End of training
Browse files- all_results.json +6 -6
- runs/May11_12-39-33_gcp/events.out.tfevents.1715431914.gcp.2381.1 +2 -2
- test_results.json +6 -6
- trainer_state.json +171 -540
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.7677419354838709,
|
| 4 |
+
"eval_loss": 0.5838326811790466,
|
| 5 |
+
"eval_runtime": 30.1878,
|
| 6 |
+
"eval_samples_per_second": 5.135,
|
| 7 |
+
"eval_steps_per_second": 1.292
|
| 8 |
}
|
runs/May11_12-39-33_gcp/events.out.tfevents.1715431914.gcp.2381.1
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51e09bfd475d44338ed825d1c7d03467d0f4481b80fea4f762ffe60cbfff08d9
|
| 3 |
+
size 734
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.7677419354838709,
|
| 4 |
+
"eval_loss": 0.5838326811790466,
|
| 5 |
+
"eval_runtime": 30.1878,
|
| 6 |
+
"eval_samples_per_second": 5.135,
|
| 7 |
+
"eval_steps_per_second": 1.292
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,662 +1,293 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 9.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 9.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 4.
|
| 57 |
-
"loss":
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"eval_accuracy": 0.
|
| 63 |
-
"eval_loss": 1.
|
| 64 |
-
"eval_runtime":
|
| 65 |
-
"eval_samples_per_second":
|
| 66 |
-
"eval_steps_per_second":
|
| 67 |
"step": 75
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch": 1.
|
| 71 |
-
"grad_norm":
|
| 72 |
-
"learning_rate": 4.
|
| 73 |
-
"loss": 1.
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 1.
|
| 78 |
-
"grad_norm":
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 1.
|
| 85 |
-
"grad_norm":
|
| 86 |
-
"learning_rate":
|
| 87 |
-
"loss": 1.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 1.
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss":
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 1.
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 1.
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 1.
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 1.
|
| 127 |
-
"eval_accuracy": 0.
|
| 128 |
-
"eval_loss": 0.
|
| 129 |
-
"eval_runtime":
|
| 130 |
-
"eval_samples_per_second":
|
| 131 |
-
"eval_steps_per_second":
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 2.
|
| 136 |
-
"grad_norm":
|
| 137 |
-
"learning_rate":
|
| 138 |
-
"loss": 0.
|
| 139 |
"step": 160
|
| 140 |
},
|
| 141 |
{
|
| 142 |
-
"epoch": 2.
|
| 143 |
-
"grad_norm":
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 2.
|
| 150 |
-
"grad_norm":
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 2.
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss": 0.
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"epoch": 2.
|
| 164 |
-
"grad_norm":
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 0.
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch": 2.
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 2.
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch": 2.
|
| 185 |
-
"eval_accuracy": 0.
|
| 186 |
-
"eval_loss": 0.
|
| 187 |
-
"eval_runtime":
|
| 188 |
-
"eval_samples_per_second":
|
| 189 |
-
"eval_steps_per_second":
|
| 190 |
"step": 225
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"epoch": 3.
|
| 194 |
-
"grad_norm": 1.
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 0.
|
| 197 |
"step": 230
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch": 3.
|
| 201 |
-
"grad_norm": 0.
|
| 202 |
-
"learning_rate":
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 240
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch": 3.
|
| 208 |
-
"grad_norm":
|
| 209 |
-
"learning_rate":
|
| 210 |
-
"loss": 0.
|
| 211 |
"step": 250
|
| 212 |
},
|
| 213 |
{
|
| 214 |
-
"epoch": 3.
|
| 215 |
-
"grad_norm": 0.
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 260
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 3.
|
| 222 |
-
"grad_norm":
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss": 0.
|
| 225 |
"step": 270
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch": 3.
|
| 229 |
-
"grad_norm": 0.
|
| 230 |
-
"learning_rate": 3.
|
| 231 |
-
"loss": 0.
|
| 232 |
"step": 280
|
| 233 |
},
|
| 234 |
{
|
| 235 |
-
"epoch": 3.
|
| 236 |
-
"grad_norm":
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 0.
|
| 239 |
"step": 290
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch": 3.
|
| 243 |
-
"grad_norm":
|
| 244 |
-
"learning_rate":
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 3.
|
| 250 |
-
"eval_accuracy": 0.
|
| 251 |
-
"eval_loss": 0.
|
| 252 |
-
"eval_runtime":
|
| 253 |
-
"eval_samples_per_second":
|
| 254 |
-
"eval_steps_per_second":
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"
|
| 260 |
-
"
|
| 261 |
-
"
|
| 262 |
-
"
|
| 263 |
-
|
| 264 |
-
|
| 265 |
-
"epoch": 4.026666666666666,
|
| 266 |
-
"grad_norm": 52.5593376159668,
|
| 267 |
-
"learning_rate": 3.185185185185185e-05,
|
| 268 |
-
"loss": 0.4402,
|
| 269 |
-
"step": 320
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 4.04,
|
| 273 |
-
"grad_norm": 0.3685360252857208,
|
| 274 |
-
"learning_rate": 3.111111111111111e-05,
|
| 275 |
-
"loss": 0.1409,
|
| 276 |
-
"step": 330
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 4.053333333333334,
|
| 280 |
-
"grad_norm": 0.13833816349506378,
|
| 281 |
-
"learning_rate": 3.037037037037037e-05,
|
| 282 |
-
"loss": 0.2763,
|
| 283 |
-
"step": 340
|
| 284 |
-
},
|
| 285 |
-
{
|
| 286 |
-
"epoch": 4.066666666666666,
|
| 287 |
-
"grad_norm": 0.11666622757911682,
|
| 288 |
-
"learning_rate": 2.962962962962963e-05,
|
| 289 |
-
"loss": 0.0503,
|
| 290 |
-
"step": 350
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 4.08,
|
| 294 |
-
"grad_norm": 0.06857863068580627,
|
| 295 |
-
"learning_rate": 2.8888888888888888e-05,
|
| 296 |
-
"loss": 0.1914,
|
| 297 |
-
"step": 360
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 4.093333333333334,
|
| 301 |
-
"grad_norm": 3.4672675132751465,
|
| 302 |
-
"learning_rate": 2.814814814814815e-05,
|
| 303 |
-
"loss": 0.0448,
|
| 304 |
-
"step": 370
|
| 305 |
-
},
|
| 306 |
-
{
|
| 307 |
-
"epoch": 4.1,
|
| 308 |
-
"eval_accuracy": 0.8571428571428571,
|
| 309 |
-
"eval_loss": 0.44399380683898926,
|
| 310 |
-
"eval_runtime": 17.9423,
|
| 311 |
-
"eval_samples_per_second": 3.901,
|
| 312 |
-
"eval_steps_per_second": 1.003,
|
| 313 |
-
"step": 375
|
| 314 |
-
},
|
| 315 |
-
{
|
| 316 |
-
"epoch": 5.006666666666667,
|
| 317 |
-
"grad_norm": 1.6939857006072998,
|
| 318 |
-
"learning_rate": 2.7407407407407408e-05,
|
| 319 |
-
"loss": 0.0308,
|
| 320 |
-
"step": 380
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 5.02,
|
| 324 |
-
"grad_norm": 1.3869863748550415,
|
| 325 |
-
"learning_rate": 2.6666666666666667e-05,
|
| 326 |
-
"loss": 0.1281,
|
| 327 |
-
"step": 390
|
| 328 |
-
},
|
| 329 |
-
{
|
| 330 |
-
"epoch": 5.033333333333333,
|
| 331 |
-
"grad_norm": 0.26335254311561584,
|
| 332 |
-
"learning_rate": 2.5925925925925925e-05,
|
| 333 |
-
"loss": 0.0254,
|
| 334 |
-
"step": 400
|
| 335 |
-
},
|
| 336 |
-
{
|
| 337 |
-
"epoch": 5.046666666666667,
|
| 338 |
-
"grad_norm": 0.060862887650728226,
|
| 339 |
-
"learning_rate": 2.5185185185185183e-05,
|
| 340 |
-
"loss": 0.0119,
|
| 341 |
-
"step": 410
|
| 342 |
-
},
|
| 343 |
-
{
|
| 344 |
-
"epoch": 5.06,
|
| 345 |
-
"grad_norm": 0.22715356945991516,
|
| 346 |
-
"learning_rate": 2.4444444444444445e-05,
|
| 347 |
-
"loss": 0.1588,
|
| 348 |
-
"step": 420
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 5.073333333333333,
|
| 352 |
-
"grad_norm": 61.806556701660156,
|
| 353 |
-
"learning_rate": 2.3703703703703707e-05,
|
| 354 |
-
"loss": 0.0864,
|
| 355 |
-
"step": 430
|
| 356 |
-
},
|
| 357 |
-
{
|
| 358 |
-
"epoch": 5.086666666666667,
|
| 359 |
-
"grad_norm": 0.05195135250687599,
|
| 360 |
-
"learning_rate": 2.2962962962962965e-05,
|
| 361 |
-
"loss": 0.073,
|
| 362 |
-
"step": 440
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"epoch": 5.1,
|
| 366 |
-
"grad_norm": 43.471065521240234,
|
| 367 |
-
"learning_rate": 2.2222222222222223e-05,
|
| 368 |
-
"loss": 0.0785,
|
| 369 |
-
"step": 450
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 5.1,
|
| 373 |
-
"eval_accuracy": 0.9,
|
| 374 |
-
"eval_loss": 0.3218167722225189,
|
| 375 |
-
"eval_runtime": 19.9327,
|
| 376 |
-
"eval_samples_per_second": 3.512,
|
| 377 |
-
"eval_steps_per_second": 0.903,
|
| 378 |
-
"step": 450
|
| 379 |
-
},
|
| 380 |
-
{
|
| 381 |
-
"epoch": 6.013333333333334,
|
| 382 |
-
"grad_norm": 0.06781550496816635,
|
| 383 |
-
"learning_rate": 2.148148148148148e-05,
|
| 384 |
-
"loss": 0.1588,
|
| 385 |
-
"step": 460
|
| 386 |
-
},
|
| 387 |
-
{
|
| 388 |
-
"epoch": 6.026666666666666,
|
| 389 |
-
"grad_norm": 0.09493797272443771,
|
| 390 |
-
"learning_rate": 2.074074074074074e-05,
|
| 391 |
-
"loss": 0.0833,
|
| 392 |
-
"step": 470
|
| 393 |
-
},
|
| 394 |
-
{
|
| 395 |
-
"epoch": 6.04,
|
| 396 |
-
"grad_norm": 1.557625412940979,
|
| 397 |
-
"learning_rate": 2e-05,
|
| 398 |
-
"loss": 0.038,
|
| 399 |
-
"step": 480
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 6.053333333333334,
|
| 403 |
-
"grad_norm": 0.041507575660943985,
|
| 404 |
-
"learning_rate": 1.925925925925926e-05,
|
| 405 |
-
"loss": 0.0782,
|
| 406 |
-
"step": 490
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 6.066666666666666,
|
| 410 |
-
"grad_norm": 0.13706688582897186,
|
| 411 |
-
"learning_rate": 1.8518518518518518e-05,
|
| 412 |
-
"loss": 0.0061,
|
| 413 |
-
"step": 500
|
| 414 |
-
},
|
| 415 |
-
{
|
| 416 |
-
"epoch": 6.08,
|
| 417 |
-
"grad_norm": 0.04828852787613869,
|
| 418 |
-
"learning_rate": 1.777777777777778e-05,
|
| 419 |
-
"loss": 0.005,
|
| 420 |
-
"step": 510
|
| 421 |
-
},
|
| 422 |
-
{
|
| 423 |
-
"epoch": 6.093333333333334,
|
| 424 |
-
"grad_norm": 0.047101154923439026,
|
| 425 |
-
"learning_rate": 1.7037037037037038e-05,
|
| 426 |
-
"loss": 0.0214,
|
| 427 |
-
"step": 520
|
| 428 |
-
},
|
| 429 |
-
{
|
| 430 |
-
"epoch": 6.1,
|
| 431 |
-
"eval_accuracy": 0.9571428571428572,
|
| 432 |
-
"eval_loss": 0.1372106671333313,
|
| 433 |
-
"eval_runtime": 19.7597,
|
| 434 |
-
"eval_samples_per_second": 3.543,
|
| 435 |
-
"eval_steps_per_second": 0.911,
|
| 436 |
-
"step": 525
|
| 437 |
-
},
|
| 438 |
-
{
|
| 439 |
-
"epoch": 7.006666666666667,
|
| 440 |
-
"grad_norm": 0.18555110692977905,
|
| 441 |
-
"learning_rate": 1.62962962962963e-05,
|
| 442 |
-
"loss": 0.0045,
|
| 443 |
-
"step": 530
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 7.02,
|
| 447 |
-
"grad_norm": 0.11104493588209152,
|
| 448 |
-
"learning_rate": 1.5555555555555555e-05,
|
| 449 |
-
"loss": 0.0273,
|
| 450 |
-
"step": 540
|
| 451 |
-
},
|
| 452 |
-
{
|
| 453 |
-
"epoch": 7.033333333333333,
|
| 454 |
-
"grad_norm": 0.04018757492303848,
|
| 455 |
-
"learning_rate": 1.4814814814814815e-05,
|
| 456 |
-
"loss": 0.1058,
|
| 457 |
-
"step": 550
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"epoch": 7.046666666666667,
|
| 461 |
-
"grad_norm": 0.04107681289315224,
|
| 462 |
-
"learning_rate": 1.4074074074074075e-05,
|
| 463 |
-
"loss": 0.0064,
|
| 464 |
-
"step": 560
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 7.06,
|
| 468 |
-
"grad_norm": 0.05029289796948433,
|
| 469 |
-
"learning_rate": 1.3333333333333333e-05,
|
| 470 |
-
"loss": 0.0708,
|
| 471 |
-
"step": 570
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 7.073333333333333,
|
| 475 |
-
"grad_norm": 0.046671733260154724,
|
| 476 |
-
"learning_rate": 1.2592592592592592e-05,
|
| 477 |
-
"loss": 0.0056,
|
| 478 |
-
"step": 580
|
| 479 |
-
},
|
| 480 |
-
{
|
| 481 |
-
"epoch": 7.086666666666667,
|
| 482 |
-
"grad_norm": 0.05804697796702385,
|
| 483 |
-
"learning_rate": 1.1851851851851853e-05,
|
| 484 |
-
"loss": 0.012,
|
| 485 |
-
"step": 590
|
| 486 |
-
},
|
| 487 |
-
{
|
| 488 |
-
"epoch": 7.1,
|
| 489 |
-
"grad_norm": 0.04299278184771538,
|
| 490 |
-
"learning_rate": 1.1111111111111112e-05,
|
| 491 |
-
"loss": 0.0039,
|
| 492 |
-
"step": 600
|
| 493 |
-
},
|
| 494 |
-
{
|
| 495 |
-
"epoch": 7.1,
|
| 496 |
-
"eval_accuracy": 0.9571428571428572,
|
| 497 |
-
"eval_loss": 0.10343696922063828,
|
| 498 |
-
"eval_runtime": 21.2334,
|
| 499 |
-
"eval_samples_per_second": 3.297,
|
| 500 |
-
"eval_steps_per_second": 0.848,
|
| 501 |
-
"step": 600
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
-
"
|
| 506 |
-
"
|
| 507 |
-
"
|
| 508 |
-
"
|
| 509 |
-
|
| 510 |
-
|
| 511 |
-
"epoch": 8.026666666666667,
|
| 512 |
-
"grad_norm": 0.0486493855714798,
|
| 513 |
-
"learning_rate": 9.62962962962963e-06,
|
| 514 |
-
"loss": 0.0037,
|
| 515 |
-
"step": 620
|
| 516 |
-
},
|
| 517 |
-
{
|
| 518 |
-
"epoch": 8.04,
|
| 519 |
-
"grad_norm": 0.03539065271615982,
|
| 520 |
-
"learning_rate": 8.88888888888889e-06,
|
| 521 |
-
"loss": 0.1036,
|
| 522 |
-
"step": 630
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 8.053333333333333,
|
| 526 |
-
"grad_norm": 0.030678970739245415,
|
| 527 |
-
"learning_rate": 8.14814814814815e-06,
|
| 528 |
-
"loss": 0.0034,
|
| 529 |
-
"step": 640
|
| 530 |
-
},
|
| 531 |
-
{
|
| 532 |
-
"epoch": 8.066666666666666,
|
| 533 |
-
"grad_norm": 0.03648629039525986,
|
| 534 |
-
"learning_rate": 7.4074074074074075e-06,
|
| 535 |
-
"loss": 0.0557,
|
| 536 |
-
"step": 650
|
| 537 |
-
},
|
| 538 |
-
{
|
| 539 |
-
"epoch": 8.08,
|
| 540 |
-
"grad_norm": 0.036471571773290634,
|
| 541 |
-
"learning_rate": 6.666666666666667e-06,
|
| 542 |
-
"loss": 0.0984,
|
| 543 |
-
"step": 660
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 8.093333333333334,
|
| 547 |
-
"grad_norm": 0.05204283446073532,
|
| 548 |
-
"learning_rate": 5.925925925925927e-06,
|
| 549 |
-
"loss": 0.0048,
|
| 550 |
-
"step": 670
|
| 551 |
-
},
|
| 552 |
-
{
|
| 553 |
-
"epoch": 8.1,
|
| 554 |
-
"eval_accuracy": 0.9142857142857143,
|
| 555 |
-
"eval_loss": 0.15946045517921448,
|
| 556 |
-
"eval_runtime": 18.3528,
|
| 557 |
-
"eval_samples_per_second": 3.814,
|
| 558 |
-
"eval_steps_per_second": 0.981,
|
| 559 |
-
"step": 675
|
| 560 |
-
},
|
| 561 |
-
{
|
| 562 |
-
"epoch": 9.006666666666666,
|
| 563 |
-
"grad_norm": 0.0372748076915741,
|
| 564 |
-
"learning_rate": 5.185185185185185e-06,
|
| 565 |
-
"loss": 0.0034,
|
| 566 |
-
"step": 680
|
| 567 |
-
},
|
| 568 |
-
{
|
| 569 |
-
"epoch": 9.02,
|
| 570 |
-
"grad_norm": 0.043480608612298965,
|
| 571 |
-
"learning_rate": 4.444444444444445e-06,
|
| 572 |
-
"loss": 0.0033,
|
| 573 |
-
"step": 690
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"epoch": 9.033333333333333,
|
| 577 |
-
"grad_norm": 0.034943871200084686,
|
| 578 |
-
"learning_rate": 3.7037037037037037e-06,
|
| 579 |
-
"loss": 0.0034,
|
| 580 |
-
"step": 700
|
| 581 |
-
},
|
| 582 |
-
{
|
| 583 |
-
"epoch": 9.046666666666667,
|
| 584 |
-
"grad_norm": 0.05171409249305725,
|
| 585 |
-
"learning_rate": 2.9629629629629633e-06,
|
| 586 |
-
"loss": 0.0036,
|
| 587 |
-
"step": 710
|
| 588 |
-
},
|
| 589 |
-
{
|
| 590 |
-
"epoch": 9.06,
|
| 591 |
-
"grad_norm": 0.035231392830610275,
|
| 592 |
-
"learning_rate": 2.2222222222222225e-06,
|
| 593 |
-
"loss": 0.0031,
|
| 594 |
-
"step": 720
|
| 595 |
-
},
|
| 596 |
-
{
|
| 597 |
-
"epoch": 9.073333333333334,
|
| 598 |
-
"grad_norm": 0.06839217245578766,
|
| 599 |
-
"learning_rate": 1.4814814814814817e-06,
|
| 600 |
-
"loss": 0.0111,
|
| 601 |
-
"step": 730
|
| 602 |
-
},
|
| 603 |
-
{
|
| 604 |
-
"epoch": 9.086666666666666,
|
| 605 |
-
"grad_norm": 0.032866183668375015,
|
| 606 |
-
"learning_rate": 7.407407407407408e-07,
|
| 607 |
-
"loss": 0.0041,
|
| 608 |
-
"step": 740
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch":
|
| 612 |
-
"
|
| 613 |
-
"
|
| 614 |
-
"
|
| 615 |
-
"
|
| 616 |
-
|
| 617 |
-
|
| 618 |
-
"epoch": 9.1,
|
| 619 |
-
"eval_accuracy": 0.9571428571428572,
|
| 620 |
-
"eval_loss": 0.10069848597049713,
|
| 621 |
-
"eval_runtime": 20.7858,
|
| 622 |
-
"eval_samples_per_second": 3.368,
|
| 623 |
-
"eval_steps_per_second": 0.866,
|
| 624 |
-
"step": 750
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch": 9.1,
|
| 628 |
-
"step": 750,
|
| 629 |
-
"total_flos": 3.73846047031296e+18,
|
| 630 |
-
"train_loss": 0.4401957253267368,
|
| 631 |
-
"train_runtime": 1973.9568,
|
| 632 |
-
"train_samples_per_second": 1.52,
|
| 633 |
-
"train_steps_per_second": 0.38
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"epoch": 9.1,
|
| 637 |
-
"eval_accuracy": 0.9225806451612903,
|
| 638 |
-
"eval_loss": 0.2622383236885071,
|
| 639 |
-
"eval_runtime": 45.5213,
|
| 640 |
-
"eval_samples_per_second": 3.405,
|
| 641 |
-
"eval_steps_per_second": 0.857,
|
| 642 |
-
"step": 750
|
| 643 |
-
},
|
| 644 |
-
{
|
| 645 |
-
"epoch": 9.1,
|
| 646 |
-
"eval_accuracy": 0.9225806451612903,
|
| 647 |
-
"eval_loss": 0.2622382938861847,
|
| 648 |
-
"eval_runtime": 44.9977,
|
| 649 |
-
"eval_samples_per_second": 3.445,
|
| 650 |
-
"eval_steps_per_second": 0.867,
|
| 651 |
-
"step": 750
|
| 652 |
}
|
| 653 |
],
|
| 654 |
"logging_steps": 10,
|
| 655 |
-
"max_steps":
|
| 656 |
"num_input_tokens_seen": 0,
|
| 657 |
"num_train_epochs": 9223372036854775807,
|
| 658 |
"save_steps": 500,
|
| 659 |
-
"total_flos":
|
| 660 |
"train_batch_size": 4,
|
| 661 |
"trial_name": null,
|
| 662 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8428571428571429,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-150",
|
| 4 |
+
"epoch": 3.25,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03333333333333333,
|
| 13 |
+
"grad_norm": 9.030272483825684,
|
| 14 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 15 |
+
"loss": 2.3471,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06666666666666667,
|
| 20 |
+
"grad_norm": 9.653508186340332,
|
| 21 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 22 |
+
"loss": 2.2728,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 9.514510154724121,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 2.2423,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.13333333333333333,
|
| 34 |
+
"grad_norm": 9.425768852233887,
|
| 35 |
+
"learning_rate": 4.814814814814815e-05,
|
| 36 |
+
"loss": 2.0797,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"grad_norm": 12.643138885498047,
|
| 42 |
+
"learning_rate": 4.62962962962963e-05,
|
| 43 |
+
"loss": 2.1012,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.2,
|
| 48 |
+
"grad_norm": 9.729035377502441,
|
| 49 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 50 |
+
"loss": 1.7497,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.23333333333333334,
|
| 55 |
+
"grad_norm": 9.487822532653809,
|
| 56 |
+
"learning_rate": 4.259259259259259e-05,
|
| 57 |
+
"loss": 1.7455,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.25,
|
| 62 |
+
"eval_accuracy": 0.5857142857142857,
|
| 63 |
+
"eval_loss": 1.3021408319473267,
|
| 64 |
+
"eval_runtime": 16.6801,
|
| 65 |
+
"eval_samples_per_second": 4.197,
|
| 66 |
+
"eval_steps_per_second": 1.079,
|
| 67 |
"step": 75
|
| 68 |
},
|
| 69 |
{
|
| 70 |
+
"epoch": 1.0166666666666666,
|
| 71 |
+
"grad_norm": 9.826342582702637,
|
| 72 |
+
"learning_rate": 4.074074074074074e-05,
|
| 73 |
+
"loss": 1.3037,
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 1.05,
|
| 78 |
+
"grad_norm": 10.802420616149902,
|
| 79 |
+
"learning_rate": 3.888888888888889e-05,
|
| 80 |
+
"loss": 1.128,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 1.0833333333333333,
|
| 85 |
+
"grad_norm": 15.328178405761719,
|
| 86 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 87 |
+
"loss": 1.1334,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 1.1166666666666667,
|
| 92 |
+
"grad_norm": 9.453879356384277,
|
| 93 |
+
"learning_rate": 3.518518518518519e-05,
|
| 94 |
+
"loss": 0.8507,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 1.15,
|
| 99 |
+
"grad_norm": 10.099981307983398,
|
| 100 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 101 |
+
"loss": 0.8617,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.1833333333333333,
|
| 106 |
+
"grad_norm": 21.905996322631836,
|
| 107 |
+
"learning_rate": 3.148148148148148e-05,
|
| 108 |
+
"loss": 0.6073,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 1.2166666666666668,
|
| 113 |
+
"grad_norm": 8.017053604125977,
|
| 114 |
+
"learning_rate": 2.962962962962963e-05,
|
| 115 |
+
"loss": 0.4951,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 1.25,
|
| 120 |
+
"grad_norm": 3.5494937896728516,
|
| 121 |
+
"learning_rate": 2.777777777777778e-05,
|
| 122 |
+
"loss": 0.9476,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 1.25,
|
| 127 |
+
"eval_accuracy": 0.8428571428571429,
|
| 128 |
+
"eval_loss": 0.5081102252006531,
|
| 129 |
+
"eval_runtime": 16.5404,
|
| 130 |
+
"eval_samples_per_second": 4.232,
|
| 131 |
+
"eval_steps_per_second": 1.088,
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 2.033333333333333,
|
| 136 |
+
"grad_norm": 11.868717193603516,
|
| 137 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 138 |
+
"loss": 0.2692,
|
| 139 |
"step": 160
|
| 140 |
},
|
| 141 |
{
|
| 142 |
+
"epoch": 2.066666666666667,
|
| 143 |
+
"grad_norm": 11.16838550567627,
|
| 144 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 145 |
+
"loss": 0.4435,
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 2.1,
|
| 150 |
+
"grad_norm": 4.307368755340576,
|
| 151 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 152 |
+
"loss": 0.3437,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 2.1333333333333333,
|
| 157 |
+
"grad_norm": 14.703262329101562,
|
| 158 |
+
"learning_rate": 2.037037037037037e-05,
|
| 159 |
+
"loss": 0.4501,
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 2.1666666666666665,
|
| 164 |
+
"grad_norm": 3.9073095321655273,
|
| 165 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 166 |
+
"loss": 0.3173,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 2.2,
|
| 171 |
+
"grad_norm": 7.474828720092773,
|
| 172 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 173 |
+
"loss": 0.1711,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 2.2333333333333334,
|
| 178 |
+
"grad_norm": 1.4244682788848877,
|
| 179 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 180 |
+
"loss": 0.3926,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 2.25,
|
| 185 |
+
"eval_accuracy": 0.7714285714285715,
|
| 186 |
+
"eval_loss": 0.38518500328063965,
|
| 187 |
+
"eval_runtime": 16.889,
|
| 188 |
+
"eval_samples_per_second": 4.145,
|
| 189 |
+
"eval_steps_per_second": 1.066,
|
| 190 |
"step": 225
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"epoch": 3.0166666666666666,
|
| 194 |
+
"grad_norm": 1.0246777534484863,
|
| 195 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 196 |
+
"loss": 0.1841,
|
| 197 |
"step": 230
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 3.05,
|
| 201 |
+
"grad_norm": 0.22415301203727722,
|
| 202 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 203 |
+
"loss": 0.12,
|
| 204 |
"step": 240
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 3.0833333333333335,
|
| 208 |
+
"grad_norm": 7.395457744598389,
|
| 209 |
+
"learning_rate": 9.259259259259259e-06,
|
| 210 |
+
"loss": 0.1242,
|
| 211 |
"step": 250
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 3.1166666666666667,
|
| 215 |
+
"grad_norm": 0.42695850133895874,
|
| 216 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 217 |
+
"loss": 0.1291,
|
| 218 |
"step": 260
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 3.15,
|
| 222 |
+
"grad_norm": 2.703252077102661,
|
| 223 |
+
"learning_rate": 5.555555555555556e-06,
|
| 224 |
+
"loss": 0.1147,
|
| 225 |
"step": 270
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 3.183333333333333,
|
| 229 |
+
"grad_norm": 0.6641141772270203,
|
| 230 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 231 |
+
"loss": 0.2027,
|
| 232 |
"step": 280
|
| 233 |
},
|
| 234 |
{
|
| 235 |
+
"epoch": 3.216666666666667,
|
| 236 |
+
"grad_norm": 1.9637229442596436,
|
| 237 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 238 |
+
"loss": 0.2594,
|
| 239 |
"step": 290
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 3.25,
|
| 243 |
+
"grad_norm": 0.8269719481468201,
|
| 244 |
+
"learning_rate": 0.0,
|
| 245 |
+
"loss": 0.2059,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 3.25,
|
| 250 |
+
"eval_accuracy": 0.8428571428571429,
|
| 251 |
+
"eval_loss": 0.39890938997268677,
|
| 252 |
+
"eval_runtime": 16.6405,
|
| 253 |
+
"eval_samples_per_second": 4.207,
|
| 254 |
+
"eval_steps_per_second": 1.082,
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 3.25,
|
| 259 |
+
"step": 300,
|
| 260 |
+
"total_flos": 1.495384188125184e+18,
|
| 261 |
+
"train_loss": 0.8531251009305318,
|
| 262 |
+
"train_runtime": 689.5393,
|
| 263 |
+
"train_samples_per_second": 1.74,
|
| 264 |
+
"train_steps_per_second": 0.435
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 3.25,
|
| 268 |
+
"eval_accuracy": 0.7677419354838709,
|
| 269 |
+
"eval_loss": 0.5838326811790466,
|
| 270 |
+
"eval_runtime": 31.2901,
|
| 271 |
+
"eval_samples_per_second": 4.954,
|
| 272 |
+
"eval_steps_per_second": 1.246,
|
| 273 |
+
"step": 300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 3.25,
|
| 277 |
+
"eval_accuracy": 0.7677419354838709,
|
| 278 |
+
"eval_loss": 0.5838326811790466,
|
| 279 |
+
"eval_runtime": 30.1878,
|
| 280 |
+
"eval_samples_per_second": 5.135,
|
| 281 |
+
"eval_steps_per_second": 1.292,
|
| 282 |
+
"step": 300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 283 |
}
|
| 284 |
],
|
| 285 |
"logging_steps": 10,
|
| 286 |
+
"max_steps": 300,
|
| 287 |
"num_input_tokens_seen": 0,
|
| 288 |
"num_train_epochs": 9223372036854775807,
|
| 289 |
"save_steps": 500,
|
| 290 |
+
"total_flos": 1.495384188125184e+18,
|
| 291 |
"train_batch_size": 4,
|
| 292 |
"trial_name": null,
|
| 293 |
"trial_params": null
|