Upload folder using huggingface_hub
Browse files- adapter_config.json +1 -1
- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +952 -608
- training_args.bin +1 -1
adapter_config.json
CHANGED
|
@@ -25,9 +25,9 @@
|
|
| 25 |
"rank_pattern": {},
|
| 26 |
"revision": null,
|
| 27 |
"target_modules": [
|
| 28 |
-
"k",
|
| 29 |
"q",
|
| 30 |
"v",
|
|
|
|
| 31 |
"o"
|
| 32 |
],
|
| 33 |
"target_parameters": null,
|
|
|
|
| 25 |
"rank_pattern": {},
|
| 26 |
"revision": null,
|
| 27 |
"target_modules": [
|
|
|
|
| 28 |
"q",
|
| 29 |
"v",
|
| 30 |
+
"k",
|
| 31 |
"o"
|
| 32 |
],
|
| 33 |
"target_parameters": null,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151086440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7467c239bc34c2324f3560e13264327d4bd97b580e627e4ab0a4731660b1230f
|
| 3 |
size 151086440
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 302477347
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53cd4d9fc3658485a8cfa00b7320dca6049c62c03fa0383662ac975d9981a327
|
| 3 |
size 302477347
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d2726325cf5cde50c6459d972222470bf1aea5454617340c20d7d559be948f8
|
| 3 |
size 14645
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f100df5eb82d022e2a5fb55dbdd699ce767bec963c9ef9b797c33c1ac74583f6
|
| 3 |
size 1465
|
trainer_state.json
CHANGED
|
@@ -1,1396 +1,1740 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": null,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.0002476166893648632,
|
| 14 |
-
"grad_norm": 0.
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
"loss": 3.0743,
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02476166893648632,
|
| 21 |
-
"grad_norm": 0.
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 2.
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.04952333787297264,
|
| 28 |
-
"grad_norm": 0.
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 200
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.07428500680945896,
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
"step": 300
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.09904667574594528,
|
| 42 |
-
"grad_norm": 1.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 400
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.1238083446824316,
|
| 49 |
-
"grad_norm": 1.
|
| 50 |
-
"learning_rate": 4.
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 500
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1238083446824316,
|
| 56 |
-
"eval_loss": 0.
|
| 57 |
-
"eval_runtime":
|
| 58 |
-
"eval_samples_per_second":
|
| 59 |
-
"eval_steps_per_second":
|
| 60 |
"step": 500
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 0.14857001361891792,
|
| 64 |
-
"grad_norm": 1.
|
| 65 |
-
"learning_rate": 4.
|
| 66 |
-
"loss": 0.
|
| 67 |
"step": 600
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 0.17333168255540424,
|
| 71 |
-
"grad_norm": 1.
|
| 72 |
-
"learning_rate": 4.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 700
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.19809335149189056,
|
| 78 |
-
"grad_norm": 1.
|
| 79 |
-
"learning_rate": 4.
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 800
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.22285502042837688,
|
| 85 |
-
"grad_norm": 1.
|
| 86 |
-
"learning_rate": 4.
|
| 87 |
-
"loss": 0.
|
| 88 |
"step": 900
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.2476166893648632,
|
| 92 |
-
"grad_norm": 1.
|
| 93 |
-
"learning_rate": 4.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 1000
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.2476166893648632,
|
| 99 |
-
"eval_loss": 0.
|
| 100 |
-
"eval_runtime":
|
| 101 |
-
"eval_samples_per_second":
|
| 102 |
-
"eval_steps_per_second":
|
| 103 |
"step": 1000
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 0.2723783583013495,
|
| 107 |
-
"grad_norm": 1.
|
| 108 |
-
"learning_rate": 4.
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 1100
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.29714002723783584,
|
| 114 |
-
"grad_norm": 1.
|
| 115 |
-
"learning_rate": 4.
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 1200
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 0.3219016961743221,
|
| 121 |
-
"grad_norm": 1.
|
| 122 |
-
"learning_rate": 4.
|
| 123 |
-
"loss": 0.
|
| 124 |
"step": 1300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 0.3466633651108085,
|
| 128 |
-
"grad_norm": 1.
|
| 129 |
-
"learning_rate": 4.
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 1400
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 0.37142503404729477,
|
| 135 |
-
"grad_norm": 1.
|
| 136 |
-
"learning_rate": 4.
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 1500
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 0.37142503404729477,
|
| 142 |
-
"eval_loss": 0.
|
| 143 |
-
"eval_runtime":
|
| 144 |
-
"eval_samples_per_second":
|
| 145 |
-
"eval_steps_per_second":
|
| 146 |
"step": 1500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.3961867029837811,
|
| 150 |
-
"grad_norm": 1.
|
| 151 |
-
"learning_rate": 4.
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 1600
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.4209483719202674,
|
| 157 |
-
"grad_norm": 1.
|
| 158 |
-
"learning_rate": 4.
|
| 159 |
-
"loss": 0.
|
| 160 |
"step": 1700
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.44571004085675375,
|
| 164 |
-
"grad_norm": 1.
|
| 165 |
-
"learning_rate": 4.
|
| 166 |
-
"loss": 0.
|
| 167 |
"step": 1800
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.47047170979324004,
|
| 171 |
-
"grad_norm": 1.
|
| 172 |
-
"learning_rate": 4.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 1900
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.4952333787297264,
|
| 178 |
-
"grad_norm": 1.
|
| 179 |
-
"learning_rate": 4.
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 2000
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.4952333787297264,
|
| 185 |
-
"eval_loss": 0.
|
| 186 |
-
"eval_runtime":
|
| 187 |
-
"eval_samples_per_second":
|
| 188 |
-
"eval_steps_per_second":
|
| 189 |
"step": 2000
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 0.5199950476662127,
|
| 193 |
-
"grad_norm": 1.
|
| 194 |
-
"learning_rate": 4.
|
| 195 |
-
"loss": 0.
|
| 196 |
"step": 2100
|
| 197 |
},
|
| 198 |
{
|
| 199 |
"epoch": 0.544756716602699,
|
| 200 |
-
"grad_norm": 1.
|
| 201 |
-
"learning_rate": 4.
|
| 202 |
-
"loss": 0.
|
| 203 |
"step": 2200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.5695183855391853,
|
| 207 |
-
"grad_norm": 1.
|
| 208 |
-
"learning_rate": 4.
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 2300
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 0.5942800544756717,
|
| 214 |
-
"grad_norm": 1.
|
| 215 |
-
"learning_rate": 4.
|
| 216 |
-
"loss": 0.
|
| 217 |
"step": 2400
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 0.619041723412158,
|
| 221 |
-
"grad_norm": 1.
|
| 222 |
-
"learning_rate": 4.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 2500
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.619041723412158,
|
| 228 |
-
"eval_loss": 0.
|
| 229 |
-
"eval_runtime":
|
| 230 |
-
"eval_samples_per_second":
|
| 231 |
-
"eval_steps_per_second": 3.
|
| 232 |
"step": 2500
|
| 233 |
},
|
| 234 |
{
|
| 235 |
"epoch": 0.6438033923486443,
|
| 236 |
-
"grad_norm": 1.
|
| 237 |
-
"learning_rate": 4.
|
| 238 |
-
"loss": 0.
|
| 239 |
"step": 2600
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.6685650612851306,
|
| 243 |
-
"grad_norm": 1.
|
| 244 |
-
"learning_rate": 4.
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 2700
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 0.693326730221617,
|
| 250 |
-
"grad_norm": 1.
|
| 251 |
-
"learning_rate": 4.
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 2800
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 0.7180883991581033,
|
| 257 |
-
"grad_norm": 1.
|
| 258 |
-
"learning_rate": 4.
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 2900
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 0.7428500680945895,
|
| 264 |
-
"grad_norm": 1.
|
| 265 |
-
"learning_rate": 4.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 3000
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 0.7428500680945895,
|
| 271 |
-
"eval_loss": 0.
|
| 272 |
-
"eval_runtime":
|
| 273 |
-
"eval_samples_per_second":
|
| 274 |
-
"eval_steps_per_second":
|
| 275 |
"step": 3000
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7676117370310759,
|
| 279 |
-
"grad_norm": 1.
|
| 280 |
-
"learning_rate": 4.
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 3100
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7923734059675622,
|
| 286 |
-
"grad_norm": 1.
|
| 287 |
-
"learning_rate": 4.
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 3200
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.8171350749040486,
|
| 293 |
-
"grad_norm": 1.
|
| 294 |
-
"learning_rate": 4.
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 3300
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.8418967438405348,
|
| 300 |
-
"grad_norm": 1.
|
| 301 |
-
"learning_rate": 4.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 3400
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.8666584127770212,
|
| 307 |
-
"grad_norm": 1.
|
| 308 |
-
"learning_rate": 4.
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 3500
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.8666584127770212,
|
| 314 |
-
"eval_loss": 0.
|
| 315 |
-
"eval_runtime":
|
| 316 |
-
"eval_samples_per_second":
|
| 317 |
-
"eval_steps_per_second":
|
| 318 |
"step": 3500
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.8914200817135075,
|
| 322 |
-
"grad_norm": 1.
|
| 323 |
-
"learning_rate": 4.
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 3600
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.9161817506499939,
|
| 329 |
-
"grad_norm": 1.
|
| 330 |
-
"learning_rate": 4.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 3700
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.9409434195864801,
|
| 336 |
-
"grad_norm": 1.
|
| 337 |
-
"learning_rate": 4.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 3800
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.9657050885229664,
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate": 4.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 3900
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.9904667574594528,
|
| 350 |
-
"grad_norm": 1.
|
| 351 |
-
"learning_rate": 4.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 4000
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.9904667574594528,
|
| 357 |
-
"eval_loss": 0.
|
| 358 |
-
"eval_runtime":
|
| 359 |
-
"eval_samples_per_second":
|
| 360 |
-
"eval_steps_per_second":
|
| 361 |
"step": 4000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 1.0151046180512566,
|
| 365 |
-
"grad_norm": 1.
|
| 366 |
-
"learning_rate": 4.
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 4100
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 1.0398662869877429,
|
| 372 |
-
"grad_norm": 1.
|
| 373 |
-
"learning_rate": 4.
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 4200
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 1.0646279559242293,
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate":
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 4300
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 1.0893896248607156,
|
| 386 |
-
"grad_norm": 1.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 4400
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 1.114151293797202,
|
| 393 |
-
"grad_norm":
|
| 394 |
-
"learning_rate": 3.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 4500
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 1.114151293797202,
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime":
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second":
|
| 404 |
"step": 4500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 1.1389129627336882,
|
| 408 |
-
"grad_norm": 0.
|
| 409 |
-
"learning_rate": 3.
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 4600
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 1.1636746316701745,
|
| 415 |
-
"grad_norm": 1.
|
| 416 |
-
"learning_rate": 3.
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 4700
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 1.188436300606661,
|
| 422 |
-
"grad_norm":
|
| 423 |
-
"learning_rate": 3.
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 4800
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.2131979695431472,
|
| 429 |
-
"grad_norm": 1.
|
| 430 |
-
"learning_rate": 3.
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 4900
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 1.2379596384796336,
|
| 436 |
-
"grad_norm":
|
| 437 |
-
"learning_rate": 3.
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 5000
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 1.2379596384796336,
|
| 443 |
-
"eval_loss": 0.
|
| 444 |
-
"eval_runtime":
|
| 445 |
-
"eval_samples_per_second":
|
| 446 |
-
"eval_steps_per_second":
|
| 447 |
"step": 5000
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 1.2627213074161199,
|
| 451 |
-
"grad_norm": 1.
|
| 452 |
-
"learning_rate": 3.
|
| 453 |
-
"loss": 0.
|
| 454 |
"step": 5100
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 1.2874829763526061,
|
| 458 |
-
"grad_norm": 1.
|
| 459 |
-
"learning_rate": 3.
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 5200
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 1.3122446452890926,
|
| 465 |
-
"grad_norm": 1.
|
| 466 |
-
"learning_rate": 3.
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 5300
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 1.3370063142255788,
|
| 472 |
-
"grad_norm": 1.
|
| 473 |
-
"learning_rate": 3.
|
| 474 |
-
"loss": 0.
|
| 475 |
"step": 5400
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.361767983162065,
|
| 479 |
-
"grad_norm": 0.
|
| 480 |
-
"learning_rate": 3.
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 5500
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.361767983162065,
|
| 486 |
-
"eval_loss": 0.
|
| 487 |
-
"eval_runtime":
|
| 488 |
-
"eval_samples_per_second":
|
| 489 |
-
"eval_steps_per_second":
|
| 490 |
"step": 5500
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.3865296520985515,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate": 3.
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 5600
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.4112913210350377,
|
| 501 |
-
"grad_norm": 1.
|
| 502 |
-
"learning_rate": 3.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 5700
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 1.436052989971524,
|
| 508 |
-
"grad_norm": 1.
|
| 509 |
-
"learning_rate": 3.
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 5800
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 1.4608146589080104,
|
| 515 |
-
"grad_norm": 1.
|
| 516 |
-
"learning_rate": 3.
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 5900
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 1.4855763278444967,
|
| 522 |
-
"grad_norm": 1.
|
| 523 |
-
"learning_rate": 3.
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 6000
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 1.4855763278444967,
|
| 529 |
-
"eval_loss": 0.
|
| 530 |
-
"eval_runtime":
|
| 531 |
-
"eval_samples_per_second":
|
| 532 |
-
"eval_steps_per_second":
|
| 533 |
"step": 6000
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 1.510337996780983,
|
| 537 |
-
"grad_norm": 1.
|
| 538 |
-
"learning_rate": 3.
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 6100
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 1.5350996657174694,
|
| 544 |
-
"grad_norm": 0.
|
| 545 |
-
"learning_rate": 3.
|
| 546 |
"loss": 0.4584,
|
| 547 |
"step": 6200
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 1.5598613346539558,
|
| 551 |
-
"grad_norm": 1.
|
| 552 |
-
"learning_rate": 3.
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 6300
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 1.5846230035904418,
|
| 558 |
-
"grad_norm": 0.
|
| 559 |
-
"learning_rate": 3.
|
| 560 |
-
"loss": 0.
|
| 561 |
"step": 6400
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 1.6093846725269283,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
-
"learning_rate": 3.
|
| 567 |
-
"loss": 0.
|
| 568 |
"step": 6500
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 1.6093846725269283,
|
| 572 |
-
"eval_loss": 0.
|
| 573 |
-
"eval_runtime":
|
| 574 |
-
"eval_samples_per_second":
|
| 575 |
-
"eval_steps_per_second":
|
| 576 |
"step": 6500
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 1.6341463414634148,
|
| 580 |
-
"grad_norm":
|
| 581 |
-
"learning_rate": 3.
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 6600
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 1.658908010399901,
|
| 587 |
-
"grad_norm": 1.
|
| 588 |
-
"learning_rate": 3.
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 6700
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 1.6836696793363872,
|
| 594 |
-
"grad_norm": 1.
|
| 595 |
-
"learning_rate": 3.
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 6800
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 1.7084313482728737,
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate": 3.
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 6900
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 1.73319301720936,
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate": 3.
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 7000
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 1.73319301720936,
|
| 615 |
-
"eval_loss": 0.
|
| 616 |
-
"eval_runtime":
|
| 617 |
-
"eval_samples_per_second":
|
| 618 |
-
"eval_steps_per_second":
|
| 619 |
"step": 7000
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 1.7579546861458462,
|
| 623 |
-
"grad_norm": 1.
|
| 624 |
-
"learning_rate": 3.
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 7100
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 1.7827163550823326,
|
| 630 |
-
"grad_norm": 1.
|
| 631 |
-
"learning_rate": 3.
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 7200
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 1.8074780240188189,
|
| 637 |
-
"grad_norm": 0.
|
| 638 |
-
"learning_rate": 3.
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 7300
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 1.832239692955305,
|
| 644 |
-
"grad_norm": 0.
|
| 645 |
-
"learning_rate": 3.
|
| 646 |
-
"loss": 0.
|
| 647 |
"step": 7400
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 1.8570013618917915,
|
| 651 |
-
"grad_norm": 1.
|
| 652 |
-
"learning_rate": 3.
|
| 653 |
-
"loss": 0.
|
| 654 |
"step": 7500
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 1.8570013618917915,
|
| 658 |
-
"eval_loss": 0.
|
| 659 |
-
"eval_runtime":
|
| 660 |
-
"eval_samples_per_second":
|
| 661 |
-
"eval_steps_per_second": 3.
|
| 662 |
"step": 7500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 1.8817630308282778,
|
| 666 |
-
"grad_norm": 0.
|
| 667 |
-
"learning_rate": 3.
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 7600
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 1.906524699764764,
|
| 673 |
-
"grad_norm": 1.
|
| 674 |
-
"learning_rate": 3.
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 7700
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 1.9312863687012505,
|
| 680 |
-
"grad_norm": 0.
|
| 681 |
-
"learning_rate": 3.
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 7800
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 1.956048037637737,
|
| 687 |
-
"grad_norm": 1.
|
| 688 |
-
"learning_rate": 3.
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 7900
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 1.980809706574223,
|
| 694 |
-
"grad_norm":
|
| 695 |
-
"learning_rate": 3.
|
| 696 |
-
"loss": 0.
|
| 697 |
"step": 8000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 1.980809706574223,
|
| 701 |
-
"eval_loss": 0.
|
| 702 |
-
"eval_runtime":
|
| 703 |
-
"eval_samples_per_second":
|
| 704 |
-
"eval_steps_per_second": 3.
|
| 705 |
"step": 8000
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 2.005447567166027,
|
| 709 |
-
"grad_norm": 1.
|
| 710 |
-
"learning_rate": 3.
|
| 711 |
-
"loss": 0.
|
| 712 |
"step": 8100
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 2.0302092361025132,
|
| 716 |
-
"grad_norm": 1.
|
| 717 |
-
"learning_rate": 3.
|
| 718 |
-
"loss": 0.
|
| 719 |
"step": 8200
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 2.0549709050389997,
|
| 723 |
-
"grad_norm": 1.
|
| 724 |
-
"learning_rate":
|
| 725 |
-
"loss": 0.
|
| 726 |
"step": 8300
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 2.0797325739754857,
|
| 730 |
-
"grad_norm": 0.
|
| 731 |
-
"learning_rate": 2.
|
| 732 |
-
"loss": 0.
|
| 733 |
"step": 8400
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 2.104494242911972,
|
| 737 |
-
"grad_norm": 1.
|
| 738 |
-
"learning_rate": 2.
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 8500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 2.104494242911972,
|
| 744 |
-
"eval_loss": 0.
|
| 745 |
-
"eval_runtime":
|
| 746 |
-
"eval_samples_per_second":
|
| 747 |
-
"eval_steps_per_second": 3.
|
| 748 |
"step": 8500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 2.1292559118484586,
|
| 752 |
-
"grad_norm": 0.
|
| 753 |
-
"learning_rate": 2.
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 8600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 2.154017580784945,
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate": 2.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 8700
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 2.178779249721431,
|
| 766 |
-
"grad_norm": 1.
|
| 767 |
-
"learning_rate": 2.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 8800
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 2.2035409186579176,
|
| 773 |
-
"grad_norm": 1.
|
| 774 |
-
"learning_rate": 2.
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 8900
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 2.228302587594404,
|
| 780 |
-
"grad_norm": 1.
|
| 781 |
-
"learning_rate": 2.
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 9000
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 2.228302587594404,
|
| 787 |
-
"eval_loss": 0.
|
| 788 |
-
"eval_runtime":
|
| 789 |
-
"eval_samples_per_second":
|
| 790 |
-
"eval_steps_per_second": 3.
|
| 791 |
"step": 9000
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 2.25306425653089,
|
| 795 |
-
"grad_norm": 0.
|
| 796 |
-
"learning_rate": 2.
|
| 797 |
-
"loss": 0.
|
| 798 |
"step": 9100
|
| 799 |
},
|
| 800 |
{
|
| 801 |
"epoch": 2.2778259254673765,
|
| 802 |
-
"grad_norm": 1.
|
| 803 |
-
"learning_rate": 2.
|
| 804 |
-
"loss": 0.
|
| 805 |
"step": 9200
|
| 806 |
},
|
| 807 |
{
|
| 808 |
"epoch": 2.302587594403863,
|
| 809 |
-
"grad_norm": 0.
|
| 810 |
-
"learning_rate": 2.
|
| 811 |
-
"loss": 0.
|
| 812 |
"step": 9300
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 2.327349263340349,
|
| 816 |
-
"grad_norm": 1.
|
| 817 |
-
"learning_rate": 2.
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 9400
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 2.3521109322768354,
|
| 823 |
-
"grad_norm": 1.
|
| 824 |
-
"learning_rate": 2.
|
| 825 |
-
"loss": 0.
|
| 826 |
"step": 9500
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 2.3521109322768354,
|
| 830 |
-
"eval_loss": 0.
|
| 831 |
-
"eval_runtime":
|
| 832 |
-
"eval_samples_per_second":
|
| 833 |
-
"eval_steps_per_second": 3.
|
| 834 |
"step": 9500
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 2.376872601213322,
|
| 838 |
-
"grad_norm": 1.
|
| 839 |
-
"learning_rate": 2.
|
| 840 |
-
"loss": 0.
|
| 841 |
"step": 9600
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 2.401634270149808,
|
| 845 |
-
"grad_norm": 1.
|
| 846 |
-
"learning_rate": 2.
|
| 847 |
-
"loss": 0.
|
| 848 |
"step": 9700
|
| 849 |
},
|
| 850 |
{
|
| 851 |
"epoch": 2.4263959390862944,
|
| 852 |
-
"grad_norm":
|
| 853 |
-
"learning_rate": 2.
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 9800
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 2.451157608022781,
|
| 859 |
-
"grad_norm": 0.
|
| 860 |
-
"learning_rate": 2.
|
| 861 |
-
"loss": 0.
|
| 862 |
"step": 9900
|
| 863 |
},
|
| 864 |
{
|
| 865 |
"epoch": 2.4759192769592673,
|
| 866 |
-
"grad_norm": 1.
|
| 867 |
-
"learning_rate": 2.
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 10000
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 2.4759192769592673,
|
| 873 |
-
"eval_loss": 0.
|
| 874 |
-
"eval_runtime":
|
| 875 |
-
"eval_samples_per_second":
|
| 876 |
-
"eval_steps_per_second": 3.
|
| 877 |
"step": 10000
|
| 878 |
},
|
| 879 |
{
|
| 880 |
"epoch": 2.5006809458957533,
|
| 881 |
-
"grad_norm": 1.
|
| 882 |
-
"learning_rate": 2.
|
| 883 |
-
"loss": 0.
|
| 884 |
"step": 10100
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 2.5254426148322398,
|
| 888 |
-
"grad_norm": 1.
|
| 889 |
-
"learning_rate": 2.
|
| 890 |
-
"loss": 0.
|
| 891 |
"step": 10200
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 2.5502042837687258,
|
| 895 |
-
"grad_norm": 1.
|
| 896 |
-
"learning_rate": 2.
|
| 897 |
-
"loss": 0.
|
| 898 |
"step": 10300
|
| 899 |
},
|
| 900 |
{
|
| 901 |
"epoch": 2.5749659527052122,
|
| 902 |
-
"grad_norm": 1.
|
| 903 |
-
"learning_rate": 2.
|
| 904 |
-
"loss": 0.
|
| 905 |
"step": 10400
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 2.5997276216416987,
|
| 909 |
-
"grad_norm": 1.
|
| 910 |
-
"learning_rate": 2.
|
| 911 |
-
"loss": 0.
|
| 912 |
"step": 10500
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 2.5997276216416987,
|
| 916 |
-
"eval_loss": 0.
|
| 917 |
-
"eval_runtime":
|
| 918 |
-
"eval_samples_per_second":
|
| 919 |
-
"eval_steps_per_second": 3.
|
| 920 |
"step": 10500
|
| 921 |
},
|
| 922 |
{
|
| 923 |
"epoch": 2.624489290578185,
|
| 924 |
-
"grad_norm": 1.
|
| 925 |
-
"learning_rate": 2.
|
| 926 |
-
"loss": 0.
|
| 927 |
"step": 10600
|
| 928 |
},
|
| 929 |
{
|
| 930 |
"epoch": 2.649250959514671,
|
| 931 |
-
"grad_norm":
|
| 932 |
-
"learning_rate": 2.
|
| 933 |
-
"loss": 0.
|
| 934 |
"step": 10700
|
| 935 |
},
|
| 936 |
{
|
| 937 |
"epoch": 2.6740126284511576,
|
| 938 |
-
"grad_norm": 0.
|
| 939 |
-
"learning_rate": 2.
|
| 940 |
"loss": 0.4321,
|
| 941 |
"step": 10800
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 2.698774297387644,
|
| 945 |
-
"grad_norm": 1.
|
| 946 |
-
"learning_rate": 2.
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 10900
|
| 949 |
},
|
| 950 |
{
|
| 951 |
"epoch": 2.72353596632413,
|
| 952 |
-
"grad_norm": 0.
|
| 953 |
-
"learning_rate": 2.
|
| 954 |
-
"loss": 0.
|
| 955 |
"step": 11000
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 2.72353596632413,
|
| 959 |
-
"eval_loss": 0.
|
| 960 |
-
"eval_runtime":
|
| 961 |
-
"eval_samples_per_second":
|
| 962 |
-
"eval_steps_per_second": 3.
|
| 963 |
"step": 11000
|
| 964 |
},
|
| 965 |
{
|
| 966 |
"epoch": 2.7482976352606165,
|
| 967 |
-
"grad_norm": 1.
|
| 968 |
-
"learning_rate": 2.
|
| 969 |
"loss": 0.4253,
|
| 970 |
"step": 11100
|
| 971 |
},
|
| 972 |
{
|
| 973 |
"epoch": 2.773059304197103,
|
| 974 |
-
"grad_norm": 1.
|
| 975 |
-
"learning_rate": 2.
|
| 976 |
-
"loss": 0.
|
| 977 |
"step": 11200
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 2.7978209731335895,
|
| 981 |
-
"grad_norm": 1.
|
| 982 |
-
"learning_rate": 2.
|
| 983 |
-
"loss": 0.
|
| 984 |
"step": 11300
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 2.8225826420700755,
|
| 988 |
-
"grad_norm": 1.
|
| 989 |
-
"learning_rate": 2.
|
| 990 |
-
"loss": 0.
|
| 991 |
"step": 11400
|
| 992 |
},
|
| 993 |
{
|
| 994 |
"epoch": 2.847344311006562,
|
| 995 |
-
"grad_norm": 1.
|
| 996 |
-
"learning_rate": 2.
|
| 997 |
-
"loss": 0.
|
| 998 |
"step": 11500
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 2.847344311006562,
|
| 1002 |
-
"eval_loss": 0.
|
| 1003 |
-
"eval_runtime":
|
| 1004 |
-
"eval_samples_per_second":
|
| 1005 |
-
"eval_steps_per_second": 3.
|
| 1006 |
"step": 11500
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
"epoch": 2.872105979943048,
|
| 1010 |
-
"grad_norm": 1.
|
| 1011 |
-
"learning_rate": 2.
|
| 1012 |
-
"loss": 0.
|
| 1013 |
"step": 11600
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 2.8968676488795344,
|
| 1017 |
-
"grad_norm":
|
| 1018 |
-
"learning_rate": 2.
|
| 1019 |
-
"loss": 0.
|
| 1020 |
"step": 11700
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
"epoch": 2.921629317816021,
|
| 1024 |
-
"grad_norm": 0.
|
| 1025 |
-
"learning_rate": 2.
|
| 1026 |
-
"loss": 0.
|
| 1027 |
"step": 11800
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 2.9463909867525073,
|
| 1031 |
-
"grad_norm": 1.
|
| 1032 |
-
"learning_rate": 2.
|
| 1033 |
-
"loss": 0.
|
| 1034 |
"step": 11900
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
"epoch": 2.9711526556889933,
|
| 1038 |
-
"grad_norm": 0.
|
| 1039 |
-
"learning_rate": 2.
|
| 1040 |
-
"loss": 0.
|
| 1041 |
"step": 12000
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 2.9711526556889933,
|
| 1045 |
-
"eval_loss": 0.
|
| 1046 |
-
"eval_runtime":
|
| 1047 |
-
"eval_samples_per_second":
|
| 1048 |
-
"eval_steps_per_second":
|
| 1049 |
"step": 12000
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 2.99591432462548,
|
| 1053 |
-
"grad_norm": 1.
|
| 1054 |
-
"learning_rate": 2.
|
| 1055 |
-
"loss": 0.
|
| 1056 |
"step": 12100
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 3.0205521852172836,
|
| 1060 |
-
"grad_norm":
|
| 1061 |
-
"learning_rate": 2.
|
| 1062 |
-
"loss": 0.
|
| 1063 |
"step": 12200
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 3.04531385415377,
|
| 1067 |
-
"grad_norm": 0.
|
| 1068 |
-
"learning_rate":
|
| 1069 |
-
"loss": 0.
|
| 1070 |
"step": 12300
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
"epoch": 3.070075523090256,
|
| 1074 |
-
"grad_norm": 0.
|
| 1075 |
-
"learning_rate": 1.
|
| 1076 |
-
"loss": 0.
|
| 1077 |
"step": 12400
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
"epoch": 3.0948371920267426,
|
| 1081 |
-
"grad_norm": 1.
|
| 1082 |
-
"learning_rate": 1.
|
| 1083 |
-
"loss": 0.
|
| 1084 |
"step": 12500
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 3.0948371920267426,
|
| 1088 |
-
"eval_loss": 0.
|
| 1089 |
-
"eval_runtime":
|
| 1090 |
-
"eval_samples_per_second":
|
| 1091 |
-
"eval_steps_per_second":
|
| 1092 |
"step": 12500
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 3.119598860963229,
|
| 1096 |
-
"grad_norm":
|
| 1097 |
-
"learning_rate": 1.
|
| 1098 |
-
"loss": 0.
|
| 1099 |
"step": 12600
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 3.144360529899715,
|
| 1103 |
-
"grad_norm": 1.
|
| 1104 |
-
"learning_rate": 1.
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 12700
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 3.1691221988362015,
|
| 1110 |
-
"grad_norm": 1.
|
| 1111 |
-
"learning_rate": 1.
|
| 1112 |
-
"loss": 0.
|
| 1113 |
"step": 12800
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 3.193883867772688,
|
| 1117 |
-
"grad_norm": 1.
|
| 1118 |
-
"learning_rate": 1.
|
| 1119 |
-
"loss": 0.
|
| 1120 |
"step": 12900
|
| 1121 |
},
|
| 1122 |
{
|
| 1123 |
"epoch": 3.2186455367091744,
|
| 1124 |
-
"grad_norm": 1.
|
| 1125 |
-
"learning_rate": 1.
|
| 1126 |
-
"loss": 0.
|
| 1127 |
"step": 13000
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
"epoch": 3.2186455367091744,
|
| 1131 |
-
"eval_loss": 0.
|
| 1132 |
-
"eval_runtime":
|
| 1133 |
-
"eval_samples_per_second":
|
| 1134 |
-
"eval_steps_per_second":
|
| 1135 |
"step": 13000
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
"epoch": 3.2434072056456604,
|
| 1139 |
-
"grad_norm": 1.
|
| 1140 |
-
"learning_rate": 1.
|
| 1141 |
-
"loss": 0.
|
| 1142 |
"step": 13100
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
"epoch": 3.268168874582147,
|
| 1146 |
-
"grad_norm": 0.
|
| 1147 |
-
"learning_rate": 1.
|
| 1148 |
-
"loss": 0.
|
| 1149 |
"step": 13200
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
"epoch": 3.2929305435186333,
|
| 1153 |
-
"grad_norm": 0.
|
| 1154 |
-
"learning_rate": 1.
|
| 1155 |
-
"loss": 0.
|
| 1156 |
"step": 13300
|
| 1157 |
},
|
| 1158 |
{
|
| 1159 |
"epoch": 3.3176922124551194,
|
| 1160 |
-
"grad_norm": 1.
|
| 1161 |
-
"learning_rate": 1.
|
| 1162 |
-
"loss": 0.
|
| 1163 |
"step": 13400
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
"epoch": 3.342453881391606,
|
| 1167 |
-
"grad_norm": 0.
|
| 1168 |
-
"learning_rate": 1.
|
| 1169 |
"loss": 0.4111,
|
| 1170 |
"step": 13500
|
| 1171 |
},
|
| 1172 |
{
|
| 1173 |
"epoch": 3.342453881391606,
|
| 1174 |
-
"eval_loss": 0.
|
| 1175 |
-
"eval_runtime":
|
| 1176 |
-
"eval_samples_per_second":
|
| 1177 |
-
"eval_steps_per_second":
|
| 1178 |
"step": 13500
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 3.3672155503280923,
|
| 1182 |
-
"grad_norm": 0.
|
| 1183 |
-
"learning_rate": 1.
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 13600
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 3.3919772192645783,
|
| 1189 |
-
"grad_norm": 1.
|
| 1190 |
-
"learning_rate": 1.
|
| 1191 |
-
"loss": 0.
|
| 1192 |
"step": 13700
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 3.4167388882010647,
|
| 1196 |
-
"grad_norm": 0.
|
| 1197 |
-
"learning_rate": 1.
|
| 1198 |
-
"loss": 0.
|
| 1199 |
"step": 13800
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 3.441500557137551,
|
| 1203 |
-
"grad_norm": 1.
|
| 1204 |
-
"learning_rate": 1.
|
| 1205 |
-
"loss": 0.
|
| 1206 |
"step": 13900
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 3.4662622260740372,
|
| 1210 |
-
"grad_norm": 1.
|
| 1211 |
-
"learning_rate": 1.
|
| 1212 |
-
"loss": 0.
|
| 1213 |
"step": 14000
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 3.4662622260740372,
|
| 1217 |
-
"eval_loss": 0.
|
| 1218 |
-
"eval_runtime":
|
| 1219 |
-
"eval_samples_per_second":
|
| 1220 |
-
"eval_steps_per_second":
|
| 1221 |
"step": 14000
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
"epoch": 3.4910238950105237,
|
| 1225 |
-
"grad_norm": 0.
|
| 1226 |
-
"learning_rate": 1.
|
| 1227 |
-
"loss": 0.
|
| 1228 |
"step": 14100
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
"epoch": 3.51578556394701,
|
| 1232 |
-
"grad_norm": 1.
|
| 1233 |
-
"learning_rate": 1.
|
| 1234 |
-
"loss": 0.
|
| 1235 |
"step": 14200
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 3.5405472328834966,
|
| 1239 |
-
"grad_norm":
|
| 1240 |
-
"learning_rate": 1.
|
| 1241 |
-
"loss": 0.
|
| 1242 |
"step": 14300
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
"epoch": 3.5653089018199826,
|
| 1246 |
-
"grad_norm": 1.
|
| 1247 |
-
"learning_rate": 1.
|
| 1248 |
-
"loss": 0.
|
| 1249 |
"step": 14400
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
"epoch": 3.590070570756469,
|
| 1253 |
-
"grad_norm": 1.
|
| 1254 |
-
"learning_rate": 1.
|
| 1255 |
-
"loss": 0.
|
| 1256 |
"step": 14500
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 3.590070570756469,
|
| 1260 |
-
"eval_loss": 0.
|
| 1261 |
-
"eval_runtime":
|
| 1262 |
-
"eval_samples_per_second":
|
| 1263 |
-
"eval_steps_per_second":
|
| 1264 |
"step": 14500
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
"epoch": 3.614832239692955,
|
| 1268 |
-
"grad_norm":
|
| 1269 |
-
"learning_rate": 1.
|
| 1270 |
-
"loss": 0.
|
| 1271 |
"step": 14600
|
| 1272 |
},
|
| 1273 |
{
|
| 1274 |
"epoch": 3.6395939086294415,
|
| 1275 |
-
"grad_norm": 1.
|
| 1276 |
-
"learning_rate": 1.
|
| 1277 |
-
"loss": 0.
|
| 1278 |
"step": 14700
|
| 1279 |
},
|
| 1280 |
{
|
| 1281 |
"epoch": 3.664355577565928,
|
| 1282 |
-
"grad_norm": 0.
|
| 1283 |
-
"learning_rate": 1.
|
| 1284 |
-
"loss": 0.
|
| 1285 |
"step": 14800
|
| 1286 |
},
|
| 1287 |
{
|
| 1288 |
"epoch": 3.6891172465024145,
|
| 1289 |
-
"grad_norm": 0.
|
| 1290 |
-
"learning_rate": 1.
|
| 1291 |
-
"loss": 0.
|
| 1292 |
"step": 14900
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
"epoch": 3.7138789154389005,
|
| 1296 |
-
"grad_norm": 1.
|
| 1297 |
-
"learning_rate": 1.
|
| 1298 |
-
"loss": 0.
|
| 1299 |
"step": 15000
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 3.7138789154389005,
|
| 1303 |
-
"eval_loss": 0.
|
| 1304 |
-
"eval_runtime":
|
| 1305 |
-
"eval_samples_per_second":
|
| 1306 |
-
"eval_steps_per_second":
|
| 1307 |
"step": 15000
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
"epoch": 3.738640584375387,
|
| 1311 |
-
"grad_norm": 1.
|
| 1312 |
-
"learning_rate": 1.
|
| 1313 |
-
"loss": 0.
|
| 1314 |
"step": 15100
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
"epoch": 3.763402253311873,
|
| 1318 |
-
"grad_norm": 1.
|
| 1319 |
-
"learning_rate": 1.
|
| 1320 |
-
"loss": 0.
|
| 1321 |
"step": 15200
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 3.7881639222483594,
|
| 1325 |
-
"grad_norm": 1.
|
| 1326 |
-
"learning_rate": 1.
|
| 1327 |
-
"loss": 0.
|
| 1328 |
"step": 15300
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
"epoch": 3.812925591184846,
|
| 1332 |
-
"grad_norm": 1.
|
| 1333 |
-
"learning_rate": 1.
|
| 1334 |
-
"loss": 0.
|
| 1335 |
"step": 15400
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
"epoch": 3.8376872601213323,
|
| 1339 |
-
"grad_norm": 1.
|
| 1340 |
-
"learning_rate": 1.
|
| 1341 |
-
"loss": 0.
|
| 1342 |
"step": 15500
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 3.8376872601213323,
|
| 1346 |
-
"eval_loss": 0.
|
| 1347 |
-
"eval_runtime":
|
| 1348 |
-
"eval_samples_per_second":
|
| 1349 |
-
"eval_steps_per_second":
|
| 1350 |
"step": 15500
|
| 1351 |
},
|
| 1352 |
{
|
| 1353 |
"epoch": 3.8624489290578183,
|
| 1354 |
-
"grad_norm": 0.
|
| 1355 |
-
"learning_rate": 1.
|
| 1356 |
-
"loss": 0.
|
| 1357 |
"step": 15600
|
| 1358 |
},
|
| 1359 |
{
|
| 1360 |
"epoch": 3.887210597994305,
|
| 1361 |
-
"grad_norm": 1.
|
| 1362 |
-
"learning_rate": 1.
|
| 1363 |
-
"loss": 0.
|
| 1364 |
"step": 15700
|
| 1365 |
},
|
| 1366 |
{
|
| 1367 |
"epoch": 3.9119722669307913,
|
| 1368 |
-
"grad_norm": 1.
|
| 1369 |
-
"learning_rate": 1.
|
| 1370 |
"loss": 0.406,
|
| 1371 |
"step": 15800
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
"epoch": 3.9367339358672773,
|
| 1375 |
-
"grad_norm": 1.
|
| 1376 |
-
"learning_rate": 1.
|
| 1377 |
-
"loss": 0.
|
| 1378 |
"step": 15900
|
| 1379 |
},
|
| 1380 |
{
|
| 1381 |
"epoch": 3.9614956048037637,
|
| 1382 |
-
"grad_norm": 0.
|
| 1383 |
-
"learning_rate": 1.
|
| 1384 |
-
"loss": 0.
|
| 1385 |
"step": 16000
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
"epoch": 3.9614956048037637,
|
| 1389 |
-
"eval_loss": 0.
|
| 1390 |
-
"eval_runtime":
|
| 1391 |
-
"eval_samples_per_second":
|
| 1392 |
-
"eval_steps_per_second": 3.
|
| 1393 |
"step": 16000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1394 |
}
|
| 1395 |
],
|
| 1396 |
"logging_steps": 100,
|
|
@@ -1410,7 +1754,7 @@
|
|
| 1410 |
"attributes": {}
|
| 1411 |
}
|
| 1412 |
},
|
| 1413 |
-
"total_flos":
|
| 1414 |
"train_batch_size": 8,
|
| 1415 |
"trial_name": null,
|
| 1416 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": null,
|
| 3 |
+
"best_metric": 0.3671196699142456,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.951838553918535,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 20000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.0002476166893648632,
|
| 14 |
+
"grad_norm": 0.48706236481666565,
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
"loss": 3.0743,
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.02476166893648632,
|
| 21 |
+
"grad_norm": 0.6543678045272827,
|
| 22 |
+
"learning_rate": 9.900000000000002e-06,
|
| 23 |
+
"loss": 2.8614,
|
| 24 |
"step": 100
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.04952333787297264,
|
| 28 |
+
"grad_norm": 0.5323843359947205,
|
| 29 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 30 |
+
"loss": 1.8779,
|
| 31 |
"step": 200
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.07428500680945896,
|
| 35 |
+
"grad_norm": 0.9197348952293396,
|
| 36 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 37 |
+
"loss": 1.2173,
|
| 38 |
"step": 300
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.09904667574594528,
|
| 42 |
+
"grad_norm": 1.113196849822998,
|
| 43 |
+
"learning_rate": 3.99e-05,
|
| 44 |
+
"loss": 0.9351,
|
| 45 |
"step": 400
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.1238083446824316,
|
| 49 |
+
"grad_norm": 1.1514325141906738,
|
| 50 |
+
"learning_rate": 4.99e-05,
|
| 51 |
+
"loss": 0.8237,
|
| 52 |
"step": 500
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.1238083446824316,
|
| 56 |
+
"eval_loss": 0.6713243126869202,
|
| 57 |
+
"eval_runtime": 303.9033,
|
| 58 |
+
"eval_samples_per_second": 47.249,
|
| 59 |
+
"eval_steps_per_second": 2.955,
|
| 60 |
"step": 500
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 0.14857001361891792,
|
| 64 |
+
"grad_norm": 1.1924649477005005,
|
| 65 |
+
"learning_rate": 4.974866717440975e-05,
|
| 66 |
+
"loss": 0.7509,
|
| 67 |
"step": 600
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 0.17333168255540424,
|
| 71 |
+
"grad_norm": 1.0876811742782593,
|
| 72 |
+
"learning_rate": 4.9494795633409495e-05,
|
| 73 |
+
"loss": 0.6963,
|
| 74 |
"step": 700
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.19809335149189056,
|
| 78 |
+
"grad_norm": 1.4002293348312378,
|
| 79 |
+
"learning_rate": 4.924092409240924e-05,
|
| 80 |
+
"loss": 0.6837,
|
| 81 |
"step": 800
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.22285502042837688,
|
| 85 |
+
"grad_norm": 1.2242807149887085,
|
| 86 |
+
"learning_rate": 4.898705255140899e-05,
|
| 87 |
+
"loss": 0.6686,
|
| 88 |
"step": 900
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.2476166893648632,
|
| 92 |
+
"grad_norm": 1.3704895973205566,
|
| 93 |
+
"learning_rate": 4.873318101040874e-05,
|
| 94 |
+
"loss": 0.6462,
|
| 95 |
"step": 1000
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.2476166893648632,
|
| 99 |
+
"eval_loss": 0.5437047481536865,
|
| 100 |
+
"eval_runtime": 309.4401,
|
| 101 |
+
"eval_samples_per_second": 46.403,
|
| 102 |
+
"eval_steps_per_second": 2.902,
|
| 103 |
"step": 1000
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 0.2723783583013495,
|
| 107 |
+
"grad_norm": 1.2968043088912964,
|
| 108 |
+
"learning_rate": 4.847930946940848e-05,
|
| 109 |
+
"loss": 0.6301,
|
| 110 |
"step": 1100
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.29714002723783584,
|
| 114 |
+
"grad_norm": 1.4131654500961304,
|
| 115 |
+
"learning_rate": 4.822543792840823e-05,
|
| 116 |
+
"loss": 0.6256,
|
| 117 |
"step": 1200
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 0.3219016961743221,
|
| 121 |
+
"grad_norm": 1.3621524572372437,
|
| 122 |
+
"learning_rate": 4.7971566387407974e-05,
|
| 123 |
+
"loss": 0.6179,
|
| 124 |
"step": 1300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 0.3466633651108085,
|
| 128 |
+
"grad_norm": 1.4688136577606201,
|
| 129 |
+
"learning_rate": 4.771769484640772e-05,
|
| 130 |
+
"loss": 0.5963,
|
| 131 |
"step": 1400
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 0.37142503404729477,
|
| 135 |
+
"grad_norm": 1.3867251873016357,
|
| 136 |
+
"learning_rate": 4.746382330540747e-05,
|
| 137 |
+
"loss": 0.5867,
|
| 138 |
"step": 1500
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 0.37142503404729477,
|
| 142 |
+
"eval_loss": 0.5053889155387878,
|
| 143 |
+
"eval_runtime": 301.1479,
|
| 144 |
+
"eval_samples_per_second": 47.681,
|
| 145 |
+
"eval_steps_per_second": 2.982,
|
| 146 |
"step": 1500
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.3961867029837811,
|
| 150 |
+
"grad_norm": 1.018188714981079,
|
| 151 |
+
"learning_rate": 4.720995176440721e-05,
|
| 152 |
+
"loss": 0.5854,
|
| 153 |
"step": 1600
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.4209483719202674,
|
| 157 |
+
"grad_norm": 1.3094606399536133,
|
| 158 |
+
"learning_rate": 4.6956080223406954e-05,
|
| 159 |
+
"loss": 0.567,
|
| 160 |
"step": 1700
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.44571004085675375,
|
| 164 |
+
"grad_norm": 1.3288716077804565,
|
| 165 |
+
"learning_rate": 4.6702208682406704e-05,
|
| 166 |
+
"loss": 0.5629,
|
| 167 |
"step": 1800
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.47047170979324004,
|
| 171 |
+
"grad_norm": 1.2919931411743164,
|
| 172 |
+
"learning_rate": 4.6448337141406453e-05,
|
| 173 |
+
"loss": 0.5568,
|
| 174 |
"step": 1900
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.4952333787297264,
|
| 178 |
+
"grad_norm": 1.161441445350647,
|
| 179 |
+
"learning_rate": 4.6194465600406197e-05,
|
| 180 |
+
"loss": 0.5574,
|
| 181 |
"step": 2000
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.4952333787297264,
|
| 185 |
+
"eval_loss": 0.4834918975830078,
|
| 186 |
+
"eval_runtime": 308.7731,
|
| 187 |
+
"eval_samples_per_second": 46.503,
|
| 188 |
+
"eval_steps_per_second": 2.908,
|
| 189 |
"step": 2000
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 0.5199950476662127,
|
| 193 |
+
"grad_norm": 1.0919381380081177,
|
| 194 |
+
"learning_rate": 4.5940594059405946e-05,
|
| 195 |
+
"loss": 0.5525,
|
| 196 |
"step": 2100
|
| 197 |
},
|
| 198 |
{
|
| 199 |
"epoch": 0.544756716602699,
|
| 200 |
+
"grad_norm": 1.302264928817749,
|
| 201 |
+
"learning_rate": 4.568672251840569e-05,
|
| 202 |
+
"loss": 0.5486,
|
| 203 |
"step": 2200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.5695183855391853,
|
| 207 |
+
"grad_norm": 1.3316988945007324,
|
| 208 |
+
"learning_rate": 4.543285097740543e-05,
|
| 209 |
+
"loss": 0.5435,
|
| 210 |
"step": 2300
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 0.5942800544756717,
|
| 214 |
+
"grad_norm": 1.16013765335083,
|
| 215 |
+
"learning_rate": 4.517897943640518e-05,
|
| 216 |
+
"loss": 0.537,
|
| 217 |
"step": 2400
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 0.619041723412158,
|
| 221 |
+
"grad_norm": 1.0240144729614258,
|
| 222 |
+
"learning_rate": 4.4925107895404926e-05,
|
| 223 |
+
"loss": 0.5351,
|
| 224 |
"step": 2500
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.619041723412158,
|
| 228 |
+
"eval_loss": 0.46248629689216614,
|
| 229 |
+
"eval_runtime": 298.9584,
|
| 230 |
+
"eval_samples_per_second": 48.03,
|
| 231 |
+
"eval_steps_per_second": 3.004,
|
| 232 |
"step": 2500
|
| 233 |
},
|
| 234 |
{
|
| 235 |
"epoch": 0.6438033923486443,
|
| 236 |
+
"grad_norm": 1.403138279914856,
|
| 237 |
+
"learning_rate": 4.467123635440467e-05,
|
| 238 |
+
"loss": 0.529,
|
| 239 |
"step": 2600
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.6685650612851306,
|
| 243 |
+
"grad_norm": 1.0086416006088257,
|
| 244 |
+
"learning_rate": 4.441736481340442e-05,
|
| 245 |
+
"loss": 0.5179,
|
| 246 |
"step": 2700
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 0.693326730221617,
|
| 250 |
+
"grad_norm": 1.0099183320999146,
|
| 251 |
+
"learning_rate": 4.416349327240417e-05,
|
| 252 |
+
"loss": 0.5159,
|
| 253 |
"step": 2800
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 0.7180883991581033,
|
| 257 |
+
"grad_norm": 1.066780686378479,
|
| 258 |
+
"learning_rate": 4.390962173140391e-05,
|
| 259 |
+
"loss": 0.5184,
|
| 260 |
"step": 2900
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 0.7428500680945895,
|
| 264 |
+
"grad_norm": 1.1880680322647095,
|
| 265 |
+
"learning_rate": 4.365575019040366e-05,
|
| 266 |
+
"loss": 0.5132,
|
| 267 |
"step": 3000
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 0.7428500680945895,
|
| 271 |
+
"eval_loss": 0.4524896740913391,
|
| 272 |
+
"eval_runtime": 309.4036,
|
| 273 |
+
"eval_samples_per_second": 46.409,
|
| 274 |
+
"eval_steps_per_second": 2.902,
|
| 275 |
"step": 3000
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7676117370310759,
|
| 279 |
+
"grad_norm": 1.128923773765564,
|
| 280 |
+
"learning_rate": 4.3401878649403405e-05,
|
| 281 |
+
"loss": 0.5132,
|
| 282 |
"step": 3100
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7923734059675622,
|
| 286 |
+
"grad_norm": 1.0441526174545288,
|
| 287 |
+
"learning_rate": 4.314800710840315e-05,
|
| 288 |
+
"loss": 0.5024,
|
| 289 |
"step": 3200
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.8171350749040486,
|
| 293 |
+
"grad_norm": 1.3435834646224976,
|
| 294 |
+
"learning_rate": 4.28941355674029e-05,
|
| 295 |
+
"loss": 0.4999,
|
| 296 |
"step": 3300
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.8418967438405348,
|
| 300 |
+
"grad_norm": 1.0047725439071655,
|
| 301 |
+
"learning_rate": 4.264026402640264e-05,
|
| 302 |
+
"loss": 0.5025,
|
| 303 |
"step": 3400
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.8666584127770212,
|
| 307 |
+
"grad_norm": 1.1918917894363403,
|
| 308 |
+
"learning_rate": 4.238639248540239e-05,
|
| 309 |
+
"loss": 0.5002,
|
| 310 |
"step": 3500
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.8666584127770212,
|
| 314 |
+
"eval_loss": 0.44193407893180847,
|
| 315 |
+
"eval_runtime": 307.4102,
|
| 316 |
+
"eval_samples_per_second": 46.71,
|
| 317 |
+
"eval_steps_per_second": 2.921,
|
| 318 |
"step": 3500
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 0.8914200817135075,
|
| 322 |
+
"grad_norm": 1.3518158197402954,
|
| 323 |
+
"learning_rate": 4.2132520944402134e-05,
|
| 324 |
+
"loss": 0.4972,
|
| 325 |
"step": 3600
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 0.9161817506499939,
|
| 329 |
+
"grad_norm": 1.2649939060211182,
|
| 330 |
+
"learning_rate": 4.1878649403401884e-05,
|
| 331 |
+
"loss": 0.4935,
|
| 332 |
"step": 3700
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 0.9409434195864801,
|
| 336 |
+
"grad_norm": 1.1180412769317627,
|
| 337 |
+
"learning_rate": 4.162477786240163e-05,
|
| 338 |
+
"loss": 0.5021,
|
| 339 |
"step": 3800
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 0.9657050885229664,
|
| 343 |
+
"grad_norm": 0.9952785968780518,
|
| 344 |
+
"learning_rate": 4.137090632140137e-05,
|
| 345 |
+
"loss": 0.4995,
|
| 346 |
"step": 3900
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 0.9904667574594528,
|
| 350 |
+
"grad_norm": 1.1129059791564941,
|
| 351 |
+
"learning_rate": 4.111703478040112e-05,
|
| 352 |
+
"loss": 0.4944,
|
| 353 |
"step": 4000
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 0.9904667574594528,
|
| 357 |
+
"eval_loss": 0.4326762557029724,
|
| 358 |
+
"eval_runtime": 305.7402,
|
| 359 |
+
"eval_samples_per_second": 46.965,
|
| 360 |
+
"eval_steps_per_second": 2.937,
|
| 361 |
"step": 4000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 1.0151046180512566,
|
| 365 |
+
"grad_norm": 1.2399227619171143,
|
| 366 |
+
"learning_rate": 4.086316323940086e-05,
|
| 367 |
+
"loss": 0.4828,
|
| 368 |
"step": 4100
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 1.0398662869877429,
|
| 372 |
+
"grad_norm": 1.2356698513031006,
|
| 373 |
+
"learning_rate": 4.0609291698400606e-05,
|
| 374 |
+
"loss": 0.4818,
|
| 375 |
"step": 4200
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 1.0646279559242293,
|
| 379 |
+
"grad_norm": 1.016559362411499,
|
| 380 |
+
"learning_rate": 4.0355420157400356e-05,
|
| 381 |
+
"loss": 0.4777,
|
| 382 |
"step": 4300
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 1.0893896248607156,
|
| 386 |
+
"grad_norm": 1.1003488302230835,
|
| 387 |
+
"learning_rate": 4.0101548616400106e-05,
|
| 388 |
+
"loss": 0.4818,
|
| 389 |
"step": 4400
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 1.114151293797202,
|
| 393 |
+
"grad_norm": 1.0257238149642944,
|
| 394 |
+
"learning_rate": 3.984767707539985e-05,
|
| 395 |
+
"loss": 0.4804,
|
| 396 |
"step": 4500
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 1.114151293797202,
|
| 400 |
+
"eval_loss": 0.4317300021648407,
|
| 401 |
+
"eval_runtime": 301.8975,
|
| 402 |
+
"eval_samples_per_second": 47.563,
|
| 403 |
+
"eval_steps_per_second": 2.975,
|
| 404 |
"step": 4500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 1.1389129627336882,
|
| 408 |
+
"grad_norm": 0.9789583683013916,
|
| 409 |
+
"learning_rate": 3.95938055343996e-05,
|
| 410 |
+
"loss": 0.4765,
|
| 411 |
"step": 4600
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 1.1636746316701745,
|
| 415 |
+
"grad_norm": 1.2558050155639648,
|
| 416 |
+
"learning_rate": 3.933993399339934e-05,
|
| 417 |
+
"loss": 0.481,
|
| 418 |
"step": 4700
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 1.188436300606661,
|
| 422 |
+
"grad_norm": 1.0205676555633545,
|
| 423 |
+
"learning_rate": 3.9086062452399085e-05,
|
| 424 |
+
"loss": 0.4826,
|
| 425 |
"step": 4800
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.2131979695431472,
|
| 429 |
+
"grad_norm": 1.0989702939987183,
|
| 430 |
+
"learning_rate": 3.8832190911398835e-05,
|
| 431 |
+
"loss": 0.4716,
|
| 432 |
"step": 4900
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 1.2379596384796336,
|
| 436 |
+
"grad_norm": 1.031387448310852,
|
| 437 |
+
"learning_rate": 3.857831937039858e-05,
|
| 438 |
+
"loss": 0.4704,
|
| 439 |
"step": 5000
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 1.2379596384796336,
|
| 443 |
+
"eval_loss": 0.4228954017162323,
|
| 444 |
+
"eval_runtime": 306.4121,
|
| 445 |
+
"eval_samples_per_second": 46.862,
|
| 446 |
+
"eval_steps_per_second": 2.931,
|
| 447 |
"step": 5000
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 1.2627213074161199,
|
| 451 |
+
"grad_norm": 1.123007893562317,
|
| 452 |
+
"learning_rate": 3.832444782939832e-05,
|
| 453 |
+
"loss": 0.4688,
|
| 454 |
"step": 5100
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 1.2874829763526061,
|
| 458 |
+
"grad_norm": 1.0141407251358032,
|
| 459 |
+
"learning_rate": 3.807057628839807e-05,
|
| 460 |
+
"loss": 0.4678,
|
| 461 |
"step": 5200
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 1.3122446452890926,
|
| 465 |
+
"grad_norm": 1.0771005153656006,
|
| 466 |
+
"learning_rate": 3.781670474739782e-05,
|
| 467 |
+
"loss": 0.4659,
|
| 468 |
"step": 5300
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 1.3370063142255788,
|
| 472 |
+
"grad_norm": 1.2410191297531128,
|
| 473 |
+
"learning_rate": 3.7562833206397564e-05,
|
| 474 |
+
"loss": 0.4725,
|
| 475 |
"step": 5400
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.361767983162065,
|
| 479 |
+
"grad_norm": 0.9323300123214722,
|
| 480 |
+
"learning_rate": 3.7308961665397314e-05,
|
| 481 |
+
"loss": 0.4662,
|
| 482 |
"step": 5500
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.361767983162065,
|
| 486 |
+
"eval_loss": 0.418260782957077,
|
| 487 |
+
"eval_runtime": 305.893,
|
| 488 |
+
"eval_samples_per_second": 46.941,
|
| 489 |
+
"eval_steps_per_second": 2.936,
|
| 490 |
"step": 5500
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.3865296520985515,
|
| 494 |
+
"grad_norm": 0.9075552225112915,
|
| 495 |
+
"learning_rate": 3.705509012439706e-05,
|
| 496 |
+
"loss": 0.469,
|
| 497 |
"step": 5600
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.4112913210350377,
|
| 501 |
+
"grad_norm": 1.095397710800171,
|
| 502 |
+
"learning_rate": 3.68012185833968e-05,
|
| 503 |
+
"loss": 0.4696,
|
| 504 |
"step": 5700
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 1.436052989971524,
|
| 508 |
+
"grad_norm": 1.2005417346954346,
|
| 509 |
+
"learning_rate": 3.654734704239655e-05,
|
| 510 |
+
"loss": 0.4553,
|
| 511 |
"step": 5800
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 1.4608146589080104,
|
| 515 |
+
"grad_norm": 1.0341594219207764,
|
| 516 |
+
"learning_rate": 3.6293475501396294e-05,
|
| 517 |
+
"loss": 0.4632,
|
| 518 |
"step": 5900
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 1.4855763278444967,
|
| 522 |
+
"grad_norm": 1.0710721015930176,
|
| 523 |
+
"learning_rate": 3.603960396039604e-05,
|
| 524 |
+
"loss": 0.4676,
|
| 525 |
"step": 6000
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 1.4855763278444967,
|
| 529 |
+
"eval_loss": 0.41258370876312256,
|
| 530 |
+
"eval_runtime": 333.4873,
|
| 531 |
+
"eval_samples_per_second": 43.057,
|
| 532 |
+
"eval_steps_per_second": 2.693,
|
| 533 |
"step": 6000
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 1.510337996780983,
|
| 537 |
+
"grad_norm": 1.0697613954544067,
|
| 538 |
+
"learning_rate": 3.5785732419395787e-05,
|
| 539 |
+
"loss": 0.4653,
|
| 540 |
"step": 6100
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 1.5350996657174694,
|
| 544 |
+
"grad_norm": 0.9342092275619507,
|
| 545 |
+
"learning_rate": 3.5531860878395536e-05,
|
| 546 |
"loss": 0.4584,
|
| 547 |
"step": 6200
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 1.5598613346539558,
|
| 551 |
+
"grad_norm": 1.18372642993927,
|
| 552 |
+
"learning_rate": 3.527798933739528e-05,
|
| 553 |
+
"loss": 0.4646,
|
| 554 |
"step": 6300
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 1.5846230035904418,
|
| 558 |
+
"grad_norm": 0.9926603436470032,
|
| 559 |
+
"learning_rate": 3.502411779639503e-05,
|
| 560 |
+
"loss": 0.4523,
|
| 561 |
"step": 6400
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 1.6093846725269283,
|
| 565 |
+
"grad_norm": 0.8520728945732117,
|
| 566 |
+
"learning_rate": 3.477024625539477e-05,
|
| 567 |
+
"loss": 0.4508,
|
| 568 |
"step": 6500
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 1.6093846725269283,
|
| 572 |
+
"eval_loss": 0.4122717082500458,
|
| 573 |
+
"eval_runtime": 337.4066,
|
| 574 |
+
"eval_samples_per_second": 42.557,
|
| 575 |
+
"eval_steps_per_second": 2.661,
|
| 576 |
"step": 6500
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 1.6341463414634148,
|
| 580 |
+
"grad_norm": 1.0143983364105225,
|
| 581 |
+
"learning_rate": 3.4516374714394516e-05,
|
| 582 |
+
"loss": 0.4612,
|
| 583 |
"step": 6600
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 1.658908010399901,
|
| 587 |
+
"grad_norm": 1.0106840133666992,
|
| 588 |
+
"learning_rate": 3.426250317339426e-05,
|
| 589 |
+
"loss": 0.4684,
|
| 590 |
"step": 6700
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 1.6836696793363872,
|
| 594 |
+
"grad_norm": 1.0055443048477173,
|
| 595 |
+
"learning_rate": 3.400863163239401e-05,
|
| 596 |
+
"loss": 0.4568,
|
| 597 |
"step": 6800
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 1.7084313482728737,
|
| 601 |
+
"grad_norm": 0.9551578760147095,
|
| 602 |
+
"learning_rate": 3.375476009139376e-05,
|
| 603 |
+
"loss": 0.4604,
|
| 604 |
"step": 6900
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 1.73319301720936,
|
| 608 |
+
"grad_norm": 0.9932017922401428,
|
| 609 |
+
"learning_rate": 3.35008885503935e-05,
|
| 610 |
+
"loss": 0.4472,
|
| 611 |
"step": 7000
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 1.73319301720936,
|
| 615 |
+
"eval_loss": 0.40953782200813293,
|
| 616 |
+
"eval_runtime": 334.3565,
|
| 617 |
+
"eval_samples_per_second": 42.945,
|
| 618 |
+
"eval_steps_per_second": 2.686,
|
| 619 |
"step": 7000
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 1.7579546861458462,
|
| 623 |
+
"grad_norm": 1.0452265739440918,
|
| 624 |
+
"learning_rate": 3.324701700939325e-05,
|
| 625 |
+
"loss": 0.4604,
|
| 626 |
"step": 7100
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 1.7827163550823326,
|
| 630 |
+
"grad_norm": 1.0083004236221313,
|
| 631 |
+
"learning_rate": 3.2993145468392995e-05,
|
| 632 |
+
"loss": 0.4507,
|
| 633 |
"step": 7200
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 1.8074780240188189,
|
| 637 |
+
"grad_norm": 0.9387707710266113,
|
| 638 |
+
"learning_rate": 3.273927392739274e-05,
|
| 639 |
+
"loss": 0.4516,
|
| 640 |
"step": 7300
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 1.832239692955305,
|
| 644 |
+
"grad_norm": 0.9475572109222412,
|
| 645 |
+
"learning_rate": 3.248540238639249e-05,
|
| 646 |
+
"loss": 0.4444,
|
| 647 |
"step": 7400
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 1.8570013618917915,
|
| 651 |
+
"grad_norm": 1.093817949295044,
|
| 652 |
+
"learning_rate": 3.223153084539223e-05,
|
| 653 |
+
"loss": 0.456,
|
| 654 |
"step": 7500
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 1.8570013618917915,
|
| 658 |
+
"eval_loss": 0.4029190242290497,
|
| 659 |
+
"eval_runtime": 292.0631,
|
| 660 |
+
"eval_samples_per_second": 49.164,
|
| 661 |
+
"eval_steps_per_second": 3.075,
|
| 662 |
"step": 7500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 1.8817630308282778,
|
| 666 |
+
"grad_norm": 0.9310826063156128,
|
| 667 |
+
"learning_rate": 3.1977659304391974e-05,
|
| 668 |
+
"loss": 0.4552,
|
| 669 |
"step": 7600
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 1.906524699764764,
|
| 673 |
+
"grad_norm": 1.0708472728729248,
|
| 674 |
+
"learning_rate": 3.1723787763391724e-05,
|
| 675 |
+
"loss": 0.4442,
|
| 676 |
"step": 7700
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 1.9312863687012505,
|
| 680 |
+
"grad_norm": 0.9762539267539978,
|
| 681 |
+
"learning_rate": 3.1469916222391474e-05,
|
| 682 |
+
"loss": 0.438,
|
| 683 |
"step": 7800
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 1.956048037637737,
|
| 687 |
+
"grad_norm": 1.1194584369659424,
|
| 688 |
+
"learning_rate": 3.121604468139122e-05,
|
| 689 |
+
"loss": 0.4475,
|
| 690 |
"step": 7900
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 1.980809706574223,
|
| 694 |
+
"grad_norm": 1.001720905303955,
|
| 695 |
+
"learning_rate": 3.096217314039097e-05,
|
| 696 |
+
"loss": 0.4355,
|
| 697 |
"step": 8000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 1.980809706574223,
|
| 701 |
+
"eval_loss": 0.40090999007225037,
|
| 702 |
+
"eval_runtime": 290.1583,
|
| 703 |
+
"eval_samples_per_second": 49.487,
|
| 704 |
+
"eval_steps_per_second": 3.095,
|
| 705 |
"step": 8000
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 2.005447567166027,
|
| 709 |
+
"grad_norm": 1.2274198532104492,
|
| 710 |
+
"learning_rate": 3.070830159939071e-05,
|
| 711 |
+
"loss": 0.4395,
|
| 712 |
"step": 8100
|
| 713 |
},
|
| 714 |
{
|
| 715 |
"epoch": 2.0302092361025132,
|
| 716 |
+
"grad_norm": 1.008782148361206,
|
| 717 |
+
"learning_rate": 3.0454430058390453e-05,
|
| 718 |
+
"loss": 0.4386,
|
| 719 |
"step": 8200
|
| 720 |
},
|
| 721 |
{
|
| 722 |
"epoch": 2.0549709050389997,
|
| 723 |
+
"grad_norm": 1.110425591468811,
|
| 724 |
+
"learning_rate": 3.0200558517390203e-05,
|
| 725 |
+
"loss": 0.4427,
|
| 726 |
"step": 8300
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 2.0797325739754857,
|
| 730 |
+
"grad_norm": 0.944615364074707,
|
| 731 |
+
"learning_rate": 2.994668697638995e-05,
|
| 732 |
+
"loss": 0.4367,
|
| 733 |
"step": 8400
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 2.104494242911972,
|
| 737 |
+
"grad_norm": 1.0133123397827148,
|
| 738 |
+
"learning_rate": 2.9692815435389693e-05,
|
| 739 |
+
"loss": 0.4426,
|
| 740 |
"step": 8500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 2.104494242911972,
|
| 744 |
+
"eval_loss": 0.3965131342411041,
|
| 745 |
+
"eval_runtime": 295.1479,
|
| 746 |
+
"eval_samples_per_second": 48.65,
|
| 747 |
+
"eval_steps_per_second": 3.043,
|
| 748 |
"step": 8500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 2.1292559118484586,
|
| 752 |
+
"grad_norm": 0.7603266835212708,
|
| 753 |
+
"learning_rate": 2.9438943894389443e-05,
|
| 754 |
+
"loss": 0.4388,
|
| 755 |
"step": 8600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 2.154017580784945,
|
| 759 |
+
"grad_norm": 0.928263783454895,
|
| 760 |
+
"learning_rate": 2.9185072353389186e-05,
|
| 761 |
+
"loss": 0.4386,
|
| 762 |
"step": 8700
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 2.178779249721431,
|
| 766 |
+
"grad_norm": 1.0597220659255981,
|
| 767 |
+
"learning_rate": 2.8931200812388932e-05,
|
| 768 |
+
"loss": 0.4281,
|
| 769 |
"step": 8800
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 2.2035409186579176,
|
| 773 |
+
"grad_norm": 1.0807472467422485,
|
| 774 |
+
"learning_rate": 2.8677329271388682e-05,
|
| 775 |
+
"loss": 0.4312,
|
| 776 |
"step": 8900
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 2.228302587594404,
|
| 780 |
+
"grad_norm": 1.0931870937347412,
|
| 781 |
+
"learning_rate": 2.8423457730388425e-05,
|
| 782 |
+
"loss": 0.4369,
|
| 783 |
"step": 9000
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 2.228302587594404,
|
| 787 |
+
"eval_loss": 0.39540666341781616,
|
| 788 |
+
"eval_runtime": 292.4821,
|
| 789 |
+
"eval_samples_per_second": 49.094,
|
| 790 |
+
"eval_steps_per_second": 3.07,
|
| 791 |
"step": 9000
|
| 792 |
},
|
| 793 |
{
|
| 794 |
"epoch": 2.25306425653089,
|
| 795 |
+
"grad_norm": 0.9188655614852905,
|
| 796 |
+
"learning_rate": 2.816958618938817e-05,
|
| 797 |
+
"loss": 0.4404,
|
| 798 |
"step": 9100
|
| 799 |
},
|
| 800 |
{
|
| 801 |
"epoch": 2.2778259254673765,
|
| 802 |
+
"grad_norm": 1.0746827125549316,
|
| 803 |
+
"learning_rate": 2.7915714648387918e-05,
|
| 804 |
+
"loss": 0.4395,
|
| 805 |
"step": 9200
|
| 806 |
},
|
| 807 |
{
|
| 808 |
"epoch": 2.302587594403863,
|
| 809 |
+
"grad_norm": 0.9879487156867981,
|
| 810 |
+
"learning_rate": 2.7661843107387665e-05,
|
| 811 |
+
"loss": 0.4209,
|
| 812 |
"step": 9300
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 2.327349263340349,
|
| 816 |
+
"grad_norm": 1.0951037406921387,
|
| 817 |
+
"learning_rate": 2.7407971566387408e-05,
|
| 818 |
+
"loss": 0.431,
|
| 819 |
"step": 9400
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 2.3521109322768354,
|
| 823 |
+
"grad_norm": 1.1300798654556274,
|
| 824 |
+
"learning_rate": 2.7154100025387158e-05,
|
| 825 |
+
"loss": 0.4286,
|
| 826 |
"step": 9500
|
| 827 |
},
|
| 828 |
{
|
| 829 |
"epoch": 2.3521109322768354,
|
| 830 |
+
"eval_loss": 0.3942405581474304,
|
| 831 |
+
"eval_runtime": 294.7824,
|
| 832 |
+
"eval_samples_per_second": 48.711,
|
| 833 |
+
"eval_steps_per_second": 3.046,
|
| 834 |
"step": 9500
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 2.376872601213322,
|
| 838 |
+
"grad_norm": 1.2054657936096191,
|
| 839 |
+
"learning_rate": 2.69002284843869e-05,
|
| 840 |
+
"loss": 0.433,
|
| 841 |
"step": 9600
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 2.401634270149808,
|
| 845 |
+
"grad_norm": 1.1896750926971436,
|
| 846 |
+
"learning_rate": 2.6646356943386647e-05,
|
| 847 |
+
"loss": 0.4295,
|
| 848 |
"step": 9700
|
| 849 |
},
|
| 850 |
{
|
| 851 |
"epoch": 2.4263959390862944,
|
| 852 |
+
"grad_norm": 1.0336116552352905,
|
| 853 |
+
"learning_rate": 2.639248540238639e-05,
|
| 854 |
+
"loss": 0.4345,
|
| 855 |
"step": 9800
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 2.451157608022781,
|
| 859 |
+
"grad_norm": 0.9962900876998901,
|
| 860 |
+
"learning_rate": 2.613861386138614e-05,
|
| 861 |
+
"loss": 0.4219,
|
| 862 |
"step": 9900
|
| 863 |
},
|
| 864 |
{
|
| 865 |
"epoch": 2.4759192769592673,
|
| 866 |
+
"grad_norm": 1.2592706680297852,
|
| 867 |
+
"learning_rate": 2.5884742320385884e-05,
|
| 868 |
+
"loss": 0.44,
|
| 869 |
"step": 10000
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 2.4759192769592673,
|
| 873 |
+
"eval_loss": 0.39048588275909424,
|
| 874 |
+
"eval_runtime": 294.6469,
|
| 875 |
+
"eval_samples_per_second": 48.733,
|
| 876 |
+
"eval_steps_per_second": 3.048,
|
| 877 |
"step": 10000
|
| 878 |
},
|
| 879 |
{
|
| 880 |
"epoch": 2.5006809458957533,
|
| 881 |
+
"grad_norm": 1.25287663936615,
|
| 882 |
+
"learning_rate": 2.563087077938563e-05,
|
| 883 |
+
"loss": 0.4331,
|
| 884 |
"step": 10100
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 2.5254426148322398,
|
| 888 |
+
"grad_norm": 1.2010307312011719,
|
| 889 |
+
"learning_rate": 2.537699923838538e-05,
|
| 890 |
+
"loss": 0.427,
|
| 891 |
"step": 10200
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 2.5502042837687258,
|
| 895 |
+
"grad_norm": 1.06023108959198,
|
| 896 |
+
"learning_rate": 2.5123127697385123e-05,
|
| 897 |
+
"loss": 0.4281,
|
| 898 |
"step": 10300
|
| 899 |
},
|
| 900 |
{
|
| 901 |
"epoch": 2.5749659527052122,
|
| 902 |
+
"grad_norm": 1.094394326210022,
|
| 903 |
+
"learning_rate": 2.486925615638487e-05,
|
| 904 |
+
"loss": 0.4237,
|
| 905 |
"step": 10400
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 2.5997276216416987,
|
| 909 |
+
"grad_norm": 1.039994716644287,
|
| 910 |
+
"learning_rate": 2.461538461538462e-05,
|
| 911 |
+
"loss": 0.4222,
|
| 912 |
"step": 10500
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 2.5997276216416987,
|
| 916 |
+
"eval_loss": 0.3886333405971527,
|
| 917 |
+
"eval_runtime": 298.9241,
|
| 918 |
+
"eval_samples_per_second": 48.036,
|
| 919 |
+
"eval_steps_per_second": 3.004,
|
| 920 |
"step": 10500
|
| 921 |
},
|
| 922 |
{
|
| 923 |
"epoch": 2.624489290578185,
|
| 924 |
+
"grad_norm": 1.0885295867919922,
|
| 925 |
+
"learning_rate": 2.4361513074384363e-05,
|
| 926 |
+
"loss": 0.4297,
|
| 927 |
"step": 10600
|
| 928 |
},
|
| 929 |
{
|
| 930 |
"epoch": 2.649250959514671,
|
| 931 |
+
"grad_norm": 0.9978443384170532,
|
| 932 |
+
"learning_rate": 2.410764153338411e-05,
|
| 933 |
+
"loss": 0.4251,
|
| 934 |
"step": 10700
|
| 935 |
},
|
| 936 |
{
|
| 937 |
"epoch": 2.6740126284511576,
|
| 938 |
+
"grad_norm": 0.8032694458961487,
|
| 939 |
+
"learning_rate": 2.3853769992383852e-05,
|
| 940 |
"loss": 0.4321,
|
| 941 |
"step": 10800
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 2.698774297387644,
|
| 945 |
+
"grad_norm": 1.1382070779800415,
|
| 946 |
+
"learning_rate": 2.3599898451383602e-05,
|
| 947 |
+
"loss": 0.4296,
|
| 948 |
"step": 10900
|
| 949 |
},
|
| 950 |
{
|
| 951 |
"epoch": 2.72353596632413,
|
| 952 |
+
"grad_norm": 0.9287987947463989,
|
| 953 |
+
"learning_rate": 2.334602691038335e-05,
|
| 954 |
+
"loss": 0.417,
|
| 955 |
"step": 11000
|
| 956 |
},
|
| 957 |
{
|
| 958 |
"epoch": 2.72353596632413,
|
| 959 |
+
"eval_loss": 0.3885912001132965,
|
| 960 |
+
"eval_runtime": 299.1382,
|
| 961 |
+
"eval_samples_per_second": 48.001,
|
| 962 |
+
"eval_steps_per_second": 3.002,
|
| 963 |
"step": 11000
|
| 964 |
},
|
| 965 |
{
|
| 966 |
"epoch": 2.7482976352606165,
|
| 967 |
+
"grad_norm": 1.0950629711151123,
|
| 968 |
+
"learning_rate": 2.3092155369383092e-05,
|
| 969 |
"loss": 0.4253,
|
| 970 |
"step": 11100
|
| 971 |
},
|
| 972 |
{
|
| 973 |
"epoch": 2.773059304197103,
|
| 974 |
+
"grad_norm": 1.0151338577270508,
|
| 975 |
+
"learning_rate": 2.2838283828382838e-05,
|
| 976 |
+
"loss": 0.4311,
|
| 977 |
"step": 11200
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 2.7978209731335895,
|
| 981 |
+
"grad_norm": 1.0384260416030884,
|
| 982 |
+
"learning_rate": 2.2584412287382585e-05,
|
| 983 |
+
"loss": 0.4277,
|
| 984 |
"step": 11300
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 2.8225826420700755,
|
| 988 |
+
"grad_norm": 1.0595632791519165,
|
| 989 |
+
"learning_rate": 2.233054074638233e-05,
|
| 990 |
+
"loss": 0.4174,
|
| 991 |
"step": 11400
|
| 992 |
},
|
| 993 |
{
|
| 994 |
"epoch": 2.847344311006562,
|
| 995 |
+
"grad_norm": 1.081904411315918,
|
| 996 |
+
"learning_rate": 2.2076669205382078e-05,
|
| 997 |
+
"loss": 0.4244,
|
| 998 |
"step": 11500
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 2.847344311006562,
|
| 1002 |
+
"eval_loss": 0.3848166763782501,
|
| 1003 |
+
"eval_runtime": 298.8323,
|
| 1004 |
+
"eval_samples_per_second": 48.05,
|
| 1005 |
+
"eval_steps_per_second": 3.005,
|
| 1006 |
"step": 11500
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
"epoch": 2.872105979943048,
|
| 1010 |
+
"grad_norm": 1.3593488931655884,
|
| 1011 |
+
"learning_rate": 2.1822797664381824e-05,
|
| 1012 |
+
"loss": 0.4236,
|
| 1013 |
"step": 11600
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 2.8968676488795344,
|
| 1017 |
+
"grad_norm": 0.9958997368812561,
|
| 1018 |
+
"learning_rate": 2.156892612338157e-05,
|
| 1019 |
+
"loss": 0.4272,
|
| 1020 |
"step": 11700
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
"epoch": 2.921629317816021,
|
| 1024 |
+
"grad_norm": 0.9125483632087708,
|
| 1025 |
+
"learning_rate": 2.1315054582381317e-05,
|
| 1026 |
+
"loss": 0.4195,
|
| 1027 |
"step": 11800
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 2.9463909867525073,
|
| 1031 |
+
"grad_norm": 1.026393175125122,
|
| 1032 |
+
"learning_rate": 2.1061183041381064e-05,
|
| 1033 |
+
"loss": 0.4216,
|
| 1034 |
"step": 11900
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
"epoch": 2.9711526556889933,
|
| 1038 |
+
"grad_norm": 0.984633207321167,
|
| 1039 |
+
"learning_rate": 2.0807311500380807e-05,
|
| 1040 |
+
"loss": 0.4197,
|
| 1041 |
"step": 12000
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
"epoch": 2.9711526556889933,
|
| 1045 |
+
"eval_loss": 0.3836701214313507,
|
| 1046 |
+
"eval_runtime": 299.7207,
|
| 1047 |
+
"eval_samples_per_second": 47.908,
|
| 1048 |
+
"eval_steps_per_second": 2.996,
|
| 1049 |
"step": 12000
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
"epoch": 2.99591432462548,
|
| 1053 |
+
"grad_norm": 1.1759393215179443,
|
| 1054 |
+
"learning_rate": 2.0553439959380553e-05,
|
| 1055 |
+
"loss": 0.4182,
|
| 1056 |
"step": 12100
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
"epoch": 3.0205521852172836,
|
| 1060 |
+
"grad_norm": 0.9976751208305359,
|
| 1061 |
+
"learning_rate": 2.02995684183803e-05,
|
| 1062 |
+
"loss": 0.4176,
|
| 1063 |
"step": 12200
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
"epoch": 3.04531385415377,
|
| 1067 |
+
"grad_norm": 0.9128820896148682,
|
| 1068 |
+
"learning_rate": 2.0045696877380047e-05,
|
| 1069 |
+
"loss": 0.4115,
|
| 1070 |
"step": 12300
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
"epoch": 3.070075523090256,
|
| 1074 |
+
"grad_norm": 0.8917282819747925,
|
| 1075 |
+
"learning_rate": 1.9791825336379793e-05,
|
| 1076 |
+
"loss": 0.4068,
|
| 1077 |
"step": 12400
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
"epoch": 3.0948371920267426,
|
| 1081 |
+
"grad_norm": 1.04193115234375,
|
| 1082 |
+
"learning_rate": 1.9537953795379536e-05,
|
| 1083 |
+
"loss": 0.4138,
|
| 1084 |
"step": 12500
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
"epoch": 3.0948371920267426,
|
| 1088 |
+
"eval_loss": 0.3802282214164734,
|
| 1089 |
+
"eval_runtime": 303.6383,
|
| 1090 |
+
"eval_samples_per_second": 47.29,
|
| 1091 |
+
"eval_steps_per_second": 2.957,
|
| 1092 |
"step": 12500
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 3.119598860963229,
|
| 1096 |
+
"grad_norm": 1.0125975608825684,
|
| 1097 |
+
"learning_rate": 1.9284082254379286e-05,
|
| 1098 |
+
"loss": 0.4074,
|
| 1099 |
"step": 12600
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 3.144360529899715,
|
| 1103 |
+
"grad_norm": 1.0177041292190552,
|
| 1104 |
+
"learning_rate": 1.9030210713379033e-05,
|
| 1105 |
+
"loss": 0.4179,
|
| 1106 |
"step": 12700
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 3.1691221988362015,
|
| 1110 |
+
"grad_norm": 1.2167311906814575,
|
| 1111 |
+
"learning_rate": 1.8776339172378776e-05,
|
| 1112 |
+
"loss": 0.415,
|
| 1113 |
"step": 12800
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 3.193883867772688,
|
| 1117 |
+
"grad_norm": 1.1083834171295166,
|
| 1118 |
+
"learning_rate": 1.8522467631378522e-05,
|
| 1119 |
+
"loss": 0.417,
|
| 1120 |
"step": 12900
|
| 1121 |
},
|
| 1122 |
{
|
| 1123 |
"epoch": 3.2186455367091744,
|
| 1124 |
+
"grad_norm": 1.3251594305038452,
|
| 1125 |
+
"learning_rate": 1.8268596090378272e-05,
|
| 1126 |
+
"loss": 0.4096,
|
| 1127 |
"step": 13000
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
"epoch": 3.2186455367091744,
|
| 1131 |
+
"eval_loss": 0.37971043586730957,
|
| 1132 |
+
"eval_runtime": 307.1968,
|
| 1133 |
+
"eval_samples_per_second": 46.742,
|
| 1134 |
+
"eval_steps_per_second": 2.923,
|
| 1135 |
"step": 13000
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
"epoch": 3.2434072056456604,
|
| 1139 |
+
"grad_norm": 1.1293381452560425,
|
| 1140 |
+
"learning_rate": 1.8014724549378015e-05,
|
| 1141 |
+
"loss": 0.4033,
|
| 1142 |
"step": 13100
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
"epoch": 3.268168874582147,
|
| 1146 |
+
"grad_norm": 0.8698910474777222,
|
| 1147 |
+
"learning_rate": 1.7760853008377762e-05,
|
| 1148 |
+
"loss": 0.414,
|
| 1149 |
"step": 13200
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
"epoch": 3.2929305435186333,
|
| 1153 |
+
"grad_norm": 0.8111106753349304,
|
| 1154 |
+
"learning_rate": 1.7506981467377508e-05,
|
| 1155 |
+
"loss": 0.4205,
|
| 1156 |
"step": 13300
|
| 1157 |
},
|
| 1158 |
{
|
| 1159 |
"epoch": 3.3176922124551194,
|
| 1160 |
+
"grad_norm": 1.0510135889053345,
|
| 1161 |
+
"learning_rate": 1.7253109926377255e-05,
|
| 1162 |
+
"loss": 0.4193,
|
| 1163 |
"step": 13400
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
"epoch": 3.342453881391606,
|
| 1167 |
+
"grad_norm": 0.9042615294456482,
|
| 1168 |
+
"learning_rate": 1.6999238385377e-05,
|
| 1169 |
"loss": 0.4111,
|
| 1170 |
"step": 13500
|
| 1171 |
},
|
| 1172 |
{
|
| 1173 |
"epoch": 3.342453881391606,
|
| 1174 |
+
"eval_loss": 0.38024312257766724,
|
| 1175 |
+
"eval_runtime": 334.2385,
|
| 1176 |
+
"eval_samples_per_second": 42.96,
|
| 1177 |
+
"eval_steps_per_second": 2.687,
|
| 1178 |
"step": 13500
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 3.3672155503280923,
|
| 1182 |
+
"grad_norm": 0.9368965029716492,
|
| 1183 |
+
"learning_rate": 1.6745366844376748e-05,
|
| 1184 |
+
"loss": 0.4124,
|
| 1185 |
"step": 13600
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 3.3919772192645783,
|
| 1189 |
+
"grad_norm": 1.1354291439056396,
|
| 1190 |
+
"learning_rate": 1.649149530337649e-05,
|
| 1191 |
+
"loss": 0.4144,
|
| 1192 |
"step": 13700
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
"epoch": 3.4167388882010647,
|
| 1196 |
+
"grad_norm": 0.862594723701477,
|
| 1197 |
+
"learning_rate": 1.6237623762376237e-05,
|
| 1198 |
+
"loss": 0.4039,
|
| 1199 |
"step": 13800
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 3.441500557137551,
|
| 1203 |
+
"grad_norm": 1.2077276706695557,
|
| 1204 |
+
"learning_rate": 1.5983752221375984e-05,
|
| 1205 |
+
"loss": 0.4163,
|
| 1206 |
"step": 13900
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
"epoch": 3.4662622260740372,
|
| 1210 |
+
"grad_norm": 1.1101346015930176,
|
| 1211 |
+
"learning_rate": 1.572988068037573e-05,
|
| 1212 |
+
"loss": 0.4103,
|
| 1213 |
"step": 14000
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 3.4662622260740372,
|
| 1217 |
+
"eval_loss": 0.37751057744026184,
|
| 1218 |
+
"eval_runtime": 344.048,
|
| 1219 |
+
"eval_samples_per_second": 41.735,
|
| 1220 |
+
"eval_steps_per_second": 2.61,
|
| 1221 |
"step": 14000
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
"epoch": 3.4910238950105237,
|
| 1225 |
+
"grad_norm": 0.9908251166343689,
|
| 1226 |
+
"learning_rate": 1.5476009139375477e-05,
|
| 1227 |
+
"loss": 0.4116,
|
| 1228 |
"step": 14100
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
"epoch": 3.51578556394701,
|
| 1232 |
+
"grad_norm": 1.144132375717163,
|
| 1233 |
+
"learning_rate": 1.5222137598375222e-05,
|
| 1234 |
+
"loss": 0.41,
|
| 1235 |
"step": 14200
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 3.5405472328834966,
|
| 1239 |
+
"grad_norm": 1.0292690992355347,
|
| 1240 |
+
"learning_rate": 1.4968266057374968e-05,
|
| 1241 |
+
"loss": 0.4157,
|
| 1242 |
"step": 14300
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
"epoch": 3.5653089018199826,
|
| 1246 |
+
"grad_norm": 1.2643660306930542,
|
| 1247 |
+
"learning_rate": 1.4714394516374716e-05,
|
| 1248 |
+
"loss": 0.4081,
|
| 1249 |
"step": 14400
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
"epoch": 3.590070570756469,
|
| 1253 |
+
"grad_norm": 1.3031563758850098,
|
| 1254 |
+
"learning_rate": 1.446052297537446e-05,
|
| 1255 |
+
"loss": 0.4098,
|
| 1256 |
"step": 14500
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 3.590070570756469,
|
| 1260 |
+
"eval_loss": 0.3778412640094757,
|
| 1261 |
+
"eval_runtime": 330.1121,
|
| 1262 |
+
"eval_samples_per_second": 43.497,
|
| 1263 |
+
"eval_steps_per_second": 2.72,
|
| 1264 |
"step": 14500
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
"epoch": 3.614832239692955,
|
| 1268 |
+
"grad_norm": 1.0413382053375244,
|
| 1269 |
+
"learning_rate": 1.4206651434374208e-05,
|
| 1270 |
+
"loss": 0.4075,
|
| 1271 |
"step": 14600
|
| 1272 |
},
|
| 1273 |
{
|
| 1274 |
"epoch": 3.6395939086294415,
|
| 1275 |
+
"grad_norm": 1.1236363649368286,
|
| 1276 |
+
"learning_rate": 1.3952779893373954e-05,
|
| 1277 |
+
"loss": 0.4049,
|
| 1278 |
"step": 14700
|
| 1279 |
},
|
| 1280 |
{
|
| 1281 |
"epoch": 3.664355577565928,
|
| 1282 |
+
"grad_norm": 0.9670480489730835,
|
| 1283 |
+
"learning_rate": 1.3698908352373699e-05,
|
| 1284 |
+
"loss": 0.4066,
|
| 1285 |
"step": 14800
|
| 1286 |
},
|
| 1287 |
{
|
| 1288 |
"epoch": 3.6891172465024145,
|
| 1289 |
+
"grad_norm": 0.8969297409057617,
|
| 1290 |
+
"learning_rate": 1.3445036811373446e-05,
|
| 1291 |
+
"loss": 0.4156,
|
| 1292 |
"step": 14900
|
| 1293 |
},
|
| 1294 |
{
|
| 1295 |
"epoch": 3.7138789154389005,
|
| 1296 |
+
"grad_norm": 1.0945312976837158,
|
| 1297 |
+
"learning_rate": 1.3191165270373194e-05,
|
| 1298 |
+
"loss": 0.4133,
|
| 1299 |
"step": 15000
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
"epoch": 3.7138789154389005,
|
| 1303 |
+
"eval_loss": 0.3742968738079071,
|
| 1304 |
+
"eval_runtime": 331.5339,
|
| 1305 |
+
"eval_samples_per_second": 43.311,
|
| 1306 |
+
"eval_steps_per_second": 2.709,
|
| 1307 |
"step": 15000
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
"epoch": 3.738640584375387,
|
| 1311 |
+
"grad_norm": 1.036347508430481,
|
| 1312 |
+
"learning_rate": 1.2937293729372937e-05,
|
| 1313 |
+
"loss": 0.4099,
|
| 1314 |
"step": 15100
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
"epoch": 3.763402253311873,
|
| 1318 |
+
"grad_norm": 1.0111111402511597,
|
| 1319 |
+
"learning_rate": 1.2683422188372685e-05,
|
| 1320 |
+
"loss": 0.4115,
|
| 1321 |
"step": 15200
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
"epoch": 3.7881639222483594,
|
| 1325 |
+
"grad_norm": 1.1293095350265503,
|
| 1326 |
+
"learning_rate": 1.242955064737243e-05,
|
| 1327 |
+
"loss": 0.4075,
|
| 1328 |
"step": 15300
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
"epoch": 3.812925591184846,
|
| 1332 |
+
"grad_norm": 1.1474101543426514,
|
| 1333 |
+
"learning_rate": 1.2175679106372177e-05,
|
| 1334 |
+
"loss": 0.4182,
|
| 1335 |
"step": 15400
|
| 1336 |
},
|
| 1337 |
{
|
| 1338 |
"epoch": 3.8376872601213323,
|
| 1339 |
+
"grad_norm": 1.1107712984085083,
|
| 1340 |
+
"learning_rate": 1.1921807565371923e-05,
|
| 1341 |
+
"loss": 0.4081,
|
| 1342 |
"step": 15500
|
| 1343 |
},
|
| 1344 |
{
|
| 1345 |
"epoch": 3.8376872601213323,
|
| 1346 |
+
"eval_loss": 0.37404417991638184,
|
| 1347 |
+
"eval_runtime": 315.6431,
|
| 1348 |
+
"eval_samples_per_second": 45.491,
|
| 1349 |
+
"eval_steps_per_second": 2.845,
|
| 1350 |
"step": 15500
|
| 1351 |
},
|
| 1352 |
{
|
| 1353 |
"epoch": 3.8624489290578183,
|
| 1354 |
+
"grad_norm": 0.9492923021316528,
|
| 1355 |
+
"learning_rate": 1.166793602437167e-05,
|
| 1356 |
+
"loss": 0.3983,
|
| 1357 |
"step": 15600
|
| 1358 |
},
|
| 1359 |
{
|
| 1360 |
"epoch": 3.887210597994305,
|
| 1361 |
+
"grad_norm": 1.1327282190322876,
|
| 1362 |
+
"learning_rate": 1.1414064483371414e-05,
|
| 1363 |
+
"loss": 0.4043,
|
| 1364 |
"step": 15700
|
| 1365 |
},
|
| 1366 |
{
|
| 1367 |
"epoch": 3.9119722669307913,
|
| 1368 |
+
"grad_norm": 1.0399820804595947,
|
| 1369 |
+
"learning_rate": 1.1160192942371161e-05,
|
| 1370 |
"loss": 0.406,
|
| 1371 |
"step": 15800
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
"epoch": 3.9367339358672773,
|
| 1375 |
+
"grad_norm": 1.1408737897872925,
|
| 1376 |
+
"learning_rate": 1.0906321401370907e-05,
|
| 1377 |
+
"loss": 0.4065,
|
| 1378 |
"step": 15900
|
| 1379 |
},
|
| 1380 |
{
|
| 1381 |
"epoch": 3.9614956048037637,
|
| 1382 |
+
"grad_norm": 0.9349102973937988,
|
| 1383 |
+
"learning_rate": 1.0652449860370652e-05,
|
| 1384 |
+
"loss": 0.4056,
|
| 1385 |
"step": 16000
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
"epoch": 3.9614956048037637,
|
| 1389 |
+
"eval_loss": 0.37279391288757324,
|
| 1390 |
+
"eval_runtime": 294.1835,
|
| 1391 |
+
"eval_samples_per_second": 48.81,
|
| 1392 |
+
"eval_steps_per_second": 3.053,
|
| 1393 |
"step": 16000
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 3.98625727374025,
|
| 1397 |
+
"grad_norm": 0.951724112033844,
|
| 1398 |
+
"learning_rate": 1.0398578319370399e-05,
|
| 1399 |
+
"loss": 0.4056,
|
| 1400 |
+
"step": 16100
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 4.010895134332054,
|
| 1404 |
+
"grad_norm": 1.131766438484192,
|
| 1405 |
+
"learning_rate": 1.0144706778370145e-05,
|
| 1406 |
+
"loss": 0.3989,
|
| 1407 |
+
"step": 16200
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 4.03565680326854,
|
| 1411 |
+
"grad_norm": 1.0067100524902344,
|
| 1412 |
+
"learning_rate": 9.890835237369892e-06,
|
| 1413 |
+
"loss": 0.4026,
|
| 1414 |
+
"step": 16300
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 4.0604184722050265,
|
| 1418 |
+
"grad_norm": 1.0153299570083618,
|
| 1419 |
+
"learning_rate": 9.636963696369637e-06,
|
| 1420 |
+
"loss": 0.4013,
|
| 1421 |
+
"step": 16400
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 4.085180141141513,
|
| 1425 |
+
"grad_norm": 1.2241690158843994,
|
| 1426 |
+
"learning_rate": 9.383092155369383e-06,
|
| 1427 |
+
"loss": 0.4,
|
| 1428 |
+
"step": 16500
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 4.085180141141513,
|
| 1432 |
+
"eval_loss": 0.3727540373802185,
|
| 1433 |
+
"eval_runtime": 296.4387,
|
| 1434 |
+
"eval_samples_per_second": 48.438,
|
| 1435 |
+
"eval_steps_per_second": 3.029,
|
| 1436 |
+
"step": 16500
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 4.109941810077999,
|
| 1440 |
+
"grad_norm": 0.9304781556129456,
|
| 1441 |
+
"learning_rate": 9.12922061436913e-06,
|
| 1442 |
+
"loss": 0.3969,
|
| 1443 |
+
"step": 16600
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 4.134703479014486,
|
| 1447 |
+
"grad_norm": 1.1999999284744263,
|
| 1448 |
+
"learning_rate": 8.875349073368876e-06,
|
| 1449 |
+
"loss": 0.3921,
|
| 1450 |
+
"step": 16700
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 4.159465147950971,
|
| 1454 |
+
"grad_norm": 1.0765935182571411,
|
| 1455 |
+
"learning_rate": 8.621477532368621e-06,
|
| 1456 |
+
"loss": 0.409,
|
| 1457 |
+
"step": 16800
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 4.184226816887458,
|
| 1461 |
+
"grad_norm": 0.886246383190155,
|
| 1462 |
+
"learning_rate": 8.367605991368369e-06,
|
| 1463 |
+
"loss": 0.4019,
|
| 1464 |
+
"step": 16900
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 4.208988485823944,
|
| 1468 |
+
"grad_norm": 1.0984737873077393,
|
| 1469 |
+
"learning_rate": 8.113734450368114e-06,
|
| 1470 |
+
"loss": 0.3991,
|
| 1471 |
+
"step": 17000
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 4.208988485823944,
|
| 1475 |
+
"eval_loss": 0.3709471821784973,
|
| 1476 |
+
"eval_runtime": 298.2727,
|
| 1477 |
+
"eval_samples_per_second": 48.141,
|
| 1478 |
+
"eval_steps_per_second": 3.011,
|
| 1479 |
+
"step": 17000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.233750154760431,
|
| 1483 |
+
"grad_norm": 1.0152359008789062,
|
| 1484 |
+
"learning_rate": 7.85986290936786e-06,
|
| 1485 |
+
"loss": 0.3962,
|
| 1486 |
+
"step": 17100
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.258511823696917,
|
| 1490 |
+
"grad_norm": 1.1482372283935547,
|
| 1491 |
+
"learning_rate": 7.605991368367606e-06,
|
| 1492 |
+
"loss": 0.4,
|
| 1493 |
+
"step": 17200
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.283273492633404,
|
| 1497 |
+
"grad_norm": 0.9904143810272217,
|
| 1498 |
+
"learning_rate": 7.352119827367353e-06,
|
| 1499 |
+
"loss": 0.3918,
|
| 1500 |
+
"step": 17300
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.30803516156989,
|
| 1504 |
+
"grad_norm": 1.0268852710723877,
|
| 1505 |
+
"learning_rate": 7.098248286367098e-06,
|
| 1506 |
+
"loss": 0.4049,
|
| 1507 |
+
"step": 17400
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.332796830506376,
|
| 1511 |
+
"grad_norm": 1.005434274673462,
|
| 1512 |
+
"learning_rate": 6.844376745366844e-06,
|
| 1513 |
+
"loss": 0.4009,
|
| 1514 |
+
"step": 17500
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.332796830506376,
|
| 1518 |
+
"eval_loss": 0.3704555034637451,
|
| 1519 |
+
"eval_runtime": 298.2902,
|
| 1520 |
+
"eval_samples_per_second": 48.138,
|
| 1521 |
+
"eval_steps_per_second": 3.01,
|
| 1522 |
+
"step": 17500
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 4.357558499442862,
|
| 1526 |
+
"grad_norm": 1.2603648900985718,
|
| 1527 |
+
"learning_rate": 6.590505204366591e-06,
|
| 1528 |
+
"loss": 0.4073,
|
| 1529 |
+
"step": 17600
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 4.382320168379349,
|
| 1533 |
+
"grad_norm": 1.231177568435669,
|
| 1534 |
+
"learning_rate": 6.336633663366337e-06,
|
| 1535 |
+
"loss": 0.4009,
|
| 1536 |
+
"step": 17700
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 4.407081837315835,
|
| 1540 |
+
"grad_norm": 1.0092039108276367,
|
| 1541 |
+
"learning_rate": 6.082762122366083e-06,
|
| 1542 |
+
"loss": 0.3984,
|
| 1543 |
+
"step": 17800
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 4.431843506252322,
|
| 1547 |
+
"grad_norm": 1.1094638109207153,
|
| 1548 |
+
"learning_rate": 5.828890581365829e-06,
|
| 1549 |
+
"loss": 0.4037,
|
| 1550 |
+
"step": 17900
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 4.456605175188808,
|
| 1554 |
+
"grad_norm": 1.2728139162063599,
|
| 1555 |
+
"learning_rate": 5.575019040365576e-06,
|
| 1556 |
+
"loss": 0.3981,
|
| 1557 |
+
"step": 18000
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 4.456605175188808,
|
| 1561 |
+
"eval_loss": 0.3691825568675995,
|
| 1562 |
+
"eval_runtime": 300.7735,
|
| 1563 |
+
"eval_samples_per_second": 47.74,
|
| 1564 |
+
"eval_steps_per_second": 2.986,
|
| 1565 |
+
"step": 18000
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 4.481366844125294,
|
| 1569 |
+
"grad_norm": 0.9797173738479614,
|
| 1570 |
+
"learning_rate": 5.321147499365321e-06,
|
| 1571 |
+
"loss": 0.3858,
|
| 1572 |
+
"step": 18100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 4.50612851306178,
|
| 1576 |
+
"grad_norm": 1.1960067749023438,
|
| 1577 |
+
"learning_rate": 5.067275958365068e-06,
|
| 1578 |
+
"loss": 0.4028,
|
| 1579 |
+
"step": 18200
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 4.5308901819982665,
|
| 1583 |
+
"grad_norm": 1.067478895187378,
|
| 1584 |
+
"learning_rate": 4.8134044173648135e-06,
|
| 1585 |
+
"loss": 0.4064,
|
| 1586 |
+
"step": 18300
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 4.555651850934753,
|
| 1590 |
+
"grad_norm": 1.0445499420166016,
|
| 1591 |
+
"learning_rate": 4.55953287636456e-06,
|
| 1592 |
+
"loss": 0.3965,
|
| 1593 |
+
"step": 18400
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 4.5804135198712395,
|
| 1597 |
+
"grad_norm": 1.035979986190796,
|
| 1598 |
+
"learning_rate": 4.3056613353643065e-06,
|
| 1599 |
+
"loss": 0.3909,
|
| 1600 |
+
"step": 18500
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 4.5804135198712395,
|
| 1604 |
+
"eval_loss": 0.3688410818576813,
|
| 1605 |
+
"eval_runtime": 305.0552,
|
| 1606 |
+
"eval_samples_per_second": 47.07,
|
| 1607 |
+
"eval_steps_per_second": 2.944,
|
| 1608 |
+
"step": 18500
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 4.605175188807726,
|
| 1612 |
+
"grad_norm": 0.9728735685348511,
|
| 1613 |
+
"learning_rate": 4.051789794364052e-06,
|
| 1614 |
+
"loss": 0.3943,
|
| 1615 |
+
"step": 18600
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 4.629936857744212,
|
| 1619 |
+
"grad_norm": 1.086190938949585,
|
| 1620 |
+
"learning_rate": 3.7979182533637982e-06,
|
| 1621 |
+
"loss": 0.3992,
|
| 1622 |
+
"step": 18700
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 4.654698526680698,
|
| 1626 |
+
"grad_norm": 1.103594183921814,
|
| 1627 |
+
"learning_rate": 3.544046712363544e-06,
|
| 1628 |
+
"loss": 0.3891,
|
| 1629 |
+
"step": 18800
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 4.679460195617184,
|
| 1633 |
+
"grad_norm": 1.0151804685592651,
|
| 1634 |
+
"learning_rate": 3.2901751713632904e-06,
|
| 1635 |
+
"loss": 0.4009,
|
| 1636 |
+
"step": 18900
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 4.704221864553671,
|
| 1640 |
+
"grad_norm": 1.141107201576233,
|
| 1641 |
+
"learning_rate": 3.0363036303630365e-06,
|
| 1642 |
+
"loss": 0.4028,
|
| 1643 |
+
"step": 19000
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 4.704221864553671,
|
| 1647 |
+
"eval_loss": 0.36849796772003174,
|
| 1648 |
+
"eval_runtime": 296.4027,
|
| 1649 |
+
"eval_samples_per_second": 48.444,
|
| 1650 |
+
"eval_steps_per_second": 3.03,
|
| 1651 |
+
"step": 19000
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 4.728983533490157,
|
| 1655 |
+
"grad_norm": 1.0555781126022339,
|
| 1656 |
+
"learning_rate": 2.7824320893627826e-06,
|
| 1657 |
+
"loss": 0.4033,
|
| 1658 |
+
"step": 19100
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 4.753745202426644,
|
| 1662 |
+
"grad_norm": 1.1756597757339478,
|
| 1663 |
+
"learning_rate": 2.5285605483625287e-06,
|
| 1664 |
+
"loss": 0.4022,
|
| 1665 |
+
"step": 19200
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 4.77850687136313,
|
| 1669 |
+
"grad_norm": 0.9200191497802734,
|
| 1670 |
+
"learning_rate": 2.2746890073622748e-06,
|
| 1671 |
+
"loss": 0.3987,
|
| 1672 |
+
"step": 19300
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 4.803268540299616,
|
| 1676 |
+
"grad_norm": 1.1391457319259644,
|
| 1677 |
+
"learning_rate": 2.0208174663620213e-06,
|
| 1678 |
+
"loss": 0.3961,
|
| 1679 |
+
"step": 19400
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 4.828030209236102,
|
| 1683 |
+
"grad_norm": 1.0987917184829712,
|
| 1684 |
+
"learning_rate": 1.7669459253617672e-06,
|
| 1685 |
+
"loss": 0.3903,
|
| 1686 |
+
"step": 19500
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 4.828030209236102,
|
| 1690 |
+
"eval_loss": 0.367659330368042,
|
| 1691 |
+
"eval_runtime": 300.2959,
|
| 1692 |
+
"eval_samples_per_second": 47.816,
|
| 1693 |
+
"eval_steps_per_second": 2.99,
|
| 1694 |
+
"step": 19500
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 4.852791878172589,
|
| 1698 |
+
"grad_norm": 0.9229596257209778,
|
| 1699 |
+
"learning_rate": 1.5130743843615132e-06,
|
| 1700 |
+
"loss": 0.3976,
|
| 1701 |
+
"step": 19600
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 4.877553547109075,
|
| 1705 |
+
"grad_norm": 0.9962015151977539,
|
| 1706 |
+
"learning_rate": 1.2592028433612593e-06,
|
| 1707 |
+
"loss": 0.3957,
|
| 1708 |
+
"step": 19700
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 4.902315216045562,
|
| 1712 |
+
"grad_norm": 1.4311672449111938,
|
| 1713 |
+
"learning_rate": 1.0053313023610054e-06,
|
| 1714 |
+
"loss": 0.3918,
|
| 1715 |
+
"step": 19800
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 4.927076884982048,
|
| 1719 |
+
"grad_norm": 1.1852339506149292,
|
| 1720 |
+
"learning_rate": 7.514597613607515e-07,
|
| 1721 |
+
"loss": 0.3989,
|
| 1722 |
+
"step": 19900
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 4.951838553918535,
|
| 1726 |
+
"grad_norm": 1.072943091392517,
|
| 1727 |
+
"learning_rate": 4.975882203604976e-07,
|
| 1728 |
+
"loss": 0.393,
|
| 1729 |
+
"step": 20000
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 4.951838553918535,
|
| 1733 |
+
"eval_loss": 0.3671196699142456,
|
| 1734 |
+
"eval_runtime": 304.303,
|
| 1735 |
+
"eval_samples_per_second": 47.187,
|
| 1736 |
+
"eval_steps_per_second": 2.951,
|
| 1737 |
+
"step": 20000
|
| 1738 |
}
|
| 1739 |
],
|
| 1740 |
"logging_steps": 100,
|
|
|
|
| 1754 |
"attributes": {}
|
| 1755 |
}
|
| 1756 |
},
|
| 1757 |
+
"total_flos": 7.046592561946024e+21,
|
| 1758 |
"train_batch_size": 8,
|
| 1759 |
"trial_name": null,
|
| 1760 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5905
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34f3d9eddddeae9f03bab89c311988daae534bc143c2b2742a73d1f5370d53f7
|
| 3 |
size 5905
|