_rate": 7.116203095934893e-06,
|
| 840 |
+
"loss": 1.3507,
|
| 841 |
+
"num_tokens": 11324900.0,
|
| 842 |
+
"step": 104
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.8203125,
|
| 846 |
+
"grad_norm": 0.7482654517365287,
|
| 847 |
+
"learning_rate": 7.062873637801692e-06,
|
| 848 |
+
"loss": 1.3618,
|
| 849 |
+
"num_tokens": 11418145.0,
|
| 850 |
+
"step": 105
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.828125,
|
| 854 |
+
"grad_norm": 0.6746963223299707,
|
| 855 |
+
"learning_rate": 7.009293387227938e-06,
|
| 856 |
+
"loss": 1.2971,
|
| 857 |
+
"num_tokens": 11528904.0,
|
| 858 |
+
"step": 106
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.8359375,
|
| 862 |
+
"grad_norm": 0.7459275146699014,
|
| 863 |
+
"learning_rate": 6.955470942171572e-06,
|
| 864 |
+
"loss": 1.3051,
|
| 865 |
+
"num_tokens": 11636989.0,
|
| 866 |
+
"step": 107
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.84375,
|
| 870 |
+
"grad_norm": 0.618423083483337,
|
| 871 |
+
"learning_rate": 6.901414939455194e-06,
|
| 872 |
+
"loss": 1.3726,
|
| 873 |
+
"num_tokens": 11791980.0,
|
| 874 |
+
"step": 108
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.8515625,
|
| 878 |
+
"grad_norm": 0.7102667517353592,
|
| 879 |
+
"learning_rate": 6.847134053380113e-06,
|
| 880 |
+
"loss": 1.3057,
|
| 881 |
+
"num_tokens": 11896266.0,
|
| 882 |
+
"step": 109
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.859375,
|
| 886 |
+
"grad_norm": 0.7013500432834089,
|
| 887 |
+
"learning_rate": 6.792636994334402e-06,
|
| 888 |
+
"loss": 1.3337,
|
| 889 |
+
"num_tokens": 11999684.0,
|
| 890 |
+
"step": 110
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.8671875,
|
| 894 |
+
"grad_norm": 0.7196266576143713,
|
| 895 |
+
"learning_rate": 6.737932507395152e-06,
|
| 896 |
+
"loss": 1.3373,
|
| 897 |
+
"num_tokens": 12100486.0,
|
| 898 |
+
"step": 111
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.875,
|
| 902 |
+
"grad_norm": 0.888079086902341,
|
| 903 |
+
"learning_rate": 6.683029370925151e-06,
|
| 904 |
+
"loss": 0.9967,
|
| 905 |
+
"num_tokens": 12158163.0,
|
| 906 |
+
"step": 112
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.8828125,
|
| 910 |
+
"grad_norm": 0.7250291787301237,
|
| 911 |
+
"learning_rate": 6.627936395164243e-06,
|
| 912 |
+
"loss": 1.328,
|
| 913 |
+
"num_tokens": 12266150.0,
|
| 914 |
+
"step": 113
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.890625,
|
| 918 |
+
"grad_norm": 0.5904999446539592,
|
| 919 |
+
"learning_rate": 6.5726624208155464e-06,
|
| 920 |
+
"loss": 1.4128,
|
| 921 |
+
"num_tokens": 12430017.0,
|
| 922 |
+
"step": 114
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.8984375,
|
| 926 |
+
"grad_norm": 0.6877510213661672,
|
| 927 |
+
"learning_rate": 6.517216317626808e-06,
|
| 928 |
+
"loss": 1.3124,
|
| 929 |
+
"num_tokens": 12532303.0,
|
| 930 |
+
"step": 115
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.90625,
|
| 934 |
+
"grad_norm": 0.758351513684432,
|
| 935 |
+
"learning_rate": 6.461606982967079e-06,
|
| 936 |
+
"loss": 1.3801,
|
| 937 |
+
"num_tokens": 12638112.0,
|
| 938 |
+
"step": 116
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.9140625,
|
| 942 |
+
"grad_norm": 0.6902978101709654,
|
| 943 |
+
"learning_rate": 6.405843340398971e-06,
|
| 944 |
+
"loss": 1.3237,
|
| 945 |
+
"num_tokens": 12754845.0,
|
| 946 |
+
"step": 117
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.921875,
|
| 950 |
+
"grad_norm": 0.8617953565270738,
|
| 951 |
+
"learning_rate": 6.349934338246695e-06,
|
| 952 |
+
"loss": 1.0111,
|
| 953 |
+
"num_tokens": 12813594.0,
|
| 954 |
+
"step": 118
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.9296875,
|
| 958 |
+
"grad_norm": 0.7268931542452745,
|
| 959 |
+
"learning_rate": 6.293888948160139e-06,
|
| 960 |
+
"loss": 1.2918,
|
| 961 |
+
"num_tokens": 12925583.0,
|
| 962 |
+
"step": 119
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.9375,
|
| 966 |
+
"grad_norm": 0.8673307866686834,
|
| 967 |
+
"learning_rate": 6.237716163675198e-06,
|
| 968 |
+
"loss": 0.9952,
|
| 969 |
+
"num_tokens": 12980566.0,
|
| 970 |
+
"step": 120
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.9453125,
|
| 974 |
+
"grad_norm": 0.7071724815702097,
|
| 975 |
+
"learning_rate": 6.181424998770595e-06,
|
| 976 |
+
"loss": 1.4215,
|
| 977 |
+
"num_tokens": 13099629.0,
|
| 978 |
+
"step": 121
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.953125,
|
| 982 |
+
"grad_norm": 0.8450840500466258,
|
| 983 |
+
"learning_rate": 6.1250244864214115e-06,
|
| 984 |
+
"loss": 1.0342,
|
| 985 |
+
"num_tokens": 13173257.0,
|
| 986 |
+
"step": 122
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.9609375,
|
| 990 |
+
"grad_norm": 0.7659617932211072,
|
| 991 |
+
"learning_rate": 6.0685236771495824e-06,
|
| 992 |
+
"loss": 1.1823,
|
| 993 |
+
"num_tokens": 13256058.0,
|
| 994 |
+
"step": 123
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.96875,
|
| 998 |
+
"grad_norm": 0.7179060564107081,
|
| 999 |
+
"learning_rate": 6.011931637571569e-06,
|
| 1000 |
+
"loss": 1.3471,
|
| 1001 |
+
"num_tokens": 13362128.0,
|
| 1002 |
+
"step": 124
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.9765625,
|
| 1006 |
+
"grad_norm": 0.8023511514218298,
|
| 1007 |
+
"learning_rate": 5.955257448943446e-06,
|
| 1008 |
+
"loss": 1.3169,
|
| 1009 |
+
"num_tokens": 13451943.0,
|
| 1010 |
+
"step": 125
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.984375,
|
| 1014 |
+
"grad_norm": 0.6882367769200813,
|
| 1015 |
+
"learning_rate": 5.8985102057036514e-06,
|
| 1016 |
+
"loss": 1.2749,
|
| 1017 |
+
"num_tokens": 13573663.0,
|
| 1018 |
+
"step": 126
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.9921875,
|
| 1022 |
+
"grad_norm": 0.7699074192862526,
|
| 1023 |
+
"learning_rate": 5.841699014013609e-06,
|
| 1024 |
+
"loss": 1.2876,
|
| 1025 |
+
"num_tokens": 13669614.0,
|
| 1026 |
+
"step": 127
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.0,
|
| 1030 |
+
"grad_norm": 0.7018143284467923,
|
| 1031 |
+
"learning_rate": 5.784832990296467e-06,
|
| 1032 |
+
"loss": 1.3114,
|
| 1033 |
+
"num_tokens": 13780944.0,
|
| 1034 |
+
"step": 128
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.0,
|
| 1038 |
+
"eval_loss": 0.7009232640266418,
|
| 1039 |
+
"eval_num_tokens": 13780944.0,
|
| 1040 |
+
"eval_runtime": 27.2149,
|
| 1041 |
+
"eval_samples_per_second": 16.756,
|
| 1042 |
+
"eval_steps_per_second": 4.189,
|
| 1043 |
+
"step": 128
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.0078125,
|
| 1047 |
+
"grad_norm": 0.9277434150672625,
|
| 1048 |
+
"learning_rate": 5.727921259774208e-06,
|
| 1049 |
+
"loss": 0.9785,
|
| 1050 |
+
"num_tokens": 13862626.0,
|
| 1051 |
+
"step": 129
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.015625,
|
| 1055 |
+
"grad_norm": 0.8426700840019798,
|
| 1056 |
+
"learning_rate": 5.670972955003328e-06,
|
| 1057 |
+
"loss": 1.1936,
|
| 1058 |
+
"num_tokens": 13965025.0,
|
| 1059 |
+
"step": 130
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.0234375,
|
| 1063 |
+
"grad_norm": 0.7573302593010723,
|
| 1064 |
+
"learning_rate": 5.613997214409346e-06,
|
| 1065 |
+
"loss": 1.203,
|
| 1066 |
+
"num_tokens": 14064644.0,
|
| 1067 |
+
"step": 131
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.03125,
|
| 1071 |
+
"grad_norm": 0.8585365328217087,
|
| 1072 |
+
"learning_rate": 5.5570031808203815e-06,
|
| 1073 |
+
"loss": 1.033,
|
| 1074 |
+
"num_tokens": 14121209.0,
|
| 1075 |
+
"step": 132
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.0390625,
|
| 1079 |
+
"grad_norm": 0.9369697369854478,
|
| 1080 |
+
"learning_rate": 5.500000000000001e-06,
|
| 1081 |
+
"loss": 1.2324,
|
| 1082 |
+
"num_tokens": 14232960.0,
|
| 1083 |
+
"step": 133
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.046875,
|
| 1087 |
+
"grad_norm": 0.7383491458478716,
|
| 1088 |
+
"learning_rate": 5.442996819179621e-06,
|
| 1089 |
+
"loss": 1.1334,
|
| 1090 |
+
"num_tokens": 14341337.0,
|
| 1091 |
+
"step": 134
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.0546875,
|
| 1095 |
+
"grad_norm": 0.8465782691184206,
|
| 1096 |
+
"learning_rate": 5.386002785590656e-06,
|
| 1097 |
+
"loss": 1.0652,
|
| 1098 |
+
"num_tokens": 14456878.0,
|
| 1099 |
+
"step": 135
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 1.0625,
|
| 1103 |
+
"grad_norm": 0.9009393512851309,
|
| 1104 |
+
"learning_rate": 5.3290270449966744e-06,
|
| 1105 |
+
"loss": 1.2604,
|
| 1106 |
+
"num_tokens": 14585468.0,
|
| 1107 |
+
"step": 136
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 1.0703125,
|
| 1111 |
+
"grad_norm": 0.838124950301535,
|
| 1112 |
+
"learning_rate": 5.272078740225793e-06,
|
| 1113 |
+
"loss": 1.2772,
|
| 1114 |
+
"num_tokens": 14720477.0,
|
| 1115 |
+
"step": 137
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.078125,
|
| 1119 |
+
"grad_norm": 0.7256188151049398,
|
| 1120 |
+
"learning_rate": 5.215167009703534e-06,
|
| 1121 |
+
"loss": 1.2161,
|
| 1122 |
+
"num_tokens": 14838111.0,
|
| 1123 |
+
"step": 138
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0859375,
|
| 1127 |
+
"grad_norm": 0.7713434296231342,
|
| 1128 |
+
"learning_rate": 5.1583009859863935e-06,
|
| 1129 |
+
"loss": 0.9921,
|
| 1130 |
+
"num_tokens": 14926950.0,
|
| 1131 |
+
"step": 139
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 1.09375,
|
| 1135 |
+
"grad_norm": 0.762366143119882,
|
| 1136 |
+
"learning_rate": 5.101489794296351e-06,
|
| 1137 |
+
"loss": 1.2182,
|
| 1138 |
+
"num_tokens": 15039803.0,
|
| 1139 |
+
"step": 140
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 1.1015625,
|
| 1143 |
+
"grad_norm": 0.7492048938741827,
|
| 1144 |
+
"learning_rate": 5.0447425510565564e-06,
|
| 1145 |
+
"loss": 1.173,
|
| 1146 |
+
"num_tokens": 15163203.0,
|
| 1147 |
+
"step": 141
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 1.109375,
|
| 1151 |
+
"grad_norm": 0.7255895354289637,
|
| 1152 |
+
"learning_rate": 4.988068362428434e-06,
|
| 1153 |
+
"loss": 1.0622,
|
| 1154 |
+
"num_tokens": 15266653.0,
|
| 1155 |
+
"step": 142
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 1.1171875,
|
| 1159 |
+
"grad_norm": 0.8692747446204775,
|
| 1160 |
+
"learning_rate": 4.931476322850421e-06,
|
| 1161 |
+
"loss": 0.9836,
|
| 1162 |
+
"num_tokens": 15346014.0,
|
| 1163 |
+
"step": 143
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 1.125,
|
| 1167 |
+
"grad_norm": 0.7396407961822765,
|
| 1168 |
+
"learning_rate": 4.87497551357859e-06,
|
| 1169 |
+
"loss": 1.1517,
|
| 1170 |
+
"num_tokens": 15455580.0,
|
| 1171 |
+
"step": 144
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.1328125,
|
| 1175 |
+
"grad_norm": 0.8310271632576728,
|
| 1176 |
+
"learning_rate": 4.818575001229406e-06,
|
| 1177 |
+
"loss": 1.136,
|
| 1178 |
+
"num_tokens": 15550989.0,
|
| 1179 |
+
"step": 145
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.140625,
|
| 1183 |
+
"grad_norm": 0.7973683671474303,
|
| 1184 |
+
"learning_rate": 4.762283836324803e-06,
|
| 1185 |
+
"loss": 1.1527,
|
| 1186 |
+
"num_tokens": 15635545.0,
|
| 1187 |
+
"step": 146
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.1484375,
|
| 1191 |
+
"grad_norm": 0.8655663089377099,
|
| 1192 |
+
"learning_rate": 4.706111051839864e-06,
|
| 1193 |
+
"loss": 0.9628,
|
| 1194 |
+
"num_tokens": 15698956.0,
|
| 1195 |
+
"step": 147
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 1.15625,
|
| 1199 |
+
"grad_norm": 0.8502520248721129,
|
| 1200 |
+
"learning_rate": 4.650065661753308e-06,
|
| 1201 |
+
"loss": 1.1624,
|
| 1202 |
+
"num_tokens": 15796035.0,
|
| 1203 |
+
"step": 148
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 1.1640625,
|
| 1207 |
+
"grad_norm": 0.8658130889912838,
|
| 1208 |
+
"learning_rate": 4.594156659601029e-06,
|
| 1209 |
+
"loss": 1.2634,
|
| 1210 |
+
"num_tokens": 15924303.0,
|
| 1211 |
+
"step": 149
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 1.171875,
|
| 1215 |
+
"grad_norm": 0.7506462285703888,
|
| 1216 |
+
"learning_rate": 4.538393017032922e-06,
|
| 1217 |
+
"loss": 1.2321,
|
| 1218 |
+
"num_tokens": 16056972.0,
|
| 1219 |
+
"step": 150
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.1796875,
|
| 1223 |
+
"grad_norm": 0.8570350856283379,
|
| 1224 |
+
"learning_rate": 4.482783682373194e-06,
|
| 1225 |
+
"loss": 1.2996,
|
| 1226 |
+
"num_tokens": 16160258.0,
|
| 1227 |
+
"step": 151
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.1875,
|
| 1231 |
+
"grad_norm": 0.7338679107741399,
|
| 1232 |
+
"learning_rate": 4.427337579184454e-06,
|
| 1233 |
+
"loss": 0.9762,
|
| 1234 |
+
"num_tokens": 16267191.0,
|
| 1235 |
+
"step": 152
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.1953125,
|
| 1239 |
+
"grad_norm": 0.7437223517311451,
|
| 1240 |
+
"learning_rate": 4.372063604835758e-06,
|
| 1241 |
+
"loss": 1.3102,
|
| 1242 |
+
"num_tokens": 16382899.0,
|
| 1243 |
+
"step": 153
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 1.203125,
|
| 1247 |
+
"grad_norm": 0.8078737210439417,
|
| 1248 |
+
"learning_rate": 4.316970629074849e-06,
|
| 1249 |
+
"loss": 1.1478,
|
| 1250 |
+
"num_tokens": 16476965.0,
|
| 1251 |
+
"step": 154
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 1.2109375,
|
| 1255 |
+
"grad_norm": 0.8010981968110408,
|
| 1256 |
+
"learning_rate": 4.262067492604849e-06,
|
| 1257 |
+
"loss": 1.1609,
|
| 1258 |
+
"num_tokens": 16576222.0,
|
| 1259 |
+
"step": 155
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 1.21875,
|
| 1263 |
+
"grad_norm": 0.7362430691273234,
|
| 1264 |
+
"learning_rate": 4.207363005665599e-06,
|
| 1265 |
+
"loss": 1.157,
|
| 1266 |
+
"num_tokens": 16682803.0,
|
| 1267 |
+
"step": 156
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 1.2265625,
|
| 1271 |
+
"grad_norm": 0.9102779009227082,
|
| 1272 |
+
"learning_rate": 4.15286594661989e-06,
|
| 1273 |
+
"loss": 0.8062,
|
| 1274 |
+
"num_tokens": 16769408.0,
|
| 1275 |
+
"step": 157
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 1.234375,
|
| 1279 |
+
"grad_norm": 0.7956227835975876,
|
| 1280 |
+
"learning_rate": 4.098585060544809e-06,
|
| 1281 |
+
"loss": 1.06,
|
| 1282 |
+
"num_tokens": 16869766.0,
|
| 1283 |
+
"step": 158
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.2421875,
|
| 1287 |
+
"grad_norm": 0.7684293753314241,
|
| 1288 |
+
"learning_rate": 4.044529057828429e-06,
|
| 1289 |
+
"loss": 1.1677,
|
| 1290 |
+
"num_tokens": 16985619.0,
|
| 1291 |
+
"step": 159
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.25,
|
| 1295 |
+
"grad_norm": 0.6797479500551905,
|
| 1296 |
+
"learning_rate": 3.990706612772064e-06,
|
| 1297 |
+
"loss": 1.211,
|
| 1298 |
+
"num_tokens": 17116813.0,
|
| 1299 |
+
"step": 160
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 1.2578125,
|
| 1303 |
+
"grad_norm": 0.9459450362001841,
|
| 1304 |
+
"learning_rate": 3.93712636219831e-06,
|
| 1305 |
+
"loss": 1.1156,
|
| 1306 |
+
"num_tokens": 17187827.0,
|
| 1307 |
+
"step": 161
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.265625,
|
| 1311 |
+
"grad_norm": 0.9418468652000539,
|
| 1312 |
+
"learning_rate": 3.883796904065108e-06,
|
| 1313 |
+
"loss": 0.8726,
|
| 1314 |
+
"num_tokens": 17249379.0,
|
| 1315 |
+
"step": 162
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 1.2734375,
|
| 1319 |
+
"grad_norm": 0.8294637183420516,
|
| 1320 |
+
"learning_rate": 3.8307267960860465e-06,
|
| 1321 |
+
"loss": 1.2131,
|
| 1322 |
+
"num_tokens": 17348117.0,
|
| 1323 |
+
"step": 163
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 1.28125,
|
| 1327 |
+
"grad_norm": 0.7228054402968582,
|
| 1328 |
+
"learning_rate": 3.777924554357096e-06,
|
| 1329 |
+
"loss": 1.1848,
|
| 1330 |
+
"num_tokens": 17457417.0,
|
| 1331 |
+
"step": 164
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 1.2890625,
|
| 1335 |
+
"grad_norm": 0.8359037722888478,
|
| 1336 |
+
"learning_rate": 3.7253986519900674e-06,
|
| 1337 |
+
"loss": 0.9503,
|
| 1338 |
+
"num_tokens": 17538753.0,
|
| 1339 |
+
"step": 165
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.296875,
|
| 1343 |
+
"grad_norm": 0.8150259943001315,
|
| 1344 |
+
"learning_rate": 3.673157517752919e-06,
|
| 1345 |
+
"loss": 0.9567,
|
| 1346 |
+
"num_tokens": 17620828.0,
|
| 1347 |
+
"step": 166
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.3046875,
|
| 1351 |
+
"grad_norm": 0.8755788643911117,
|
| 1352 |
+
"learning_rate": 3.6212095347172173e-06,
|
| 1353 |
+
"loss": 1.0949,
|
| 1354 |
+
"num_tokens": 17723084.0,
|
| 1355 |
+
"step": 167
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.3125,
|
| 1359 |
+
"grad_norm": 0.7426900340474549,
|
| 1360 |
+
"learning_rate": 3.569563038912909e-06,
|
| 1361 |
+
"loss": 1.2034,
|
| 1362 |
+
"num_tokens": 17847281.0,
|
| 1363 |
+
"step": 168
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.3203125,
|
| 1367 |
+
"grad_norm": 0.8892884344254952,
|
| 1368 |
+
"learning_rate": 3.5182263179906462e-06,
|
| 1369 |
+
"loss": 0.8077,
|
| 1370 |
+
"num_tokens": 17918615.0,
|
| 1371 |
+
"step": 169
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 1.328125,
|
| 1375 |
+
"grad_norm": 0.7124398883240389,
|
| 1376 |
+
"learning_rate": 3.4672076098918884e-06,
|
| 1377 |
+
"loss": 1.2433,
|
| 1378 |
+
"num_tokens": 18043464.0,
|
| 1379 |
+
"step": 170
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.3359375,
|
| 1383 |
+
"grad_norm": 0.7074740196277697,
|
| 1384 |
+
"learning_rate": 3.4165151015269573e-06,
|
| 1385 |
+
"loss": 1.1956,
|
| 1386 |
+
"num_tokens": 18151113.0,
|
| 1387 |
+
"step": 171
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 1.34375,
|
| 1391 |
+
"grad_norm": 0.77924697095352,
|
| 1392 |
+
"learning_rate": 3.3661569274612977e-06,
|
| 1393 |
+
"loss": 1.1756,
|
| 1394 |
+
"num_tokens": 18267805.0,
|
| 1395 |
+
"step": 172
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.3515625,
|
| 1399 |
+
"grad_norm": 0.7626070992304894,
|
| 1400 |
+
"learning_rate": 3.316141168610136e-06,
|
| 1401 |
+
"loss": 1.0063,
|
| 1402 |
+
"num_tokens": 18374627.0,
|
| 1403 |
+
"step": 173
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.359375,
|
| 1407 |
+
"grad_norm": 0.7156575703865125,
|
| 1408 |
+
"learning_rate": 3.2664758509417438e-06,
|
| 1409 |
+
"loss": 1.22,
|
| 1410 |
+
"num_tokens": 18496755.0,
|
| 1411 |
+
"step": 174
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 1.3671875,
|
| 1415 |
+
"grad_norm": 0.7930011630053356,
|
| 1416 |
+
"learning_rate": 3.217168944189518e-06,
|
| 1417 |
+
"loss": 1.2138,
|
| 1418 |
+
"num_tokens": 18599518.0,
|
| 1419 |
+
"step": 175
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.375,
|
| 1423 |
+
"grad_norm": 0.6786193439223048,
|
| 1424 |
+
"learning_rate": 3.1682283605730927e-06,
|
| 1425 |
+
"loss": 1.2775,
|
| 1426 |
+
"num_tokens": 18747130.0,
|
| 1427 |
+
"step": 176
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 1.3828125,
|
| 1431 |
+
"grad_norm": 0.8821840368890882,
|
| 1432 |
+
"learning_rate": 3.119661953528671e-06,
|
| 1433 |
+
"loss": 0.9483,
|
| 1434 |
+
"num_tokens": 18805587.0,
|
| 1435 |
+
"step": 177
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 1.390625,
|
| 1439 |
+
"grad_norm": 0.8125417620146214,
|
| 1440 |
+
"learning_rate": 3.071477516448792e-06,
|
| 1441 |
+
"loss": 1.1767,
|
| 1442 |
+
"num_tokens": 18894242.0,
|
| 1443 |
+
"step": 178
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 1.3984375,
|
| 1447 |
+
"grad_norm": 0.7043700363763821,
|
| 1448 |
+
"learning_rate": 3.0236827814317395e-06,
|
| 1449 |
+
"loss": 1.1893,
|
| 1450 |
+
"num_tokens": 19026696.0,
|
| 1451 |
+
"step": 179
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.40625,
|
| 1455 |
+
"grad_norm": 0.6910910972329093,
|
| 1456 |
+
"learning_rate": 2.976285418040781e-06,
|
| 1457 |
+
"loss": 1.2393,
|
| 1458 |
+
"num_tokens": 19155994.0,
|
| 1459 |
+
"step": 180
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.4140625,
|
| 1463 |
+
"grad_norm": 0.7997604311217631,
|
| 1464 |
+
"learning_rate": 2.9292930320734337e-06,
|
| 1465 |
+
"loss": 0.8965,
|
| 1466 |
+
"num_tokens": 19237754.0,
|
| 1467 |
+
"step": 181
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 1.421875,
|
| 1471 |
+
"grad_norm": 0.737295630916884,
|
| 1472 |
+
"learning_rate": 2.8827131643409878e-06,
|
| 1473 |
+
"loss": 1.1404,
|
| 1474 |
+
"num_tokens": 19346674.0,
|
| 1475 |
+
"step": 182
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 1.4296875,
|
| 1479 |
+
"grad_norm": 0.9455758188729559,
|
| 1480 |
+
"learning_rate": 2.8365532894584226e-06,
|
| 1481 |
+
"loss": 1.2053,
|
| 1482 |
+
"num_tokens": 19464630.0,
|
| 1483 |
+
"step": 183
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 1.4375,
|
| 1487 |
+
"grad_norm": 0.6946850131583951,
|
| 1488 |
+
"learning_rate": 2.790820814644976e-06,
|
| 1489 |
+
"loss": 1.217,
|
| 1490 |
+
"num_tokens": 19582966.0,
|
| 1491 |
+
"step": 184
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.4453125,
|
| 1495 |
+
"grad_norm": 0.8540798342673012,
|
| 1496 |
+
"learning_rate": 2.745523078535517e-06,
|
| 1497 |
+
"loss": 1.0924,
|
| 1498 |
+
"num_tokens": 19661796.0,
|
| 1499 |
+
"step": 185
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 1.453125,
|
| 1503 |
+
"grad_norm": 0.8561589833214078,
|
| 1504 |
+
"learning_rate": 2.7006673500029186e-06,
|
| 1505 |
+
"loss": 0.9604,
|
| 1506 |
+
"num_tokens": 19745502.0,
|
| 1507 |
+
"step": 186
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 1.4609375,
|
| 1511 |
+
"grad_norm": 0.8029178215471267,
|
| 1512 |
+
"learning_rate": 2.6562608269916323e-06,
|
| 1513 |
+
"loss": 1.1846,
|
| 1514 |
+
"num_tokens": 19853861.0,
|
| 1515 |
+
"step": 187
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.46875,
|
| 1519 |
+
"grad_norm": 0.7134473646385301,
|
| 1520 |
+
"learning_rate": 2.612310635362646e-06,
|
| 1521 |
+
"loss": 1.1809,
|
| 1522 |
+
"num_tokens": 19974435.0,
|
| 1523 |
+
"step": 188
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 1.4765625,
|
| 1527 |
+
"grad_norm": 0.6794409173033282,
|
| 1528 |
+
"learning_rate": 2.568823827750001e-06,
|
| 1529 |
+
"loss": 1.2213,
|
| 1530 |
+
"num_tokens": 20092837.0,
|
| 1531 |
+
"step": 189
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 1.484375,
|
| 1535 |
+
"grad_norm": 0.890753556898069,
|
| 1536 |
+
"learning_rate": 2.5258073824290674e-06,
|
| 1537 |
+
"loss": 0.948,
|
| 1538 |
+
"num_tokens": 20169103.0,
|
| 1539 |
+
"step": 190
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 1.4921875,
|
| 1543 |
+
"grad_norm": 0.8337646328005143,
|
| 1544 |
+
"learning_rate": 2.4832682021967468e-06,
|
| 1545 |
+
"loss": 0.745,
|
| 1546 |
+
"num_tokens": 20248099.0,
|
| 1547 |
+
"step": 191
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 1.5,
|
| 1551 |
+
"grad_norm": 0.8155358573114214,
|
| 1552 |
+
"learning_rate": 2.4412131132637924e-06,
|
| 1553 |
+
"loss": 0.9907,
|
| 1554 |
+
"num_tokens": 20328280.0,
|
| 1555 |
+
"step": 192
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 1.5078125,
|
| 1559 |
+
"grad_norm": 0.7441099106710569,
|
| 1560 |
+
"learning_rate": 2.3996488641594104e-06,
|
| 1561 |
+
"loss": 1.1861,
|
| 1562 |
+
"num_tokens": 20433613.0,
|
| 1563 |
+
"step": 193
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 1.515625,
|
| 1567 |
+
"grad_norm": 0.7954231657805638,
|
| 1568 |
+
"learning_rate": 2.3585821246483352e-06,
|
| 1569 |
+
"loss": 1.1572,
|
| 1570 |
+
"num_tokens": 20547999.0,
|
| 1571 |
+
"step": 194
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.5234375,
|
| 1575 |
+
"grad_norm": 0.7661074186012893,
|
| 1576 |
+
"learning_rate": 2.3180194846605367e-06,
|
| 1577 |
+
"loss": 1.142,
|
| 1578 |
+
"num_tokens": 20671661.0,
|
| 1579 |
+
"step": 195
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 1.53125,
|
| 1583 |
+
"grad_norm": 0.7383604199046713,
|
| 1584 |
+
"learning_rate": 2.2779674532337472e-06,
|
| 1585 |
+
"loss": 1.132,
|
| 1586 |
+
"num_tokens": 20778597.0,
|
| 1587 |
+
"step": 196
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 1.5390625,
|
| 1591 |
+
"grad_norm": 0.6966007241629179,
|
| 1592 |
+
"learning_rate": 2.2384324574689614e-06,
|
| 1593 |
+
"loss": 1.1958,
|
| 1594 |
+
"num_tokens": 20898133.0,
|
| 1595 |
+
"step": 197
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 1.546875,
|
| 1599 |
+
"grad_norm": 0.6892884301659002,
|
| 1600 |
+
"learning_rate": 2.1994208414990847e-06,
|
| 1601 |
+
"loss": 1.1541,
|
| 1602 |
+
"num_tokens": 21025731.0,
|
| 1603 |
+
"step": 198
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.5546875,
|
| 1607 |
+
"grad_norm": 0.6880802437639095,
|
| 1608 |
+
"learning_rate": 2.160938865470903e-06,
|
| 1609 |
+
"loss": 1.1468,
|
| 1610 |
+
"num_tokens": 21149643.0,
|
| 1611 |
+
"step": 199
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 1.5625,
|
| 1615 |
+
"grad_norm": 0.7366918694461689,
|
| 1616 |
+
"learning_rate": 2.1229927045405193e-06,
|
| 1617 |
+
"loss": 1.1744,
|
| 1618 |
+
"num_tokens": 21271207.0,
|
| 1619 |
+
"step": 200
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 1.5703125,
|
| 1623 |
+
"grad_norm": 0.7554102928121678,
|
| 1624 |
+
"learning_rate": 2.085588447882441e-06,
|
| 1625 |
+
"loss": 0.9801,
|
| 1626 |
+
"num_tokens": 21380280.0,
|
| 1627 |
+
"step": 201
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.578125,
|
| 1631 |
+
"grad_norm": 0.7444571406816006,
|
| 1632 |
+
"learning_rate": 2.04873209771245e-06,
|
| 1633 |
+
"loss": 1.2081,
|
| 1634 |
+
"num_tokens": 21494423.0,
|
| 1635 |
+
"step": 202
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 1.5859375,
|
| 1639 |
+
"grad_norm": 0.7181886597109618,
|
| 1640 |
+
"learning_rate": 2.012429568324436e-06,
|
| 1641 |
+
"loss": 1.2704,
|
| 1642 |
+
"num_tokens": 21643937.0,
|
| 1643 |
+
"step": 203
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 1.59375,
|
| 1647 |
+
"grad_norm": 0.7293100242658389,
|
| 1648 |
+
"learning_rate": 1.9766866851413414e-06,
|
| 1649 |
+
"loss": 1.2193,
|
| 1650 |
+
"num_tokens": 21761981.0,
|
| 1651 |
+
"step": 204
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 1.6015625,
|
| 1655 |
+
"grad_norm": 0.8200918180171405,
|
| 1656 |
+
"learning_rate": 1.9415091837803574e-06,
|
| 1657 |
+
"loss": 1.0795,
|
| 1658 |
+
"num_tokens": 21856131.0,
|
| 1659 |
+
"step": 205
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 1.609375,
|
| 1663 |
+
"grad_norm": 0.6876526309331059,
|
| 1664 |
+
"learning_rate": 1.9069027091325396e-06,
|
| 1665 |
+
"loss": 1.2016,
|
| 1666 |
+
"num_tokens": 21990368.0,
|
| 1667 |
+
"step": 206
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 1.6171875,
|
| 1671 |
+
"grad_norm": 0.7042559011362667,
|
| 1672 |
+
"learning_rate": 1.8728728144569785e-06,
|
| 1673 |
+
"loss": 1.1942,
|
| 1674 |
+
"num_tokens": 22102017.0,
|
| 1675 |
+
"step": 207
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 1.625,
|
| 1679 |
+
"grad_norm": 0.7605159991190399,
|
| 1680 |
+
"learning_rate": 1.8394249604896738e-06,
|
| 1681 |
+
"loss": 1.1616,
|
| 1682 |
+
"num_tokens": 22212512.0,
|
| 1683 |
+
"step": 208
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.6328125,
|
| 1687 |
+
"grad_norm": 0.780998142368562,
|
| 1688 |
+
"learning_rate": 1.8065645145672578e-06,
|
| 1689 |
+
"loss": 1.1456,
|
| 1690 |
+
"num_tokens": 22318836.0,
|
| 1691 |
+
"step": 209
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 1.640625,
|
| 1695 |
+
"grad_norm": 0.8098855920402751,
|
| 1696 |
+
"learning_rate": 1.7742967497657043e-06,
|
| 1697 |
+
"loss": 1.0363,
|
| 1698 |
+
"num_tokens": 22415745.0,
|
| 1699 |
+
"step": 210
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 1.6484375,
|
| 1703 |
+
"grad_norm": 0.8107814930537132,
|
| 1704 |
+
"learning_rate": 1.7426268440541678e-06,
|
| 1705 |
+
"loss": 1.2396,
|
| 1706 |
+
"num_tokens": 22531690.0,
|
| 1707 |
+
"step": 211
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 1.65625,
|
| 1711 |
+
"grad_norm": 0.7202126685434231,
|
| 1712 |
+
"learning_rate": 1.7115598794640731e-06,
|
| 1713 |
+
"loss": 1.2456,
|
| 1714 |
+
"num_tokens": 22652863.0,
|
| 1715 |
+
"step": 212
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 1.6640625,
|
| 1719 |
+
"grad_norm": 0.8480447667587903,
|
| 1720 |
+
"learning_rate": 1.6811008412736207e-06,
|
| 1721 |
+
"loss": 0.9323,
|
| 1722 |
+
"num_tokens": 22729271.0,
|
| 1723 |
+
"step": 213
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 1.671875,
|
| 1727 |
+
"grad_norm": 0.769590389444158,
|
| 1728 |
+
"learning_rate": 1.651254617207791e-06,
|
| 1729 |
+
"loss": 1.0984,
|
| 1730 |
+
"num_tokens": 22830436.0,
|
| 1731 |
+
"step": 214
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 1.6796875,
|
| 1735 |
+
"grad_norm": 0.765964114943066,
|
| 1736 |
+
"learning_rate": 1.622025996654027e-06,
|
| 1737 |
+
"loss": 1.1783,
|
| 1738 |
+
"num_tokens": 22930310.0,
|
| 1739 |
+
"step": 215
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.6875,
|
| 1743 |
+
"grad_norm": 0.799158835678557,
|
| 1744 |
+
"learning_rate": 1.5934196698936822e-06,
|
| 1745 |
+
"loss": 1.1953,
|
| 1746 |
+
"num_tokens": 23039737.0,
|
| 1747 |
+
"step": 216
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 1.6953125,
|
| 1751 |
+
"grad_norm": 0.9711730350830016,
|
| 1752 |
+
"learning_rate": 1.5654402273493808e-06,
|
| 1753 |
+
"loss": 0.9642,
|
| 1754 |
+
"num_tokens": 23148702.0,
|
| 1755 |
+
"step": 217
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 1.703125,
|
| 1759 |
+
"grad_norm": 0.8334824198417734,
|
| 1760 |
+
"learning_rate": 1.5380921588483931e-06,
|
| 1761 |
+
"loss": 1.1427,
|
| 1762 |
+
"num_tokens": 23238392.0,
|
| 1763 |
+
"step": 218
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 1.7109375,
|
| 1767 |
+
"grad_norm": 0.7164710982007709,
|
| 1768 |
+
"learning_rate": 1.5113798529021636e-06,
|
| 1769 |
+
"loss": 1.1317,
|
| 1770 |
+
"num_tokens": 23351550.0,
|
| 1771 |
+
"step": 219
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.71875,
|
| 1775 |
+
"grad_norm": 0.6965309778041131,
|
| 1776 |
+
"learning_rate": 1.485307596002087e-06,
|
| 1777 |
+
"loss": 1.1853,
|
| 1778 |
+
"num_tokens": 23465589.0,
|
| 1779 |
+
"step": 220
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 1.7265625,
|
| 1783 |
+
"grad_norm": 0.7769795565876408,
|
| 1784 |
+
"learning_rate": 1.459879571931663e-06,
|
| 1785 |
+
"loss": 1.1107,
|
| 1786 |
+
"num_tokens": 23564701.0,
|
| 1787 |
+
"step": 221
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 1.734375,
|
| 1791 |
+
"grad_norm": 0.8790213964926503,
|
| 1792 |
+
"learning_rate": 1.4350998610951303e-06,
|
| 1793 |
+
"loss": 0.9538,
|
| 1794 |
+
"num_tokens": 23641310.0,
|
| 1795 |
+
"step": 222
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.7421875,
|
| 1799 |
+
"grad_norm": 0.8553804908202591,
|
| 1800 |
+
"learning_rate": 1.410972439862688e-06,
|
| 1801 |
+
"loss": 0.9867,
|
| 1802 |
+
"num_tokens": 23749553.0,
|
| 1803 |
+
"step": 223
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.75,
|
| 1807 |
+
"grad_norm": 0.8032893501540129,
|
| 1808 |
+
"learning_rate": 1.38750117993241e-06,
|
| 1809 |
+
"loss": 1.1523,
|
| 1810 |
+
"num_tokens": 23846533.0,
|
| 1811 |
+
"step": 224
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 1.7578125,
|
| 1815 |
+
"grad_norm": 0.6687621039844708,
|
| 1816 |
+
"learning_rate": 1.3646898477089627e-06,
|
| 1817 |
+
"loss": 1.1897,
|
| 1818 |
+
"num_tokens": 23990878.0,
|
| 1819 |
+
"step": 225
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 1.765625,
|
| 1823 |
+
"grad_norm": 0.8204213823139216,
|
| 1824 |
+
"learning_rate": 1.3425421036992098e-06,
|
| 1825 |
+
"loss": 1.1202,
|
| 1826 |
+
"num_tokens": 24091995.0,
|
| 1827 |
+
"step": 226
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 1.7734375,
|
| 1831 |
+
"grad_norm": 0.7391316889122852,
|
| 1832 |
+
"learning_rate": 1.3210615019248208e-06,
|
| 1833 |
+
"loss": 1.1463,
|
| 1834 |
+
"num_tokens": 24208784.0,
|
| 1835 |
+
"step": 227
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 1.78125,
|
| 1839 |
+
"grad_norm": 0.83630026769931,
|
| 1840 |
+
"learning_rate": 1.3002514893519555e-06,
|
| 1841 |
+
"loss": 0.8598,
|
| 1842 |
+
"num_tokens": 24291362.0,
|
| 1843 |
+
"step": 228
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 1.7890625,
|
| 1847 |
+
"grad_norm": 0.8629314035810544,
|
| 1848 |
+
"learning_rate": 1.2801154053381386e-06,
|
| 1849 |
+
"loss": 1.2049,
|
| 1850 |
+
"num_tokens": 24399097.0,
|
| 1851 |
+
"step": 229
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.796875,
|
| 1855 |
+
"grad_norm": 0.700068994516527,
|
| 1856 |
+
"learning_rate": 1.260656481096393e-06,
|
| 1857 |
+
"loss": 0.9999,
|
| 1858 |
+
"num_tokens": 24538394.0,
|
| 1859 |
+
"step": 230
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 1.8046875,
|
| 1863 |
+
"grad_norm": 0.6421961129868629,
|
| 1864 |
+
"learning_rate": 1.2418778391767342e-06,
|
| 1865 |
+
"loss": 1.2093,
|
| 1866 |
+
"num_tokens": 24679007.0,
|
| 1867 |
+
"step": 231
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 1.8125,
|
| 1871 |
+
"grad_norm": 0.7936130108169558,
|
| 1872 |
+
"learning_rate": 1.2237824929650973e-06,
|
| 1873 |
+
"loss": 1.1381,
|
| 1874 |
+
"num_tokens": 24767202.0,
|
| 1875 |
+
"step": 232
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 1.8203125,
|
| 1879 |
+
"grad_norm": 0.677254485643581,
|
| 1880 |
+
"learning_rate": 1.2063733461997804e-06,
|
| 1881 |
+
"loss": 1.3,
|
| 1882 |
+
"num_tokens": 24919506.0,
|
| 1883 |
+
"step": 233
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.828125,
|
| 1887 |
+
"grad_norm": 0.723105804403935,
|
| 1888 |
+
"learning_rate": 1.1896531925054892e-06,
|
| 1889 |
+
"loss": 1.006,
|
| 1890 |
+
"num_tokens": 25050216.0,
|
| 1891 |
+
"step": 234
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 1.8359375,
|
| 1895 |
+
"grad_norm": 0.6551730939800575,
|
| 1896 |
+
"learning_rate": 1.1736247149450443e-06,
|
| 1897 |
+
"loss": 1.2558,
|
| 1898 |
+
"num_tokens": 25218932.0,
|
| 1899 |
+
"step": 235
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 1.84375,
|
| 1903 |
+
"grad_norm": 0.7980141619699178,
|
| 1904 |
+
"learning_rate": 1.1582904855888347e-06,
|
| 1905 |
+
"loss": 1.1867,
|
| 1906 |
+
"num_tokens": 25331518.0,
|
| 1907 |
+
"step": 236
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.8515625,
|
| 1911 |
+
"grad_norm": 0.7428422637257986,
|
| 1912 |
+
"learning_rate": 1.1436529651020813e-06,
|
| 1913 |
+
"loss": 1.117,
|
| 1914 |
+
"num_tokens": 25433593.0,
|
| 1915 |
+
"step": 237
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 1.859375,
|
| 1919 |
+
"grad_norm": 0.7862456181911465,
|
| 1920 |
+
"learning_rate": 1.129714502349975e-06,
|
| 1921 |
+
"loss": 0.9308,
|
| 1922 |
+
"num_tokens": 25537107.0,
|
| 1923 |
+
"step": 238
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 1.8671875,
|
| 1927 |
+
"grad_norm": 0.6812248397173684,
|
| 1928 |
+
"learning_rate": 1.116477334020759e-06,
|
| 1929 |
+
"loss": 1.176,
|
| 1930 |
+
"num_tokens": 25663580.0,
|
| 1931 |
+
"step": 239
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 1.875,
|
| 1935 |
+
"grad_norm": 0.669230252389982,
|
| 1936 |
+
"learning_rate": 1.1039435842668095e-06,
|
| 1937 |
+
"loss": 1.2768,
|
| 1938 |
+
"num_tokens": 25808549.0,
|
| 1939 |
+
"step": 240
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 1.8828125,
|
| 1943 |
+
"grad_norm": 0.7205241998404597,
|
| 1944 |
+
"learning_rate": 1.092115264363775e-06,
|
| 1945 |
+
"loss": 1.0597,
|
| 1946 |
+
"num_tokens": 25934778.0,
|
| 1947 |
+
"step": 241
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 1.890625,
|
| 1951 |
+
"grad_norm": 0.9103634896519411,
|
| 1952 |
+
"learning_rate": 1.080994272387831e-06,
|
| 1953 |
+
"loss": 0.8861,
|
| 1954 |
+
"num_tokens": 26001231.0,
|
| 1955 |
+
"step": 242
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 1.8984375,
|
| 1959 |
+
"grad_norm": 0.7209364795151052,
|
| 1960 |
+
"learning_rate": 1.0705823929110962e-06,
|
| 1961 |
+
"loss": 1.1561,
|
| 1962 |
+
"num_tokens": 26106051.0,
|
| 1963 |
+
"step": 243
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.90625,
|
| 1967 |
+
"grad_norm": 0.778694389353822,
|
| 1968 |
+
"learning_rate": 1.060881296715265e-06,
|
| 1969 |
+
"loss": 1.181,
|
| 1970 |
+
"num_tokens": 26210742.0,
|
| 1971 |
+
"step": 244
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 1.9140625,
|
| 1975 |
+
"grad_norm": 0.7758610155301724,
|
| 1976 |
+
"learning_rate": 1.0518925405234988e-06,
|
| 1977 |
+
"loss": 1.1989,
|
| 1978 |
+
"num_tokens": 26324416.0,
|
| 1979 |
+
"step": 245
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 1.921875,
|
| 1983 |
+
"grad_norm": 0.8487994142264144,
|
| 1984 |
+
"learning_rate": 1.0436175667506216e-06,
|
| 1985 |
+
"loss": 0.8846,
|
| 1986 |
+
"num_tokens": 26408914.0,
|
| 1987 |
+
"step": 246
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 1.9296875,
|
| 1991 |
+
"grad_norm": 0.8697814558131984,
|
| 1992 |
+
"learning_rate": 1.0360577032716565e-06,
|
| 1993 |
+
"loss": 1.0735,
|
| 1994 |
+
"num_tokens": 26479130.0,
|
| 1995 |
+
"step": 247
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 1.9375,
|
| 1999 |
+
"grad_norm": 0.7798945211987117,
|
| 2000 |
+
"learning_rate": 1.029214163208744e-06,
|
| 2001 |
+
"loss": 1.1957,
|
| 2002 |
+
"num_tokens": 26598719.0,
|
| 2003 |
+
"step": 248
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 1.9453125,
|
| 2007 |
+
"grad_norm": 0.7427049429468312,
|
| 2008 |
+
"learning_rate": 1.0230880447364719e-06,
|
| 2009 |
+
"loss": 1.1566,
|
| 2010 |
+
"num_tokens": 26704847.0,
|
| 2011 |
+
"step": 249
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 1.953125,
|
| 2015 |
+
"grad_norm": 0.6761729959695535,
|
| 2016 |
+
"learning_rate": 1.017680330905656e-06,
|
| 2017 |
+
"loss": 1.2619,
|
| 2018 |
+
"num_tokens": 26846721.0,
|
| 2019 |
+
"step": 250
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.9609375,
|
| 2023 |
+
"grad_norm": 0.7475739004508511,
|
| 2024 |
+
"learning_rate": 1.0129918894855868e-06,
|
| 2025 |
+
"loss": 1.1663,
|
| 2026 |
+
"num_tokens": 26970177.0,
|
| 2027 |
+
"step": 251
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 1.96875,
|
| 2031 |
+
"grad_norm": 0.7296678764365407,
|
| 2032 |
+
"learning_rate": 1.009023472824783e-06,
|
| 2033 |
+
"loss": 1.1539,
|
| 2034 |
+
"num_tokens": 27081433.0,
|
| 2035 |
+
"step": 252
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 1.9765625,
|
| 2039 |
+
"grad_norm": 0.6746325329726267,
|
| 2040 |
+
"learning_rate": 1.0057757177302628e-06,
|
| 2041 |
+
"loss": 1.2187,
|
| 2042 |
+
"num_tokens": 27211598.0,
|
| 2043 |
+
"step": 253
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 1.984375,
|
| 2047 |
+
"grad_norm": 0.8338233963854741,
|
| 2048 |
+
"learning_rate": 1.003249145365353e-06,
|
| 2049 |
+
"loss": 1.0428,
|
| 2050 |
+
"num_tokens": 27295798.0,
|
| 2051 |
+
"step": 254
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 1.9921875,
|
| 2055 |
+
"grad_norm": 0.8204506315224706,
|
| 2056 |
+
"learning_rate": 1.0014441611660628e-06,
|
| 2057 |
+
"loss": 1.1588,
|
| 2058 |
+
"num_tokens": 27392305.0,
|
| 2059 |
+
"step": 255
|
| 2060 |
+
},
|
| 2061 |
{
|
| 2062 |
"epoch": 2.0,
|
| 2063 |
+
"grad_norm": 0.6380605222559861,
|
| 2064 |
+
"learning_rate": 1.0003610547760215e-06,
|
| 2065 |
+
"loss": 1.2573,
|
| 2066 |
+
"num_tokens": 27553671.0,
|
| 2067 |
+
"step": 256
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.0,
|
| 2071 |
+
"eval_loss": 0.6671378016471863,
|
| 2072 |
+
"eval_num_tokens": 27553671.0,
|
| 2073 |
+
"eval_runtime": 26.6823,
|
| 2074 |
+
"eval_samples_per_second": 17.09,
|
| 2075 |
+
"eval_steps_per_second": 4.272,
|
| 2076 |
+
"step": 256
|
| 2077 |
},
|
| 2078 |
{
|
| 2079 |
"epoch": 2.0,
|
| 2080 |
+
"step": 256,
|
| 2081 |
+
"total_flos": 90995475070976.0,
|
| 2082 |
+
"train_loss": 1.261348171858117,
|
| 2083 |
+
"train_runtime": 1909.4501,
|
| 2084 |
+
"train_samples_per_second": 4.289,
|
| 2085 |
+
"train_steps_per_second": 0.134
|
| 2086 |
}
|
| 2087 |
],
|
| 2088 |
"logging_steps": 1,
|
| 2089 |
+
"max_steps": 256,
|
| 2090 |
"num_input_tokens_seen": 0,
|
| 2091 |
"num_train_epochs": 2,
|
| 2092 |
"save_steps": 500,
|
|
|
|
| 2102 |
"attributes": {}
|
| 2103 |
}
|
| 2104 |
},
|
| 2105 |
+
"total_flos": 90995475070976.0,
|
| 2106 |
+
"train_batch_size": 8,
|
| 2107 |
"trial_name": null,
|
| 2108 |
"trial_params": null
|
| 2109 |
}
|