End of training
Browse files
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: other
|
| 4 |
base_model: facebook/mask2former-swin-base-IN21k-ade-semantic
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: mask2former-finetuned-ER-Mito-LD3
|
|
@@ -14,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# mask2former-finetuned-ER-Mito-LD3
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [facebook/mask2former-swin-base-IN21k-ade-semantic](https://huggingface.co/facebook/mask2former-swin-base-IN21k-ade-semantic) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 39.
|
| 20 |
- Dummy: 1.0
|
| 21 |
|
| 22 |
## Model description
|
|
|
|
| 3 |
license: other
|
| 4 |
base_model: facebook/mask2former-swin-base-IN21k-ade-semantic
|
| 5 |
tags:
|
| 6 |
+
- image-segmentation
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: mask2former-finetuned-ER-Mito-LD3
|
|
|
|
| 16 |
|
| 17 |
# mask2former-finetuned-ER-Mito-LD3
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [facebook/mask2former-swin-base-IN21k-ade-semantic](https://huggingface.co/facebook/mask2former-swin-base-IN21k-ade-semantic) on the Dnq2025/Mask2former_Pretrain dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 39.9236
|
| 22 |
- Dummy: 1.0
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_dummy": 1.0,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 2.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"total_flos": 1.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
"eval_dummy": 1.0,
|
| 4 |
+
"eval_loss": 39.92361831665039,
|
| 5 |
+
"eval_runtime": 17.0469,
|
| 6 |
+
"eval_samples_per_second": 2.816,
|
| 7 |
+
"eval_steps_per_second": 0.704,
|
| 8 |
+
"total_flos": 1.4631500418239693e+19,
|
| 9 |
+
"train_loss": 16.309644344832545,
|
| 10 |
+
"train_runtime": 27199.2721,
|
| 11 |
+
"train_samples_per_second": 1.897,
|
| 12 |
+
"train_steps_per_second": 0.474
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_dummy": 1.0,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 2.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
"eval_dummy": 1.0,
|
| 4 |
+
"eval_loss": 39.92361831665039,
|
| 5 |
+
"eval_runtime": 17.0469,
|
| 6 |
+
"eval_samples_per_second": 2.816,
|
| 7 |
+
"eval_steps_per_second": 0.704
|
| 8 |
}
|
runs/Mar27_08-43-10_cdr2539.int.cedar.computecanada.ca/events.out.tfevents.1743117495.cdr2539.int.cedar.computecanada.ca.2316099.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca0b1280ead51acad1dc481185973289077fe063d3e3d99f1c5fa45ab41ed821
|
| 3 |
+
size 408
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"total_flos": 1.4631500418239693e+19,
|
| 4 |
+
"train_loss": 16.309644344832545,
|
| 5 |
+
"train_runtime": 27199.2721,
|
| 6 |
+
"train_samples_per_second": 1.897,
|
| 7 |
+
"train_steps_per_second": 0.474
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 12900,
|
| 8 |
"is_hyper_param_search": false,
|
|
@@ -10,2047 +10,1822 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"epoch": 0.
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss":
|
| 17 |
"step": 100
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
"eval_dummy": 1.0,
|
| 22 |
-
"eval_loss":
|
| 23 |
-
"eval_runtime":
|
| 24 |
-
"eval_samples_per_second": 2.
|
| 25 |
-
"eval_steps_per_second": 0.
|
| 26 |
-
"step":
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 1.
|
| 30 |
-
"grad_norm":
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 200
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.0,
|
| 37 |
"eval_dummy": 1.0,
|
| 38 |
-
"eval_loss":
|
| 39 |
-
"eval_runtime":
|
| 40 |
-
"eval_samples_per_second": 2.
|
| 41 |
-
"eval_steps_per_second": 0.
|
| 42 |
-
"step":
|
| 43 |
},
|
| 44 |
{
|
| 45 |
-
"epoch": 2.
|
| 46 |
-
"grad_norm":
|
| 47 |
-
"learning_rate":
|
| 48 |
-
"loss":
|
| 49 |
"step": 300
|
| 50 |
},
|
| 51 |
{
|
| 52 |
"epoch": 3.0,
|
| 53 |
"eval_dummy": 1.0,
|
| 54 |
-
"eval_loss":
|
| 55 |
-
"eval_runtime":
|
| 56 |
-
"eval_samples_per_second": 2.
|
| 57 |
-
"eval_steps_per_second": 0.
|
| 58 |
-
"step":
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 3.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch":
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
-
"
|
| 72 |
-
"eval_samples_per_second": 2.833,
|
| 73 |
-
"eval_steps_per_second": 0.59,
|
| 74 |
-
"step": 416
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 4.8076923076923075,
|
| 78 |
-
"grad_norm": 112.39118194580078,
|
| 79 |
-
"learning_rate": 9.612790697674419e-05,
|
| 80 |
-
"loss": 24.7501,
|
| 81 |
"step": 500
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch":
|
| 85 |
"eval_dummy": 1.0,
|
| 86 |
-
"eval_loss":
|
| 87 |
-
"eval_runtime":
|
| 88 |
-
"eval_samples_per_second": 2.
|
| 89 |
-
"eval_steps_per_second": 0.
|
| 90 |
-
"step":
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"grad_norm":
|
| 95 |
-
"learning_rate":
|
| 96 |
-
"loss":
|
| 97 |
"step": 600
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch":
|
| 101 |
"eval_dummy": 1.0,
|
| 102 |
-
"eval_loss":
|
| 103 |
-
"eval_runtime":
|
| 104 |
-
"eval_samples_per_second": 2.
|
| 105 |
-
"eval_steps_per_second": 0.
|
| 106 |
-
"step":
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch":
|
| 110 |
-
"grad_norm":
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss":
|
| 113 |
"step": 700
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
"eval_dummy": 1.0,
|
| 118 |
-
"eval_loss":
|
| 119 |
-
"eval_runtime": 17.
|
| 120 |
-
"eval_samples_per_second": 2.
|
| 121 |
-
"eval_steps_per_second": 0.
|
| 122 |
-
"step":
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss":
|
| 129 |
"step": 800
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch":
|
| 133 |
-
"
|
| 134 |
-
"
|
| 135 |
-
"
|
| 136 |
-
"eval_samples_per_second": 2.952,
|
| 137 |
-
"eval_steps_per_second": 0.615,
|
| 138 |
-
"step": 832
|
| 139 |
-
},
|
| 140 |
-
{
|
| 141 |
-
"epoch": 8.653846153846153,
|
| 142 |
-
"grad_norm": 69.99388122558594,
|
| 143 |
-
"learning_rate": 9.303023255813954e-05,
|
| 144 |
-
"loss": 20.373,
|
| 145 |
"step": 900
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch":
|
| 149 |
"eval_dummy": 1.0,
|
| 150 |
-
"eval_loss":
|
| 151 |
-
"eval_runtime":
|
| 152 |
-
"eval_samples_per_second": 2.
|
| 153 |
-
"eval_steps_per_second": 0.
|
| 154 |
-
"step":
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"epoch":
|
| 158 |
-
"grad_norm":
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss":
|
| 161 |
"step": 1000
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
"eval_dummy": 1.0,
|
| 166 |
-
"eval_loss":
|
| 167 |
-
"eval_runtime":
|
| 168 |
-
"eval_samples_per_second": 2.
|
| 169 |
-
"eval_steps_per_second": 0.
|
| 170 |
-
"step":
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
"step": 1100
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
"eval_dummy": 1.0,
|
| 182 |
-
"eval_loss":
|
| 183 |
-
"eval_runtime":
|
| 184 |
-
"eval_samples_per_second": 2.
|
| 185 |
-
"eval_steps_per_second": 0.
|
| 186 |
-
"step":
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch":
|
| 190 |
-
"grad_norm":
|
| 191 |
-
"learning_rate":
|
| 192 |
-
"loss":
|
| 193 |
"step": 1200
|
| 194 |
},
|
| 195 |
{
|
| 196 |
-
"epoch":
|
| 197 |
"eval_dummy": 1.0,
|
| 198 |
-
"eval_loss":
|
| 199 |
-
"eval_runtime":
|
| 200 |
-
"eval_samples_per_second": 2.
|
| 201 |
-
"eval_steps_per_second": 0.
|
| 202 |
-
"step":
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch":
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss":
|
| 209 |
"step": 1300
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"
|
| 214 |
-
"
|
| 215 |
-
"
|
| 216 |
-
"eval_samples_per_second": 2.853,
|
| 217 |
-
"eval_steps_per_second": 0.594,
|
| 218 |
-
"step": 1352
|
| 219 |
-
},
|
| 220 |
-
{
|
| 221 |
-
"epoch": 13.461538461538462,
|
| 222 |
-
"grad_norm": 33.62112808227539,
|
| 223 |
-
"learning_rate": 8.915813953488373e-05,
|
| 224 |
-
"loss": 18.3621,
|
| 225 |
"step": 1400
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch":
|
| 229 |
"eval_dummy": 1.0,
|
| 230 |
-
"eval_loss":
|
| 231 |
-
"eval_runtime":
|
| 232 |
-
"eval_samples_per_second": 2.
|
| 233 |
-
"eval_steps_per_second": 0.
|
| 234 |
-
"step":
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss":
|
| 241 |
"step": 1500
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
"eval_dummy": 1.0,
|
| 246 |
-
"eval_loss":
|
| 247 |
-
"eval_runtime":
|
| 248 |
-
"eval_samples_per_second": 2.
|
| 249 |
-
"eval_steps_per_second": 0.
|
| 250 |
-
"step":
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch":
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss":
|
| 257 |
"step": 1600
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
"eval_dummy": 1.0,
|
| 262 |
-
"eval_loss":
|
| 263 |
-
"eval_runtime":
|
| 264 |
-
"eval_samples_per_second": 2.
|
| 265 |
-
"eval_steps_per_second": 0.
|
| 266 |
-
"step":
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch":
|
| 270 |
-
"grad_norm":
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss":
|
| 273 |
"step": 1700
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
"
|
| 280 |
-
"eval_samples_per_second": 2.719,
|
| 281 |
-
"eval_steps_per_second": 0.566,
|
| 282 |
-
"step": 1768
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 17.307692307692307,
|
| 286 |
-
"grad_norm": 31.138574600219727,
|
| 287 |
-
"learning_rate": 8.606046511627907e-05,
|
| 288 |
-
"loss": 17.8402,
|
| 289 |
"step": 1800
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch":
|
| 293 |
"eval_dummy": 1.0,
|
| 294 |
-
"eval_loss":
|
| 295 |
-
"eval_runtime":
|
| 296 |
-
"eval_samples_per_second": 2.
|
| 297 |
-
"eval_steps_per_second": 0.
|
| 298 |
-
"step":
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch":
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss":
|
| 305 |
"step": 1900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
"eval_dummy": 1.0,
|
| 310 |
-
"eval_loss":
|
| 311 |
-
"eval_runtime":
|
| 312 |
-
"eval_samples_per_second": 2.
|
| 313 |
-
"eval_steps_per_second": 0.
|
| 314 |
-
"step":
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch":
|
| 318 |
-
"grad_norm":
|
| 319 |
-
"learning_rate":
|
| 320 |
-
"loss":
|
| 321 |
"step": 2000
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
"eval_dummy": 1.0,
|
| 326 |
-
"eval_loss":
|
| 327 |
-
"eval_runtime":
|
| 328 |
-
"eval_samples_per_second": 2.
|
| 329 |
-
"eval_steps_per_second": 0.
|
| 330 |
-
"step":
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch":
|
| 334 |
-
"grad_norm":
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"loss":
|
| 337 |
"step": 2100
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
"eval_dummy": 1.0,
|
| 342 |
-
"eval_loss":
|
| 343 |
-
"eval_runtime":
|
| 344 |
-
"eval_samples_per_second": 2.
|
| 345 |
-
"eval_steps_per_second": 0.
|
| 346 |
-
"step":
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss":
|
| 353 |
"step": 2200
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"
|
| 358 |
-
"
|
| 359 |
-
"
|
| 360 |
-
"eval_samples_per_second": 2.942,
|
| 361 |
-
"eval_steps_per_second": 0.613,
|
| 362 |
-
"step": 2288
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"epoch": 22.115384615384617,
|
| 366 |
-
"grad_norm": 23.20075798034668,
|
| 367 |
-
"learning_rate": 8.218837209302326e-05,
|
| 368 |
-
"loss": 16.6797,
|
| 369 |
"step": 2300
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
"eval_dummy": 1.0,
|
| 374 |
-
"eval_loss":
|
| 375 |
-
"eval_runtime": 17.
|
| 376 |
-
"eval_samples_per_second": 2.
|
| 377 |
-
"eval_steps_per_second": 0.
|
| 378 |
-
"step":
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"grad_norm":
|
| 383 |
-
"learning_rate":
|
| 384 |
-
"loss":
|
| 385 |
"step": 2400
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
"eval_dummy": 1.0,
|
| 390 |
-
"eval_loss":
|
| 391 |
-
"eval_runtime":
|
| 392 |
-
"eval_samples_per_second": 2.
|
| 393 |
-
"eval_steps_per_second": 0.
|
| 394 |
-
"step":
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch":
|
| 398 |
-
"grad_norm":
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss":
|
| 401 |
"step": 2500
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"
|
| 406 |
-
"
|
| 407 |
-
"
|
| 408 |
-
"
|
|
|
|
|
|
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch":
|
| 412 |
-
"
|
| 413 |
-
"
|
| 414 |
-
"
|
| 415 |
-
"eval_samples_per_second": 2.885,
|
| 416 |
-
"eval_steps_per_second": 0.601,
|
| 417 |
"step": 2600
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate":
|
| 423 |
-
"loss":
|
| 424 |
"step": 2700
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
"eval_dummy": 1.0,
|
| 429 |
-
"eval_loss":
|
| 430 |
-
"eval_runtime":
|
| 431 |
-
"eval_samples_per_second": 2.
|
| 432 |
-
"eval_steps_per_second": 0.
|
| 433 |
-
"step":
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss":
|
| 440 |
"step": 2800
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch":
|
| 444 |
"eval_dummy": 1.0,
|
| 445 |
-
"eval_loss":
|
| 446 |
-
"eval_runtime":
|
| 447 |
-
"eval_samples_per_second": 2.
|
| 448 |
-
"eval_steps_per_second": 0.
|
| 449 |
-
"step":
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"grad_norm":
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss":
|
| 456 |
"step": 2900
|
| 457 |
},
|
| 458 |
{
|
| 459 |
-
"epoch":
|
| 460 |
"eval_dummy": 1.0,
|
| 461 |
-
"eval_loss":
|
| 462 |
-
"eval_runtime": 17.
|
| 463 |
-
"eval_samples_per_second": 2.
|
| 464 |
-
"eval_steps_per_second": 0.
|
| 465 |
-
"step":
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"grad_norm":
|
| 470 |
-
"learning_rate":
|
| 471 |
-
"loss":
|
| 472 |
"step": 3000
|
| 473 |
},
|
| 474 |
{
|
| 475 |
-
"epoch":
|
| 476 |
"eval_dummy": 1.0,
|
| 477 |
-
"eval_loss":
|
| 478 |
-
"eval_runtime": 17.
|
| 479 |
-
"eval_samples_per_second": 2.
|
| 480 |
-
"eval_steps_per_second": 0.
|
| 481 |
-
"step":
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"grad_norm":
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss":
|
| 488 |
"step": 3100
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"
|
| 493 |
-
"
|
| 494 |
-
"
|
| 495 |
-
"eval_samples_per_second": 2.88,
|
| 496 |
-
"eval_steps_per_second": 0.6,
|
| 497 |
-
"step": 3120
|
| 498 |
-
},
|
| 499 |
-
{
|
| 500 |
-
"epoch": 30.76923076923077,
|
| 501 |
-
"grad_norm": 23.02336883544922,
|
| 502 |
-
"learning_rate": 7.52186046511628e-05,
|
| 503 |
-
"loss": 15.0833,
|
| 504 |
"step": 3200
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch":
|
| 508 |
"eval_dummy": 1.0,
|
| 509 |
-
"eval_loss":
|
| 510 |
-
"eval_runtime":
|
| 511 |
-
"eval_samples_per_second": 2.
|
| 512 |
-
"eval_steps_per_second": 0.
|
| 513 |
-
"step":
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch":
|
| 517 |
-
"grad_norm": 24.
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss":
|
| 520 |
"step": 3300
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch":
|
| 524 |
"eval_dummy": 1.0,
|
| 525 |
-
"eval_loss": 28.
|
| 526 |
-
"eval_runtime":
|
| 527 |
-
"eval_samples_per_second": 2.
|
| 528 |
-
"eval_steps_per_second": 0.
|
| 529 |
-
"step":
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"epoch":
|
| 533 |
-
"grad_norm":
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss":
|
| 536 |
"step": 3400
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
"eval_dummy": 1.0,
|
| 541 |
-
"eval_loss": 28.
|
| 542 |
-
"eval_runtime":
|
| 543 |
-
"eval_samples_per_second": 2.
|
| 544 |
-
"eval_steps_per_second": 0.
|
| 545 |
-
"step":
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"grad_norm":
|
| 550 |
-
"learning_rate":
|
| 551 |
-
"loss":
|
| 552 |
"step": 3500
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
| 556 |
-
"
|
| 557 |
-
"
|
| 558 |
-
"
|
| 559 |
-
"eval_samples_per_second": 2.799,
|
| 560 |
-
"eval_steps_per_second": 0.583,
|
| 561 |
-
"step": 3536
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 34.61538461538461,
|
| 565 |
-
"grad_norm": 34.329986572265625,
|
| 566 |
-
"learning_rate": 7.212093023255815e-05,
|
| 567 |
-
"loss": 14.7177,
|
| 568 |
"step": 3600
|
| 569 |
},
|
| 570 |
{
|
| 571 |
-
"epoch":
|
| 572 |
"eval_dummy": 1.0,
|
| 573 |
-
"eval_loss":
|
| 574 |
-
"eval_runtime": 17.
|
| 575 |
-
"eval_samples_per_second": 2.
|
| 576 |
-
"eval_steps_per_second": 0.
|
| 577 |
-
"step":
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch":
|
| 581 |
-
"grad_norm":
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss":
|
| 584 |
"step": 3700
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
"eval_dummy": 1.0,
|
| 589 |
-
"eval_loss":
|
| 590 |
-
"eval_runtime":
|
| 591 |
-
"eval_samples_per_second": 2.
|
| 592 |
-
"eval_steps_per_second": 0.
|
| 593 |
-
"step":
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch":
|
| 597 |
-
"grad_norm":
|
| 598 |
-
"learning_rate":
|
| 599 |
-
"loss":
|
| 600 |
"step": 3800
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch":
|
| 604 |
"eval_dummy": 1.0,
|
| 605 |
-
"eval_loss":
|
| 606 |
-
"eval_runtime": 16.
|
| 607 |
-
"eval_samples_per_second": 2.
|
| 608 |
-
"eval_steps_per_second": 0.
|
| 609 |
-
"step":
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch":
|
| 613 |
-
"grad_norm":
|
| 614 |
-
"learning_rate":
|
| 615 |
-
"loss":
|
| 616 |
"step": 3900
|
| 617 |
},
|
| 618 |
{
|
| 619 |
-
"epoch":
|
| 620 |
"eval_dummy": 1.0,
|
| 621 |
-
"eval_loss":
|
| 622 |
-
"eval_runtime":
|
| 623 |
-
"eval_samples_per_second": 2.
|
| 624 |
-
"eval_steps_per_second": 0.
|
| 625 |
-
"step":
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch":
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss":
|
| 632 |
"step": 4000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"
|
| 637 |
-
"
|
| 638 |
-
"
|
| 639 |
-
"eval_samples_per_second": 2.871,
|
| 640 |
-
"eval_steps_per_second": 0.598,
|
| 641 |
-
"step": 4056
|
| 642 |
-
},
|
| 643 |
-
{
|
| 644 |
-
"epoch": 39.42307692307692,
|
| 645 |
-
"grad_norm": 24.0490665435791,
|
| 646 |
-
"learning_rate": 6.824883720930233e-05,
|
| 647 |
-
"loss": 13.7545,
|
| 648 |
"step": 4100
|
| 649 |
},
|
| 650 |
{
|
| 651 |
-
"epoch":
|
| 652 |
"eval_dummy": 1.0,
|
| 653 |
-
"eval_loss":
|
| 654 |
-
"eval_runtime": 16.
|
| 655 |
-
"eval_samples_per_second": 2.
|
| 656 |
-
"eval_steps_per_second": 0.
|
| 657 |
-
"step":
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch":
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate":
|
| 663 |
-
"loss":
|
| 664 |
"step": 4200
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"epoch":
|
| 668 |
"eval_dummy": 1.0,
|
| 669 |
-
"eval_loss": 29.
|
| 670 |
-
"eval_runtime":
|
| 671 |
-
"eval_samples_per_second": 2.
|
| 672 |
-
"eval_steps_per_second": 0.
|
| 673 |
-
"step":
|
| 674 |
},
|
| 675 |
{
|
| 676 |
-
"epoch":
|
| 677 |
-
"grad_norm":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss":
|
| 680 |
"step": 4300
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
"eval_dummy": 1.0,
|
| 685 |
-
"eval_loss":
|
| 686 |
-
"eval_runtime": 16.
|
| 687 |
-
"eval_samples_per_second": 2.
|
| 688 |
-
"eval_steps_per_second": 0.
|
| 689 |
-
"step":
|
| 690 |
},
|
| 691 |
{
|
| 692 |
-
"epoch":
|
| 693 |
-
"grad_norm":
|
| 694 |
-
"learning_rate":
|
| 695 |
-
"loss":
|
| 696 |
"step": 4400
|
| 697 |
},
|
| 698 |
{
|
| 699 |
-
"epoch":
|
| 700 |
-
"
|
| 701 |
-
"
|
| 702 |
-
"
|
| 703 |
-
"eval_samples_per_second": 2.847,
|
| 704 |
-
"eval_steps_per_second": 0.593,
|
| 705 |
-
"step": 4472
|
| 706 |
-
},
|
| 707 |
-
{
|
| 708 |
-
"epoch": 43.26923076923077,
|
| 709 |
-
"grad_norm": 21.939144134521484,
|
| 710 |
-
"learning_rate": 6.515116279069768e-05,
|
| 711 |
-
"loss": 13.6644,
|
| 712 |
"step": 4500
|
| 713 |
},
|
| 714 |
{
|
| 715 |
-
"epoch":
|
| 716 |
"eval_dummy": 1.0,
|
| 717 |
-
"eval_loss": 29.
|
| 718 |
-
"eval_runtime":
|
| 719 |
-
"eval_samples_per_second": 2.
|
| 720 |
-
"eval_steps_per_second": 0.
|
| 721 |
-
"step":
|
| 722 |
},
|
| 723 |
{
|
| 724 |
-
"epoch":
|
| 725 |
-
"grad_norm":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss":
|
| 728 |
"step": 4600
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
"eval_dummy": 1.0,
|
| 733 |
-
"eval_loss":
|
| 734 |
-
"eval_runtime": 16.
|
| 735 |
-
"eval_samples_per_second": 2.
|
| 736 |
-
"eval_steps_per_second": 0.
|
| 737 |
-
"step":
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch":
|
| 741 |
-
"grad_norm":
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss":
|
| 744 |
"step": 4700
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch":
|
| 748 |
"eval_dummy": 1.0,
|
| 749 |
-
"eval_loss": 29.
|
| 750 |
-
"eval_runtime": 17.
|
| 751 |
-
"eval_samples_per_second": 2.
|
| 752 |
-
"eval_steps_per_second": 0.
|
| 753 |
-
"step":
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch":
|
| 757 |
-
"grad_norm":
|
| 758 |
-
"learning_rate":
|
| 759 |
-
"loss":
|
| 760 |
"step": 4800
|
| 761 |
},
|
| 762 |
{
|
| 763 |
-
"epoch":
|
| 764 |
-
"
|
| 765 |
-
"
|
| 766 |
-
"
|
| 767 |
-
"eval_samples_per_second": 2.872,
|
| 768 |
-
"eval_steps_per_second": 0.598,
|
| 769 |
-
"step": 4888
|
| 770 |
-
},
|
| 771 |
-
{
|
| 772 |
-
"epoch": 47.11538461538461,
|
| 773 |
-
"grad_norm": 16.147850036621094,
|
| 774 |
-
"learning_rate": 6.205348837209302e-05,
|
| 775 |
-
"loss": 12.8073,
|
| 776 |
"step": 4900
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch":
|
| 780 |
"eval_dummy": 1.0,
|
| 781 |
-
"eval_loss":
|
| 782 |
-
"eval_runtime": 17.
|
| 783 |
-
"eval_samples_per_second": 2.
|
| 784 |
-
"eval_steps_per_second": 0.
|
| 785 |
-
"step":
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"epoch":
|
| 789 |
-
"grad_norm":
|
| 790 |
-
"learning_rate":
|
| 791 |
-
"loss":
|
| 792 |
"step": 5000
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch":
|
| 796 |
"eval_dummy": 1.0,
|
| 797 |
-
"eval_loss":
|
| 798 |
-
"eval_runtime":
|
| 799 |
-
"eval_samples_per_second": 2.
|
| 800 |
-
"eval_steps_per_second": 0.
|
| 801 |
-
"step":
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch":
|
| 805 |
-
"grad_norm":
|
| 806 |
-
"learning_rate":
|
| 807 |
-
"loss":
|
| 808 |
"step": 5100
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch":
|
| 812 |
-
"
|
| 813 |
-
"
|
| 814 |
-
"
|
| 815 |
-
"
|
|
|
|
|
|
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"
|
| 820 |
-
"
|
| 821 |
-
"
|
| 822 |
-
"eval_samples_per_second": 2.857,
|
| 823 |
-
"eval_steps_per_second": 0.595,
|
| 824 |
"step": 5200
|
| 825 |
},
|
| 826 |
{
|
| 827 |
-
"epoch":
|
| 828 |
-
"
|
| 829 |
-
"
|
| 830 |
-
"
|
| 831 |
-
"
|
|
|
|
|
|
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch":
|
| 835 |
-
"
|
| 836 |
-
"
|
| 837 |
-
"
|
| 838 |
-
"
|
| 839 |
-
"eval_steps_per_second": 0.61,
|
| 840 |
-
"step": 5304
|
| 841 |
},
|
| 842 |
{
|
| 843 |
-
"epoch":
|
| 844 |
-
"grad_norm":
|
| 845 |
-
"learning_rate":
|
| 846 |
-
"loss":
|
| 847 |
"step": 5400
|
| 848 |
},
|
| 849 |
{
|
| 850 |
-
"epoch":
|
| 851 |
"eval_dummy": 1.0,
|
| 852 |
-
"eval_loss": 31.
|
| 853 |
-
"eval_runtime":
|
| 854 |
-
"eval_samples_per_second": 2.
|
| 855 |
-
"eval_steps_per_second": 0.
|
| 856 |
-
"step":
|
| 857 |
},
|
| 858 |
{
|
| 859 |
-
"epoch":
|
| 860 |
-
"grad_norm":
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss":
|
| 863 |
"step": 5500
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch":
|
| 867 |
"eval_dummy": 1.0,
|
| 868 |
-
"eval_loss":
|
| 869 |
-
"eval_runtime":
|
| 870 |
-
"eval_samples_per_second": 2.
|
| 871 |
-
"eval_steps_per_second": 0.
|
| 872 |
-
"step":
|
| 873 |
},
|
| 874 |
{
|
| 875 |
-
"epoch":
|
| 876 |
-
"grad_norm":
|
| 877 |
-
"learning_rate":
|
| 878 |
-
"loss":
|
| 879 |
"step": 5600
|
| 880 |
},
|
| 881 |
{
|
| 882 |
-
"epoch":
|
| 883 |
"eval_dummy": 1.0,
|
| 884 |
-
"eval_loss":
|
| 885 |
-
"eval_runtime":
|
| 886 |
-
"eval_samples_per_second": 2.
|
| 887 |
-
"eval_steps_per_second": 0.
|
| 888 |
-
"step":
|
| 889 |
},
|
| 890 |
{
|
| 891 |
-
"epoch":
|
| 892 |
-
"grad_norm":
|
| 893 |
-
"learning_rate":
|
| 894 |
-
"loss":
|
| 895 |
"step": 5700
|
| 896 |
},
|
| 897 |
{
|
| 898 |
-
"epoch":
|
| 899 |
-
"
|
| 900 |
-
"
|
| 901 |
-
"
|
| 902 |
-
"eval_samples_per_second": 2.91,
|
| 903 |
-
"eval_steps_per_second": 0.606,
|
| 904 |
-
"step": 5720
|
| 905 |
-
},
|
| 906 |
-
{
|
| 907 |
-
"epoch": 55.76923076923077,
|
| 908 |
-
"grad_norm": 19.976686477661133,
|
| 909 |
-
"learning_rate": 5.5083720930232566e-05,
|
| 910 |
-
"loss": 12.8723,
|
| 911 |
"step": 5800
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
"eval_dummy": 1.0,
|
| 916 |
-
"eval_loss": 30.
|
| 917 |
-
"eval_runtime":
|
| 918 |
-
"eval_samples_per_second": 2.
|
| 919 |
-
"eval_steps_per_second": 0.
|
| 920 |
-
"step":
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"epoch":
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"learning_rate":
|
| 926 |
-
"loss":
|
| 927 |
"step": 5900
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch":
|
| 931 |
"eval_dummy": 1.0,
|
| 932 |
-
"eval_loss": 30.
|
| 933 |
-
"eval_runtime":
|
| 934 |
-
"eval_samples_per_second": 2.
|
| 935 |
-
"eval_steps_per_second": 0.
|
| 936 |
-
"step":
|
| 937 |
},
|
| 938 |
{
|
| 939 |
-
"epoch":
|
| 940 |
-
"grad_norm":
|
| 941 |
-
"learning_rate":
|
| 942 |
-
"loss":
|
| 943 |
"step": 6000
|
| 944 |
},
|
| 945 |
{
|
| 946 |
-
"epoch":
|
| 947 |
"eval_dummy": 1.0,
|
| 948 |
-
"eval_loss":
|
| 949 |
-
"eval_runtime":
|
| 950 |
-
"eval_samples_per_second": 2.
|
| 951 |
-
"eval_steps_per_second": 0.
|
| 952 |
-
"step":
|
| 953 |
},
|
| 954 |
{
|
| 955 |
-
"epoch":
|
| 956 |
-
"grad_norm":
|
| 957 |
-
"learning_rate":
|
| 958 |
-
"loss":
|
| 959 |
"step": 6100
|
| 960 |
},
|
| 961 |
{
|
| 962 |
-
"epoch":
|
| 963 |
"eval_dummy": 1.0,
|
| 964 |
-
"eval_loss":
|
| 965 |
-
"eval_runtime":
|
| 966 |
-
"eval_samples_per_second": 2.
|
| 967 |
-
"eval_steps_per_second": 0.
|
| 968 |
-
"step":
|
| 969 |
},
|
| 970 |
{
|
| 971 |
-
"epoch":
|
| 972 |
-
"grad_norm":
|
| 973 |
-
"learning_rate":
|
| 974 |
-
"loss":
|
| 975 |
"step": 6200
|
| 976 |
},
|
| 977 |
{
|
| 978 |
-
"epoch":
|
| 979 |
-
"
|
| 980 |
-
"
|
| 981 |
-
"
|
| 982 |
-
"eval_samples_per_second": 2.878,
|
| 983 |
-
"eval_steps_per_second": 0.6,
|
| 984 |
-
"step": 6240
|
| 985 |
-
},
|
| 986 |
-
{
|
| 987 |
-
"epoch": 60.57692307692308,
|
| 988 |
-
"grad_norm": 34.31249237060547,
|
| 989 |
-
"learning_rate": 5.121162790697675e-05,
|
| 990 |
-
"loss": 12.7717,
|
| 991 |
"step": 6300
|
| 992 |
},
|
| 993 |
{
|
| 994 |
-
"epoch":
|
| 995 |
"eval_dummy": 1.0,
|
| 996 |
-
"eval_loss":
|
| 997 |
-
"eval_runtime":
|
| 998 |
-
"eval_samples_per_second": 2.
|
| 999 |
-
"eval_steps_per_second": 0.
|
| 1000 |
-
"step":
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
-
"epoch":
|
| 1004 |
-
"grad_norm":
|
| 1005 |
-
"learning_rate":
|
| 1006 |
-
"loss":
|
| 1007 |
"step": 6400
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
-
"epoch":
|
| 1011 |
"eval_dummy": 1.0,
|
| 1012 |
-
"eval_loss":
|
| 1013 |
-
"eval_runtime": 16.
|
| 1014 |
-
"eval_samples_per_second": 2.
|
| 1015 |
-
"eval_steps_per_second": 0.
|
| 1016 |
-
"step":
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch":
|
| 1020 |
-
"grad_norm":
|
| 1021 |
-
"learning_rate":
|
| 1022 |
-
"loss":
|
| 1023 |
"step": 6500
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
"eval_dummy": 1.0,
|
| 1028 |
-
"eval_loss":
|
| 1029 |
-
"eval_runtime":
|
| 1030 |
-
"eval_samples_per_second": 2.
|
| 1031 |
-
"eval_steps_per_second": 0.
|
| 1032 |
-
"step":
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
-
"epoch":
|
| 1036 |
-
"grad_norm":
|
| 1037 |
-
"learning_rate":
|
| 1038 |
-
"loss":
|
| 1039 |
"step": 6600
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
-
"epoch":
|
| 1043 |
-
"
|
| 1044 |
-
"
|
| 1045 |
-
"
|
| 1046 |
-
"eval_samples_per_second": 2.886,
|
| 1047 |
-
"eval_steps_per_second": 0.601,
|
| 1048 |
-
"step": 6656
|
| 1049 |
-
},
|
| 1050 |
-
{
|
| 1051 |
-
"epoch": 64.42307692307692,
|
| 1052 |
-
"grad_norm": 20.313173294067383,
|
| 1053 |
-
"learning_rate": 4.811395348837209e-05,
|
| 1054 |
-
"loss": 11.9496,
|
| 1055 |
"step": 6700
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
-
"epoch":
|
| 1059 |
"eval_dummy": 1.0,
|
| 1060 |
-
"eval_loss":
|
| 1061 |
-
"eval_runtime":
|
| 1062 |
-
"eval_samples_per_second": 2.
|
| 1063 |
-
"eval_steps_per_second": 0.
|
| 1064 |
-
"step":
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
-
"epoch":
|
| 1068 |
-
"grad_norm":
|
| 1069 |
-
"learning_rate":
|
| 1070 |
-
"loss":
|
| 1071 |
"step": 6800
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
-
"epoch":
|
| 1075 |
"eval_dummy": 1.0,
|
| 1076 |
-
"eval_loss": 31.
|
| 1077 |
-
"eval_runtime":
|
| 1078 |
-
"eval_samples_per_second": 2.
|
| 1079 |
-
"eval_steps_per_second": 0.
|
| 1080 |
-
"step":
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
-
"epoch":
|
| 1084 |
-
"grad_norm":
|
| 1085 |
-
"learning_rate":
|
| 1086 |
-
"loss": 12.
|
| 1087 |
"step": 6900
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch":
|
| 1091 |
"eval_dummy": 1.0,
|
| 1092 |
-
"eval_loss":
|
| 1093 |
-
"eval_runtime":
|
| 1094 |
-
"eval_samples_per_second": 2.
|
| 1095 |
-
"eval_steps_per_second": 0.
|
| 1096 |
-
"step":
|
| 1097 |
},
|
| 1098 |
{
|
| 1099 |
-
"epoch":
|
| 1100 |
-
"grad_norm":
|
| 1101 |
-
"learning_rate":
|
| 1102 |
-
"loss":
|
| 1103 |
"step": 7000
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
-
"epoch":
|
| 1107 |
"eval_dummy": 1.0,
|
| 1108 |
-
"eval_loss":
|
| 1109 |
-
"eval_runtime":
|
| 1110 |
-
"eval_samples_per_second": 2.
|
| 1111 |
-
"eval_steps_per_second": 0.
|
| 1112 |
-
"step":
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"epoch":
|
| 1116 |
-
"grad_norm":
|
| 1117 |
-
"learning_rate":
|
| 1118 |
-
"loss":
|
| 1119 |
"step": 7100
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"
|
| 1124 |
-
"
|
| 1125 |
-
"
|
| 1126 |
-
"eval_samples_per_second": 2.86,
|
| 1127 |
-
"eval_steps_per_second": 0.596,
|
| 1128 |
-
"step": 7176
|
| 1129 |
-
},
|
| 1130 |
-
{
|
| 1131 |
-
"epoch": 69.23076923076923,
|
| 1132 |
-
"grad_norm": 18.128597259521484,
|
| 1133 |
-
"learning_rate": 4.4241860465116286e-05,
|
| 1134 |
-
"loss": 12.2058,
|
| 1135 |
"step": 7200
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
-
"epoch":
|
| 1139 |
"eval_dummy": 1.0,
|
| 1140 |
-
"eval_loss":
|
| 1141 |
-
"eval_runtime": 17.
|
| 1142 |
-
"eval_samples_per_second": 2.
|
| 1143 |
-
"eval_steps_per_second": 0.
|
| 1144 |
-
"step":
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
-
"epoch":
|
| 1148 |
-
"grad_norm":
|
| 1149 |
-
"learning_rate":
|
| 1150 |
-
"loss": 12.
|
| 1151 |
"step": 7300
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
-
"epoch":
|
| 1155 |
"eval_dummy": 1.0,
|
| 1156 |
-
"eval_loss":
|
| 1157 |
-
"eval_runtime":
|
| 1158 |
-
"eval_samples_per_second": 2.
|
| 1159 |
-
"eval_steps_per_second": 0.
|
| 1160 |
-
"step":
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
-
"epoch":
|
| 1164 |
-
"grad_norm": 9.
|
| 1165 |
-
"learning_rate":
|
| 1166 |
-
"loss":
|
| 1167 |
"step": 7400
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
-
"epoch":
|
| 1171 |
"eval_dummy": 1.0,
|
| 1172 |
-
"eval_loss":
|
| 1173 |
-
"eval_runtime":
|
| 1174 |
-
"eval_samples_per_second": 2.
|
| 1175 |
-
"eval_steps_per_second": 0.
|
| 1176 |
-
"step":
|
| 1177 |
},
|
| 1178 |
{
|
| 1179 |
-
"epoch":
|
| 1180 |
-
"grad_norm":
|
| 1181 |
-
"learning_rate":
|
| 1182 |
-
"loss": 12.
|
| 1183 |
"step": 7500
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
-
"epoch":
|
| 1187 |
-
"
|
| 1188 |
-
"
|
| 1189 |
-
"
|
| 1190 |
-
"eval_samples_per_second": 2.823,
|
| 1191 |
-
"eval_steps_per_second": 0.588,
|
| 1192 |
-
"step": 7592
|
| 1193 |
-
},
|
| 1194 |
-
{
|
| 1195 |
-
"epoch": 73.07692307692308,
|
| 1196 |
-
"grad_norm": 16.92377281188965,
|
| 1197 |
-
"learning_rate": 4.114418604651163e-05,
|
| 1198 |
-
"loss": 11.9852,
|
| 1199 |
"step": 7600
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
-
"epoch":
|
| 1203 |
"eval_dummy": 1.0,
|
| 1204 |
-
"eval_loss":
|
| 1205 |
-
"eval_runtime":
|
| 1206 |
-
"eval_samples_per_second": 2.
|
| 1207 |
-
"eval_steps_per_second": 0.
|
| 1208 |
-
"step":
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
-
"epoch":
|
| 1212 |
-
"grad_norm":
|
| 1213 |
-
"learning_rate":
|
| 1214 |
-
"loss":
|
| 1215 |
"step": 7700
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
-
"epoch":
|
| 1219 |
-
"
|
| 1220 |
-
"
|
| 1221 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1222 |
"step": 7800
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
-
"epoch":
|
| 1226 |
"eval_dummy": 1.0,
|
| 1227 |
-
"eval_loss":
|
| 1228 |
-
"eval_runtime":
|
| 1229 |
-
"eval_samples_per_second": 2.
|
| 1230 |
-
"eval_steps_per_second": 0.
|
| 1231 |
-
"step":
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
-
"epoch":
|
| 1235 |
-
"grad_norm":
|
| 1236 |
-
"learning_rate":
|
| 1237 |
-
"loss":
|
| 1238 |
"step": 7900
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
-
"epoch":
|
| 1242 |
"eval_dummy": 1.0,
|
| 1243 |
-
"eval_loss":
|
| 1244 |
-
"eval_runtime":
|
| 1245 |
-
"eval_samples_per_second": 2.
|
| 1246 |
-
"eval_steps_per_second": 0.
|
| 1247 |
-
"step":
|
| 1248 |
},
|
| 1249 |
{
|
| 1250 |
-
"epoch":
|
| 1251 |
-
"grad_norm":
|
| 1252 |
-
"learning_rate":
|
| 1253 |
-
"loss":
|
| 1254 |
"step": 8000
|
| 1255 |
},
|
| 1256 |
{
|
| 1257 |
-
"epoch":
|
| 1258 |
-
"
|
| 1259 |
-
"
|
| 1260 |
-
"
|
| 1261 |
-
"eval_samples_per_second": 2.826,
|
| 1262 |
-
"eval_steps_per_second": 0.589,
|
| 1263 |
-
"step": 8008
|
| 1264 |
-
},
|
| 1265 |
-
{
|
| 1266 |
-
"epoch": 77.88461538461539,
|
| 1267 |
-
"grad_norm": 16.33148193359375,
|
| 1268 |
-
"learning_rate": 3.727209302325582e-05,
|
| 1269 |
-
"loss": 11.5474,
|
| 1270 |
"step": 8100
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"epoch":
|
| 1274 |
"eval_dummy": 1.0,
|
| 1275 |
-
"eval_loss":
|
| 1276 |
-
"eval_runtime":
|
| 1277 |
-
"eval_samples_per_second": 2.
|
| 1278 |
-
"eval_steps_per_second": 0.
|
| 1279 |
-
"step":
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
-
"epoch":
|
| 1283 |
-
"grad_norm":
|
| 1284 |
-
"learning_rate":
|
| 1285 |
-
"loss":
|
| 1286 |
"step": 8200
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
-
"epoch":
|
| 1290 |
"eval_dummy": 1.0,
|
| 1291 |
-
"eval_loss":
|
| 1292 |
-
"eval_runtime":
|
| 1293 |
-
"eval_samples_per_second": 2.
|
| 1294 |
-
"eval_steps_per_second": 0.
|
| 1295 |
-
"step":
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
-
"epoch":
|
| 1299 |
-
"grad_norm":
|
| 1300 |
-
"learning_rate":
|
| 1301 |
-
"loss":
|
| 1302 |
"step": 8300
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
-
"epoch":
|
| 1306 |
"eval_dummy": 1.0,
|
| 1307 |
-
"eval_loss":
|
| 1308 |
-
"eval_runtime":
|
| 1309 |
-
"eval_samples_per_second": 2.
|
| 1310 |
-
"eval_steps_per_second": 0.
|
| 1311 |
-
"step":
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
-
"epoch":
|
| 1315 |
-
"grad_norm":
|
| 1316 |
-
"learning_rate":
|
| 1317 |
-
"loss": 11.
|
| 1318 |
"step": 8400
|
| 1319 |
},
|
| 1320 |
{
|
| 1321 |
-
"epoch":
|
| 1322 |
-
"
|
| 1323 |
-
"
|
| 1324 |
-
"
|
| 1325 |
-
"eval_samples_per_second": 2.875,
|
| 1326 |
-
"eval_steps_per_second": 0.599,
|
| 1327 |
-
"step": 8424
|
| 1328 |
-
},
|
| 1329 |
-
{
|
| 1330 |
-
"epoch": 81.73076923076923,
|
| 1331 |
-
"grad_norm": 26.05837631225586,
|
| 1332 |
-
"learning_rate": 3.417441860465117e-05,
|
| 1333 |
-
"loss": 11.6578,
|
| 1334 |
"step": 8500
|
| 1335 |
},
|
| 1336 |
{
|
| 1337 |
-
"epoch":
|
| 1338 |
"eval_dummy": 1.0,
|
| 1339 |
-
"eval_loss":
|
| 1340 |
-
"eval_runtime":
|
| 1341 |
-
"eval_samples_per_second": 2.
|
| 1342 |
-
"eval_steps_per_second": 0.
|
| 1343 |
-
"step":
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch":
|
| 1347 |
-
"grad_norm":
|
| 1348 |
-
"learning_rate":
|
| 1349 |
-
"loss":
|
| 1350 |
"step": 8600
|
| 1351 |
},
|
| 1352 |
{
|
| 1353 |
-
"epoch":
|
| 1354 |
"eval_dummy": 1.0,
|
| 1355 |
-
"eval_loss":
|
| 1356 |
-
"eval_runtime":
|
| 1357 |
-
"eval_samples_per_second": 2.
|
| 1358 |
-
"eval_steps_per_second": 0.
|
| 1359 |
-
"step":
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
-
"epoch":
|
| 1363 |
-
"grad_norm":
|
| 1364 |
-
"learning_rate":
|
| 1365 |
-
"loss": 11.
|
| 1366 |
"step": 8700
|
| 1367 |
},
|
| 1368 |
{
|
| 1369 |
-
"epoch":
|
| 1370 |
"eval_dummy": 1.0,
|
| 1371 |
-
"eval_loss":
|
| 1372 |
-
"eval_runtime":
|
| 1373 |
-
"eval_samples_per_second": 2.
|
| 1374 |
-
"eval_steps_per_second": 0.
|
| 1375 |
-
"step":
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
-
"epoch":
|
| 1379 |
-
"grad_norm":
|
| 1380 |
-
"learning_rate":
|
| 1381 |
-
"loss": 11.
|
| 1382 |
"step": 8800
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
-
"epoch":
|
| 1386 |
-
"
|
| 1387 |
-
"
|
| 1388 |
-
"
|
| 1389 |
-
"eval_samples_per_second": 2.861,
|
| 1390 |
-
"eval_steps_per_second": 0.596,
|
| 1391 |
-
"step": 8840
|
| 1392 |
-
},
|
| 1393 |
-
{
|
| 1394 |
-
"epoch": 85.57692307692308,
|
| 1395 |
-
"grad_norm": 19.492916107177734,
|
| 1396 |
-
"learning_rate": 3.107674418604651e-05,
|
| 1397 |
-
"loss": 11.6955,
|
| 1398 |
"step": 8900
|
| 1399 |
},
|
| 1400 |
{
|
| 1401 |
-
"epoch":
|
| 1402 |
"eval_dummy": 1.0,
|
| 1403 |
-
"eval_loss":
|
| 1404 |
-
"eval_runtime":
|
| 1405 |
-
"eval_samples_per_second": 2.
|
| 1406 |
-
"eval_steps_per_second": 0.
|
| 1407 |
-
"step":
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
-
"epoch":
|
| 1411 |
-
"grad_norm":
|
| 1412 |
-
"learning_rate":
|
| 1413 |
-
"loss": 11.
|
| 1414 |
"step": 9000
|
| 1415 |
},
|
| 1416 |
{
|
| 1417 |
-
"epoch":
|
| 1418 |
"eval_dummy": 1.0,
|
| 1419 |
-
"eval_loss":
|
| 1420 |
-
"eval_runtime":
|
| 1421 |
-
"eval_samples_per_second": 2.
|
| 1422 |
-
"eval_steps_per_second": 0.
|
| 1423 |
-
"step":
|
| 1424 |
},
|
| 1425 |
{
|
| 1426 |
-
"epoch":
|
| 1427 |
-
"grad_norm":
|
| 1428 |
-
"learning_rate":
|
| 1429 |
-
"loss": 11.
|
| 1430 |
"step": 9100
|
| 1431 |
},
|
| 1432 |
{
|
| 1433 |
-
"epoch":
|
| 1434 |
"eval_dummy": 1.0,
|
| 1435 |
-
"eval_loss":
|
| 1436 |
-
"eval_runtime":
|
| 1437 |
-
"eval_samples_per_second": 2.
|
| 1438 |
-
"eval_steps_per_second": 0.
|
| 1439 |
-
"step":
|
| 1440 |
},
|
| 1441 |
{
|
| 1442 |
-
"epoch":
|
| 1443 |
-
"grad_norm":
|
| 1444 |
-
"learning_rate":
|
| 1445 |
-
"loss": 11.
|
| 1446 |
"step": 9200
|
| 1447 |
},
|
| 1448 |
{
|
| 1449 |
-
"epoch":
|
| 1450 |
"eval_dummy": 1.0,
|
| 1451 |
-
"eval_loss":
|
| 1452 |
-
"eval_runtime":
|
| 1453 |
-
"eval_samples_per_second": 2.
|
| 1454 |
-
"eval_steps_per_second": 0.
|
| 1455 |
-
"step":
|
| 1456 |
},
|
| 1457 |
{
|
| 1458 |
-
"epoch":
|
| 1459 |
-
"grad_norm":
|
| 1460 |
-
"learning_rate":
|
| 1461 |
-
"loss": 11.
|
| 1462 |
"step": 9300
|
| 1463 |
},
|
| 1464 |
{
|
| 1465 |
-
"epoch":
|
| 1466 |
-
"
|
| 1467 |
-
"
|
| 1468 |
-
"
|
| 1469 |
-
"eval_samples_per_second": 2.876,
|
| 1470 |
-
"eval_steps_per_second": 0.599,
|
| 1471 |
-
"step": 9360
|
| 1472 |
-
},
|
| 1473 |
-
{
|
| 1474 |
-
"epoch": 90.38461538461539,
|
| 1475 |
-
"grad_norm": 18.4831600189209,
|
| 1476 |
-
"learning_rate": 2.72046511627907e-05,
|
| 1477 |
-
"loss": 11.067,
|
| 1478 |
"step": 9400
|
| 1479 |
},
|
| 1480 |
{
|
| 1481 |
-
"epoch":
|
| 1482 |
"eval_dummy": 1.0,
|
| 1483 |
-
"eval_loss":
|
| 1484 |
-
"eval_runtime":
|
| 1485 |
-
"eval_samples_per_second": 2.
|
| 1486 |
-
"eval_steps_per_second": 0.
|
| 1487 |
-
"step":
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
-
"epoch":
|
| 1491 |
-
"grad_norm":
|
| 1492 |
-
"learning_rate":
|
| 1493 |
-
"loss": 11.
|
| 1494 |
"step": 9500
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
-
"epoch":
|
| 1498 |
"eval_dummy": 1.0,
|
| 1499 |
-
"eval_loss":
|
| 1500 |
-
"eval_runtime":
|
| 1501 |
-
"eval_samples_per_second": 2.
|
| 1502 |
-
"eval_steps_per_second": 0.
|
| 1503 |
-
"step":
|
| 1504 |
},
|
| 1505 |
{
|
| 1506 |
-
"epoch":
|
| 1507 |
-
"grad_norm":
|
| 1508 |
-
"learning_rate":
|
| 1509 |
-
"loss": 11.
|
| 1510 |
"step": 9600
|
| 1511 |
},
|
| 1512 |
{
|
| 1513 |
-
"epoch":
|
| 1514 |
"eval_dummy": 1.0,
|
| 1515 |
-
"eval_loss":
|
| 1516 |
-
"eval_runtime":
|
| 1517 |
-
"eval_samples_per_second": 2.
|
| 1518 |
-
"eval_steps_per_second": 0.
|
| 1519 |
-
"step":
|
| 1520 |
},
|
| 1521 |
{
|
| 1522 |
-
"epoch":
|
| 1523 |
-
"grad_norm":
|
| 1524 |
-
"learning_rate":
|
| 1525 |
-
"loss": 11.
|
| 1526 |
"step": 9700
|
| 1527 |
},
|
| 1528 |
{
|
| 1529 |
-
"epoch":
|
| 1530 |
-
"
|
| 1531 |
-
"
|
| 1532 |
-
"
|
| 1533 |
-
"eval_samples_per_second": 2.888,
|
| 1534 |
-
"eval_steps_per_second": 0.602,
|
| 1535 |
-
"step": 9776
|
| 1536 |
-
},
|
| 1537 |
-
{
|
| 1538 |
-
"epoch": 94.23076923076923,
|
| 1539 |
-
"grad_norm": 47.0684814453125,
|
| 1540 |
-
"learning_rate": 2.4106976744186043e-05,
|
| 1541 |
-
"loss": 11.2935,
|
| 1542 |
"step": 9800
|
| 1543 |
},
|
| 1544 |
{
|
| 1545 |
-
"epoch":
|
| 1546 |
"eval_dummy": 1.0,
|
| 1547 |
-
"eval_loss":
|
| 1548 |
-
"eval_runtime":
|
| 1549 |
-
"eval_samples_per_second": 2.
|
| 1550 |
-
"eval_steps_per_second": 0.
|
| 1551 |
-
"step":
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
-
"epoch":
|
| 1555 |
-
"grad_norm":
|
| 1556 |
-
"learning_rate":
|
| 1557 |
-
"loss": 11.
|
| 1558 |
"step": 9900
|
| 1559 |
},
|
| 1560 |
{
|
| 1561 |
-
"epoch":
|
| 1562 |
"eval_dummy": 1.0,
|
| 1563 |
-
"eval_loss":
|
| 1564 |
-
"eval_runtime":
|
| 1565 |
-
"eval_samples_per_second": 2.
|
| 1566 |
-
"eval_steps_per_second": 0.
|
| 1567 |
-
"step":
|
| 1568 |
},
|
| 1569 |
{
|
| 1570 |
-
"epoch":
|
| 1571 |
-
"grad_norm":
|
| 1572 |
-
"learning_rate":
|
| 1573 |
-
"loss": 11.
|
| 1574 |
"step": 10000
|
| 1575 |
},
|
| 1576 |
{
|
| 1577 |
-
"epoch":
|
| 1578 |
"eval_dummy": 1.0,
|
| 1579 |
-
"eval_loss":
|
| 1580 |
-
"eval_runtime":
|
| 1581 |
-
"eval_samples_per_second": 2.
|
| 1582 |
-
"eval_steps_per_second": 0.
|
| 1583 |
-
"step":
|
| 1584 |
},
|
| 1585 |
{
|
| 1586 |
-
"epoch":
|
| 1587 |
-
"grad_norm":
|
| 1588 |
-
"learning_rate":
|
| 1589 |
-
"loss": 11.
|
| 1590 |
"step": 10100
|
| 1591 |
},
|
| 1592 |
{
|
| 1593 |
-
"epoch":
|
| 1594 |
"eval_dummy": 1.0,
|
| 1595 |
-
"eval_loss":
|
| 1596 |
-
"eval_runtime": 16.
|
| 1597 |
-
"eval_samples_per_second": 2.
|
| 1598 |
-
"eval_steps_per_second": 0.
|
| 1599 |
-
"step":
|
| 1600 |
},
|
| 1601 |
{
|
| 1602 |
-
"epoch":
|
| 1603 |
-
"grad_norm":
|
| 1604 |
-
"learning_rate":
|
| 1605 |
-
"loss": 11.
|
| 1606 |
"step": 10200
|
| 1607 |
},
|
| 1608 |
{
|
| 1609 |
-
"epoch":
|
| 1610 |
-
"
|
| 1611 |
-
"
|
| 1612 |
-
"
|
| 1613 |
-
"
|
| 1614 |
-
"eval_steps_per_second": 0.602,
|
| 1615 |
-
"step": 10296
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
-
"epoch":
|
| 1619 |
-
"
|
| 1620 |
-
"
|
| 1621 |
-
"
|
| 1622 |
-
"
|
|
|
|
|
|
|
| 1623 |
},
|
| 1624 |
{
|
| 1625 |
-
"epoch":
|
| 1626 |
-
"grad_norm":
|
| 1627 |
-
"learning_rate":
|
| 1628 |
-
"loss": 11.
|
| 1629 |
"step": 10400
|
| 1630 |
},
|
| 1631 |
{
|
| 1632 |
-
"epoch":
|
| 1633 |
"eval_dummy": 1.0,
|
| 1634 |
-
"eval_loss":
|
| 1635 |
-
"eval_runtime":
|
| 1636 |
-
"eval_samples_per_second": 2.
|
| 1637 |
-
"eval_steps_per_second": 0.
|
| 1638 |
-
"step":
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
-
"epoch":
|
| 1642 |
-
"grad_norm":
|
| 1643 |
-
"learning_rate":
|
| 1644 |
-
"loss": 11.
|
| 1645 |
"step": 10500
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
-
"epoch":
|
| 1649 |
"eval_dummy": 1.0,
|
| 1650 |
-
"eval_loss":
|
| 1651 |
-
"eval_runtime":
|
| 1652 |
-
"eval_samples_per_second": 2.
|
| 1653 |
-
"eval_steps_per_second": 0.
|
| 1654 |
-
"step":
|
| 1655 |
},
|
| 1656 |
{
|
| 1657 |
-
"epoch":
|
| 1658 |
-
"grad_norm":
|
| 1659 |
-
"learning_rate":
|
| 1660 |
-
"loss": 10.
|
| 1661 |
"step": 10600
|
| 1662 |
},
|
| 1663 |
{
|
| 1664 |
-
"epoch":
|
| 1665 |
-
"
|
| 1666 |
-
"
|
| 1667 |
-
"
|
| 1668 |
-
"eval_samples_per_second": 2.919,
|
| 1669 |
-
"eval_steps_per_second": 0.608,
|
| 1670 |
-
"step": 10608
|
| 1671 |
-
},
|
| 1672 |
-
{
|
| 1673 |
-
"epoch": 102.88461538461539,
|
| 1674 |
-
"grad_norm": 7.563882350921631,
|
| 1675 |
-
"learning_rate": 1.713720930232558e-05,
|
| 1676 |
-
"loss": 10.9188,
|
| 1677 |
"step": 10700
|
| 1678 |
},
|
| 1679 |
{
|
| 1680 |
-
"epoch":
|
| 1681 |
"eval_dummy": 1.0,
|
| 1682 |
-
"eval_loss":
|
| 1683 |
-
"eval_runtime":
|
| 1684 |
-
"eval_samples_per_second": 2.
|
| 1685 |
-
"eval_steps_per_second": 0.
|
| 1686 |
-
"step":
|
| 1687 |
},
|
| 1688 |
{
|
| 1689 |
-
"epoch":
|
| 1690 |
-
"grad_norm":
|
| 1691 |
-
"learning_rate":
|
| 1692 |
-
"loss":
|
| 1693 |
"step": 10800
|
| 1694 |
},
|
| 1695 |
{
|
| 1696 |
-
"epoch":
|
| 1697 |
"eval_dummy": 1.0,
|
| 1698 |
-
"eval_loss":
|
| 1699 |
-
"eval_runtime":
|
| 1700 |
-
"eval_samples_per_second": 2.
|
| 1701 |
-
"eval_steps_per_second": 0.
|
| 1702 |
-
"step":
|
| 1703 |
},
|
| 1704 |
{
|
| 1705 |
-
"epoch":
|
| 1706 |
-
"grad_norm":
|
| 1707 |
-
"learning_rate":
|
| 1708 |
-
"loss": 11.
|
| 1709 |
"step": 10900
|
| 1710 |
},
|
| 1711 |
{
|
| 1712 |
-
"epoch":
|
| 1713 |
"eval_dummy": 1.0,
|
| 1714 |
-
"eval_loss":
|
| 1715 |
-
"eval_runtime": 16.
|
| 1716 |
-
"eval_samples_per_second": 2.
|
| 1717 |
-
"eval_steps_per_second": 0.
|
| 1718 |
-
"step":
|
| 1719 |
},
|
| 1720 |
{
|
| 1721 |
-
"epoch":
|
| 1722 |
-
"grad_norm":
|
| 1723 |
-
"learning_rate":
|
| 1724 |
-
"loss": 10.
|
| 1725 |
"step": 11000
|
| 1726 |
},
|
| 1727 |
{
|
| 1728 |
-
"epoch":
|
| 1729 |
"eval_dummy": 1.0,
|
| 1730 |
-
"eval_loss":
|
| 1731 |
-
"eval_runtime":
|
| 1732 |
-
"eval_samples_per_second": 2.
|
| 1733 |
-
"eval_steps_per_second": 0.
|
| 1734 |
-
"step":
|
| 1735 |
},
|
| 1736 |
{
|
| 1737 |
-
"epoch":
|
| 1738 |
-
"grad_norm":
|
| 1739 |
-
"learning_rate":
|
| 1740 |
-
"loss": 11.
|
| 1741 |
"step": 11100
|
| 1742 |
},
|
| 1743 |
{
|
| 1744 |
-
"epoch":
|
| 1745 |
-
"
|
| 1746 |
-
"
|
| 1747 |
-
"
|
| 1748 |
-
"eval_samples_per_second": 2.936,
|
| 1749 |
-
"eval_steps_per_second": 0.612,
|
| 1750 |
-
"step": 11128
|
| 1751 |
-
},
|
| 1752 |
-
{
|
| 1753 |
-
"epoch": 107.6923076923077,
|
| 1754 |
-
"grad_norm": 26.309329986572266,
|
| 1755 |
-
"learning_rate": 1.326511627906977e-05,
|
| 1756 |
-
"loss": 11.0193,
|
| 1757 |
"step": 11200
|
| 1758 |
},
|
| 1759 |
{
|
| 1760 |
-
"epoch":
|
| 1761 |
"eval_dummy": 1.0,
|
| 1762 |
-
"eval_loss":
|
| 1763 |
-
"eval_runtime":
|
| 1764 |
-
"eval_samples_per_second": 2.
|
| 1765 |
-
"eval_steps_per_second": 0.
|
| 1766 |
-
"step":
|
| 1767 |
},
|
| 1768 |
{
|
| 1769 |
-
"epoch":
|
| 1770 |
-
"grad_norm":
|
| 1771 |
-
"learning_rate":
|
| 1772 |
-
"loss":
|
| 1773 |
"step": 11300
|
| 1774 |
},
|
| 1775 |
{
|
| 1776 |
-
"epoch":
|
| 1777 |
"eval_dummy": 1.0,
|
| 1778 |
-
"eval_loss":
|
| 1779 |
-
"eval_runtime":
|
| 1780 |
-
"eval_samples_per_second": 2.
|
| 1781 |
-
"eval_steps_per_second": 0.
|
| 1782 |
-
"step":
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
-
"epoch":
|
| 1786 |
-
"grad_norm":
|
| 1787 |
-
"learning_rate":
|
| 1788 |
-
"loss":
|
| 1789 |
"step": 11400
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
-
"epoch":
|
| 1793 |
"eval_dummy": 1.0,
|
| 1794 |
-
"eval_loss":
|
| 1795 |
-
"eval_runtime": 16.
|
| 1796 |
-
"eval_samples_per_second": 2.
|
| 1797 |
-
"eval_steps_per_second": 0.
|
| 1798 |
-
"step":
|
| 1799 |
},
|
| 1800 |
{
|
| 1801 |
-
"epoch":
|
| 1802 |
-
"grad_norm":
|
| 1803 |
-
"learning_rate":
|
| 1804 |
-
"loss": 10.
|
| 1805 |
"step": 11500
|
| 1806 |
},
|
| 1807 |
{
|
| 1808 |
-
"epoch":
|
| 1809 |
-
"
|
| 1810 |
-
"
|
| 1811 |
-
"
|
| 1812 |
-
"eval_samples_per_second": 2.839,
|
| 1813 |
-
"eval_steps_per_second": 0.591,
|
| 1814 |
-
"step": 11544
|
| 1815 |
-
},
|
| 1816 |
-
{
|
| 1817 |
-
"epoch": 111.53846153846153,
|
| 1818 |
-
"grad_norm": 8.586106300354004,
|
| 1819 |
-
"learning_rate": 1.0167441860465113e-05,
|
| 1820 |
-
"loss": 10.9706,
|
| 1821 |
"step": 11600
|
| 1822 |
},
|
| 1823 |
{
|
| 1824 |
-
"epoch":
|
| 1825 |
"eval_dummy": 1.0,
|
| 1826 |
-
"eval_loss":
|
| 1827 |
-
"eval_runtime":
|
| 1828 |
-
"eval_samples_per_second": 2.
|
| 1829 |
-
"eval_steps_per_second": 0.
|
| 1830 |
-
"step":
|
| 1831 |
},
|
| 1832 |
{
|
| 1833 |
-
"epoch":
|
| 1834 |
-
"grad_norm":
|
| 1835 |
-
"learning_rate":
|
| 1836 |
-
"loss":
|
| 1837 |
"step": 11700
|
| 1838 |
},
|
| 1839 |
{
|
| 1840 |
-
"epoch":
|
| 1841 |
"eval_dummy": 1.0,
|
| 1842 |
-
"eval_loss":
|
| 1843 |
-
"eval_runtime": 16.
|
| 1844 |
-
"eval_samples_per_second": 2.
|
| 1845 |
-
"eval_steps_per_second": 0.
|
| 1846 |
-
"step":
|
| 1847 |
},
|
| 1848 |
{
|
| 1849 |
-
"epoch":
|
| 1850 |
-
"grad_norm":
|
| 1851 |
-
"learning_rate":
|
| 1852 |
-
"loss": 10.
|
| 1853 |
"step": 11800
|
| 1854 |
},
|
| 1855 |
{
|
| 1856 |
-
"epoch":
|
| 1857 |
"eval_dummy": 1.0,
|
| 1858 |
-
"eval_loss":
|
| 1859 |
-
"eval_runtime": 17.
|
| 1860 |
-
"eval_samples_per_second": 2.
|
| 1861 |
-
"eval_steps_per_second": 0.
|
| 1862 |
-
"step":
|
| 1863 |
},
|
| 1864 |
{
|
| 1865 |
-
"epoch":
|
| 1866 |
-
"grad_norm":
|
| 1867 |
-
"learning_rate":
|
| 1868 |
-
"loss": 10.
|
| 1869 |
"step": 11900
|
| 1870 |
},
|
| 1871 |
{
|
| 1872 |
-
"epoch":
|
| 1873 |
"eval_dummy": 1.0,
|
| 1874 |
-
"eval_loss":
|
| 1875 |
-
"eval_runtime":
|
| 1876 |
-
"eval_samples_per_second": 2.
|
| 1877 |
-
"eval_steps_per_second": 0.
|
| 1878 |
-
"step":
|
| 1879 |
},
|
| 1880 |
{
|
| 1881 |
-
"epoch":
|
| 1882 |
-
"grad_norm":
|
| 1883 |
-
"learning_rate":
|
| 1884 |
-
"loss": 10.
|
| 1885 |
"step": 12000
|
| 1886 |
},
|
| 1887 |
{
|
| 1888 |
-
"epoch":
|
| 1889 |
-
"
|
| 1890 |
-
"
|
| 1891 |
-
"
|
| 1892 |
-
"eval_samples_per_second": 2.857,
|
| 1893 |
-
"eval_steps_per_second": 0.595,
|
| 1894 |
-
"step": 12064
|
| 1895 |
-
},
|
| 1896 |
-
{
|
| 1897 |
-
"epoch": 116.34615384615384,
|
| 1898 |
-
"grad_norm": 10.545084953308105,
|
| 1899 |
-
"learning_rate": 6.295348837209302e-06,
|
| 1900 |
-
"loss": 10.4932,
|
| 1901 |
"step": 12100
|
| 1902 |
},
|
| 1903 |
{
|
| 1904 |
-
"epoch":
|
| 1905 |
"eval_dummy": 1.0,
|
| 1906 |
-
"eval_loss":
|
| 1907 |
-
"eval_runtime": 17.
|
| 1908 |
-
"eval_samples_per_second": 2.
|
| 1909 |
-
"eval_steps_per_second": 0.
|
| 1910 |
-
"step":
|
| 1911 |
},
|
| 1912 |
{
|
| 1913 |
-
"epoch":
|
| 1914 |
-
"grad_norm":
|
| 1915 |
-
"learning_rate":
|
| 1916 |
-
"loss": 10.
|
| 1917 |
"step": 12200
|
| 1918 |
},
|
| 1919 |
{
|
| 1920 |
-
"epoch":
|
| 1921 |
"eval_dummy": 1.0,
|
| 1922 |
-
"eval_loss":
|
| 1923 |
-
"eval_runtime": 16.
|
| 1924 |
-
"eval_samples_per_second": 2.
|
| 1925 |
-
"eval_steps_per_second": 0.
|
| 1926 |
-
"step":
|
| 1927 |
},
|
| 1928 |
{
|
| 1929 |
-
"epoch":
|
| 1930 |
-
"grad_norm":
|
| 1931 |
-
"learning_rate":
|
| 1932 |
-
"loss": 10.
|
| 1933 |
"step": 12300
|
| 1934 |
},
|
| 1935 |
{
|
| 1936 |
-
"epoch":
|
| 1937 |
"eval_dummy": 1.0,
|
| 1938 |
-
"eval_loss":
|
| 1939 |
-
"eval_runtime": 17.
|
| 1940 |
-
"eval_samples_per_second": 2.
|
| 1941 |
-
"eval_steps_per_second": 0.
|
| 1942 |
-
"step":
|
| 1943 |
},
|
| 1944 |
{
|
| 1945 |
-
"epoch":
|
| 1946 |
-
"grad_norm":
|
| 1947 |
-
"learning_rate":
|
| 1948 |
-
"loss": 10.
|
| 1949 |
"step": 12400
|
| 1950 |
},
|
| 1951 |
{
|
| 1952 |
-
"epoch":
|
| 1953 |
-
"
|
| 1954 |
-
"
|
| 1955 |
-
"
|
| 1956 |
-
"eval_samples_per_second": 2.775,
|
| 1957 |
-
"eval_steps_per_second": 0.578,
|
| 1958 |
-
"step": 12480
|
| 1959 |
-
},
|
| 1960 |
-
{
|
| 1961 |
-
"epoch": 120.1923076923077,
|
| 1962 |
-
"grad_norm": 7.985867023468018,
|
| 1963 |
-
"learning_rate": 3.1976744186046562e-06,
|
| 1964 |
-
"loss": 10.6876,
|
| 1965 |
"step": 12500
|
| 1966 |
},
|
| 1967 |
{
|
| 1968 |
-
"epoch":
|
| 1969 |
"eval_dummy": 1.0,
|
| 1970 |
-
"eval_loss":
|
| 1971 |
-
"eval_runtime":
|
| 1972 |
-
"eval_samples_per_second": 2.
|
| 1973 |
-
"eval_steps_per_second": 0.
|
| 1974 |
-
"step":
|
| 1975 |
},
|
| 1976 |
{
|
| 1977 |
-
"epoch":
|
| 1978 |
-
"grad_norm":
|
| 1979 |
-
"learning_rate":
|
| 1980 |
-
"loss": 10.
|
| 1981 |
"step": 12600
|
| 1982 |
},
|
| 1983 |
{
|
| 1984 |
-
"epoch":
|
| 1985 |
"eval_dummy": 1.0,
|
| 1986 |
-
"eval_loss":
|
| 1987 |
-
"eval_runtime":
|
| 1988 |
-
"eval_samples_per_second": 2.
|
| 1989 |
-
"eval_steps_per_second": 0.
|
| 1990 |
-
"step":
|
| 1991 |
},
|
| 1992 |
{
|
| 1993 |
-
"epoch":
|
| 1994 |
-
"grad_norm":
|
| 1995 |
-
"learning_rate":
|
| 1996 |
-
"loss":
|
| 1997 |
"step": 12700
|
| 1998 |
},
|
| 1999 |
{
|
| 2000 |
-
"epoch":
|
| 2001 |
"eval_dummy": 1.0,
|
| 2002 |
-
"eval_loss":
|
| 2003 |
-
"eval_runtime": 17.
|
| 2004 |
-
"eval_samples_per_second": 2.
|
| 2005 |
-
"eval_steps_per_second": 0.
|
| 2006 |
-
"step":
|
| 2007 |
},
|
| 2008 |
{
|
| 2009 |
-
"epoch":
|
| 2010 |
-
"grad_norm":
|
| 2011 |
-
"learning_rate":
|
| 2012 |
-
"loss": 10.
|
| 2013 |
"step": 12800
|
| 2014 |
},
|
| 2015 |
{
|
| 2016 |
-
"epoch":
|
| 2017 |
-
"
|
| 2018 |
-
"eval_loss": 33.539737701416016,
|
| 2019 |
-
"eval_runtime": 16.3513,
|
| 2020 |
-
"eval_samples_per_second": 2.936,
|
| 2021 |
-
"eval_steps_per_second": 0.612,
|
| 2022 |
-
"step": 12896
|
| 2023 |
-
},
|
| 2024 |
-
{
|
| 2025 |
-
"epoch": 124.03846153846153,
|
| 2026 |
-
"grad_norm": 7.083764553070068,
|
| 2027 |
"learning_rate": 9.999999999999998e-08,
|
| 2028 |
-
"loss": 10.
|
| 2029 |
"step": 12900
|
| 2030 |
},
|
| 2031 |
{
|
| 2032 |
-
"epoch":
|
| 2033 |
"eval_dummy": 1.0,
|
| 2034 |
-
"eval_loss":
|
| 2035 |
-
"eval_runtime": 17.
|
| 2036 |
-
"eval_samples_per_second": 2.
|
| 2037 |
-
"eval_steps_per_second": 0.
|
| 2038 |
"step": 12900
|
| 2039 |
},
|
| 2040 |
{
|
| 2041 |
-
"epoch":
|
| 2042 |
"step": 12900,
|
| 2043 |
-
"total_flos": 1.
|
| 2044 |
-
"train_loss":
|
| 2045 |
-
"train_runtime":
|
| 2046 |
-
"train_samples_per_second":
|
| 2047 |
-
"train_steps_per_second": 0.
|
| 2048 |
}
|
| 2049 |
],
|
| 2050 |
"logging_steps": 100,
|
| 2051 |
"max_steps": 12900,
|
| 2052 |
"num_input_tokens_seen": 0,
|
| 2053 |
-
"num_train_epochs":
|
| 2054 |
"save_steps": 500,
|
| 2055 |
"stateful_callbacks": {
|
| 2056 |
"TrainerControl": {
|
|
@@ -2064,8 +1839,8 @@
|
|
| 2064 |
"attributes": {}
|
| 2065 |
}
|
| 2066 |
},
|
| 2067 |
-
"total_flos": 1.
|
| 2068 |
-
"train_batch_size":
|
| 2069 |
"trial_name": null,
|
| 2070 |
"trial_params": null
|
| 2071 |
}
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 100.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 12900,
|
| 8 |
"is_hyper_param_search": false,
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"epoch": 0.7751937984496124,
|
| 14 |
+
"grad_norm": 126.4291763305664,
|
| 15 |
+
"learning_rate": 0.0003969,
|
| 16 |
+
"loss": 57.794,
|
| 17 |
"step": 100
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
"eval_dummy": 1.0,
|
| 22 |
+
"eval_loss": 47.50299072265625,
|
| 23 |
+
"eval_runtime": 18.4297,
|
| 24 |
+
"eval_samples_per_second": 2.604,
|
| 25 |
+
"eval_steps_per_second": 0.651,
|
| 26 |
+
"step": 129
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 1.550387596899225,
|
| 30 |
+
"grad_norm": 110.41481018066406,
|
| 31 |
+
"learning_rate": 0.00039380000000000003,
|
| 32 |
+
"loss": 45.635,
|
| 33 |
"step": 200
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.0,
|
| 37 |
"eval_dummy": 1.0,
|
| 38 |
+
"eval_loss": 42.4621696472168,
|
| 39 |
+
"eval_runtime": 17.0743,
|
| 40 |
+
"eval_samples_per_second": 2.811,
|
| 41 |
+
"eval_steps_per_second": 0.703,
|
| 42 |
+
"step": 258
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 2.3255813953488373,
|
| 46 |
+
"grad_norm": 119.6634750366211,
|
| 47 |
+
"learning_rate": 0.0003907,
|
| 48 |
+
"loss": 43.6742,
|
| 49 |
"step": 300
|
| 50 |
},
|
| 51 |
{
|
| 52 |
"epoch": 3.0,
|
| 53 |
"eval_dummy": 1.0,
|
| 54 |
+
"eval_loss": 40.13833999633789,
|
| 55 |
+
"eval_runtime": 18.4155,
|
| 56 |
+
"eval_samples_per_second": 2.606,
|
| 57 |
+
"eval_steps_per_second": 0.652,
|
| 58 |
+
"step": 387
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 3.10077519379845,
|
| 62 |
+
"grad_norm": 118.22651672363281,
|
| 63 |
+
"learning_rate": 0.0003876,
|
| 64 |
+
"loss": 40.4744,
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 3.875968992248062,
|
| 69 |
+
"grad_norm": 134.10031127929688,
|
| 70 |
+
"learning_rate": 0.0003845,
|
| 71 |
+
"loss": 37.5286,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
"step": 500
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 4.0,
|
| 76 |
"eval_dummy": 1.0,
|
| 77 |
+
"eval_loss": 41.196353912353516,
|
| 78 |
+
"eval_runtime": 17.0535,
|
| 79 |
+
"eval_samples_per_second": 2.815,
|
| 80 |
+
"eval_steps_per_second": 0.704,
|
| 81 |
+
"step": 516
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 4.651162790697675,
|
| 85 |
+
"grad_norm": 80.17464447021484,
|
| 86 |
+
"learning_rate": 0.00038140000000000005,
|
| 87 |
+
"loss": 33.7618,
|
| 88 |
"step": 600
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 5.0,
|
| 92 |
"eval_dummy": 1.0,
|
| 93 |
+
"eval_loss": 34.43735122680664,
|
| 94 |
+
"eval_runtime": 19.3245,
|
| 95 |
+
"eval_samples_per_second": 2.484,
|
| 96 |
+
"eval_steps_per_second": 0.621,
|
| 97 |
+
"step": 645
|
| 98 |
},
|
| 99 |
{
|
| 100 |
+
"epoch": 5.426356589147287,
|
| 101 |
+
"grad_norm": 372.6417236328125,
|
| 102 |
+
"learning_rate": 0.00037830000000000003,
|
| 103 |
+
"loss": 31.5899,
|
| 104 |
"step": 700
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
"eval_dummy": 1.0,
|
| 109 |
+
"eval_loss": 39.824180603027344,
|
| 110 |
+
"eval_runtime": 17.2226,
|
| 111 |
+
"eval_samples_per_second": 2.787,
|
| 112 |
+
"eval_steps_per_second": 0.697,
|
| 113 |
+
"step": 774
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 6.2015503875969,
|
| 117 |
+
"grad_norm": 226.50611877441406,
|
| 118 |
+
"learning_rate": 0.0003752,
|
| 119 |
+
"loss": 31.2097,
|
| 120 |
"step": 800
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 6.976744186046512,
|
| 124 |
+
"grad_norm": 99.72407531738281,
|
| 125 |
+
"learning_rate": 0.0003721,
|
| 126 |
+
"loss": 29.0727,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
"step": 900
|
| 128 |
},
|
| 129 |
{
|
| 130 |
+
"epoch": 7.0,
|
| 131 |
"eval_dummy": 1.0,
|
| 132 |
+
"eval_loss": 33.32234191894531,
|
| 133 |
+
"eval_runtime": 18.5858,
|
| 134 |
+
"eval_samples_per_second": 2.583,
|
| 135 |
+
"eval_steps_per_second": 0.646,
|
| 136 |
+
"step": 903
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 7.751937984496124,
|
| 140 |
+
"grad_norm": 107.07185363769531,
|
| 141 |
+
"learning_rate": 0.000369,
|
| 142 |
+
"loss": 27.8483,
|
| 143 |
"step": 1000
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 8.0,
|
| 147 |
"eval_dummy": 1.0,
|
| 148 |
+
"eval_loss": 30.962478637695312,
|
| 149 |
+
"eval_runtime": 17.3408,
|
| 150 |
+
"eval_samples_per_second": 2.768,
|
| 151 |
+
"eval_steps_per_second": 0.692,
|
| 152 |
+
"step": 1032
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 8.527131782945737,
|
| 156 |
+
"grad_norm": 82.22681427001953,
|
| 157 |
+
"learning_rate": 0.0003659,
|
| 158 |
+
"loss": 26.0904,
|
| 159 |
"step": 1100
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 9.0,
|
| 163 |
"eval_dummy": 1.0,
|
| 164 |
+
"eval_loss": 31.708364486694336,
|
| 165 |
+
"eval_runtime": 18.5166,
|
| 166 |
+
"eval_samples_per_second": 2.592,
|
| 167 |
+
"eval_steps_per_second": 0.648,
|
| 168 |
+
"step": 1161
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 9.30232558139535,
|
| 172 |
+
"grad_norm": 184.18087768554688,
|
| 173 |
+
"learning_rate": 0.00036280000000000004,
|
| 174 |
+
"loss": 26.1043,
|
| 175 |
"step": 1200
|
| 176 |
},
|
| 177 |
{
|
| 178 |
+
"epoch": 10.0,
|
| 179 |
"eval_dummy": 1.0,
|
| 180 |
+
"eval_loss": 31.808786392211914,
|
| 181 |
+
"eval_runtime": 17.2835,
|
| 182 |
+
"eval_samples_per_second": 2.777,
|
| 183 |
+
"eval_steps_per_second": 0.694,
|
| 184 |
+
"step": 1290
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 10.077519379844961,
|
| 188 |
+
"grad_norm": 82.96527862548828,
|
| 189 |
+
"learning_rate": 0.0003597,
|
| 190 |
+
"loss": 26.1461,
|
| 191 |
"step": 1300
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 10.852713178294573,
|
| 195 |
+
"grad_norm": 76.06727600097656,
|
| 196 |
+
"learning_rate": 0.00035660000000000005,
|
| 197 |
+
"loss": 24.3038,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 198 |
"step": 1400
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 11.0,
|
| 202 |
"eval_dummy": 1.0,
|
| 203 |
+
"eval_loss": 30.336084365844727,
|
| 204 |
+
"eval_runtime": 18.0862,
|
| 205 |
+
"eval_samples_per_second": 2.654,
|
| 206 |
+
"eval_steps_per_second": 0.663,
|
| 207 |
+
"step": 1419
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 11.627906976744185,
|
| 211 |
+
"grad_norm": 57.900630950927734,
|
| 212 |
+
"learning_rate": 0.00035350000000000003,
|
| 213 |
+
"loss": 23.6493,
|
| 214 |
"step": 1500
|
| 215 |
},
|
| 216 |
{
|
| 217 |
+
"epoch": 12.0,
|
| 218 |
"eval_dummy": 1.0,
|
| 219 |
+
"eval_loss": 30.202993392944336,
|
| 220 |
+
"eval_runtime": 16.9497,
|
| 221 |
+
"eval_samples_per_second": 2.832,
|
| 222 |
+
"eval_steps_per_second": 0.708,
|
| 223 |
+
"step": 1548
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 12.4031007751938,
|
| 227 |
+
"grad_norm": 60.565650939941406,
|
| 228 |
+
"learning_rate": 0.0003504,
|
| 229 |
+
"loss": 23.9146,
|
| 230 |
"step": 1600
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 13.0,
|
| 234 |
"eval_dummy": 1.0,
|
| 235 |
+
"eval_loss": 31.08062744140625,
|
| 236 |
+
"eval_runtime": 17.9909,
|
| 237 |
+
"eval_samples_per_second": 2.668,
|
| 238 |
+
"eval_steps_per_second": 0.667,
|
| 239 |
+
"step": 1677
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 13.178294573643411,
|
| 243 |
+
"grad_norm": 75.52674865722656,
|
| 244 |
+
"learning_rate": 0.00034730000000000004,
|
| 245 |
+
"loss": 23.033,
|
| 246 |
"step": 1700
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 13.953488372093023,
|
| 250 |
+
"grad_norm": 45.41800308227539,
|
| 251 |
+
"learning_rate": 0.0003442,
|
| 252 |
+
"loss": 21.9133,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 253 |
"step": 1800
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 14.0,
|
| 257 |
"eval_dummy": 1.0,
|
| 258 |
+
"eval_loss": 31.397354125976562,
|
| 259 |
+
"eval_runtime": 17.0643,
|
| 260 |
+
"eval_samples_per_second": 2.813,
|
| 261 |
+
"eval_steps_per_second": 0.703,
|
| 262 |
+
"step": 1806
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 14.728682170542635,
|
| 266 |
+
"grad_norm": 46.21736145019531,
|
| 267 |
+
"learning_rate": 0.0003411,
|
| 268 |
+
"loss": 22.3071,
|
| 269 |
"step": 1900
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 15.0,
|
| 273 |
"eval_dummy": 1.0,
|
| 274 |
+
"eval_loss": 32.09249496459961,
|
| 275 |
+
"eval_runtime": 17.5136,
|
| 276 |
+
"eval_samples_per_second": 2.741,
|
| 277 |
+
"eval_steps_per_second": 0.685,
|
| 278 |
+
"step": 1935
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 15.503875968992247,
|
| 282 |
+
"grad_norm": 29.049968719482422,
|
| 283 |
+
"learning_rate": 0.000338,
|
| 284 |
+
"loss": 21.0819,
|
| 285 |
"step": 2000
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 16.0,
|
| 289 |
"eval_dummy": 1.0,
|
| 290 |
+
"eval_loss": 29.936742782592773,
|
| 291 |
+
"eval_runtime": 17.2024,
|
| 292 |
+
"eval_samples_per_second": 2.79,
|
| 293 |
+
"eval_steps_per_second": 0.698,
|
| 294 |
+
"step": 2064
|
| 295 |
},
|
| 296 |
{
|
| 297 |
+
"epoch": 16.27906976744186,
|
| 298 |
+
"grad_norm": 70.16988372802734,
|
| 299 |
+
"learning_rate": 0.0003349,
|
| 300 |
+
"loss": 21.0089,
|
| 301 |
"step": 2100
|
| 302 |
},
|
| 303 |
{
|
| 304 |
+
"epoch": 17.0,
|
| 305 |
"eval_dummy": 1.0,
|
| 306 |
+
"eval_loss": 30.042001724243164,
|
| 307 |
+
"eval_runtime": 17.9057,
|
| 308 |
+
"eval_samples_per_second": 2.681,
|
| 309 |
+
"eval_steps_per_second": 0.67,
|
| 310 |
+
"step": 2193
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 17.05426356589147,
|
| 314 |
+
"grad_norm": 77.12342834472656,
|
| 315 |
+
"learning_rate": 0.00033180000000000004,
|
| 316 |
+
"loss": 21.1193,
|
| 317 |
"step": 2200
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 17.829457364341085,
|
| 321 |
+
"grad_norm": 40.200958251953125,
|
| 322 |
+
"learning_rate": 0.0003287,
|
| 323 |
+
"loss": 20.9169,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"step": 2300
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 18.0,
|
| 328 |
"eval_dummy": 1.0,
|
| 329 |
+
"eval_loss": 29.293771743774414,
|
| 330 |
+
"eval_runtime": 17.1083,
|
| 331 |
+
"eval_samples_per_second": 2.806,
|
| 332 |
+
"eval_steps_per_second": 0.701,
|
| 333 |
+
"step": 2322
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 18.6046511627907,
|
| 337 |
+
"grad_norm": 31.95384979248047,
|
| 338 |
+
"learning_rate": 0.0003256,
|
| 339 |
+
"loss": 19.7935,
|
| 340 |
"step": 2400
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 19.0,
|
| 344 |
"eval_dummy": 1.0,
|
| 345 |
+
"eval_loss": 31.394454956054688,
|
| 346 |
+
"eval_runtime": 18.6297,
|
| 347 |
+
"eval_samples_per_second": 2.577,
|
| 348 |
+
"eval_steps_per_second": 0.644,
|
| 349 |
+
"step": 2451
|
| 350 |
},
|
| 351 |
{
|
| 352 |
+
"epoch": 19.37984496124031,
|
| 353 |
+
"grad_norm": 54.4798698425293,
|
| 354 |
+
"learning_rate": 0.00032250000000000003,
|
| 355 |
+
"loss": 19.8749,
|
| 356 |
"step": 2500
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 20.0,
|
| 360 |
+
"eval_dummy": 1.0,
|
| 361 |
+
"eval_loss": 29.845718383789062,
|
| 362 |
+
"eval_runtime": 17.4128,
|
| 363 |
+
"eval_samples_per_second": 2.757,
|
| 364 |
+
"eval_steps_per_second": 0.689,
|
| 365 |
+
"step": 2580
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 20.155038759689923,
|
| 369 |
+
"grad_norm": 61.0432243347168,
|
| 370 |
+
"learning_rate": 0.0003194,
|
| 371 |
+
"loss": 19.6959,
|
|
|
|
|
|
|
| 372 |
"step": 2600
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 20.930232558139537,
|
| 376 |
+
"grad_norm": 64.08226013183594,
|
| 377 |
+
"learning_rate": 0.0003163,
|
| 378 |
+
"loss": 19.2973,
|
| 379 |
"step": 2700
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 21.0,
|
| 383 |
"eval_dummy": 1.0,
|
| 384 |
+
"eval_loss": 29.071313858032227,
|
| 385 |
+
"eval_runtime": 18.6354,
|
| 386 |
+
"eval_samples_per_second": 2.576,
|
| 387 |
+
"eval_steps_per_second": 0.644,
|
| 388 |
+
"step": 2709
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 21.705426356589147,
|
| 392 |
+
"grad_norm": 54.097721099853516,
|
| 393 |
+
"learning_rate": 0.0003132,
|
| 394 |
+
"loss": 18.5436,
|
| 395 |
"step": 2800
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 22.0,
|
| 399 |
"eval_dummy": 1.0,
|
| 400 |
+
"eval_loss": 29.084577560424805,
|
| 401 |
+
"eval_runtime": 17.3593,
|
| 402 |
+
"eval_samples_per_second": 2.765,
|
| 403 |
+
"eval_steps_per_second": 0.691,
|
| 404 |
+
"step": 2838
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 22.48062015503876,
|
| 408 |
+
"grad_norm": 37.09468460083008,
|
| 409 |
+
"learning_rate": 0.00031010000000000006,
|
| 410 |
+
"loss": 18.5996,
|
| 411 |
"step": 2900
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 23.0,
|
| 415 |
"eval_dummy": 1.0,
|
| 416 |
+
"eval_loss": 29.88102149963379,
|
| 417 |
+
"eval_runtime": 17.5696,
|
| 418 |
+
"eval_samples_per_second": 2.732,
|
| 419 |
+
"eval_steps_per_second": 0.683,
|
| 420 |
+
"step": 2967
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 23.25581395348837,
|
| 424 |
+
"grad_norm": 40.900291442871094,
|
| 425 |
+
"learning_rate": 0.00030700000000000004,
|
| 426 |
+
"loss": 19.1228,
|
| 427 |
"step": 3000
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 24.0,
|
| 431 |
"eval_dummy": 1.0,
|
| 432 |
+
"eval_loss": 29.301599502563477,
|
| 433 |
+
"eval_runtime": 17.3752,
|
| 434 |
+
"eval_samples_per_second": 2.763,
|
| 435 |
+
"eval_steps_per_second": 0.691,
|
| 436 |
+
"step": 3096
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 24.031007751937985,
|
| 440 |
+
"grad_norm": 16.258365631103516,
|
| 441 |
+
"learning_rate": 0.0003039,
|
| 442 |
+
"loss": 18.2692,
|
| 443 |
"step": 3100
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 24.8062015503876,
|
| 447 |
+
"grad_norm": 53.048091888427734,
|
| 448 |
+
"learning_rate": 0.0003008,
|
| 449 |
+
"loss": 18.0519,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 450 |
"step": 3200
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 25.0,
|
| 454 |
"eval_dummy": 1.0,
|
| 455 |
+
"eval_loss": 30.71547508239746,
|
| 456 |
+
"eval_runtime": 18.0599,
|
| 457 |
+
"eval_samples_per_second": 2.658,
|
| 458 |
+
"eval_steps_per_second": 0.664,
|
| 459 |
+
"step": 3225
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 25.58139534883721,
|
| 463 |
+
"grad_norm": 24.09917640686035,
|
| 464 |
+
"learning_rate": 0.0002977,
|
| 465 |
+
"loss": 17.7073,
|
| 466 |
"step": 3300
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 26.0,
|
| 470 |
"eval_dummy": 1.0,
|
| 471 |
+
"eval_loss": 28.716806411743164,
|
| 472 |
+
"eval_runtime": 17.32,
|
| 473 |
+
"eval_samples_per_second": 2.771,
|
| 474 |
+
"eval_steps_per_second": 0.693,
|
| 475 |
+
"step": 3354
|
| 476 |
},
|
| 477 |
{
|
| 478 |
+
"epoch": 26.356589147286822,
|
| 479 |
+
"grad_norm": 86.21530151367188,
|
| 480 |
+
"learning_rate": 0.0002946,
|
| 481 |
+
"loss": 17.5055,
|
| 482 |
"step": 3400
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 27.0,
|
| 486 |
"eval_dummy": 1.0,
|
| 487 |
+
"eval_loss": 28.989931106567383,
|
| 488 |
+
"eval_runtime": 17.6249,
|
| 489 |
+
"eval_samples_per_second": 2.723,
|
| 490 |
+
"eval_steps_per_second": 0.681,
|
| 491 |
+
"step": 3483
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 27.131782945736433,
|
| 495 |
+
"grad_norm": 20.129249572753906,
|
| 496 |
+
"learning_rate": 0.0002915,
|
| 497 |
+
"loss": 18.054,
|
| 498 |
"step": 3500
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 27.906976744186046,
|
| 502 |
+
"grad_norm": 22.863677978515625,
|
| 503 |
+
"learning_rate": 0.0002884,
|
| 504 |
+
"loss": 17.4854,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 505 |
"step": 3600
|
| 506 |
},
|
| 507 |
{
|
| 508 |
+
"epoch": 28.0,
|
| 509 |
"eval_dummy": 1.0,
|
| 510 |
+
"eval_loss": 30.19437599182129,
|
| 511 |
+
"eval_runtime": 17.6593,
|
| 512 |
+
"eval_samples_per_second": 2.718,
|
| 513 |
+
"eval_steps_per_second": 0.68,
|
| 514 |
+
"step": 3612
|
| 515 |
},
|
| 516 |
{
|
| 517 |
+
"epoch": 28.68217054263566,
|
| 518 |
+
"grad_norm": 13.843172073364258,
|
| 519 |
+
"learning_rate": 0.0002853,
|
| 520 |
+
"loss": 17.0048,
|
| 521 |
"step": 3700
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 29.0,
|
| 525 |
"eval_dummy": 1.0,
|
| 526 |
+
"eval_loss": 29.28289031982422,
|
| 527 |
+
"eval_runtime": 18.1574,
|
| 528 |
+
"eval_samples_per_second": 2.644,
|
| 529 |
+
"eval_steps_per_second": 0.661,
|
| 530 |
+
"step": 3741
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 29.45736434108527,
|
| 534 |
+
"grad_norm": 20.727048873901367,
|
| 535 |
+
"learning_rate": 0.0002822,
|
| 536 |
+
"loss": 16.8731,
|
| 537 |
"step": 3800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 30.0,
|
| 541 |
"eval_dummy": 1.0,
|
| 542 |
+
"eval_loss": 30.1208438873291,
|
| 543 |
+
"eval_runtime": 16.9906,
|
| 544 |
+
"eval_samples_per_second": 2.825,
|
| 545 |
+
"eval_steps_per_second": 0.706,
|
| 546 |
+
"step": 3870
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 30.232558139534884,
|
| 550 |
+
"grad_norm": 19.721155166625977,
|
| 551 |
+
"learning_rate": 0.0002791,
|
| 552 |
+
"loss": 16.683,
|
| 553 |
"step": 3900
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 31.0,
|
| 557 |
"eval_dummy": 1.0,
|
| 558 |
+
"eval_loss": 30.758291244506836,
|
| 559 |
+
"eval_runtime": 17.7849,
|
| 560 |
+
"eval_samples_per_second": 2.699,
|
| 561 |
+
"eval_steps_per_second": 0.675,
|
| 562 |
+
"step": 3999
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 31.007751937984494,
|
| 566 |
+
"grad_norm": 25.496213912963867,
|
| 567 |
+
"learning_rate": 0.00027600000000000004,
|
| 568 |
+
"loss": 16.9178,
|
| 569 |
"step": 4000
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 31.782945736434108,
|
| 573 |
+
"grad_norm": 26.640628814697266,
|
| 574 |
+
"learning_rate": 0.0002729,
|
| 575 |
+
"loss": 16.6109,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 576 |
"step": 4100
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 32.0,
|
| 580 |
"eval_dummy": 1.0,
|
| 581 |
+
"eval_loss": 30.623199462890625,
|
| 582 |
+
"eval_runtime": 16.843,
|
| 583 |
+
"eval_samples_per_second": 2.85,
|
| 584 |
+
"eval_steps_per_second": 0.712,
|
| 585 |
+
"step": 4128
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 32.55813953488372,
|
| 589 |
+
"grad_norm": 25.067975997924805,
|
| 590 |
+
"learning_rate": 0.0002698,
|
| 591 |
+
"loss": 15.8261,
|
| 592 |
"step": 4200
|
| 593 |
},
|
| 594 |
{
|
| 595 |
+
"epoch": 33.0,
|
| 596 |
"eval_dummy": 1.0,
|
| 597 |
+
"eval_loss": 29.416189193725586,
|
| 598 |
+
"eval_runtime": 17.6752,
|
| 599 |
+
"eval_samples_per_second": 2.716,
|
| 600 |
+
"eval_steps_per_second": 0.679,
|
| 601 |
+
"step": 4257
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 33.333333333333336,
|
| 605 |
+
"grad_norm": 30.74283218383789,
|
| 606 |
+
"learning_rate": 0.00026670000000000003,
|
| 607 |
+
"loss": 16.9002,
|
| 608 |
"step": 4300
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 34.0,
|
| 612 |
"eval_dummy": 1.0,
|
| 613 |
+
"eval_loss": 30.438751220703125,
|
| 614 |
+
"eval_runtime": 16.8785,
|
| 615 |
+
"eval_samples_per_second": 2.844,
|
| 616 |
+
"eval_steps_per_second": 0.711,
|
| 617 |
+
"step": 4386
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 34.10852713178294,
|
| 621 |
+
"grad_norm": 72.6050033569336,
|
| 622 |
+
"learning_rate": 0.0002636,
|
| 623 |
+
"loss": 15.7742,
|
| 624 |
"step": 4400
|
| 625 |
},
|
| 626 |
{
|
| 627 |
+
"epoch": 34.883720930232556,
|
| 628 |
+
"grad_norm": 93.86290740966797,
|
| 629 |
+
"learning_rate": 0.00026050000000000004,
|
| 630 |
+
"loss": 16.3081,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 631 |
"step": 4500
|
| 632 |
},
|
| 633 |
{
|
| 634 |
+
"epoch": 35.0,
|
| 635 |
"eval_dummy": 1.0,
|
| 636 |
+
"eval_loss": 29.97564697265625,
|
| 637 |
+
"eval_runtime": 17.4518,
|
| 638 |
+
"eval_samples_per_second": 2.75,
|
| 639 |
+
"eval_steps_per_second": 0.688,
|
| 640 |
+
"step": 4515
|
| 641 |
},
|
| 642 |
{
|
| 643 |
+
"epoch": 35.65891472868217,
|
| 644 |
+
"grad_norm": 24.93766975402832,
|
| 645 |
+
"learning_rate": 0.0002574,
|
| 646 |
+
"loss": 15.4745,
|
| 647 |
"step": 4600
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 36.0,
|
| 651 |
"eval_dummy": 1.0,
|
| 652 |
+
"eval_loss": 28.821380615234375,
|
| 653 |
+
"eval_runtime": 16.9764,
|
| 654 |
+
"eval_samples_per_second": 2.827,
|
| 655 |
+
"eval_steps_per_second": 0.707,
|
| 656 |
+
"step": 4644
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 36.434108527131784,
|
| 660 |
+
"grad_norm": 33.70745849609375,
|
| 661 |
+
"learning_rate": 0.0002543,
|
| 662 |
+
"loss": 15.938,
|
| 663 |
"step": 4700
|
| 664 |
},
|
| 665 |
{
|
| 666 |
+
"epoch": 37.0,
|
| 667 |
"eval_dummy": 1.0,
|
| 668 |
+
"eval_loss": 29.100107192993164,
|
| 669 |
+
"eval_runtime": 17.5981,
|
| 670 |
+
"eval_samples_per_second": 2.728,
|
| 671 |
+
"eval_steps_per_second": 0.682,
|
| 672 |
+
"step": 4773
|
| 673 |
},
|
| 674 |
{
|
| 675 |
+
"epoch": 37.2093023255814,
|
| 676 |
+
"grad_norm": 59.88523864746094,
|
| 677 |
+
"learning_rate": 0.00025120000000000003,
|
| 678 |
+
"loss": 14.9862,
|
| 679 |
"step": 4800
|
| 680 |
},
|
| 681 |
{
|
| 682 |
+
"epoch": 37.98449612403101,
|
| 683 |
+
"grad_norm": 20.979228973388672,
|
| 684 |
+
"learning_rate": 0.0002481,
|
| 685 |
+
"loss": 15.9947,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 686 |
"step": 4900
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 38.0,
|
| 690 |
"eval_dummy": 1.0,
|
| 691 |
+
"eval_loss": 31.053319931030273,
|
| 692 |
+
"eval_runtime": 17.0472,
|
| 693 |
+
"eval_samples_per_second": 2.816,
|
| 694 |
+
"eval_steps_per_second": 0.704,
|
| 695 |
+
"step": 4902
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 38.75968992248062,
|
| 699 |
+
"grad_norm": 17.90158462524414,
|
| 700 |
+
"learning_rate": 0.000245,
|
| 701 |
+
"loss": 15.2328,
|
| 702 |
"step": 5000
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 39.0,
|
| 706 |
"eval_dummy": 1.0,
|
| 707 |
+
"eval_loss": 31.62113380432129,
|
| 708 |
+
"eval_runtime": 17.318,
|
| 709 |
+
"eval_samples_per_second": 2.772,
|
| 710 |
+
"eval_steps_per_second": 0.693,
|
| 711 |
+
"step": 5031
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 39.53488372093023,
|
| 715 |
+
"grad_norm": 33.11941909790039,
|
| 716 |
+
"learning_rate": 0.00024190000000000003,
|
| 717 |
+
"loss": 15.202,
|
| 718 |
"step": 5100
|
| 719 |
},
|
| 720 |
{
|
| 721 |
+
"epoch": 40.0,
|
| 722 |
+
"eval_dummy": 1.0,
|
| 723 |
+
"eval_loss": 33.138301849365234,
|
| 724 |
+
"eval_runtime": 17.0128,
|
| 725 |
+
"eval_samples_per_second": 2.821,
|
| 726 |
+
"eval_steps_per_second": 0.705,
|
| 727 |
+
"step": 5160
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 40.310077519379846,
|
| 731 |
+
"grad_norm": 15.685113906860352,
|
| 732 |
+
"learning_rate": 0.0002388,
|
| 733 |
+
"loss": 15.0583,
|
|
|
|
|
|
|
| 734 |
"step": 5200
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 41.0,
|
| 738 |
+
"eval_dummy": 1.0,
|
| 739 |
+
"eval_loss": 31.408859252929688,
|
| 740 |
+
"eval_runtime": 17.9066,
|
| 741 |
+
"eval_samples_per_second": 2.681,
|
| 742 |
+
"eval_steps_per_second": 0.67,
|
| 743 |
+
"step": 5289
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 41.08527131782946,
|
| 747 |
+
"grad_norm": 20.353235244750977,
|
| 748 |
+
"learning_rate": 0.00023569999999999998,
|
| 749 |
+
"loss": 14.7257,
|
| 750 |
+
"step": 5300
|
|
|
|
|
|
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"epoch": 41.86046511627907,
|
| 754 |
+
"grad_norm": 22.713470458984375,
|
| 755 |
+
"learning_rate": 0.00023259999999999996,
|
| 756 |
+
"loss": 14.573,
|
| 757 |
"step": 5400
|
| 758 |
},
|
| 759 |
{
|
| 760 |
+
"epoch": 42.0,
|
| 761 |
"eval_dummy": 1.0,
|
| 762 |
+
"eval_loss": 31.568130493164062,
|
| 763 |
+
"eval_runtime": 17.7042,
|
| 764 |
+
"eval_samples_per_second": 2.711,
|
| 765 |
+
"eval_steps_per_second": 0.678,
|
| 766 |
+
"step": 5418
|
| 767 |
},
|
| 768 |
{
|
| 769 |
+
"epoch": 42.63565891472868,
|
| 770 |
+
"grad_norm": 24.60871696472168,
|
| 771 |
+
"learning_rate": 0.0002295,
|
| 772 |
+
"loss": 14.7401,
|
| 773 |
"step": 5500
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 43.0,
|
| 777 |
"eval_dummy": 1.0,
|
| 778 |
+
"eval_loss": 30.554765701293945,
|
| 779 |
+
"eval_runtime": 18.0539,
|
| 780 |
+
"eval_samples_per_second": 2.659,
|
| 781 |
+
"eval_steps_per_second": 0.665,
|
| 782 |
+
"step": 5547
|
| 783 |
},
|
| 784 |
{
|
| 785 |
+
"epoch": 43.41085271317829,
|
| 786 |
+
"grad_norm": 36.38352966308594,
|
| 787 |
+
"learning_rate": 0.0002264,
|
| 788 |
+
"loss": 14.6052,
|
| 789 |
"step": 5600
|
| 790 |
},
|
| 791 |
{
|
| 792 |
+
"epoch": 44.0,
|
| 793 |
"eval_dummy": 1.0,
|
| 794 |
+
"eval_loss": 31.39527702331543,
|
| 795 |
+
"eval_runtime": 17.3086,
|
| 796 |
+
"eval_samples_per_second": 2.773,
|
| 797 |
+
"eval_steps_per_second": 0.693,
|
| 798 |
+
"step": 5676
|
| 799 |
},
|
| 800 |
{
|
| 801 |
+
"epoch": 44.18604651162791,
|
| 802 |
+
"grad_norm": 14.869057655334473,
|
| 803 |
+
"learning_rate": 0.00022330000000000003,
|
| 804 |
+
"loss": 13.9636,
|
| 805 |
"step": 5700
|
| 806 |
},
|
| 807 |
{
|
| 808 |
+
"epoch": 44.96124031007752,
|
| 809 |
+
"grad_norm": 12.379744529724121,
|
| 810 |
+
"learning_rate": 0.0002202,
|
| 811 |
+
"loss": 14.1299,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 812 |
"step": 5800
|
| 813 |
},
|
| 814 |
{
|
| 815 |
+
"epoch": 45.0,
|
| 816 |
"eval_dummy": 1.0,
|
| 817 |
+
"eval_loss": 30.81528663635254,
|
| 818 |
+
"eval_runtime": 17.5777,
|
| 819 |
+
"eval_samples_per_second": 2.731,
|
| 820 |
+
"eval_steps_per_second": 0.683,
|
| 821 |
+
"step": 5805
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 45.736434108527135,
|
| 825 |
+
"grad_norm": 51.40928649902344,
|
| 826 |
+
"learning_rate": 0.00021710000000000005,
|
| 827 |
+
"loss": 13.6851,
|
| 828 |
"step": 5900
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 46.0,
|
| 832 |
"eval_dummy": 1.0,
|
| 833 |
+
"eval_loss": 30.969324111938477,
|
| 834 |
+
"eval_runtime": 17.5744,
|
| 835 |
+
"eval_samples_per_second": 2.731,
|
| 836 |
+
"eval_steps_per_second": 0.683,
|
| 837 |
+
"step": 5934
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 46.51162790697674,
|
| 841 |
+
"grad_norm": 13.089680671691895,
|
| 842 |
+
"learning_rate": 0.00021400000000000002,
|
| 843 |
+
"loss": 14.6677,
|
| 844 |
"step": 6000
|
| 845 |
},
|
| 846 |
{
|
| 847 |
+
"epoch": 47.0,
|
| 848 |
"eval_dummy": 1.0,
|
| 849 |
+
"eval_loss": 31.936065673828125,
|
| 850 |
+
"eval_runtime": 18.038,
|
| 851 |
+
"eval_samples_per_second": 2.661,
|
| 852 |
+
"eval_steps_per_second": 0.665,
|
| 853 |
+
"step": 6063
|
| 854 |
},
|
| 855 |
{
|
| 856 |
+
"epoch": 47.286821705426355,
|
| 857 |
+
"grad_norm": 15.198484420776367,
|
| 858 |
+
"learning_rate": 0.0002109,
|
| 859 |
+
"loss": 13.6493,
|
| 860 |
"step": 6100
|
| 861 |
},
|
| 862 |
{
|
| 863 |
+
"epoch": 48.0,
|
| 864 |
"eval_dummy": 1.0,
|
| 865 |
+
"eval_loss": 34.3327751159668,
|
| 866 |
+
"eval_runtime": 17.0231,
|
| 867 |
+
"eval_samples_per_second": 2.82,
|
| 868 |
+
"eval_steps_per_second": 0.705,
|
| 869 |
+
"step": 6192
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 48.06201550387597,
|
| 873 |
+
"grad_norm": 17.197790145874023,
|
| 874 |
+
"learning_rate": 0.00020780000000000004,
|
| 875 |
+
"loss": 13.7191,
|
| 876 |
"step": 6200
|
| 877 |
},
|
| 878 |
{
|
| 879 |
+
"epoch": 48.83720930232558,
|
| 880 |
+
"grad_norm": 22.198528289794922,
|
| 881 |
+
"learning_rate": 0.00020470000000000002,
|
| 882 |
+
"loss": 14.166,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 883 |
"step": 6300
|
| 884 |
},
|
| 885 |
{
|
| 886 |
+
"epoch": 49.0,
|
| 887 |
"eval_dummy": 1.0,
|
| 888 |
+
"eval_loss": 32.62310791015625,
|
| 889 |
+
"eval_runtime": 17.8418,
|
| 890 |
+
"eval_samples_per_second": 2.69,
|
| 891 |
+
"eval_steps_per_second": 0.673,
|
| 892 |
+
"step": 6321
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 49.6124031007752,
|
| 896 |
+
"grad_norm": 12.192609786987305,
|
| 897 |
+
"learning_rate": 0.00020160000000000002,
|
| 898 |
+
"loss": 13.7388,
|
| 899 |
"step": 6400
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 50.0,
|
| 903 |
"eval_dummy": 1.0,
|
| 904 |
+
"eval_loss": 33.17361831665039,
|
| 905 |
+
"eval_runtime": 16.9796,
|
| 906 |
+
"eval_samples_per_second": 2.827,
|
| 907 |
+
"eval_steps_per_second": 0.707,
|
| 908 |
+
"step": 6450
|
| 909 |
},
|
| 910 |
{
|
| 911 |
+
"epoch": 50.3875968992248,
|
| 912 |
+
"grad_norm": 24.24190330505371,
|
| 913 |
+
"learning_rate": 0.0001985,
|
| 914 |
+
"loss": 13.0849,
|
| 915 |
"step": 6500
|
| 916 |
},
|
| 917 |
{
|
| 918 |
+
"epoch": 51.0,
|
| 919 |
"eval_dummy": 1.0,
|
| 920 |
+
"eval_loss": 34.95216369628906,
|
| 921 |
+
"eval_runtime": 17.6852,
|
| 922 |
+
"eval_samples_per_second": 2.714,
|
| 923 |
+
"eval_steps_per_second": 0.679,
|
| 924 |
+
"step": 6579
|
| 925 |
},
|
| 926 |
{
|
| 927 |
+
"epoch": 51.16279069767442,
|
| 928 |
+
"grad_norm": 10.653921127319336,
|
| 929 |
+
"learning_rate": 0.0001954,
|
| 930 |
+
"loss": 13.7478,
|
| 931 |
"step": 6600
|
| 932 |
},
|
| 933 |
{
|
| 934 |
+
"epoch": 51.93798449612403,
|
| 935 |
+
"grad_norm": 12.344590187072754,
|
| 936 |
+
"learning_rate": 0.00019229999999999999,
|
| 937 |
+
"loss": 13.2502,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 938 |
"step": 6700
|
| 939 |
},
|
| 940 |
{
|
| 941 |
+
"epoch": 52.0,
|
| 942 |
"eval_dummy": 1.0,
|
| 943 |
+
"eval_loss": 35.79899215698242,
|
| 944 |
+
"eval_runtime": 17.0885,
|
| 945 |
+
"eval_samples_per_second": 2.809,
|
| 946 |
+
"eval_steps_per_second": 0.702,
|
| 947 |
+
"step": 6708
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 52.713178294573645,
|
| 951 |
+
"grad_norm": 11.102241516113281,
|
| 952 |
+
"learning_rate": 0.0001892,
|
| 953 |
+
"loss": 13.5116,
|
| 954 |
"step": 6800
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 53.0,
|
| 958 |
"eval_dummy": 1.0,
|
| 959 |
+
"eval_loss": 31.57374382019043,
|
| 960 |
+
"eval_runtime": 17.6271,
|
| 961 |
+
"eval_samples_per_second": 2.723,
|
| 962 |
+
"eval_steps_per_second": 0.681,
|
| 963 |
+
"step": 6837
|
| 964 |
},
|
| 965 |
{
|
| 966 |
+
"epoch": 53.48837209302326,
|
| 967 |
+
"grad_norm": 10.652983665466309,
|
| 968 |
+
"learning_rate": 0.00018610000000000002,
|
| 969 |
+
"loss": 12.6993,
|
| 970 |
"step": 6900
|
| 971 |
},
|
| 972 |
{
|
| 973 |
+
"epoch": 54.0,
|
| 974 |
"eval_dummy": 1.0,
|
| 975 |
+
"eval_loss": 33.26504898071289,
|
| 976 |
+
"eval_runtime": 17.4525,
|
| 977 |
+
"eval_samples_per_second": 2.75,
|
| 978 |
+
"eval_steps_per_second": 0.688,
|
| 979 |
+
"step": 6966
|
| 980 |
},
|
| 981 |
{
|
| 982 |
+
"epoch": 54.263565891472865,
|
| 983 |
+
"grad_norm": 12.211697578430176,
|
| 984 |
+
"learning_rate": 0.00018300000000000003,
|
| 985 |
+
"loss": 13.3602,
|
| 986 |
"step": 7000
|
| 987 |
},
|
| 988 |
{
|
| 989 |
+
"epoch": 55.0,
|
| 990 |
"eval_dummy": 1.0,
|
| 991 |
+
"eval_loss": 34.891380310058594,
|
| 992 |
+
"eval_runtime": 18.9822,
|
| 993 |
+
"eval_samples_per_second": 2.529,
|
| 994 |
+
"eval_steps_per_second": 0.632,
|
| 995 |
+
"step": 7095
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 55.03875968992248,
|
| 999 |
+
"grad_norm": 14.056374549865723,
|
| 1000 |
+
"learning_rate": 0.0001799,
|
| 1001 |
+
"loss": 12.9955,
|
| 1002 |
"step": 7100
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
+
"epoch": 55.81395348837209,
|
| 1006 |
+
"grad_norm": 10.69999885559082,
|
| 1007 |
+
"learning_rate": 0.00017680000000000001,
|
| 1008 |
+
"loss": 12.9585,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1009 |
"step": 7200
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"epoch": 56.0,
|
| 1013 |
"eval_dummy": 1.0,
|
| 1014 |
+
"eval_loss": 35.98616409301758,
|
| 1015 |
+
"eval_runtime": 17.2599,
|
| 1016 |
+
"eval_samples_per_second": 2.781,
|
| 1017 |
+
"eval_steps_per_second": 0.695,
|
| 1018 |
+
"step": 7224
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
+
"epoch": 56.58914728682171,
|
| 1022 |
+
"grad_norm": 7.194685459136963,
|
| 1023 |
+
"learning_rate": 0.00017370000000000002,
|
| 1024 |
+
"loss": 12.7434,
|
| 1025 |
"step": 7300
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 57.0,
|
| 1029 |
"eval_dummy": 1.0,
|
| 1030 |
+
"eval_loss": 34.91057205200195,
|
| 1031 |
+
"eval_runtime": 18.4913,
|
| 1032 |
+
"eval_samples_per_second": 2.596,
|
| 1033 |
+
"eval_steps_per_second": 0.649,
|
| 1034 |
+
"step": 7353
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
+
"epoch": 57.36434108527132,
|
| 1038 |
+
"grad_norm": 9.769908905029297,
|
| 1039 |
+
"learning_rate": 0.0001706,
|
| 1040 |
+
"loss": 12.7299,
|
| 1041 |
"step": 7400
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
+
"epoch": 58.0,
|
| 1045 |
"eval_dummy": 1.0,
|
| 1046 |
+
"eval_loss": 34.010562896728516,
|
| 1047 |
+
"eval_runtime": 17.0454,
|
| 1048 |
+
"eval_samples_per_second": 2.816,
|
| 1049 |
+
"eval_steps_per_second": 0.704,
|
| 1050 |
+
"step": 7482
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 58.13953488372093,
|
| 1054 |
+
"grad_norm": 18.091665267944336,
|
| 1055 |
+
"learning_rate": 0.0001675,
|
| 1056 |
+
"loss": 12.3929,
|
| 1057 |
"step": 7500
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
+
"epoch": 58.91472868217054,
|
| 1061 |
+
"grad_norm": 8.427603721618652,
|
| 1062 |
+
"learning_rate": 0.00016439999999999998,
|
| 1063 |
+
"loss": 12.717,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1064 |
"step": 7600
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 59.0,
|
| 1068 |
"eval_dummy": 1.0,
|
| 1069 |
+
"eval_loss": 36.35882568359375,
|
| 1070 |
+
"eval_runtime": 18.0781,
|
| 1071 |
+
"eval_samples_per_second": 2.655,
|
| 1072 |
+
"eval_steps_per_second": 0.664,
|
| 1073 |
+
"step": 7611
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 59.689922480620154,
|
| 1077 |
+
"grad_norm": 15.918642044067383,
|
| 1078 |
+
"learning_rate": 0.00016130000000000002,
|
| 1079 |
+
"loss": 12.0563,
|
| 1080 |
"step": 7700
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
+
"epoch": 60.0,
|
| 1084 |
+
"eval_dummy": 1.0,
|
| 1085 |
+
"eval_loss": 35.09232711791992,
|
| 1086 |
+
"eval_runtime": 16.9066,
|
| 1087 |
+
"eval_samples_per_second": 2.839,
|
| 1088 |
+
"eval_steps_per_second": 0.71,
|
| 1089 |
+
"step": 7740
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 60.46511627906977,
|
| 1093 |
+
"grad_norm": 13.870895385742188,
|
| 1094 |
+
"learning_rate": 0.00015820000000000002,
|
| 1095 |
+
"loss": 13.012,
|
| 1096 |
"step": 7800
|
| 1097 |
},
|
| 1098 |
{
|
| 1099 |
+
"epoch": 61.0,
|
| 1100 |
"eval_dummy": 1.0,
|
| 1101 |
+
"eval_loss": 38.73225402832031,
|
| 1102 |
+
"eval_runtime": 17.9846,
|
| 1103 |
+
"eval_samples_per_second": 2.669,
|
| 1104 |
+
"eval_steps_per_second": 0.667,
|
| 1105 |
+
"step": 7869
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
+
"epoch": 61.24031007751938,
|
| 1109 |
+
"grad_norm": 7.798965930938721,
|
| 1110 |
+
"learning_rate": 0.00015510000000000003,
|
| 1111 |
+
"loss": 12.2878,
|
| 1112 |
"step": 7900
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
+
"epoch": 62.0,
|
| 1116 |
"eval_dummy": 1.0,
|
| 1117 |
+
"eval_loss": 34.9967155456543,
|
| 1118 |
+
"eval_runtime": 17.0439,
|
| 1119 |
+
"eval_samples_per_second": 2.816,
|
| 1120 |
+
"eval_steps_per_second": 0.704,
|
| 1121 |
+
"step": 7998
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
+
"epoch": 62.01550387596899,
|
| 1125 |
+
"grad_norm": 8.46688461303711,
|
| 1126 |
+
"learning_rate": 0.000152,
|
| 1127 |
+
"loss": 12.3515,
|
| 1128 |
"step": 8000
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
+
"epoch": 62.7906976744186,
|
| 1132 |
+
"grad_norm": 9.745466232299805,
|
| 1133 |
+
"learning_rate": 0.00014890000000000001,
|
| 1134 |
+
"loss": 12.2794,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1135 |
"step": 8100
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
+
"epoch": 63.0,
|
| 1139 |
"eval_dummy": 1.0,
|
| 1140 |
+
"eval_loss": 37.55772399902344,
|
| 1141 |
+
"eval_runtime": 18.0451,
|
| 1142 |
+
"eval_samples_per_second": 2.66,
|
| 1143 |
+
"eval_steps_per_second": 0.665,
|
| 1144 |
+
"step": 8127
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
+
"epoch": 63.565891472868216,
|
| 1148 |
+
"grad_norm": 7.328401565551758,
|
| 1149 |
+
"learning_rate": 0.0001458,
|
| 1150 |
+
"loss": 12.4147,
|
| 1151 |
"step": 8200
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
+
"epoch": 64.0,
|
| 1155 |
"eval_dummy": 1.0,
|
| 1156 |
+
"eval_loss": 37.27333068847656,
|
| 1157 |
+
"eval_runtime": 19.2621,
|
| 1158 |
+
"eval_samples_per_second": 2.492,
|
| 1159 |
+
"eval_steps_per_second": 0.623,
|
| 1160 |
+
"step": 8256
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
+
"epoch": 64.34108527131782,
|
| 1164 |
+
"grad_norm": 12.89833927154541,
|
| 1165 |
+
"learning_rate": 0.0001427,
|
| 1166 |
+
"loss": 12.0032,
|
| 1167 |
"step": 8300
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
+
"epoch": 65.0,
|
| 1171 |
"eval_dummy": 1.0,
|
| 1172 |
+
"eval_loss": 35.3015022277832,
|
| 1173 |
+
"eval_runtime": 17.8838,
|
| 1174 |
+
"eval_samples_per_second": 2.684,
|
| 1175 |
+
"eval_steps_per_second": 0.671,
|
| 1176 |
+
"step": 8385
|
| 1177 |
},
|
| 1178 |
{
|
| 1179 |
+
"epoch": 65.11627906976744,
|
| 1180 |
+
"grad_norm": 15.308392524719238,
|
| 1181 |
+
"learning_rate": 0.00013959999999999998,
|
| 1182 |
+
"loss": 11.7392,
|
| 1183 |
"step": 8400
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
+
"epoch": 65.89147286821705,
|
| 1187 |
+
"grad_norm": 12.101038932800293,
|
| 1188 |
+
"learning_rate": 0.00013650000000000004,
|
| 1189 |
+
"loss": 12.2793,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1190 |
"step": 8500
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
+
"epoch": 66.0,
|
| 1194 |
"eval_dummy": 1.0,
|
| 1195 |
+
"eval_loss": 35.280582427978516,
|
| 1196 |
+
"eval_runtime": 17.6628,
|
| 1197 |
+
"eval_samples_per_second": 2.718,
|
| 1198 |
+
"eval_steps_per_second": 0.679,
|
| 1199 |
+
"step": 8514
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
+
"epoch": 66.66666666666667,
|
| 1203 |
+
"grad_norm": 12.754354476928711,
|
| 1204 |
+
"learning_rate": 0.00013340000000000002,
|
| 1205 |
+
"loss": 12.2309,
|
| 1206 |
"step": 8600
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
+
"epoch": 67.0,
|
| 1210 |
"eval_dummy": 1.0,
|
| 1211 |
+
"eval_loss": 36.24875259399414,
|
| 1212 |
+
"eval_runtime": 17.2522,
|
| 1213 |
+
"eval_samples_per_second": 2.782,
|
| 1214 |
+
"eval_steps_per_second": 0.696,
|
| 1215 |
+
"step": 8643
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
+
"epoch": 67.44186046511628,
|
| 1219 |
+
"grad_norm": 9.756113052368164,
|
| 1220 |
+
"learning_rate": 0.00013030000000000002,
|
| 1221 |
+
"loss": 11.7082,
|
| 1222 |
"step": 8700
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"epoch": 68.0,
|
| 1226 |
"eval_dummy": 1.0,
|
| 1227 |
+
"eval_loss": 35.66865158081055,
|
| 1228 |
+
"eval_runtime": 18.2695,
|
| 1229 |
+
"eval_samples_per_second": 2.627,
|
| 1230 |
+
"eval_steps_per_second": 0.657,
|
| 1231 |
+
"step": 8772
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
+
"epoch": 68.21705426356588,
|
| 1235 |
+
"grad_norm": 9.372435569763184,
|
| 1236 |
+
"learning_rate": 0.0001272,
|
| 1237 |
+
"loss": 11.5136,
|
| 1238 |
"step": 8800
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
+
"epoch": 68.9922480620155,
|
| 1242 |
+
"grad_norm": 15.044282913208008,
|
| 1243 |
+
"learning_rate": 0.0001241,
|
| 1244 |
+
"loss": 11.8694,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1245 |
"step": 8900
|
| 1246 |
},
|
| 1247 |
{
|
| 1248 |
+
"epoch": 69.0,
|
| 1249 |
"eval_dummy": 1.0,
|
| 1250 |
+
"eval_loss": 36.04698944091797,
|
| 1251 |
+
"eval_runtime": 17.3888,
|
| 1252 |
+
"eval_samples_per_second": 2.76,
|
| 1253 |
+
"eval_steps_per_second": 0.69,
|
| 1254 |
+
"step": 8901
|
| 1255 |
},
|
| 1256 |
{
|
| 1257 |
+
"epoch": 69.76744186046511,
|
| 1258 |
+
"grad_norm": 9.250027656555176,
|
| 1259 |
+
"learning_rate": 0.000121,
|
| 1260 |
+
"loss": 11.782,
|
| 1261 |
"step": 9000
|
| 1262 |
},
|
| 1263 |
{
|
| 1264 |
+
"epoch": 70.0,
|
| 1265 |
"eval_dummy": 1.0,
|
| 1266 |
+
"eval_loss": 35.40549087524414,
|
| 1267 |
+
"eval_runtime": 18.0194,
|
| 1268 |
+
"eval_samples_per_second": 2.664,
|
| 1269 |
+
"eval_steps_per_second": 0.666,
|
| 1270 |
+
"step": 9030
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
+
"epoch": 70.54263565891473,
|
| 1274 |
+
"grad_norm": 16.64485740661621,
|
| 1275 |
+
"learning_rate": 0.00011789999999999999,
|
| 1276 |
+
"loss": 11.6254,
|
| 1277 |
"step": 9100
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 71.0,
|
| 1281 |
"eval_dummy": 1.0,
|
| 1282 |
+
"eval_loss": 36.70663070678711,
|
| 1283 |
+
"eval_runtime": 17.3613,
|
| 1284 |
+
"eval_samples_per_second": 2.765,
|
| 1285 |
+
"eval_steps_per_second": 0.691,
|
| 1286 |
+
"step": 9159
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
+
"epoch": 71.31782945736434,
|
| 1290 |
+
"grad_norm": 15.693510055541992,
|
| 1291 |
+
"learning_rate": 0.00011479999999999997,
|
| 1292 |
+
"loss": 11.5873,
|
| 1293 |
"step": 9200
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
+
"epoch": 72.0,
|
| 1297 |
"eval_dummy": 1.0,
|
| 1298 |
+
"eval_loss": 36.10844421386719,
|
| 1299 |
+
"eval_runtime": 17.5839,
|
| 1300 |
+
"eval_samples_per_second": 2.73,
|
| 1301 |
+
"eval_steps_per_second": 0.682,
|
| 1302 |
+
"step": 9288
|
| 1303 |
},
|
| 1304 |
{
|
| 1305 |
+
"epoch": 72.09302325581395,
|
| 1306 |
+
"grad_norm": 7.485771179199219,
|
| 1307 |
+
"learning_rate": 0.00011170000000000003,
|
| 1308 |
+
"loss": 11.6159,
|
| 1309 |
"step": 9300
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
+
"epoch": 72.86821705426357,
|
| 1313 |
+
"grad_norm": 15.41925048828125,
|
| 1314 |
+
"learning_rate": 0.00010860000000000004,
|
| 1315 |
+
"loss": 11.6251,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1316 |
"step": 9400
|
| 1317 |
},
|
| 1318 |
{
|
| 1319 |
+
"epoch": 73.0,
|
| 1320 |
"eval_dummy": 1.0,
|
| 1321 |
+
"eval_loss": 38.29316329956055,
|
| 1322 |
+
"eval_runtime": 17.0634,
|
| 1323 |
+
"eval_samples_per_second": 2.813,
|
| 1324 |
+
"eval_steps_per_second": 0.703,
|
| 1325 |
+
"step": 9417
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
+
"epoch": 73.64341085271317,
|
| 1329 |
+
"grad_norm": 16.74988555908203,
|
| 1330 |
+
"learning_rate": 0.00010550000000000002,
|
| 1331 |
+
"loss": 11.4589,
|
| 1332 |
"step": 9500
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
+
"epoch": 74.0,
|
| 1336 |
"eval_dummy": 1.0,
|
| 1337 |
+
"eval_loss": 36.55695724487305,
|
| 1338 |
+
"eval_runtime": 17.9041,
|
| 1339 |
+
"eval_samples_per_second": 2.681,
|
| 1340 |
+
"eval_steps_per_second": 0.67,
|
| 1341 |
+
"step": 9546
|
| 1342 |
},
|
| 1343 |
{
|
| 1344 |
+
"epoch": 74.4186046511628,
|
| 1345 |
+
"grad_norm": 146.4043426513672,
|
| 1346 |
+
"learning_rate": 0.00010240000000000001,
|
| 1347 |
+
"loss": 11.7378,
|
| 1348 |
"step": 9600
|
| 1349 |
},
|
| 1350 |
{
|
| 1351 |
+
"epoch": 75.0,
|
| 1352 |
"eval_dummy": 1.0,
|
| 1353 |
+
"eval_loss": 35.988651275634766,
|
| 1354 |
+
"eval_runtime": 17.0167,
|
| 1355 |
+
"eval_samples_per_second": 2.821,
|
| 1356 |
+
"eval_steps_per_second": 0.705,
|
| 1357 |
+
"step": 9675
|
| 1358 |
},
|
| 1359 |
{
|
| 1360 |
+
"epoch": 75.1937984496124,
|
| 1361 |
+
"grad_norm": 10.800848960876465,
|
| 1362 |
+
"learning_rate": 9.93e-05,
|
| 1363 |
+
"loss": 11.4043,
|
| 1364 |
"step": 9700
|
| 1365 |
},
|
| 1366 |
{
|
| 1367 |
+
"epoch": 75.96899224806202,
|
| 1368 |
+
"grad_norm": 9.41010570526123,
|
| 1369 |
+
"learning_rate": 9.62e-05,
|
| 1370 |
+
"loss": 11.4933,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1371 |
"step": 9800
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
+
"epoch": 76.0,
|
| 1375 |
"eval_dummy": 1.0,
|
| 1376 |
+
"eval_loss": 36.47134017944336,
|
| 1377 |
+
"eval_runtime": 17.8569,
|
| 1378 |
+
"eval_samples_per_second": 2.688,
|
| 1379 |
+
"eval_steps_per_second": 0.672,
|
| 1380 |
+
"step": 9804
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
+
"epoch": 76.74418604651163,
|
| 1384 |
+
"grad_norm": 19.61960220336914,
|
| 1385 |
+
"learning_rate": 9.31e-05,
|
| 1386 |
+
"loss": 11.2566,
|
| 1387 |
"step": 9900
|
| 1388 |
},
|
| 1389 |
{
|
| 1390 |
+
"epoch": 77.0,
|
| 1391 |
"eval_dummy": 1.0,
|
| 1392 |
+
"eval_loss": 36.96221923828125,
|
| 1393 |
+
"eval_runtime": 17.0325,
|
| 1394 |
+
"eval_samples_per_second": 2.818,
|
| 1395 |
+
"eval_steps_per_second": 0.705,
|
| 1396 |
+
"step": 9933
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
+
"epoch": 77.51937984496124,
|
| 1400 |
+
"grad_norm": 9.528326034545898,
|
| 1401 |
+
"learning_rate": 8.999999999999999e-05,
|
| 1402 |
+
"loss": 11.25,
|
| 1403 |
"step": 10000
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
+
"epoch": 78.0,
|
| 1407 |
"eval_dummy": 1.0,
|
| 1408 |
+
"eval_loss": 37.10159683227539,
|
| 1409 |
+
"eval_runtime": 18.225,
|
| 1410 |
+
"eval_samples_per_second": 2.634,
|
| 1411 |
+
"eval_steps_per_second": 0.658,
|
| 1412 |
+
"step": 10062
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
+
"epoch": 78.29457364341086,
|
| 1416 |
+
"grad_norm": 8.064875602722168,
|
| 1417 |
+
"learning_rate": 8.690000000000003e-05,
|
| 1418 |
+
"loss": 11.2962,
|
| 1419 |
"step": 10100
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
+
"epoch": 79.0,
|
| 1423 |
"eval_dummy": 1.0,
|
| 1424 |
+
"eval_loss": 37.87105178833008,
|
| 1425 |
+
"eval_runtime": 16.9948,
|
| 1426 |
+
"eval_samples_per_second": 2.824,
|
| 1427 |
+
"eval_steps_per_second": 0.706,
|
| 1428 |
+
"step": 10191
|
| 1429 |
},
|
| 1430 |
{
|
| 1431 |
+
"epoch": 79.06976744186046,
|
| 1432 |
+
"grad_norm": 10.451505661010742,
|
| 1433 |
+
"learning_rate": 8.380000000000002e-05,
|
| 1434 |
+
"loss": 11.1642,
|
| 1435 |
"step": 10200
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
+
"epoch": 79.84496124031008,
|
| 1439 |
+
"grad_norm": 7.318461894989014,
|
| 1440 |
+
"learning_rate": 8.070000000000001e-05,
|
| 1441 |
+
"loss": 11.0868,
|
| 1442 |
+
"step": 10300
|
|
|
|
|
|
|
| 1443 |
},
|
| 1444 |
{
|
| 1445 |
+
"epoch": 80.0,
|
| 1446 |
+
"eval_dummy": 1.0,
|
| 1447 |
+
"eval_loss": 38.571414947509766,
|
| 1448 |
+
"eval_runtime": 18.0923,
|
| 1449 |
+
"eval_samples_per_second": 2.653,
|
| 1450 |
+
"eval_steps_per_second": 0.663,
|
| 1451 |
+
"step": 10320
|
| 1452 |
},
|
| 1453 |
{
|
| 1454 |
+
"epoch": 80.62015503875969,
|
| 1455 |
+
"grad_norm": 10.301888465881348,
|
| 1456 |
+
"learning_rate": 7.760000000000002e-05,
|
| 1457 |
+
"loss": 11.2786,
|
| 1458 |
"step": 10400
|
| 1459 |
},
|
| 1460 |
{
|
| 1461 |
+
"epoch": 81.0,
|
| 1462 |
"eval_dummy": 1.0,
|
| 1463 |
+
"eval_loss": 38.1493034362793,
|
| 1464 |
+
"eval_runtime": 17.0167,
|
| 1465 |
+
"eval_samples_per_second": 2.821,
|
| 1466 |
+
"eval_steps_per_second": 0.705,
|
| 1467 |
+
"step": 10449
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
+
"epoch": 81.3953488372093,
|
| 1471 |
+
"grad_norm": 8.667201042175293,
|
| 1472 |
+
"learning_rate": 7.450000000000001e-05,
|
| 1473 |
+
"loss": 11.1528,
|
| 1474 |
"step": 10500
|
| 1475 |
},
|
| 1476 |
{
|
| 1477 |
+
"epoch": 82.0,
|
| 1478 |
"eval_dummy": 1.0,
|
| 1479 |
+
"eval_loss": 39.0099983215332,
|
| 1480 |
+
"eval_runtime": 17.9494,
|
| 1481 |
+
"eval_samples_per_second": 2.674,
|
| 1482 |
+
"eval_steps_per_second": 0.669,
|
| 1483 |
+
"step": 10578
|
| 1484 |
},
|
| 1485 |
{
|
| 1486 |
+
"epoch": 82.17054263565892,
|
| 1487 |
+
"grad_norm": 5.117663860321045,
|
| 1488 |
+
"learning_rate": 7.14e-05,
|
| 1489 |
+
"loss": 10.9299,
|
| 1490 |
"step": 10600
|
| 1491 |
},
|
| 1492 |
{
|
| 1493 |
+
"epoch": 82.94573643410853,
|
| 1494 |
+
"grad_norm": 5.9621806144714355,
|
| 1495 |
+
"learning_rate": 6.829999999999999e-05,
|
| 1496 |
+
"loss": 11.089,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1497 |
"step": 10700
|
| 1498 |
},
|
| 1499 |
{
|
| 1500 |
+
"epoch": 83.0,
|
| 1501 |
"eval_dummy": 1.0,
|
| 1502 |
+
"eval_loss": 38.5473518371582,
|
| 1503 |
+
"eval_runtime": 17.0039,
|
| 1504 |
+
"eval_samples_per_second": 2.823,
|
| 1505 |
+
"eval_steps_per_second": 0.706,
|
| 1506 |
+
"step": 10707
|
| 1507 |
},
|
| 1508 |
{
|
| 1509 |
+
"epoch": 83.72093023255815,
|
| 1510 |
+
"grad_norm": 6.17501974105835,
|
| 1511 |
+
"learning_rate": 6.519999999999999e-05,
|
| 1512 |
+
"loss": 10.954,
|
| 1513 |
"step": 10800
|
| 1514 |
},
|
| 1515 |
{
|
| 1516 |
+
"epoch": 84.0,
|
| 1517 |
"eval_dummy": 1.0,
|
| 1518 |
+
"eval_loss": 38.940486907958984,
|
| 1519 |
+
"eval_runtime": 17.721,
|
| 1520 |
+
"eval_samples_per_second": 2.709,
|
| 1521 |
+
"eval_steps_per_second": 0.677,
|
| 1522 |
+
"step": 10836
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
+
"epoch": 84.49612403100775,
|
| 1526 |
+
"grad_norm": 31.69377326965332,
|
| 1527 |
+
"learning_rate": 6.210000000000003e-05,
|
| 1528 |
+
"loss": 11.0157,
|
| 1529 |
"step": 10900
|
| 1530 |
},
|
| 1531 |
{
|
| 1532 |
+
"epoch": 85.0,
|
| 1533 |
"eval_dummy": 1.0,
|
| 1534 |
+
"eval_loss": 39.3872184753418,
|
| 1535 |
+
"eval_runtime": 16.9062,
|
| 1536 |
+
"eval_samples_per_second": 2.839,
|
| 1537 |
+
"eval_steps_per_second": 0.71,
|
| 1538 |
+
"step": 10965
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
+
"epoch": 85.27131782945736,
|
| 1542 |
+
"grad_norm": 7.1002984046936035,
|
| 1543 |
+
"learning_rate": 5.9000000000000025e-05,
|
| 1544 |
+
"loss": 10.9849,
|
| 1545 |
"step": 11000
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
+
"epoch": 86.0,
|
| 1549 |
"eval_dummy": 1.0,
|
| 1550 |
+
"eval_loss": 39.4875373840332,
|
| 1551 |
+
"eval_runtime": 17.7347,
|
| 1552 |
+
"eval_samples_per_second": 2.707,
|
| 1553 |
+
"eval_steps_per_second": 0.677,
|
| 1554 |
+
"step": 11094
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
+
"epoch": 86.04651162790698,
|
| 1558 |
+
"grad_norm": 13.370129585266113,
|
| 1559 |
+
"learning_rate": 5.590000000000002e-05,
|
| 1560 |
+
"loss": 11.0614,
|
| 1561 |
"step": 11100
|
| 1562 |
},
|
| 1563 |
{
|
| 1564 |
+
"epoch": 86.82170542635659,
|
| 1565 |
+
"grad_norm": 5.192051887512207,
|
| 1566 |
+
"learning_rate": 5.28e-05,
|
| 1567 |
+
"loss": 10.5423,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1568 |
"step": 11200
|
| 1569 |
},
|
| 1570 |
{
|
| 1571 |
+
"epoch": 87.0,
|
| 1572 |
"eval_dummy": 1.0,
|
| 1573 |
+
"eval_loss": 39.11787796020508,
|
| 1574 |
+
"eval_runtime": 16.9675,
|
| 1575 |
+
"eval_samples_per_second": 2.829,
|
| 1576 |
+
"eval_steps_per_second": 0.707,
|
| 1577 |
+
"step": 11223
|
| 1578 |
},
|
| 1579 |
{
|
| 1580 |
+
"epoch": 87.59689922480621,
|
| 1581 |
+
"grad_norm": 5.747579097747803,
|
| 1582 |
+
"learning_rate": 4.97e-05,
|
| 1583 |
+
"loss": 11.1968,
|
| 1584 |
"step": 11300
|
| 1585 |
},
|
| 1586 |
{
|
| 1587 |
+
"epoch": 88.0,
|
| 1588 |
"eval_dummy": 1.0,
|
| 1589 |
+
"eval_loss": 39.4084358215332,
|
| 1590 |
+
"eval_runtime": 17.9374,
|
| 1591 |
+
"eval_samples_per_second": 2.676,
|
| 1592 |
+
"eval_steps_per_second": 0.669,
|
| 1593 |
+
"step": 11352
|
| 1594 |
},
|
| 1595 |
{
|
| 1596 |
+
"epoch": 88.37209302325581,
|
| 1597 |
+
"grad_norm": 11.57238483428955,
|
| 1598 |
+
"learning_rate": 4.66e-05,
|
| 1599 |
+
"loss": 10.6376,
|
| 1600 |
"step": 11400
|
| 1601 |
},
|
| 1602 |
{
|
| 1603 |
+
"epoch": 89.0,
|
| 1604 |
"eval_dummy": 1.0,
|
| 1605 |
+
"eval_loss": 39.82176971435547,
|
| 1606 |
+
"eval_runtime": 16.9422,
|
| 1607 |
+
"eval_samples_per_second": 2.833,
|
| 1608 |
+
"eval_steps_per_second": 0.708,
|
| 1609 |
+
"step": 11481
|
| 1610 |
},
|
| 1611 |
{
|
| 1612 |
+
"epoch": 89.14728682170542,
|
| 1613 |
+
"grad_norm": 8.44890308380127,
|
| 1614 |
+
"learning_rate": 4.3499999999999993e-05,
|
| 1615 |
+
"loss": 10.8035,
|
| 1616 |
"step": 11500
|
| 1617 |
},
|
| 1618 |
{
|
| 1619 |
+
"epoch": 89.92248062015504,
|
| 1620 |
+
"grad_norm": 7.732810974121094,
|
| 1621 |
+
"learning_rate": 4.0399999999999986e-05,
|
| 1622 |
+
"loss": 10.7131,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1623 |
"step": 11600
|
| 1624 |
},
|
| 1625 |
{
|
| 1626 |
+
"epoch": 90.0,
|
| 1627 |
"eval_dummy": 1.0,
|
| 1628 |
+
"eval_loss": 39.25526428222656,
|
| 1629 |
+
"eval_runtime": 18.1455,
|
| 1630 |
+
"eval_samples_per_second": 2.645,
|
| 1631 |
+
"eval_steps_per_second": 0.661,
|
| 1632 |
+
"step": 11610
|
| 1633 |
},
|
| 1634 |
{
|
| 1635 |
+
"epoch": 90.69767441860465,
|
| 1636 |
+
"grad_norm": 6.144818305969238,
|
| 1637 |
+
"learning_rate": 3.7300000000000026e-05,
|
| 1638 |
+
"loss": 10.8252,
|
| 1639 |
"step": 11700
|
| 1640 |
},
|
| 1641 |
{
|
| 1642 |
+
"epoch": 91.0,
|
| 1643 |
"eval_dummy": 1.0,
|
| 1644 |
+
"eval_loss": 39.136837005615234,
|
| 1645 |
+
"eval_runtime": 16.9741,
|
| 1646 |
+
"eval_samples_per_second": 2.828,
|
| 1647 |
+
"eval_steps_per_second": 0.707,
|
| 1648 |
+
"step": 11739
|
| 1649 |
},
|
| 1650 |
{
|
| 1651 |
+
"epoch": 91.47286821705427,
|
| 1652 |
+
"grad_norm": 4.7243475914001465,
|
| 1653 |
+
"learning_rate": 3.420000000000002e-05,
|
| 1654 |
+
"loss": 10.6456,
|
| 1655 |
"step": 11800
|
| 1656 |
},
|
| 1657 |
{
|
| 1658 |
+
"epoch": 92.0,
|
| 1659 |
"eval_dummy": 1.0,
|
| 1660 |
+
"eval_loss": 38.91936111450195,
|
| 1661 |
+
"eval_runtime": 17.9915,
|
| 1662 |
+
"eval_samples_per_second": 2.668,
|
| 1663 |
+
"eval_steps_per_second": 0.667,
|
| 1664 |
+
"step": 11868
|
| 1665 |
},
|
| 1666 |
{
|
| 1667 |
+
"epoch": 92.24806201550388,
|
| 1668 |
+
"grad_norm": 13.374404907226562,
|
| 1669 |
+
"learning_rate": 3.110000000000002e-05,
|
| 1670 |
+
"loss": 10.8488,
|
| 1671 |
"step": 11900
|
| 1672 |
},
|
| 1673 |
{
|
| 1674 |
+
"epoch": 93.0,
|
| 1675 |
"eval_dummy": 1.0,
|
| 1676 |
+
"eval_loss": 39.595462799072266,
|
| 1677 |
+
"eval_runtime": 16.9478,
|
| 1678 |
+
"eval_samples_per_second": 2.832,
|
| 1679 |
+
"eval_steps_per_second": 0.708,
|
| 1680 |
+
"step": 11997
|
| 1681 |
},
|
| 1682 |
{
|
| 1683 |
+
"epoch": 93.02325581395348,
|
| 1684 |
+
"grad_norm": 9.406952857971191,
|
| 1685 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 1686 |
+
"loss": 10.5219,
|
| 1687 |
"step": 12000
|
| 1688 |
},
|
| 1689 |
{
|
| 1690 |
+
"epoch": 93.7984496124031,
|
| 1691 |
+
"grad_norm": 5.360720634460449,
|
| 1692 |
+
"learning_rate": 2.49e-05,
|
| 1693 |
+
"loss": 10.8675,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1694 |
"step": 12100
|
| 1695 |
},
|
| 1696 |
{
|
| 1697 |
+
"epoch": 94.0,
|
| 1698 |
"eval_dummy": 1.0,
|
| 1699 |
+
"eval_loss": 39.47597885131836,
|
| 1700 |
+
"eval_runtime": 17.7005,
|
| 1701 |
+
"eval_samples_per_second": 2.712,
|
| 1702 |
+
"eval_steps_per_second": 0.678,
|
| 1703 |
+
"step": 12126
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
+
"epoch": 94.57364341085271,
|
| 1707 |
+
"grad_norm": 3.855013132095337,
|
| 1708 |
+
"learning_rate": 2.1799999999999995e-05,
|
| 1709 |
+
"loss": 10.4757,
|
| 1710 |
"step": 12200
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
+
"epoch": 95.0,
|
| 1714 |
"eval_dummy": 1.0,
|
| 1715 |
+
"eval_loss": 40.484397888183594,
|
| 1716 |
+
"eval_runtime": 16.9799,
|
| 1717 |
+
"eval_samples_per_second": 2.827,
|
| 1718 |
+
"eval_steps_per_second": 0.707,
|
| 1719 |
+
"step": 12255
|
| 1720 |
},
|
| 1721 |
{
|
| 1722 |
+
"epoch": 95.34883720930233,
|
| 1723 |
+
"grad_norm": 7.718498229980469,
|
| 1724 |
+
"learning_rate": 1.8699999999999987e-05,
|
| 1725 |
+
"loss": 10.3191,
|
| 1726 |
"step": 12300
|
| 1727 |
},
|
| 1728 |
{
|
| 1729 |
+
"epoch": 96.0,
|
| 1730 |
"eval_dummy": 1.0,
|
| 1731 |
+
"eval_loss": 39.06733322143555,
|
| 1732 |
+
"eval_runtime": 17.7268,
|
| 1733 |
+
"eval_samples_per_second": 2.708,
|
| 1734 |
+
"eval_steps_per_second": 0.677,
|
| 1735 |
+
"step": 12384
|
| 1736 |
},
|
| 1737 |
{
|
| 1738 |
+
"epoch": 96.12403100775194,
|
| 1739 |
+
"grad_norm": 3.9156548976898193,
|
| 1740 |
+
"learning_rate": 1.5599999999999983e-05,
|
| 1741 |
+
"loss": 10.6169,
|
| 1742 |
"step": 12400
|
| 1743 |
},
|
| 1744 |
{
|
| 1745 |
+
"epoch": 96.89922480620154,
|
| 1746 |
+
"grad_norm": 3.464470148086548,
|
| 1747 |
+
"learning_rate": 1.2500000000000023e-05,
|
| 1748 |
+
"loss": 10.6073,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1749 |
"step": 12500
|
| 1750 |
},
|
| 1751 |
{
|
| 1752 |
+
"epoch": 97.0,
|
| 1753 |
"eval_dummy": 1.0,
|
| 1754 |
+
"eval_loss": 39.37672805786133,
|
| 1755 |
+
"eval_runtime": 16.9277,
|
| 1756 |
+
"eval_samples_per_second": 2.836,
|
| 1757 |
+
"eval_steps_per_second": 0.709,
|
| 1758 |
+
"step": 12513
|
| 1759 |
},
|
| 1760 |
{
|
| 1761 |
+
"epoch": 97.67441860465117,
|
| 1762 |
+
"grad_norm": 5.994311332702637,
|
| 1763 |
+
"learning_rate": 9.400000000000018e-06,
|
| 1764 |
+
"loss": 10.3038,
|
| 1765 |
"step": 12600
|
| 1766 |
},
|
| 1767 |
{
|
| 1768 |
+
"epoch": 98.0,
|
| 1769 |
"eval_dummy": 1.0,
|
| 1770 |
+
"eval_loss": 39.69685745239258,
|
| 1771 |
+
"eval_runtime": 18.0251,
|
| 1772 |
+
"eval_samples_per_second": 2.663,
|
| 1773 |
+
"eval_steps_per_second": 0.666,
|
| 1774 |
+
"step": 12642
|
| 1775 |
},
|
| 1776 |
{
|
| 1777 |
+
"epoch": 98.44961240310077,
|
| 1778 |
+
"grad_norm": 4.125715732574463,
|
| 1779 |
+
"learning_rate": 6.300000000000012e-06,
|
| 1780 |
+
"loss": 11.0709,
|
| 1781 |
"step": 12700
|
| 1782 |
},
|
| 1783 |
{
|
| 1784 |
+
"epoch": 99.0,
|
| 1785 |
"eval_dummy": 1.0,
|
| 1786 |
+
"eval_loss": 39.93254470825195,
|
| 1787 |
+
"eval_runtime": 17.9269,
|
| 1788 |
+
"eval_samples_per_second": 2.678,
|
| 1789 |
+
"eval_steps_per_second": 0.669,
|
| 1790 |
+
"step": 12771
|
| 1791 |
},
|
| 1792 |
{
|
| 1793 |
+
"epoch": 99.2248062015504,
|
| 1794 |
+
"grad_norm": 4.381137371063232,
|
| 1795 |
+
"learning_rate": 3.200000000000005e-06,
|
| 1796 |
+
"loss": 10.2398,
|
| 1797 |
"step": 12800
|
| 1798 |
},
|
| 1799 |
{
|
| 1800 |
+
"epoch": 100.0,
|
| 1801 |
+
"grad_norm": 4.022866249084473,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1802 |
"learning_rate": 9.999999999999998e-08,
|
| 1803 |
+
"loss": 10.5951,
|
| 1804 |
"step": 12900
|
| 1805 |
},
|
| 1806 |
{
|
| 1807 |
+
"epoch": 100.0,
|
| 1808 |
"eval_dummy": 1.0,
|
| 1809 |
+
"eval_loss": 39.87546157836914,
|
| 1810 |
+
"eval_runtime": 17.8005,
|
| 1811 |
+
"eval_samples_per_second": 2.697,
|
| 1812 |
+
"eval_steps_per_second": 0.674,
|
| 1813 |
"step": 12900
|
| 1814 |
},
|
| 1815 |
{
|
| 1816 |
+
"epoch": 100.0,
|
| 1817 |
"step": 12900,
|
| 1818 |
+
"total_flos": 1.4631500418239693e+19,
|
| 1819 |
+
"train_loss": 16.309644344832545,
|
| 1820 |
+
"train_runtime": 27199.2721,
|
| 1821 |
+
"train_samples_per_second": 1.897,
|
| 1822 |
+
"train_steps_per_second": 0.474
|
| 1823 |
}
|
| 1824 |
],
|
| 1825 |
"logging_steps": 100,
|
| 1826 |
"max_steps": 12900,
|
| 1827 |
"num_input_tokens_seen": 0,
|
| 1828 |
+
"num_train_epochs": 100,
|
| 1829 |
"save_steps": 500,
|
| 1830 |
"stateful_callbacks": {
|
| 1831 |
"TrainerControl": {
|
|
|
|
| 1839 |
"attributes": {}
|
| 1840 |
}
|
| 1841 |
},
|
| 1842 |
+
"total_flos": 1.4631500418239693e+19,
|
| 1843 |
+
"train_batch_size": 4,
|
| 1844 |
"trial_name": null,
|
| 1845 |
"trial_params": null
|
| 1846 |
}
|