Upload folder using huggingface_hub
Browse files- config.json +2 -2
- pytorch_model-00001-of-00006.bin +2 -2
- pytorch_model-00002-of-00006.bin +2 -2
- pytorch_model-00003-of-00006.bin +2 -2
- pytorch_model-00004-of-00006.bin +2 -2
- pytorch_model-00005-of-00006.bin +2 -2
- pytorch_model-00006-of-00006.bin +2 -2
- trainer_state.json +1585 -406
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
@@ -10,7 +10,7 @@
|
|
| 10 |
"initializer_range": 0.02,
|
| 11 |
"intermediate_size": 13824,
|
| 12 |
"max_position_embeddings": 4096,
|
| 13 |
-
"
|
| 14 |
"max_sequence_length": 4096,
|
| 15 |
"model_type": "llama",
|
| 16 |
"num_attention_heads": 40,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "airoboros-13b-gpt4",
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
|
|
| 10 |
"initializer_range": 0.02,
|
| 11 |
"intermediate_size": 13824,
|
| 12 |
"max_position_embeddings": 4096,
|
| 13 |
+
"max_seq_len": 4096,
|
| 14 |
"max_sequence_length": 4096,
|
| 15 |
"model_type": "llama",
|
| 16 |
"num_attention_heads": 40,
|
pytorch_model-00001-of-00006.bin
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 sha256:66489a87afab2b6dea92dd6891777c62fe165822a46f7ff4ac41a42c55769505
|
| 3 |
+
size 9956546443
|
pytorch_model-00002-of-00006.bin
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:aeda356e227ba84f6b6acf9bd38a5b6f00010e4d4e866afc9315659971a7d598
|
| 3 |
+
size 9940859009
|
pytorch_model-00003-of-00006.bin
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:cbce747ed27f2c8ac15efb174759baabb76407a8d5ad7316de9f92197d3cea47
|
| 3 |
+
size 9940859567
|
pytorch_model-00004-of-00006.bin
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:b58fd2a8572aef39ad551d47112c3ad4630bb5173b898643bcce32dc88f5ddce
|
| 3 |
+
size 9867417913
|
pytorch_model-00005-of-00006.bin
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:bdd348511b019d1d68b8323f50a6eb0cb30558659d9d857eafecb3faa5e1aec3
|
| 3 |
+
size 9867459649
|
pytorch_model-00006-of-00006.bin
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:c251f981ad4e8996c3d98660324ff70b0ba237e6f2ea66701d2b1f665baa1643
|
| 3 |
+
size 2490476719
|
trainer_state.json
CHANGED
|
@@ -1,817 +1,1996 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.01,
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 0.
|
| 14 |
"step": 1
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.02,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 3
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 4
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 5
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 6
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 7
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 8
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 0.
|
| 62 |
"step": 9
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 10
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 11
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 12
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 13
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 0.
|
| 92 |
"step": 14
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 15
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 16
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 17
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 18
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 19
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 20
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 21
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 0.
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 22
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 0.
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 23
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 24
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch": 0.
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 25
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 0.
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 26
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 27
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 28
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 29
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 30
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 0.
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 31
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 32
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 33
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 34
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 35
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 36
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch": 0.
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 37
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 0.
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 38
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch": 0.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 39
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 0.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 40
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 41
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 42
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 0.
|
| 264 |
-
"learning_rate":
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 43
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch": 0.
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 0.
|
| 272 |
"step": 44
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch": 0.
|
| 276 |
-
"learning_rate":
|
| 277 |
-
"loss": 0.
|
| 278 |
"step": 45
|
| 279 |
},
|
| 280 |
{
|
| 281 |
-
"epoch": 0.
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
"step": 46
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch": 0.
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 47
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 0.
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 48
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 49
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch": 0.
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
"step": 50
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch": 0.
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss": 0.
|
| 314 |
"step": 51
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch": 0.
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 52
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch": 0.
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 53
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch": 0.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 54
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch": 0.
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 55
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"learning_rate":
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 56
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch": 0.
|
| 348 |
-
"learning_rate":
|
| 349 |
-
"loss": 0.
|
| 350 |
"step": 57
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch": 0.
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 58
|
| 357 |
},
|
| 358 |
{
|
| 359 |
-
"epoch": 0.
|
| 360 |
-
"learning_rate":
|
| 361 |
-
"loss": 0.
|
| 362 |
"step": 59
|
| 363 |
},
|
| 364 |
{
|
| 365 |
-
"epoch": 0.
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 60
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 0.
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 61
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch": 0.
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 62
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"learning_rate":
|
| 385 |
-
"loss": 0.
|
| 386 |
"step": 63
|
| 387 |
},
|
| 388 |
{
|
| 389 |
-
"epoch": 0.
|
| 390 |
-
"learning_rate":
|
| 391 |
-
"loss": 0.
|
| 392 |
"step": 64
|
| 393 |
},
|
| 394 |
{
|
| 395 |
-
"epoch": 0.
|
| 396 |
-
"learning_rate":
|
| 397 |
-
"loss": 0.
|
| 398 |
"step": 65
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch": 0.
|
| 402 |
-
"learning_rate":
|
| 403 |
-
"loss": 0.
|
| 404 |
"step": 66
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch": 0.
|
| 408 |
-
"learning_rate":
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 67
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch": 0.
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 68
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch": 0.
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 69
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"learning_rate": 9.
|
| 427 |
-
"loss": 0.
|
| 428 |
"step": 70
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch": 0.
|
| 432 |
-
"learning_rate": 9.
|
| 433 |
-
"loss": 0.
|
| 434 |
"step": 71
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch": 0.
|
| 438 |
-
"learning_rate": 9.
|
| 439 |
-
"loss": 0.
|
| 440 |
"step": 72
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch": 0.
|
| 444 |
-
"learning_rate": 9.
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 73
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch": 0.
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 74
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch": 0.
|
| 456 |
-
"learning_rate":
|
| 457 |
-
"loss": 0.
|
| 458 |
"step": 75
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch": 0.
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 76
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 77
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch": 0.
|
| 474 |
-
"learning_rate":
|
| 475 |
-
"loss": 0.
|
| 476 |
"step": 78
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch": 0.
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 79
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch": 0.
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 80
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch": 0.
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 81
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch": 0.
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 82
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch": 0.
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 83
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 84
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch": 0.
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 85
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch": 0.
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 86
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch": 0.
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 87
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 0.
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 88
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch": 0.
|
| 540 |
-
"learning_rate":
|
| 541 |
-
"loss": 0.
|
| 542 |
"step": 89
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"epoch": 0.
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 90
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"learning_rate":
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 91
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch": 0.
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 92
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch": 0.
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 93
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch": 0.
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 94
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 0.
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 95
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch": 0.
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 96
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 97
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss": 0.
|
| 596 |
"step": 98
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch": 0.
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 99
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch": 0.
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 100
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 0.
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 101
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch": 0.
|
| 618 |
-
"learning_rate":
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 102
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch": 0.
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 103
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 104
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 105
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch": 0.
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 106
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch": 0.
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 107
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch": 0.
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 108
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch": 0
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 109
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 110
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 111
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 0.
|
| 680 |
"step": 112
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 113
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch":
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 114
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"learning_rate": 8.
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 115
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 116
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 117
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 118
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 119
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 120
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 121
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 122
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 123
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"learning_rate":
|
| 751 |
-
"loss": 0.
|
| 752 |
"step": 124
|
| 753 |
},
|
| 754 |
{
|
| 755 |
-
"epoch":
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 125
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch":
|
| 762 |
-
"learning_rate":
|
| 763 |
-
"loss": 0.
|
| 764 |
"step": 126
|
| 765 |
},
|
| 766 |
{
|
| 767 |
-
"epoch":
|
| 768 |
-
"learning_rate":
|
| 769 |
-
"loss": 0.
|
| 770 |
"step": 127
|
| 771 |
},
|
| 772 |
{
|
| 773 |
-
"epoch":
|
| 774 |
-
"learning_rate":
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 128
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch":
|
| 780 |
-
"learning_rate":
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 129
|
| 783 |
},
|
| 784 |
{
|
| 785 |
-
"epoch":
|
| 786 |
-
"learning_rate":
|
| 787 |
-
"loss": 0.
|
| 788 |
"step": 130
|
| 789 |
},
|
| 790 |
{
|
| 791 |
-
"epoch":
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss": 0.
|
| 794 |
"step": 131
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch": 1.
|
| 798 |
-
"learning_rate":
|
| 799 |
-
"loss": 0.
|
| 800 |
"step": 132
|
| 801 |
},
|
| 802 |
{
|
| 803 |
-
"epoch": 1.
|
| 804 |
-
"
|
| 805 |
-
"
|
| 806 |
-
"
|
| 807 |
-
|
| 808 |
-
|
| 809 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 810 |
}
|
| 811 |
],
|
| 812 |
-
"max_steps":
|
| 813 |
-
"num_train_epochs":
|
| 814 |
-
"total_flos":
|
| 815 |
"trial_name": null,
|
| 816 |
"trial_params": null
|
| 817 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0275229357798166,
|
| 5 |
+
"global_step": 330,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 5.555555555555555e-07,
|
| 13 |
+
"loss": 0.6498,
|
| 14 |
"step": 1
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 1.111111111111111e-06,
|
| 19 |
+
"loss": 0.6642,
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 25 |
+
"loss": 0.6479,
|
| 26 |
"step": 3
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 2.222222222222222e-06,
|
| 31 |
+
"loss": 0.5813,
|
| 32 |
"step": 4
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 37 |
+
"loss": 0.6306,
|
| 38 |
"step": 5
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 43 |
+
"loss": 0.6881,
|
| 44 |
"step": 6
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 3.88888888888889e-06,
|
| 49 |
+
"loss": 0.6066,
|
| 50 |
"step": 7
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 4.444444444444444e-06,
|
| 55 |
+
"loss": 0.5688,
|
| 56 |
"step": 8
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 5e-06,
|
| 61 |
+
"loss": 0.5619,
|
| 62 |
"step": 9
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.09,
|
| 66 |
+
"learning_rate": 5.555555555555557e-06,
|
| 67 |
+
"loss": 0.5144,
|
| 68 |
"step": 10
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.1,
|
| 72 |
+
"learning_rate": 6.111111111111112e-06,
|
| 73 |
+
"loss": 0.5744,
|
| 74 |
"step": 11
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.11,
|
| 78 |
+
"learning_rate": 6.666666666666667e-06,
|
| 79 |
+
"loss": 0.6051,
|
| 80 |
"step": 12
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.12,
|
| 84 |
+
"learning_rate": 7.222222222222223e-06,
|
| 85 |
+
"loss": 0.5917,
|
| 86 |
"step": 13
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.13,
|
| 90 |
+
"learning_rate": 7.77777777777778e-06,
|
| 91 |
+
"loss": 0.5239,
|
| 92 |
"step": 14
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.14,
|
| 96 |
+
"learning_rate": 8.333333333333334e-06,
|
| 97 |
+
"loss": 0.5253,
|
| 98 |
"step": 15
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.15,
|
| 102 |
+
"learning_rate": 8.888888888888888e-06,
|
| 103 |
+
"loss": 0.4143,
|
| 104 |
"step": 16
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.16,
|
| 108 |
+
"learning_rate": 9.444444444444445e-06,
|
| 109 |
+
"loss": 0.4822,
|
| 110 |
"step": 17
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.17,
|
| 114 |
+
"learning_rate": 1e-05,
|
| 115 |
+
"loss": 0.4627,
|
| 116 |
"step": 18
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.17,
|
| 120 |
+
"learning_rate": 9.999858783596665e-06,
|
| 121 |
+
"loss": 0.5389,
|
| 122 |
"step": 19
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.18,
|
| 126 |
+
"learning_rate": 9.999435142363484e-06,
|
| 127 |
+
"loss": 0.5544,
|
| 128 |
"step": 20
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.19,
|
| 132 |
+
"learning_rate": 9.998729100230497e-06,
|
| 133 |
+
"loss": 0.5013,
|
| 134 |
"step": 21
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.2,
|
| 138 |
+
"learning_rate": 9.997740697079595e-06,
|
| 139 |
+
"loss": 0.5035,
|
| 140 |
"step": 22
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.21,
|
| 144 |
+
"learning_rate": 9.99646998874227e-06,
|
| 145 |
+
"loss": 0.4924,
|
| 146 |
"step": 23
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.22,
|
| 150 |
+
"learning_rate": 9.994917046996472e-06,
|
| 151 |
+
"loss": 0.507,
|
| 152 |
"step": 24
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.23,
|
| 156 |
+
"learning_rate": 9.993081959562539e-06,
|
| 157 |
+
"loss": 0.4876,
|
| 158 |
"step": 25
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.24,
|
| 162 |
+
"learning_rate": 9.990964830098246e-06,
|
| 163 |
+
"loss": 0.555,
|
| 164 |
"step": 26
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.25,
|
| 168 |
+
"learning_rate": 9.98856577819296e-06,
|
| 169 |
+
"loss": 0.4692,
|
| 170 |
"step": 27
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.26,
|
| 174 |
+
"learning_rate": 9.985884939360873e-06,
|
| 175 |
+
"loss": 0.466,
|
| 176 |
"step": 28
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.27,
|
| 180 |
+
"learning_rate": 9.98292246503335e-06,
|
| 181 |
+
"loss": 0.4132,
|
| 182 |
"step": 29
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.28,
|
| 186 |
+
"learning_rate": 9.979678522550382e-06,
|
| 187 |
+
"loss": 0.4741,
|
| 188 |
"step": 30
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.28,
|
| 192 |
+
"learning_rate": 9.976153295151123e-06,
|
| 193 |
+
"loss": 0.5424,
|
| 194 |
"step": 31
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.29,
|
| 198 |
+
"learning_rate": 9.972346981963546e-06,
|
| 199 |
+
"loss": 0.4949,
|
| 200 |
"step": 32
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.3,
|
| 204 |
+
"learning_rate": 9.968259797993197e-06,
|
| 205 |
+
"loss": 0.5126,
|
| 206 |
"step": 33
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.31,
|
| 210 |
+
"learning_rate": 9.963891974111042e-06,
|
| 211 |
+
"loss": 0.4202,
|
| 212 |
"step": 34
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.32,
|
| 216 |
+
"learning_rate": 9.959243757040434e-06,
|
| 217 |
+
"loss": 0.4736,
|
| 218 |
"step": 35
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.33,
|
| 222 |
+
"learning_rate": 9.95431540934317e-06,
|
| 223 |
+
"loss": 0.3851,
|
| 224 |
"step": 36
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.34,
|
| 228 |
+
"learning_rate": 9.949107209404664e-06,
|
| 229 |
+
"loss": 0.5542,
|
| 230 |
"step": 37
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.35,
|
| 234 |
+
"learning_rate": 9.943619451418225e-06,
|
| 235 |
+
"loss": 0.5095,
|
| 236 |
"step": 38
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.36,
|
| 240 |
+
"learning_rate": 9.937852445368427e-06,
|
| 241 |
+
"loss": 0.4338,
|
| 242 |
"step": 39
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.37,
|
| 246 |
+
"learning_rate": 9.931806517013612e-06,
|
| 247 |
+
"loss": 0.45,
|
| 248 |
"step": 40
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.38,
|
| 252 |
+
"learning_rate": 9.925482007867485e-06,
|
| 253 |
+
"loss": 0.4735,
|
| 254 |
"step": 41
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.39,
|
| 258 |
+
"learning_rate": 9.918879275179819e-06,
|
| 259 |
+
"loss": 0.4735,
|
| 260 |
"step": 42
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 0.39,
|
| 264 |
+
"learning_rate": 9.911998691916275e-06,
|
| 265 |
+
"loss": 0.5041,
|
| 266 |
"step": 43
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 0.4,
|
| 270 |
+
"learning_rate": 9.904840646737346e-06,
|
| 271 |
+
"loss": 0.4432,
|
| 272 |
"step": 44
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 0.41,
|
| 276 |
+
"learning_rate": 9.89740554397639e-06,
|
| 277 |
+
"loss": 0.4457,
|
| 278 |
"step": 45
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 0.42,
|
| 282 |
+
"learning_rate": 9.889693803616793e-06,
|
| 283 |
+
"loss": 0.4692,
|
| 284 |
"step": 46
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 0.43,
|
| 288 |
+
"learning_rate": 9.881705861268252e-06,
|
| 289 |
+
"loss": 0.4469,
|
| 290 |
"step": 47
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.44,
|
| 294 |
+
"learning_rate": 9.873442168142158e-06,
|
| 295 |
+
"loss": 0.4389,
|
| 296 |
"step": 48
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.45,
|
| 300 |
+
"learning_rate": 9.864903191026125e-06,
|
| 301 |
+
"loss": 0.4195,
|
| 302 |
"step": 49
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 0.46,
|
| 306 |
+
"learning_rate": 9.856089412257605e-06,
|
| 307 |
+
"loss": 0.4635,
|
| 308 |
"step": 50
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 0.47,
|
| 312 |
+
"learning_rate": 9.847001329696653e-06,
|
| 313 |
+
"loss": 0.4944,
|
| 314 |
"step": 51
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 0.48,
|
| 318 |
+
"learning_rate": 9.837639456697802e-06,
|
| 319 |
+
"loss": 0.4917,
|
| 320 |
"step": 52
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 0.49,
|
| 324 |
+
"learning_rate": 9.828004322081067e-06,
|
| 325 |
+
"loss": 0.4197,
|
| 326 |
"step": 53
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 0.5,
|
| 330 |
+
"learning_rate": 9.818096470102067e-06,
|
| 331 |
+
"loss": 0.4294,
|
| 332 |
"step": 54
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 0.5,
|
| 336 |
+
"learning_rate": 9.807916460421294e-06,
|
| 337 |
+
"loss": 0.4839,
|
| 338 |
"step": 55
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.51,
|
| 342 |
+
"learning_rate": 9.797464868072489e-06,
|
| 343 |
+
"loss": 0.4918,
|
| 344 |
"step": 56
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 0.52,
|
| 348 |
+
"learning_rate": 9.78674228343016e-06,
|
| 349 |
+
"loss": 0.5003,
|
| 350 |
"step": 57
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 0.53,
|
| 354 |
+
"learning_rate": 9.775749312176249e-06,
|
| 355 |
+
"loss": 0.4864,
|
| 356 |
"step": 58
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 0.54,
|
| 360 |
+
"learning_rate": 9.764486575265893e-06,
|
| 361 |
+
"loss": 0.4977,
|
| 362 |
"step": 59
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 0.55,
|
| 366 |
+
"learning_rate": 9.752954708892379e-06,
|
| 367 |
+
"loss": 0.4125,
|
| 368 |
"step": 60
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 0.56,
|
| 372 |
+
"learning_rate": 9.741154364451179e-06,
|
| 373 |
+
"loss": 0.4442,
|
| 374 |
"step": 61
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.57,
|
| 378 |
+
"learning_rate": 9.729086208503174e-06,
|
| 379 |
+
"loss": 0.4624,
|
| 380 |
"step": 62
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.58,
|
| 384 |
+
"learning_rate": 9.716750922736998e-06,
|
| 385 |
+
"loss": 0.3754,
|
| 386 |
"step": 63
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 0.59,
|
| 390 |
+
"learning_rate": 9.704149203930522e-06,
|
| 391 |
+
"loss": 0.4218,
|
| 392 |
"step": 64
|
| 393 |
},
|
| 394 |
{
|
| 395 |
+
"epoch": 0.6,
|
| 396 |
+
"learning_rate": 9.691281763911513e-06,
|
| 397 |
+
"loss": 0.4997,
|
| 398 |
"step": 65
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 0.61,
|
| 402 |
+
"learning_rate": 9.67814932951741e-06,
|
| 403 |
+
"loss": 0.3829,
|
| 404 |
"step": 66
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 0.61,
|
| 408 |
+
"learning_rate": 9.664752642554272e-06,
|
| 409 |
+
"loss": 0.3637,
|
| 410 |
"step": 67
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.62,
|
| 414 |
+
"learning_rate": 9.651092459754879e-06,
|
| 415 |
+
"loss": 0.5271,
|
| 416 |
"step": 68
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 0.63,
|
| 420 |
+
"learning_rate": 9.637169552735985e-06,
|
| 421 |
+
"loss": 0.4425,
|
| 422 |
"step": 69
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.64,
|
| 426 |
+
"learning_rate": 9.622984707954732e-06,
|
| 427 |
+
"loss": 0.4109,
|
| 428 |
"step": 70
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 0.65,
|
| 432 |
+
"learning_rate": 9.608538726664224e-06,
|
| 433 |
+
"loss": 0.3699,
|
| 434 |
"step": 71
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 0.66,
|
| 438 |
+
"learning_rate": 9.593832424868271e-06,
|
| 439 |
+
"loss": 0.4199,
|
| 440 |
"step": 72
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 0.67,
|
| 444 |
+
"learning_rate": 9.578866633275289e-06,
|
| 445 |
+
"loss": 0.4548,
|
| 446 |
"step": 73
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 0.68,
|
| 450 |
+
"learning_rate": 9.563642197251382e-06,
|
| 451 |
+
"loss": 0.4464,
|
| 452 |
"step": 74
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.69,
|
| 456 |
+
"learning_rate": 9.548159976772593e-06,
|
| 457 |
+
"loss": 0.4286,
|
| 458 |
"step": 75
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 0.7,
|
| 462 |
+
"learning_rate": 9.532420846376316e-06,
|
| 463 |
+
"loss": 0.397,
|
| 464 |
"step": 76
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.71,
|
| 468 |
+
"learning_rate": 9.516425695111906e-06,
|
| 469 |
+
"loss": 0.4522,
|
| 470 |
"step": 77
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 0.72,
|
| 474 |
+
"learning_rate": 9.500175426490455e-06,
|
| 475 |
+
"loss": 0.4763,
|
| 476 |
"step": 78
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 0.72,
|
| 480 |
+
"learning_rate": 9.48367095843376e-06,
|
| 481 |
+
"loss": 0.4386,
|
| 482 |
"step": 79
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 0.73,
|
| 486 |
+
"learning_rate": 9.466913223222467e-06,
|
| 487 |
+
"loss": 0.411,
|
| 488 |
"step": 80
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 0.74,
|
| 492 |
+
"learning_rate": 9.449903167443415e-06,
|
| 493 |
+
"loss": 0.3719,
|
| 494 |
"step": 81
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 0.75,
|
| 498 |
+
"learning_rate": 9.432641751936162e-06,
|
| 499 |
+
"loss": 0.3693,
|
| 500 |
"step": 82
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 0.76,
|
| 504 |
+
"learning_rate": 9.415129951738713e-06,
|
| 505 |
+
"loss": 0.3955,
|
| 506 |
"step": 83
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.77,
|
| 510 |
+
"learning_rate": 9.397368756032445e-06,
|
| 511 |
+
"loss": 0.4196,
|
| 512 |
"step": 84
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 0.78,
|
| 516 |
+
"learning_rate": 9.379359168086231e-06,
|
| 517 |
+
"loss": 0.4527,
|
| 518 |
"step": 85
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 0.79,
|
| 522 |
+
"learning_rate": 9.361102205199762e-06,
|
| 523 |
+
"loss": 0.4054,
|
| 524 |
"step": 86
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.8,
|
| 528 |
+
"learning_rate": 9.34259889864609e-06,
|
| 529 |
+
"loss": 0.4001,
|
| 530 |
"step": 87
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 0.81,
|
| 534 |
+
"learning_rate": 9.32385029361338e-06,
|
| 535 |
+
"loss": 0.4189,
|
| 536 |
"step": 88
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 0.82,
|
| 540 |
+
"learning_rate": 9.304857449145858e-06,
|
| 541 |
+
"loss": 0.439,
|
| 542 |
"step": 89
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 0.83,
|
| 546 |
+
"learning_rate": 9.285621438083997e-06,
|
| 547 |
+
"loss": 0.4221,
|
| 548 |
"step": 90
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.83,
|
| 552 |
+
"learning_rate": 9.26614334700392e-06,
|
| 553 |
+
"loss": 0.4208,
|
| 554 |
"step": 91
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 0.84,
|
| 558 |
+
"learning_rate": 9.246424276156008e-06,
|
| 559 |
+
"loss": 0.4707,
|
| 560 |
"step": 92
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 0.85,
|
| 564 |
+
"learning_rate": 9.226465339402768e-06,
|
| 565 |
+
"loss": 0.3372,
|
| 566 |
"step": 93
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 0.86,
|
| 570 |
+
"learning_rate": 9.206267664155906e-06,
|
| 571 |
+
"loss": 0.3637,
|
| 572 |
"step": 94
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 0.87,
|
| 576 |
+
"learning_rate": 9.185832391312644e-06,
|
| 577 |
+
"loss": 0.4146,
|
| 578 |
"step": 95
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 0.88,
|
| 582 |
+
"learning_rate": 9.165160675191272e-06,
|
| 583 |
+
"loss": 0.393,
|
| 584 |
"step": 96
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 0.89,
|
| 588 |
+
"learning_rate": 9.144253683465953e-06,
|
| 589 |
+
"loss": 0.4712,
|
| 590 |
"step": 97
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.9,
|
| 594 |
+
"learning_rate": 9.123112597100759e-06,
|
| 595 |
+
"loss": 0.3904,
|
| 596 |
"step": 98
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 0.91,
|
| 600 |
+
"learning_rate": 9.101738610282956e-06,
|
| 601 |
+
"loss": 0.4192,
|
| 602 |
"step": 99
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 0.92,
|
| 606 |
+
"learning_rate": 9.080132930355567e-06,
|
| 607 |
+
"loss": 0.4039,
|
| 608 |
"step": 100
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 0.93,
|
| 612 |
+
"learning_rate": 9.058296777749154e-06,
|
| 613 |
+
"loss": 0.4516,
|
| 614 |
"step": 101
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 0.94,
|
| 618 |
+
"learning_rate": 9.03623138591289e-06,
|
| 619 |
+
"loss": 0.3348,
|
| 620 |
"step": 102
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 0.94,
|
| 624 |
+
"learning_rate": 9.013938001244885e-06,
|
| 625 |
+
"loss": 0.4256,
|
| 626 |
"step": 103
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 0.95,
|
| 630 |
+
"learning_rate": 8.99141788302178e-06,
|
| 631 |
+
"loss": 0.3774,
|
| 632 |
"step": 104
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.96,
|
| 636 |
+
"learning_rate": 8.968672303327614e-06,
|
| 637 |
+
"loss": 0.4471,
|
| 638 |
"step": 105
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 0.97,
|
| 642 |
+
"learning_rate": 8.94570254698197e-06,
|
| 643 |
+
"loss": 0.3771,
|
| 644 |
"step": 106
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 0.98,
|
| 648 |
+
"learning_rate": 8.922509911467395e-06,
|
| 649 |
+
"loss": 0.4137,
|
| 650 |
"step": 107
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 0.99,
|
| 654 |
+
"learning_rate": 8.899095706856122e-06,
|
| 655 |
+
"loss": 0.4269,
|
| 656 |
"step": 108
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 1.0,
|
| 660 |
+
"learning_rate": 8.875461255736055e-06,
|
| 661 |
+
"loss": 0.4102,
|
| 662 |
"step": 109
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 1.01,
|
| 666 |
+
"learning_rate": 8.851607893136065e-06,
|
| 667 |
+
"loss": 0.2619,
|
| 668 |
"step": 110
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 1.02,
|
| 672 |
+
"learning_rate": 8.827536966450584e-06,
|
| 673 |
+
"loss": 0.2864,
|
| 674 |
"step": 111
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 1.03,
|
| 678 |
+
"learning_rate": 8.803249835363486e-06,
|
| 679 |
+
"loss": 0.2712,
|
| 680 |
"step": 112
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 1.04,
|
| 684 |
+
"learning_rate": 8.778747871771293e-06,
|
| 685 |
+
"loss": 0.2186,
|
| 686 |
"step": 113
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 1.05,
|
| 690 |
+
"learning_rate": 8.754032459705672e-06,
|
| 691 |
+
"loss": 0.2585,
|
| 692 |
"step": 114
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 1.06,
|
| 696 |
+
"learning_rate": 8.729104995255265e-06,
|
| 697 |
+
"loss": 0.2711,
|
| 698 |
"step": 115
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 1.06,
|
| 702 |
+
"learning_rate": 8.703966886486819e-06,
|
| 703 |
+
"loss": 0.2801,
|
| 704 |
"step": 116
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 1.07,
|
| 708 |
+
"learning_rate": 8.67861955336566e-06,
|
| 709 |
+
"loss": 0.2586,
|
| 710 |
"step": 117
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 1.08,
|
| 714 |
+
"learning_rate": 8.65306442767547e-06,
|
| 715 |
+
"loss": 0.3134,
|
| 716 |
"step": 118
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 1.09,
|
| 720 |
+
"learning_rate": 8.627302952937431e-06,
|
| 721 |
+
"loss": 0.2067,
|
| 722 |
"step": 119
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 1.1,
|
| 726 |
+
"learning_rate": 8.601336584328659e-06,
|
| 727 |
+
"loss": 0.2637,
|
| 728 |
"step": 120
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 1.11,
|
| 732 |
+
"learning_rate": 8.575166788600031e-06,
|
| 733 |
+
"loss": 0.2445,
|
| 734 |
"step": 121
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 1.12,
|
| 738 |
+
"learning_rate": 8.548795043993316e-06,
|
| 739 |
+
"loss": 0.2456,
|
| 740 |
"step": 122
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 1.13,
|
| 744 |
+
"learning_rate": 8.522222840157687e-06,
|
| 745 |
+
"loss": 0.2963,
|
| 746 |
"step": 123
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 1.14,
|
| 750 |
+
"learning_rate": 8.495451678065563e-06,
|
| 751 |
+
"loss": 0.2781,
|
| 752 |
"step": 124
|
| 753 |
},
|
| 754 |
{
|
| 755 |
+
"epoch": 1.15,
|
| 756 |
+
"learning_rate": 8.468483069927832e-06,
|
| 757 |
+
"loss": 0.2572,
|
| 758 |
"step": 125
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 1.16,
|
| 762 |
+
"learning_rate": 8.441318539108433e-06,
|
| 763 |
+
"loss": 0.2473,
|
| 764 |
"step": 126
|
| 765 |
},
|
| 766 |
{
|
| 767 |
+
"epoch": 1.17,
|
| 768 |
+
"learning_rate": 8.413959620038306e-06,
|
| 769 |
+
"loss": 0.1962,
|
| 770 |
"step": 127
|
| 771 |
},
|
| 772 |
{
|
| 773 |
+
"epoch": 1.17,
|
| 774 |
+
"learning_rate": 8.386407858128707e-06,
|
| 775 |
+
"loss": 0.2644,
|
| 776 |
"step": 128
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 1.18,
|
| 780 |
+
"learning_rate": 8.358664809683926e-06,
|
| 781 |
+
"loss": 0.2451,
|
| 782 |
"step": 129
|
| 783 |
},
|
| 784 |
{
|
| 785 |
+
"epoch": 1.19,
|
| 786 |
+
"learning_rate": 8.330732041813367e-06,
|
| 787 |
+
"loss": 0.2801,
|
| 788 |
"step": 130
|
| 789 |
},
|
| 790 |
{
|
| 791 |
+
"epoch": 1.2,
|
| 792 |
+
"learning_rate": 8.302611132343042e-06,
|
| 793 |
+
"loss": 0.2425,
|
| 794 |
"step": 131
|
| 795 |
},
|
| 796 |
{
|
| 797 |
+
"epoch": 1.21,
|
| 798 |
+
"learning_rate": 8.274303669726427e-06,
|
| 799 |
+
"loss": 0.2113,
|
| 800 |
"step": 132
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 1.22,
|
| 804 |
+
"learning_rate": 8.245811252954741e-06,
|
| 805 |
+
"loss": 0.2918,
|
| 806 |
+
"step": 133
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.23,
|
| 810 |
+
"learning_rate": 8.217135491466636e-06,
|
| 811 |
+
"loss": 0.2356,
|
| 812 |
+
"step": 134
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.24,
|
| 816 |
+
"learning_rate": 8.18827800505727e-06,
|
| 817 |
+
"loss": 0.2809,
|
| 818 |
+
"step": 135
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.25,
|
| 822 |
+
"learning_rate": 8.15924042378682e-06,
|
| 823 |
+
"loss": 0.2559,
|
| 824 |
+
"step": 136
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.26,
|
| 828 |
+
"learning_rate": 8.130024387888402e-06,
|
| 829 |
+
"loss": 0.2446,
|
| 830 |
+
"step": 137
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.27,
|
| 834 |
+
"learning_rate": 8.100631547675417e-06,
|
| 835 |
+
"loss": 0.2314,
|
| 836 |
+
"step": 138
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.28,
|
| 840 |
+
"learning_rate": 8.071063563448341e-06,
|
| 841 |
+
"loss": 0.2729,
|
| 842 |
+
"step": 139
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.28,
|
| 846 |
+
"learning_rate": 8.041322105400923e-06,
|
| 847 |
+
"loss": 0.2017,
|
| 848 |
+
"step": 140
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.29,
|
| 852 |
+
"learning_rate": 8.01140885352586e-06,
|
| 853 |
+
"loss": 0.2195,
|
| 854 |
+
"step": 141
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.3,
|
| 858 |
+
"learning_rate": 7.981325497519892e-06,
|
| 859 |
+
"loss": 0.2934,
|
| 860 |
+
"step": 142
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.31,
|
| 864 |
+
"learning_rate": 7.951073736688348e-06,
|
| 865 |
+
"loss": 0.2598,
|
| 866 |
+
"step": 143
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.32,
|
| 870 |
+
"learning_rate": 7.920655279849173e-06,
|
| 871 |
+
"loss": 0.2776,
|
| 872 |
+
"step": 144
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.33,
|
| 876 |
+
"learning_rate": 7.890071845236395e-06,
|
| 877 |
+
"loss": 0.2622,
|
| 878 |
+
"step": 145
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.34,
|
| 882 |
+
"learning_rate": 7.859325160403073e-06,
|
| 883 |
+
"loss": 0.2621,
|
| 884 |
+
"step": 146
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.35,
|
| 888 |
+
"learning_rate": 7.8284169621237e-06,
|
| 889 |
+
"loss": 0.2499,
|
| 890 |
+
"step": 147
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.36,
|
| 894 |
+
"learning_rate": 7.797348996296116e-06,
|
| 895 |
+
"loss": 0.28,
|
| 896 |
+
"step": 148
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.37,
|
| 900 |
+
"learning_rate": 7.766123017842877e-06,
|
| 901 |
+
"loss": 0.1906,
|
| 902 |
+
"step": 149
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.38,
|
| 906 |
+
"learning_rate": 7.734740790612137e-06,
|
| 907 |
+
"loss": 0.2373,
|
| 908 |
+
"step": 150
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.39,
|
| 912 |
+
"learning_rate": 7.703204087277989e-06,
|
| 913 |
+
"loss": 0.2,
|
| 914 |
+
"step": 151
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.39,
|
| 918 |
+
"learning_rate": 7.671514689240366e-06,
|
| 919 |
+
"loss": 0.2678,
|
| 920 |
+
"step": 152
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.4,
|
| 924 |
+
"learning_rate": 7.639674386524395e-06,
|
| 925 |
+
"loss": 0.2295,
|
| 926 |
+
"step": 153
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.41,
|
| 930 |
+
"learning_rate": 7.607684977679284e-06,
|
| 931 |
+
"loss": 0.2417,
|
| 932 |
+
"step": 154
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.42,
|
| 936 |
+
"learning_rate": 7.575548269676741e-06,
|
| 937 |
+
"loss": 0.2662,
|
| 938 |
+
"step": 155
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 1.43,
|
| 942 |
+
"learning_rate": 7.543266077808893e-06,
|
| 943 |
+
"loss": 0.2645,
|
| 944 |
+
"step": 156
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 1.44,
|
| 948 |
+
"learning_rate": 7.510840225585749e-06,
|
| 949 |
+
"loss": 0.2757,
|
| 950 |
+
"step": 157
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.45,
|
| 954 |
+
"learning_rate": 7.478272544632204e-06,
|
| 955 |
+
"loss": 0.2324,
|
| 956 |
+
"step": 158
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.46,
|
| 960 |
+
"learning_rate": 7.44556487458456e-06,
|
| 961 |
+
"loss": 0.2655,
|
| 962 |
+
"step": 159
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.47,
|
| 966 |
+
"learning_rate": 7.412719062986632e-06,
|
| 967 |
+
"loss": 0.2091,
|
| 968 |
+
"step": 160
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.48,
|
| 972 |
+
"learning_rate": 7.379736965185369e-06,
|
| 973 |
+
"loss": 0.2293,
|
| 974 |
+
"step": 161
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.49,
|
| 978 |
+
"learning_rate": 7.3466204442260605e-06,
|
| 979 |
+
"loss": 0.246,
|
| 980 |
+
"step": 162
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 1.5,
|
| 984 |
+
"learning_rate": 7.313371370747104e-06,
|
| 985 |
+
"loss": 0.2362,
|
| 986 |
+
"step": 163
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.5,
|
| 990 |
+
"learning_rate": 7.279991622874319e-06,
|
| 991 |
+
"loss": 0.2074,
|
| 992 |
+
"step": 164
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.51,
|
| 996 |
+
"learning_rate": 7.24648308611489e-06,
|
| 997 |
+
"loss": 0.2035,
|
| 998 |
+
"step": 165
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.52,
|
| 1002 |
+
"learning_rate": 7.212847653250828e-06,
|
| 1003 |
+
"loss": 0.3013,
|
| 1004 |
+
"step": 166
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.53,
|
| 1008 |
+
"learning_rate": 7.1790872242320775e-06,
|
| 1009 |
+
"loss": 0.2286,
|
| 1010 |
+
"step": 167
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.54,
|
| 1014 |
+
"learning_rate": 7.145203706069183e-06,
|
| 1015 |
+
"loss": 0.2005,
|
| 1016 |
+
"step": 168
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.55,
|
| 1020 |
+
"learning_rate": 7.1111990127255684e-06,
|
| 1021 |
+
"loss": 0.2351,
|
| 1022 |
+
"step": 169
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.56,
|
| 1026 |
+
"learning_rate": 7.0770750650094335e-06,
|
| 1027 |
+
"loss": 0.2378,
|
| 1028 |
+
"step": 170
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.57,
|
| 1032 |
+
"learning_rate": 7.042833790465241e-06,
|
| 1033 |
+
"loss": 0.2557,
|
| 1034 |
+
"step": 171
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.58,
|
| 1038 |
+
"learning_rate": 7.008477123264849e-06,
|
| 1039 |
+
"loss": 0.1742,
|
| 1040 |
+
"step": 172
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.59,
|
| 1044 |
+
"learning_rate": 6.974007004098243e-06,
|
| 1045 |
+
"loss": 0.2862,
|
| 1046 |
+
"step": 173
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.6,
|
| 1050 |
+
"learning_rate": 6.939425380063924e-06,
|
| 1051 |
+
"loss": 0.2359,
|
| 1052 |
+
"step": 174
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.61,
|
| 1056 |
+
"learning_rate": 6.9047342045589224e-06,
|
| 1057 |
+
"loss": 0.2646,
|
| 1058 |
+
"step": 175
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.61,
|
| 1062 |
+
"learning_rate": 6.869935437168449e-06,
|
| 1063 |
+
"loss": 0.2223,
|
| 1064 |
+
"step": 176
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 1.62,
|
| 1068 |
+
"learning_rate": 6.835031043555211e-06,
|
| 1069 |
+
"loss": 0.2292,
|
| 1070 |
+
"step": 177
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 1.63,
|
| 1074 |
+
"learning_rate": 6.800022995348381e-06,
|
| 1075 |
+
"loss": 0.2395,
|
| 1076 |
+
"step": 178
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.64,
|
| 1080 |
+
"learning_rate": 6.76491327003222e-06,
|
| 1081 |
+
"loss": 0.2201,
|
| 1082 |
+
"step": 179
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.65,
|
| 1086 |
+
"learning_rate": 6.729703850834381e-06,
|
| 1087 |
+
"loss": 0.2071,
|
| 1088 |
+
"step": 180
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.66,
|
| 1092 |
+
"learning_rate": 6.694396726613883e-06,
|
| 1093 |
+
"loss": 0.2461,
|
| 1094 |
+
"step": 181
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.67,
|
| 1098 |
+
"learning_rate": 6.65899389174876e-06,
|
| 1099 |
+
"loss": 0.2804,
|
| 1100 |
+
"step": 182
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.68,
|
| 1104 |
+
"learning_rate": 6.6234973460234184e-06,
|
| 1105 |
+
"loss": 0.2859,
|
| 1106 |
+
"step": 183
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 1.69,
|
| 1110 |
+
"learning_rate": 6.587909094515663e-06,
|
| 1111 |
+
"loss": 0.2358,
|
| 1112 |
+
"step": 184
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.7,
|
| 1116 |
+
"learning_rate": 6.552231147483448e-06,
|
| 1117 |
+
"loss": 0.229,
|
| 1118 |
+
"step": 185
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.71,
|
| 1122 |
+
"learning_rate": 6.5164655202513135e-06,
|
| 1123 |
+
"loss": 0.2673,
|
| 1124 |
+
"step": 186
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.72,
|
| 1128 |
+
"learning_rate": 6.480614233096558e-06,
|
| 1129 |
+
"loss": 0.2169,
|
| 1130 |
+
"step": 187
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.72,
|
| 1134 |
+
"learning_rate": 6.444679311135112e-06,
|
| 1135 |
+
"loss": 0.1908,
|
| 1136 |
+
"step": 188
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.73,
|
| 1140 |
+
"learning_rate": 6.408662784207149e-06,
|
| 1141 |
+
"loss": 0.2167,
|
| 1142 |
+
"step": 189
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.74,
|
| 1146 |
+
"learning_rate": 6.372566686762427e-06,
|
| 1147 |
+
"loss": 0.2533,
|
| 1148 |
+
"step": 190
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 1.75,
|
| 1152 |
+
"learning_rate": 6.336393057745365e-06,
|
| 1153 |
+
"loss": 0.2361,
|
| 1154 |
+
"step": 191
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.76,
|
| 1158 |
+
"learning_rate": 6.300143940479881e-06,
|
| 1159 |
+
"loss": 0.2396,
|
| 1160 |
+
"step": 192
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.77,
|
| 1164 |
+
"learning_rate": 6.2638213825539595e-06,
|
| 1165 |
+
"loss": 0.2716,
|
| 1166 |
+
"step": 193
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.78,
|
| 1170 |
+
"learning_rate": 6.227427435703997e-06,
|
| 1171 |
+
"loss": 0.2187,
|
| 1172 |
+
"step": 194
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.79,
|
| 1176 |
+
"learning_rate": 6.190964155698903e-06,
|
| 1177 |
+
"loss": 0.2419,
|
| 1178 |
+
"step": 195
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.8,
|
| 1182 |
+
"learning_rate": 6.154433602223979e-06,
|
| 1183 |
+
"loss": 0.2547,
|
| 1184 |
+
"step": 196
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 1.81,
|
| 1188 |
+
"learning_rate": 6.117837838764579e-06,
|
| 1189 |
+
"loss": 0.2534,
|
| 1190 |
+
"step": 197
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 1.82,
|
| 1194 |
+
"learning_rate": 6.0811789324895365e-06,
|
| 1195 |
+
"loss": 0.255,
|
| 1196 |
+
"step": 198
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 1.83,
|
| 1200 |
+
"learning_rate": 6.044458954134411e-06,
|
| 1201 |
+
"loss": 0.2383,
|
| 1202 |
+
"step": 199
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 1.83,
|
| 1206 |
+
"learning_rate": 6.0076799778845105e-06,
|
| 1207 |
+
"loss": 0.2407,
|
| 1208 |
+
"step": 200
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.84,
|
| 1212 |
+
"learning_rate": 5.970844081257734e-06,
|
| 1213 |
+
"loss": 0.1883,
|
| 1214 |
+
"step": 201
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.85,
|
| 1218 |
+
"learning_rate": 5.933953344987215e-06,
|
| 1219 |
+
"loss": 0.3112,
|
| 1220 |
+
"step": 202
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.86,
|
| 1224 |
+
"learning_rate": 5.897009852903792e-06,
|
| 1225 |
+
"loss": 0.2424,
|
| 1226 |
+
"step": 203
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 1.87,
|
| 1230 |
+
"learning_rate": 5.860015691818292e-06,
|
| 1231 |
+
"loss": 0.2663,
|
| 1232 |
+
"step": 204
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.88,
|
| 1236 |
+
"learning_rate": 5.82297295140367e-06,
|
| 1237 |
+
"loss": 0.2616,
|
| 1238 |
+
"step": 205
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 1.89,
|
| 1242 |
+
"learning_rate": 5.78588372407695e-06,
|
| 1243 |
+
"loss": 0.2067,
|
| 1244 |
+
"step": 206
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 1.9,
|
| 1248 |
+
"learning_rate": 5.748750104881051e-06,
|
| 1249 |
+
"loss": 0.2092,
|
| 1250 |
+
"step": 207
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.91,
|
| 1254 |
+
"learning_rate": 5.711574191366427e-06,
|
| 1255 |
+
"loss": 0.2383,
|
| 1256 |
+
"step": 208
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.92,
|
| 1260 |
+
"learning_rate": 5.674358083472598e-06,
|
| 1261 |
+
"loss": 0.2377,
|
| 1262 |
+
"step": 209
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.93,
|
| 1266 |
+
"learning_rate": 5.637103883409525e-06,
|
| 1267 |
+
"loss": 0.2415,
|
| 1268 |
+
"step": 210
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 1.94,
|
| 1272 |
+
"learning_rate": 5.599813695538866e-06,
|
| 1273 |
+
"loss": 0.2478,
|
| 1274 |
+
"step": 211
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.94,
|
| 1278 |
+
"learning_rate": 5.562489626255104e-06,
|
| 1279 |
+
"loss": 0.2752,
|
| 1280 |
+
"step": 212
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.95,
|
| 1284 |
+
"learning_rate": 5.52513378386657e-06,
|
| 1285 |
+
"loss": 0.2133,
|
| 1286 |
+
"step": 213
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.96,
|
| 1290 |
+
"learning_rate": 5.487748278476342e-06,
|
| 1291 |
+
"loss": 0.2043,
|
| 1292 |
+
"step": 214
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.97,
|
| 1296 |
+
"learning_rate": 5.450335221863068e-06,
|
| 1297 |
+
"loss": 0.2664,
|
| 1298 |
+
"step": 215
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.98,
|
| 1302 |
+
"learning_rate": 5.412896727361663e-06,
|
| 1303 |
+
"loss": 0.2617,
|
| 1304 |
+
"step": 216
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.99,
|
| 1308 |
+
"learning_rate": 5.375434909743942e-06,
|
| 1309 |
+
"loss": 0.2001,
|
| 1310 |
+
"step": 217
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 2.0,
|
| 1314 |
+
"learning_rate": 5.337951885099167e-06,
|
| 1315 |
+
"loss": 0.2353,
|
| 1316 |
+
"step": 218
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 2.01,
|
| 1320 |
+
"learning_rate": 5.300449770714502e-06,
|
| 1321 |
+
"loss": 0.112,
|
| 1322 |
+
"step": 219
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.02,
|
| 1326 |
+
"learning_rate": 5.262930684955439e-06,
|
| 1327 |
+
"loss": 0.1022,
|
| 1328 |
+
"step": 220
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 2.03,
|
| 1332 |
+
"learning_rate": 5.225396747146112e-06,
|
| 1333 |
+
"loss": 0.1263,
|
| 1334 |
+
"step": 221
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.04,
|
| 1338 |
+
"learning_rate": 5.187850077449604e-06,
|
| 1339 |
+
"loss": 0.1059,
|
| 1340 |
+
"step": 222
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.05,
|
| 1344 |
+
"learning_rate": 5.150292796748174e-06,
|
| 1345 |
+
"loss": 0.1101,
|
| 1346 |
+
"step": 223
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.06,
|
| 1350 |
+
"learning_rate": 5.112727026523461e-06,
|
| 1351 |
+
"loss": 0.1274,
|
| 1352 |
+
"step": 224
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.06,
|
| 1356 |
+
"learning_rate": 5.075154888736653e-06,
|
| 1357 |
+
"loss": 0.1604,
|
| 1358 |
+
"step": 225
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 2.07,
|
| 1362 |
+
"learning_rate": 5.03757850570861e-06,
|
| 1363 |
+
"loss": 0.1067,
|
| 1364 |
+
"step": 226
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 2.08,
|
| 1368 |
+
"learning_rate": 5e-06,
|
| 1369 |
+
"loss": 0.1124,
|
| 1370 |
+
"step": 227
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.09,
|
| 1374 |
+
"learning_rate": 4.9624214942913916e-06,
|
| 1375 |
+
"loss": 0.0976,
|
| 1376 |
+
"step": 228
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.1,
|
| 1380 |
+
"learning_rate": 4.924845111263349e-06,
|
| 1381 |
+
"loss": 0.1094,
|
| 1382 |
+
"step": 229
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.11,
|
| 1386 |
+
"learning_rate": 4.88727297347654e-06,
|
| 1387 |
+
"loss": 0.1156,
|
| 1388 |
+
"step": 230
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.12,
|
| 1392 |
+
"learning_rate": 4.8497072032518274e-06,
|
| 1393 |
+
"loss": 0.1201,
|
| 1394 |
+
"step": 231
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 2.13,
|
| 1398 |
+
"learning_rate": 4.8121499225503974e-06,
|
| 1399 |
+
"loss": 0.1053,
|
| 1400 |
+
"step": 232
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 2.14,
|
| 1404 |
+
"learning_rate": 4.774603252853889e-06,
|
| 1405 |
+
"loss": 0.1344,
|
| 1406 |
+
"step": 233
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.15,
|
| 1410 |
+
"learning_rate": 4.737069315044562e-06,
|
| 1411 |
+
"loss": 0.1171,
|
| 1412 |
+
"step": 234
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.16,
|
| 1416 |
+
"learning_rate": 4.699550229285499e-06,
|
| 1417 |
+
"loss": 0.0906,
|
| 1418 |
+
"step": 235
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.17,
|
| 1422 |
+
"learning_rate": 4.662048114900837e-06,
|
| 1423 |
+
"loss": 0.1313,
|
| 1424 |
+
"step": 236
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.17,
|
| 1428 |
+
"learning_rate": 4.624565090256059e-06,
|
| 1429 |
+
"loss": 0.1096,
|
| 1430 |
+
"step": 237
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.18,
|
| 1434 |
+
"learning_rate": 4.587103272638339e-06,
|
| 1435 |
+
"loss": 0.1185,
|
| 1436 |
+
"step": 238
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 2.19,
|
| 1440 |
+
"learning_rate": 4.549664778136933e-06,
|
| 1441 |
+
"loss": 0.1372,
|
| 1442 |
+
"step": 239
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.2,
|
| 1446 |
+
"learning_rate": 4.512251721523659e-06,
|
| 1447 |
+
"loss": 0.1042,
|
| 1448 |
+
"step": 240
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 2.21,
|
| 1452 |
+
"learning_rate": 4.4748662161334335e-06,
|
| 1453 |
+
"loss": 0.1185,
|
| 1454 |
+
"step": 241
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.22,
|
| 1458 |
+
"learning_rate": 4.437510373744897e-06,
|
| 1459 |
+
"loss": 0.1448,
|
| 1460 |
+
"step": 242
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.23,
|
| 1464 |
+
"learning_rate": 4.400186304461136e-06,
|
| 1465 |
+
"loss": 0.113,
|
| 1466 |
+
"step": 243
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 2.24,
|
| 1470 |
+
"learning_rate": 4.362896116590475e-06,
|
| 1471 |
+
"loss": 0.1211,
|
| 1472 |
+
"step": 244
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.25,
|
| 1476 |
+
"learning_rate": 4.325641916527405e-06,
|
| 1477 |
+
"loss": 0.132,
|
| 1478 |
+
"step": 245
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 2.26,
|
| 1482 |
+
"learning_rate": 4.2884258086335755e-06,
|
| 1483 |
+
"loss": 0.1105,
|
| 1484 |
+
"step": 246
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 2.27,
|
| 1488 |
+
"learning_rate": 4.25124989511895e-06,
|
| 1489 |
+
"loss": 0.0987,
|
| 1490 |
+
"step": 247
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 2.28,
|
| 1494 |
+
"learning_rate": 4.214116275923051e-06,
|
| 1495 |
+
"loss": 0.1197,
|
| 1496 |
+
"step": 248
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.28,
|
| 1500 |
+
"learning_rate": 4.17702704859633e-06,
|
| 1501 |
+
"loss": 0.1334,
|
| 1502 |
+
"step": 249
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.29,
|
| 1506 |
+
"learning_rate": 4.1399843081817085e-06,
|
| 1507 |
+
"loss": 0.1468,
|
| 1508 |
+
"step": 250
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.3,
|
| 1512 |
+
"learning_rate": 4.1029901470962105e-06,
|
| 1513 |
+
"loss": 0.1161,
|
| 1514 |
+
"step": 251
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.31,
|
| 1518 |
+
"learning_rate": 4.066046655012786e-06,
|
| 1519 |
+
"loss": 0.1106,
|
| 1520 |
+
"step": 252
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 2.32,
|
| 1524 |
+
"learning_rate": 4.029155918742268e-06,
|
| 1525 |
+
"loss": 0.1234,
|
| 1526 |
+
"step": 253
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 2.33,
|
| 1530 |
+
"learning_rate": 3.992320022115492e-06,
|
| 1531 |
+
"loss": 0.1261,
|
| 1532 |
+
"step": 254
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.34,
|
| 1536 |
+
"learning_rate": 3.955541045865591e-06,
|
| 1537 |
+
"loss": 0.1011,
|
| 1538 |
+
"step": 255
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.35,
|
| 1542 |
+
"learning_rate": 3.918821067510464e-06,
|
| 1543 |
+
"loss": 0.1058,
|
| 1544 |
+
"step": 256
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.36,
|
| 1548 |
+
"learning_rate": 3.882162161235421e-06,
|
| 1549 |
+
"loss": 0.1002,
|
| 1550 |
+
"step": 257
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.37,
|
| 1554 |
+
"learning_rate": 3.845566397776022e-06,
|
| 1555 |
+
"loss": 0.1023,
|
| 1556 |
+
"step": 258
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.38,
|
| 1560 |
+
"learning_rate": 3.8090358443010993e-06,
|
| 1561 |
+
"loss": 0.1174,
|
| 1562 |
+
"step": 259
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 2.39,
|
| 1566 |
+
"learning_rate": 3.7725725642960047e-06,
|
| 1567 |
+
"loss": 0.111,
|
| 1568 |
+
"step": 260
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.39,
|
| 1572 |
+
"learning_rate": 3.7361786174460414e-06,
|
| 1573 |
+
"loss": 0.0948,
|
| 1574 |
+
"step": 261
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.4,
|
| 1578 |
+
"learning_rate": 3.6998560595201188e-06,
|
| 1579 |
+
"loss": 0.0972,
|
| 1580 |
+
"step": 262
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.41,
|
| 1584 |
+
"learning_rate": 3.6636069422546363e-06,
|
| 1585 |
+
"loss": 0.0982,
|
| 1586 |
+
"step": 263
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.42,
|
| 1590 |
+
"learning_rate": 3.627433313237576e-06,
|
| 1591 |
+
"loss": 0.1374,
|
| 1592 |
+
"step": 264
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.43,
|
| 1596 |
+
"learning_rate": 3.5913372157928515e-06,
|
| 1597 |
+
"loss": 0.0988,
|
| 1598 |
+
"step": 265
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.44,
|
| 1602 |
+
"learning_rate": 3.555320688864889e-06,
|
| 1603 |
+
"loss": 0.1281,
|
| 1604 |
+
"step": 266
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.45,
|
| 1608 |
+
"learning_rate": 3.519385766903442e-06,
|
| 1609 |
+
"loss": 0.1273,
|
| 1610 |
+
"step": 267
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 2.46,
|
| 1614 |
+
"learning_rate": 3.483534479748688e-06,
|
| 1615 |
+
"loss": 0.1256,
|
| 1616 |
+
"step": 268
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 2.47,
|
| 1620 |
+
"learning_rate": 3.447768852516554e-06,
|
| 1621 |
+
"loss": 0.1131,
|
| 1622 |
+
"step": 269
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 2.48,
|
| 1626 |
+
"learning_rate": 3.4120909054843375e-06,
|
| 1627 |
+
"loss": 0.0974,
|
| 1628 |
+
"step": 270
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.49,
|
| 1632 |
+
"learning_rate": 3.3765026539765832e-06,
|
| 1633 |
+
"loss": 0.1233,
|
| 1634 |
+
"step": 271
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.5,
|
| 1638 |
+
"learning_rate": 3.3410061082512422e-06,
|
| 1639 |
+
"loss": 0.1175,
|
| 1640 |
+
"step": 272
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.5,
|
| 1644 |
+
"learning_rate": 3.3056032733861188e-06,
|
| 1645 |
+
"loss": 0.1178,
|
| 1646 |
+
"step": 273
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.51,
|
| 1650 |
+
"learning_rate": 3.2702961491656197e-06,
|
| 1651 |
+
"loss": 0.11,
|
| 1652 |
+
"step": 274
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 2.52,
|
| 1656 |
+
"learning_rate": 3.2350867299677802e-06,
|
| 1657 |
+
"loss": 0.1091,
|
| 1658 |
+
"step": 275
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.53,
|
| 1662 |
+
"learning_rate": 3.1999770046516198e-06,
|
| 1663 |
+
"loss": 0.1146,
|
| 1664 |
+
"step": 276
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.54,
|
| 1668 |
+
"learning_rate": 3.164968956444791e-06,
|
| 1669 |
+
"loss": 0.0945,
|
| 1670 |
+
"step": 277
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.55,
|
| 1674 |
+
"learning_rate": 3.130064562831553e-06,
|
| 1675 |
+
"loss": 0.1216,
|
| 1676 |
+
"step": 278
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.56,
|
| 1680 |
+
"learning_rate": 3.0952657954410792e-06,
|
| 1681 |
+
"loss": 0.1273,
|
| 1682 |
+
"step": 279
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.57,
|
| 1686 |
+
"learning_rate": 3.0605746199360755e-06,
|
| 1687 |
+
"loss": 0.1161,
|
| 1688 |
+
"step": 280
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 2.58,
|
| 1692 |
+
"learning_rate": 3.0259929959017585e-06,
|
| 1693 |
+
"loss": 0.1395,
|
| 1694 |
+
"step": 281
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 2.59,
|
| 1698 |
+
"learning_rate": 2.991522876735154e-06,
|
| 1699 |
+
"loss": 0.1224,
|
| 1700 |
+
"step": 282
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.6,
|
| 1704 |
+
"learning_rate": 2.95716620953476e-06,
|
| 1705 |
+
"loss": 0.1052,
|
| 1706 |
+
"step": 283
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 2.61,
|
| 1710 |
+
"learning_rate": 2.9229249349905686e-06,
|
| 1711 |
+
"loss": 0.1383,
|
| 1712 |
+
"step": 284
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 2.61,
|
| 1716 |
+
"learning_rate": 2.8888009872744332e-06,
|
| 1717 |
+
"loss": 0.0985,
|
| 1718 |
+
"step": 285
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.62,
|
| 1722 |
+
"learning_rate": 2.8547962939308187e-06,
|
| 1723 |
+
"loss": 0.1164,
|
| 1724 |
+
"step": 286
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.63,
|
| 1728 |
+
"learning_rate": 2.8209127757679246e-06,
|
| 1729 |
+
"loss": 0.095,
|
| 1730 |
+
"step": 287
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 2.64,
|
| 1734 |
+
"learning_rate": 2.787152346749173e-06,
|
| 1735 |
+
"loss": 0.1308,
|
| 1736 |
+
"step": 288
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 2.65,
|
| 1740 |
+
"learning_rate": 2.7535169138851124e-06,
|
| 1741 |
+
"loss": 0.1336,
|
| 1742 |
+
"step": 289
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 2.66,
|
| 1746 |
+
"learning_rate": 2.720008377125682e-06,
|
| 1747 |
+
"loss": 0.0952,
|
| 1748 |
+
"step": 290
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 2.67,
|
| 1752 |
+
"learning_rate": 2.686628629252899e-06,
|
| 1753 |
+
"loss": 0.134,
|
| 1754 |
+
"step": 291
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 2.68,
|
| 1758 |
+
"learning_rate": 2.6533795557739407e-06,
|
| 1759 |
+
"loss": 0.1008,
|
| 1760 |
+
"step": 292
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.69,
|
| 1764 |
+
"learning_rate": 2.6202630348146323e-06,
|
| 1765 |
+
"loss": 0.0938,
|
| 1766 |
+
"step": 293
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 2.7,
|
| 1770 |
+
"learning_rate": 2.5872809370133704e-06,
|
| 1771 |
+
"loss": 0.1146,
|
| 1772 |
+
"step": 294
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 2.71,
|
| 1776 |
+
"learning_rate": 2.5544351254154407e-06,
|
| 1777 |
+
"loss": 0.0909,
|
| 1778 |
+
"step": 295
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 2.72,
|
| 1782 |
+
"learning_rate": 2.5217274553677975e-06,
|
| 1783 |
+
"loss": 0.1054,
|
| 1784 |
+
"step": 296
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 2.72,
|
| 1788 |
+
"learning_rate": 2.489159774414252e-06,
|
| 1789 |
+
"loss": 0.0909,
|
| 1790 |
+
"step": 297
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 2.73,
|
| 1794 |
+
"learning_rate": 2.4567339221911086e-06,
|
| 1795 |
+
"loss": 0.0918,
|
| 1796 |
+
"step": 298
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 2.74,
|
| 1800 |
+
"learning_rate": 2.424451730323261e-06,
|
| 1801 |
+
"loss": 0.1043,
|
| 1802 |
+
"step": 299
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.75,
|
| 1806 |
+
"learning_rate": 2.3923150223207176e-06,
|
| 1807 |
+
"loss": 0.1139,
|
| 1808 |
+
"step": 300
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.76,
|
| 1812 |
+
"learning_rate": 2.3603256134756066e-06,
|
| 1813 |
+
"loss": 0.0958,
|
| 1814 |
+
"step": 301
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 2.77,
|
| 1818 |
+
"learning_rate": 2.328485310759635e-06,
|
| 1819 |
+
"loss": 0.1115,
|
| 1820 |
+
"step": 302
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 2.78,
|
| 1824 |
+
"learning_rate": 2.296795912722014e-06,
|
| 1825 |
+
"loss": 0.092,
|
| 1826 |
+
"step": 303
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 2.79,
|
| 1830 |
+
"learning_rate": 2.265259209387867e-06,
|
| 1831 |
+
"loss": 0.1011,
|
| 1832 |
+
"step": 304
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 2.8,
|
| 1836 |
+
"learning_rate": 2.2338769821571225e-06,
|
| 1837 |
+
"loss": 0.106,
|
| 1838 |
+
"step": 305
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.81,
|
| 1842 |
+
"learning_rate": 2.202651003703885e-06,
|
| 1843 |
+
"loss": 0.1053,
|
| 1844 |
+
"step": 306
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.82,
|
| 1848 |
+
"learning_rate": 2.1715830378763025e-06,
|
| 1849 |
+
"loss": 0.1314,
|
| 1850 |
+
"step": 307
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.83,
|
| 1854 |
+
"learning_rate": 2.140674839596931e-06,
|
| 1855 |
+
"loss": 0.1252,
|
| 1856 |
+
"step": 308
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 2.83,
|
| 1860 |
+
"learning_rate": 2.109928154763606e-06,
|
| 1861 |
+
"loss": 0.1032,
|
| 1862 |
+
"step": 309
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 2.84,
|
| 1866 |
+
"learning_rate": 2.0793447201508288e-06,
|
| 1867 |
+
"loss": 0.0911,
|
| 1868 |
+
"step": 310
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 2.85,
|
| 1872 |
+
"learning_rate": 2.0489262633116536e-06,
|
| 1873 |
+
"loss": 0.0865,
|
| 1874 |
+
"step": 311
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 2.86,
|
| 1878 |
+
"learning_rate": 2.01867450248011e-06,
|
| 1879 |
+
"loss": 0.1087,
|
| 1880 |
+
"step": 312
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.87,
|
| 1884 |
+
"learning_rate": 1.9885911464741413e-06,
|
| 1885 |
+
"loss": 0.1136,
|
| 1886 |
+
"step": 313
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.88,
|
| 1890 |
+
"learning_rate": 1.9586778945990785e-06,
|
| 1891 |
+
"loss": 0.0873,
|
| 1892 |
+
"step": 314
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.89,
|
| 1896 |
+
"learning_rate": 1.928936436551661e-06,
|
| 1897 |
+
"loss": 0.1155,
|
| 1898 |
+
"step": 315
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 2.9,
|
| 1902 |
+
"learning_rate": 1.8993684523245842e-06,
|
| 1903 |
+
"loss": 0.117,
|
| 1904 |
+
"step": 316
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 2.91,
|
| 1908 |
+
"learning_rate": 1.8699756121115997e-06,
|
| 1909 |
+
"loss": 0.1083,
|
| 1910 |
+
"step": 317
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 2.92,
|
| 1914 |
+
"learning_rate": 1.8407595762131814e-06,
|
| 1915 |
+
"loss": 0.1099,
|
| 1916 |
+
"step": 318
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 2.93,
|
| 1920 |
+
"learning_rate": 1.811721994942731e-06,
|
| 1921 |
+
"loss": 0.1183,
|
| 1922 |
+
"step": 319
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 2.94,
|
| 1926 |
+
"learning_rate": 1.7828645085333645e-06,
|
| 1927 |
+
"loss": 0.1168,
|
| 1928 |
+
"step": 320
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.94,
|
| 1932 |
+
"learning_rate": 1.7541887470452606e-06,
|
| 1933 |
+
"loss": 0.0985,
|
| 1934 |
+
"step": 321
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.95,
|
| 1938 |
+
"learning_rate": 1.7256963302735752e-06,
|
| 1939 |
+
"loss": 0.1012,
|
| 1940 |
+
"step": 322
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.96,
|
| 1944 |
+
"learning_rate": 1.6973888676569594e-06,
|
| 1945 |
+
"loss": 0.1107,
|
| 1946 |
+
"step": 323
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 2.97,
|
| 1950 |
+
"learning_rate": 1.6692679581866334e-06,
|
| 1951 |
+
"loss": 0.1301,
|
| 1952 |
+
"step": 324
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 2.98,
|
| 1956 |
+
"learning_rate": 1.6413351903160763e-06,
|
| 1957 |
+
"loss": 0.1022,
|
| 1958 |
+
"step": 325
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 2.99,
|
| 1962 |
+
"learning_rate": 1.6135921418712959e-06,
|
| 1963 |
+
"loss": 0.1213,
|
| 1964 |
+
"step": 326
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 3.0,
|
| 1968 |
+
"learning_rate": 1.5860403799616951e-06,
|
| 1969 |
+
"loss": 0.1093,
|
| 1970 |
+
"step": 327
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 3.01,
|
| 1974 |
+
"learning_rate": 1.5586814608915673e-06,
|
| 1975 |
+
"loss": 0.0554,
|
| 1976 |
+
"step": 328
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 3.02,
|
| 1980 |
+
"learning_rate": 1.5315169300721694e-06,
|
| 1981 |
+
"loss": 0.084,
|
| 1982 |
+
"step": 329
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 3.03,
|
| 1986 |
+
"learning_rate": 1.5045483219344387e-06,
|
| 1987 |
+
"loss": 0.0741,
|
| 1988 |
+
"step": 330
|
| 1989 |
}
|
| 1990 |
],
|
| 1991 |
+
"max_steps": 436,
|
| 1992 |
+
"num_train_epochs": 4,
|
| 1993 |
+
"total_flos": 1.2638375430617825e+18,
|
| 1994 |
"trial_name": null,
|
| 1995 |
"trial_params": null
|
| 1996 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:230e2de2b3c1ee9465cc7ecb9d3d138cecd7230b0261d72773fc5adf1461a099
|
| 3 |
size 4091
|