Training in progress, epoch 1, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -16,8 +16,8 @@
|
|
| 16 |
"rank_pattern": {},
|
| 17 |
"revision": null,
|
| 18 |
"target_modules": [
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
],
|
| 22 |
"task_type": "SEQ_2_SEQ_LM"
|
| 23 |
}
|
|
|
|
| 16 |
"rank_pattern": {},
|
| 17 |
"revision": null,
|
| 18 |
"target_modules": [
|
| 19 |
+
"v",
|
| 20 |
+
"q"
|
| 21 |
],
|
| 22 |
"task_type": "SEQ_2_SEQ_LM"
|
| 23 |
}
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 18915040
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bd1821877d7d29077d85fe04dfea93e7292825d05231ebab51a1e126697c743
|
| 3 |
size 18915040
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2603258
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f41766bc78b6ec0c8e40003000365c7287487226ebb9545f086a8ab4a75b48ae
|
| 3 |
size 2603258
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8f8922c352ab63d1593bcb944e77d461472df192dcdc1202984e9e08f16a111
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c80ef793f3bbade49954a24290acb44d3ed9d4cc93da4ad48634f3740d8739a
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,644 +1,169 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 2.
|
| 3 |
-
"best_model_checkpoint": "dq158/pingusPongus/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": 0.02,
|
| 13 |
-
"learning_rate": 0.0004,
|
| 14 |
-
"loss": 3.2511,
|
| 15 |
-
"step": 500
|
| 16 |
-
},
|
| 17 |
{
|
| 18 |
"epoch": 0.04,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
-
"step":
|
| 22 |
-
},
|
| 23 |
-
{
|
| 24 |
-
"epoch": 0.06,
|
| 25 |
-
"learning_rate": 0.00039999828328897294,
|
| 26 |
-
"loss": 2.7835,
|
| 27 |
-
"step": 1500
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.08,
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss": 2.
|
| 33 |
-
"step":
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"epoch": 0.1,
|
| 37 |
-
"learning_rate": 0.00039999313318536277,
|
| 38 |
-
"loss": 2.6816,
|
| 39 |
-
"step": 2500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.12,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 2.
|
| 45 |
-
"step":
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.14,
|
| 49 |
-
"learning_rate": 0.00039998454977758173,
|
| 50 |
-
"loss": 2.7713,
|
| 51 |
-
"step": 3500
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.16,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss": 2.
|
| 57 |
-
"step":
|
| 58 |
-
},
|
| 59 |
-
{
|
| 60 |
-
"epoch": 0.18,
|
| 61 |
-
"learning_rate": 0.0003999725332129823,
|
| 62 |
-
"loss": 2.6873,
|
| 63 |
-
"step": 4500
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.2,
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss": 2.
|
| 69 |
-
"step":
|
| 70 |
-
},
|
| 71 |
-
{
|
| 72 |
-
"epoch": 0.22,
|
| 73 |
-
"learning_rate": 0.00039995708369785403,
|
| 74 |
-
"loss": 2.6328,
|
| 75 |
-
"step": 5500
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.24,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 2.
|
| 81 |
-
"step":
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 0.26,
|
| 85 |
-
"learning_rate": 0.0003999382014974206,
|
| 86 |
-
"loss": 2.581,
|
| 87 |
-
"step": 6500
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.28,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 2.
|
| 93 |
-
"step":
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.3,
|
| 97 |
-
"learning_rate": 0.0003999158869358347,
|
| 98 |
-
"loss": 2.6095,
|
| 99 |
-
"step": 7500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.32,
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss": 2.
|
| 105 |
-
"step":
|
| 106 |
-
},
|
| 107 |
-
{
|
| 108 |
-
"epoch": 0.34,
|
| 109 |
-
"learning_rate": 0.00039989014039617295,
|
| 110 |
-
"loss": 2.6528,
|
| 111 |
-
"step": 8500
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.36,
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 2.
|
| 117 |
-
"step":
|
| 118 |
-
},
|
| 119 |
-
{
|
| 120 |
-
"epoch": 0.38,
|
| 121 |
-
"learning_rate": 0.000399860962320429,
|
| 122 |
-
"loss": 2.6032,
|
| 123 |
-
"step": 9500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.4,
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 2.
|
| 129 |
-
"step":
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.42,
|
| 133 |
-
"learning_rate": 0.0003998283532095061,
|
| 134 |
-
"loss": 2.584,
|
| 135 |
-
"step": 10500
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.43,
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss": 2.
|
| 141 |
-
"step":
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 0.45,
|
| 145 |
-
"learning_rate": 0.00039979231362320845,
|
| 146 |
-
"loss": 2.6264,
|
| 147 |
-
"step": 11500
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 0.47,
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 2.
|
| 153 |
-
"step":
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"epoch": 0.49,
|
| 157 |
-
"learning_rate": 0.0003997528441802316,
|
| 158 |
-
"loss": 2.5442,
|
| 159 |
-
"step": 12500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 0.51,
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss": 2.
|
| 165 |
-
"step":
|
| 166 |
-
},
|
| 167 |
-
{
|
| 168 |
-
"epoch": 0.53,
|
| 169 |
-
"learning_rate": 0.0003997099455581518,
|
| 170 |
-
"loss": 2.5825,
|
| 171 |
-
"step": 13500
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 0.55,
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss": 2.
|
| 177 |
-
"step":
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 0.57,
|
| 181 |
-
"learning_rate": 0.0003996636184934145,
|
| 182 |
-
"loss": 2.5633,
|
| 183 |
-
"step": 14500
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 0.59,
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 2.
|
| 189 |
-
"step":
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 0.61,
|
| 193 |
-
"learning_rate": 0.0003996138637813214,
|
| 194 |
-
"loss": 2.5991,
|
| 195 |
-
"step": 15500
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 0.63,
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss": 2.
|
| 201 |
-
"step":
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 0.65,
|
| 205 |
-
"learning_rate": 0.00039956068227601734,
|
| 206 |
-
"loss": 2.5436,
|
| 207 |
-
"step": 16500
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 0.67,
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 2.
|
| 213 |
-
"step":
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 0.69,
|
| 217 |
-
"learning_rate": 0.00039950407489047486,
|
| 218 |
-
"loss": 2.5067,
|
| 219 |
-
"step": 17500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.71,
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss": 2.
|
| 225 |
-
"step":
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 0.73,
|
| 229 |
-
"learning_rate": 0.00039944404259647934,
|
| 230 |
-
"loss": 2.5333,
|
| 231 |
-
"step": 18500
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 0.75,
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 2.
|
| 237 |
-
"step":
|
| 238 |
-
},
|
| 239 |
-
{
|
| 240 |
-
"epoch": 0.77,
|
| 241 |
-
"learning_rate": 0.00039938058642461174,
|
| 242 |
-
"loss": 2.5632,
|
| 243 |
-
"step": 19500
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 0.79,
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss": 2.
|
| 249 |
-
"step":
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 0.81,
|
| 253 |
-
"learning_rate": 0.0003993137074642312,
|
| 254 |
-
"loss": 2.619,
|
| 255 |
-
"step": 20500
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 0.83,
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 2.
|
| 261 |
-
"step":
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 0.85,
|
| 265 |
-
"learning_rate": 0.0003992434068634561,
|
| 266 |
-
"loss": 2.5927,
|
| 267 |
-
"step": 21500
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 0.87,
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss": 2.
|
| 273 |
-
"step":
|
| 274 |
-
},
|
| 275 |
-
{
|
| 276 |
-
"epoch": 0.89,
|
| 277 |
-
"learning_rate": 0.0003991696858291447,
|
| 278 |
-
"loss": 2.5948,
|
| 279 |
-
"step": 22500
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 0.91,
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss": 2.
|
| 285 |
-
"step":
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 0.93,
|
| 289 |
-
"learning_rate": 0.0003990925456268741,
|
| 290 |
-
"loss": 2.5284,
|
| 291 |
-
"step": 23500
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 0.95,
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 2.
|
| 297 |
-
"step":
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 0.97,
|
| 301 |
-
"learning_rate": 0.0003990119875809187,
|
| 302 |
-
"loss": 2.6065,
|
| 303 |
-
"step": 24500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.99,
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 2.
|
| 309 |
-
"step":
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 1.0,
|
| 313 |
"eval_bleu": 1.0,
|
| 314 |
"eval_brevity_penalty": 1.0,
|
| 315 |
"eval_length_ratio": 1.0,
|
| 316 |
-
"eval_loss": 2.
|
| 317 |
-
"eval_precisions": [
|
| 318 |
-
1.0,
|
| 319 |
-
1.0,
|
| 320 |
-
1.0,
|
| 321 |
-
1.0
|
| 322 |
-
],
|
| 323 |
-
"eval_reference_length": 1439232,
|
| 324 |
-
"eval_runtime": 2451.7899,
|
| 325 |
-
"eval_samples_per_second": 1.147,
|
| 326 |
-
"eval_steps_per_second": 1.147,
|
| 327 |
-
"eval_translation_length": 1439232,
|
| 328 |
-
"step": 25291
|
| 329 |
-
},
|
| 330 |
-
{
|
| 331 |
-
"epoch": 1.01,
|
| 332 |
-
"learning_rate": 0.00039892801307422725,
|
| 333 |
-
"loss": 2.5875,
|
| 334 |
-
"step": 25500
|
| 335 |
-
},
|
| 336 |
-
{
|
| 337 |
-
"epoch": 1.03,
|
| 338 |
-
"learning_rate": 0.00039888474509631506,
|
| 339 |
-
"loss": 2.5277,
|
| 340 |
-
"step": 26000
|
| 341 |
-
},
|
| 342 |
-
{
|
| 343 |
-
"epoch": 1.05,
|
| 344 |
-
"learning_rate": 0.00039884062354839946,
|
| 345 |
-
"loss": 2.551,
|
| 346 |
-
"step": 26500
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 1.07,
|
| 350 |
-
"learning_rate": 0.0003987956486198406,
|
| 351 |
-
"loss": 2.5672,
|
| 352 |
-
"step": 27000
|
| 353 |
-
},
|
| 354 |
-
{
|
| 355 |
-
"epoch": 1.09,
|
| 356 |
-
"learning_rate": 0.00039874982050366095,
|
| 357 |
-
"loss": 2.5912,
|
| 358 |
-
"step": 27500
|
| 359 |
-
},
|
| 360 |
-
{
|
| 361 |
-
"epoch": 1.11,
|
| 362 |
-
"learning_rate": 0.0003987031393965449,
|
| 363 |
-
"loss": 2.4598,
|
| 364 |
-
"step": 28000
|
| 365 |
-
},
|
| 366 |
-
{
|
| 367 |
-
"epoch": 1.13,
|
| 368 |
-
"learning_rate": 0.0003986556054988376,
|
| 369 |
-
"loss": 2.4747,
|
| 370 |
-
"step": 28500
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 1.15,
|
| 374 |
-
"learning_rate": 0.0003986072190145441,
|
| 375 |
-
"loss": 2.4814,
|
| 376 |
-
"step": 29000
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 1.17,
|
| 380 |
-
"learning_rate": 0.00039855798015132866,
|
| 381 |
-
"loss": 2.5069,
|
| 382 |
-
"step": 29500
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 1.19,
|
| 386 |
-
"learning_rate": 0.0003985078891205139,
|
| 387 |
-
"loss": 2.5162,
|
| 388 |
-
"step": 30000
|
| 389 |
-
},
|
| 390 |
-
{
|
| 391 |
-
"epoch": 1.21,
|
| 392 |
-
"learning_rate": 0.0003984569461370795,
|
| 393 |
-
"loss": 2.5361,
|
| 394 |
-
"step": 30500
|
| 395 |
-
},
|
| 396 |
-
{
|
| 397 |
-
"epoch": 1.23,
|
| 398 |
-
"learning_rate": 0.0003984051514196616,
|
| 399 |
-
"loss": 2.5585,
|
| 400 |
-
"step": 31000
|
| 401 |
-
},
|
| 402 |
-
{
|
| 403 |
-
"epoch": 1.25,
|
| 404 |
-
"learning_rate": 0.0003983525051905519,
|
| 405 |
-
"loss": 2.5088,
|
| 406 |
-
"step": 31500
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 1.27,
|
| 410 |
-
"learning_rate": 0.0003982990076756965,
|
| 411 |
-
"loss": 2.5096,
|
| 412 |
-
"step": 32000
|
| 413 |
-
},
|
| 414 |
-
{
|
| 415 |
-
"epoch": 1.29,
|
| 416 |
-
"learning_rate": 0.00039824465910469526,
|
| 417 |
-
"loss": 2.5099,
|
| 418 |
-
"step": 32500
|
| 419 |
-
},
|
| 420 |
-
{
|
| 421 |
-
"epoch": 1.3,
|
| 422 |
-
"learning_rate": 0.00039818945971080026,
|
| 423 |
-
"loss": 2.556,
|
| 424 |
-
"step": 33000
|
| 425 |
-
},
|
| 426 |
-
{
|
| 427 |
-
"epoch": 1.32,
|
| 428 |
-
"learning_rate": 0.0003981334097309152,
|
| 429 |
-
"loss": 2.5494,
|
| 430 |
-
"step": 33500
|
| 431 |
-
},
|
| 432 |
-
{
|
| 433 |
-
"epoch": 1.34,
|
| 434 |
-
"learning_rate": 0.00039807650940559456,
|
| 435 |
-
"loss": 2.5276,
|
| 436 |
-
"step": 34000
|
| 437 |
-
},
|
| 438 |
-
{
|
| 439 |
-
"epoch": 1.36,
|
| 440 |
-
"learning_rate": 0.00039801875897904203,
|
| 441 |
-
"loss": 2.4754,
|
| 442 |
-
"step": 34500
|
| 443 |
-
},
|
| 444 |
-
{
|
| 445 |
-
"epoch": 1.38,
|
| 446 |
-
"learning_rate": 0.00039796015869910994,
|
| 447 |
-
"loss": 2.5322,
|
| 448 |
-
"step": 35000
|
| 449 |
-
},
|
| 450 |
-
{
|
| 451 |
-
"epoch": 1.4,
|
| 452 |
-
"learning_rate": 0.0003979007088172978,
|
| 453 |
-
"loss": 2.4941,
|
| 454 |
-
"step": 35500
|
| 455 |
-
},
|
| 456 |
-
{
|
| 457 |
-
"epoch": 1.42,
|
| 458 |
-
"learning_rate": 0.00039784040958875164,
|
| 459 |
-
"loss": 2.487,
|
| 460 |
-
"step": 36000
|
| 461 |
-
},
|
| 462 |
-
{
|
| 463 |
-
"epoch": 1.44,
|
| 464 |
-
"learning_rate": 0.0003977792612722627,
|
| 465 |
-
"loss": 2.4806,
|
| 466 |
-
"step": 36500
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 1.46,
|
| 470 |
-
"learning_rate": 0.00039771726413026607,
|
| 471 |
-
"loss": 2.556,
|
| 472 |
-
"step": 37000
|
| 473 |
-
},
|
| 474 |
-
{
|
| 475 |
-
"epoch": 1.48,
|
| 476 |
-
"learning_rate": 0.00039765441842884,
|
| 477 |
-
"loss": 2.5574,
|
| 478 |
-
"step": 37500
|
| 479 |
-
},
|
| 480 |
-
{
|
| 481 |
-
"epoch": 1.5,
|
| 482 |
-
"learning_rate": 0.00039759072443770467,
|
| 483 |
-
"loss": 2.5504,
|
| 484 |
-
"step": 38000
|
| 485 |
-
},
|
| 486 |
-
{
|
| 487 |
-
"epoch": 1.52,
|
| 488 |
-
"learning_rate": 0.00039752618243022066,
|
| 489 |
-
"loss": 2.6211,
|
| 490 |
-
"step": 38500
|
| 491 |
-
},
|
| 492 |
-
{
|
| 493 |
-
"epoch": 1.54,
|
| 494 |
-
"learning_rate": 0.0003974607926833884,
|
| 495 |
-
"loss": 2.5365,
|
| 496 |
-
"step": 39000
|
| 497 |
-
},
|
| 498 |
-
{
|
| 499 |
-
"epoch": 1.56,
|
| 500 |
-
"learning_rate": 0.0003973945554778463,
|
| 501 |
-
"loss": 2.6015,
|
| 502 |
-
"step": 39500
|
| 503 |
-
},
|
| 504 |
-
{
|
| 505 |
-
"epoch": 1.58,
|
| 506 |
-
"learning_rate": 0.00039732747109787,
|
| 507 |
-
"loss": 2.5784,
|
| 508 |
-
"step": 40000
|
| 509 |
-
},
|
| 510 |
-
{
|
| 511 |
-
"epoch": 1.6,
|
| 512 |
-
"learning_rate": 0.00039725953983137113,
|
| 513 |
-
"loss": 2.565,
|
| 514 |
-
"step": 40500
|
| 515 |
-
},
|
| 516 |
-
{
|
| 517 |
-
"epoch": 1.62,
|
| 518 |
-
"learning_rate": 0.0003971907619698959,
|
| 519 |
-
"loss": 2.5028,
|
| 520 |
-
"step": 41000
|
| 521 |
-
},
|
| 522 |
-
{
|
| 523 |
-
"epoch": 1.64,
|
| 524 |
-
"learning_rate": 0.0003971211378086238,
|
| 525 |
-
"loss": 2.5863,
|
| 526 |
-
"step": 41500
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 1.66,
|
| 530 |
-
"learning_rate": 0.0003970506676463666,
|
| 531 |
-
"loss": 2.5241,
|
| 532 |
-
"step": 42000
|
| 533 |
-
},
|
| 534 |
-
{
|
| 535 |
-
"epoch": 1.68,
|
| 536 |
-
"learning_rate": 0.000396979351785567,
|
| 537 |
-
"loss": 2.4885,
|
| 538 |
-
"step": 42500
|
| 539 |
-
},
|
| 540 |
-
{
|
| 541 |
-
"epoch": 1.7,
|
| 542 |
-
"learning_rate": 0.000396907190532297,
|
| 543 |
-
"loss": 2.5004,
|
| 544 |
-
"step": 43000
|
| 545 |
-
},
|
| 546 |
-
{
|
| 547 |
-
"epoch": 1.72,
|
| 548 |
-
"learning_rate": 0.0003968341841962571,
|
| 549 |
-
"loss": 2.5461,
|
| 550 |
-
"step": 43500
|
| 551 |
-
},
|
| 552 |
-
{
|
| 553 |
-
"epoch": 1.74,
|
| 554 |
-
"learning_rate": 0.00039676033309077443,
|
| 555 |
-
"loss": 2.5531,
|
| 556 |
-
"step": 44000
|
| 557 |
-
},
|
| 558 |
-
{
|
| 559 |
-
"epoch": 1.76,
|
| 560 |
-
"learning_rate": 0.00039668563753280197,
|
| 561 |
-
"loss": 2.5425,
|
| 562 |
-
"step": 44500
|
| 563 |
-
},
|
| 564 |
-
{
|
| 565 |
-
"epoch": 1.78,
|
| 566 |
-
"learning_rate": 0.0003966100978429168,
|
| 567 |
-
"loss": 2.529,
|
| 568 |
-
"step": 45000
|
| 569 |
-
},
|
| 570 |
-
{
|
| 571 |
-
"epoch": 1.8,
|
| 572 |
-
"learning_rate": 0.0003965337143453187,
|
| 573 |
-
"loss": 2.5401,
|
| 574 |
-
"step": 45500
|
| 575 |
-
},
|
| 576 |
-
{
|
| 577 |
-
"epoch": 1.82,
|
| 578 |
-
"learning_rate": 0.0003964564873678292,
|
| 579 |
-
"loss": 2.5465,
|
| 580 |
-
"step": 46000
|
| 581 |
-
},
|
| 582 |
-
{
|
| 583 |
-
"epoch": 1.84,
|
| 584 |
-
"learning_rate": 0.0003963784172418894,
|
| 585 |
-
"loss": 2.5185,
|
| 586 |
-
"step": 46500
|
| 587 |
-
},
|
| 588 |
-
{
|
| 589 |
-
"epoch": 1.86,
|
| 590 |
-
"learning_rate": 0.00039629950430255953,
|
| 591 |
-
"loss": 2.5568,
|
| 592 |
-
"step": 47000
|
| 593 |
-
},
|
| 594 |
-
{
|
| 595 |
-
"epoch": 1.88,
|
| 596 |
-
"learning_rate": 0.00039621974888851653,
|
| 597 |
-
"loss": 2.5297,
|
| 598 |
-
"step": 47500
|
| 599 |
-
},
|
| 600 |
-
{
|
| 601 |
-
"epoch": 1.9,
|
| 602 |
-
"learning_rate": 0.00039613915134205336,
|
| 603 |
-
"loss": 2.5155,
|
| 604 |
-
"step": 48000
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 1.92,
|
| 608 |
-
"learning_rate": 0.0003960577120090772,
|
| 609 |
-
"loss": 2.5491,
|
| 610 |
-
"step": 48500
|
| 611 |
-
},
|
| 612 |
-
{
|
| 613 |
-
"epoch": 1.94,
|
| 614 |
-
"learning_rate": 0.0003959754312391078,
|
| 615 |
-
"loss": 2.4615,
|
| 616 |
-
"step": 49000
|
| 617 |
-
},
|
| 618 |
-
{
|
| 619 |
-
"epoch": 1.96,
|
| 620 |
-
"learning_rate": 0.0003958923093852764,
|
| 621 |
-
"loss": 2.522,
|
| 622 |
-
"step": 49500
|
| 623 |
-
},
|
| 624 |
-
{
|
| 625 |
-
"epoch": 1.98,
|
| 626 |
-
"learning_rate": 0.00039580834680432385,
|
| 627 |
-
"loss": 2.5588,
|
| 628 |
-
"step": 50000
|
| 629 |
-
},
|
| 630 |
-
{
|
| 631 |
-
"epoch": 2.0,
|
| 632 |
-
"learning_rate": 0.00039572354385659925,
|
| 633 |
-
"loss": 2.492,
|
| 634 |
-
"step": 50500
|
| 635 |
-
},
|
| 636 |
-
{
|
| 637 |
-
"epoch": 2.0,
|
| 638 |
-
"eval_bleu": 1.0,
|
| 639 |
-
"eval_brevity_penalty": 1.0,
|
| 640 |
-
"eval_length_ratio": 1.0,
|
| 641 |
-
"eval_loss": 2.45813250541687,
|
| 642 |
"eval_precisions": [
|
| 643 |
1.0,
|
| 644 |
1.0,
|
|
@@ -646,18 +171,18 @@
|
|
| 646 |
1.0
|
| 647 |
],
|
| 648 |
"eval_reference_length": 1439232,
|
| 649 |
-
"eval_runtime":
|
| 650 |
-
"eval_samples_per_second": 1.
|
| 651 |
-
"eval_steps_per_second":
|
| 652 |
"eval_translation_length": 1439232,
|
| 653 |
-
"step":
|
| 654 |
}
|
| 655 |
],
|
| 656 |
"logging_steps": 500,
|
| 657 |
-
"max_steps":
|
| 658 |
"num_train_epochs": 30,
|
| 659 |
"save_steps": 500,
|
| 660 |
-
"total_flos":
|
| 661 |
"trial_name": null,
|
| 662 |
"trial_params": null
|
| 663 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.393324613571167,
|
| 3 |
+
"best_model_checkpoint": "dq158/pingusPongus/checkpoint-12646",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 12646,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 8e-05,
|
| 14 |
+
"loss": 3.4297,
|
| 15 |
+
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 7.999965623174768e-05,
|
| 20 |
+
"loss": 2.9121,
|
| 21 |
+
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 7.999862493289958e-05,
|
| 26 |
+
"loss": 2.8542,
|
| 27 |
+
"step": 1500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 7.999690612118205e-05,
|
| 32 |
+
"loss": 2.7645,
|
| 33 |
+
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.2,
|
| 37 |
+
"learning_rate": 7.999449982613875e-05,
|
| 38 |
+
"loss": 2.7235,
|
| 39 |
+
"step": 2500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.24,
|
| 43 |
+
"learning_rate": 7.99914060891301e-05,
|
| 44 |
+
"loss": 2.712,
|
| 45 |
+
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.28,
|
| 49 |
+
"learning_rate": 7.998762496333249e-05,
|
| 50 |
+
"loss": 2.6692,
|
| 51 |
+
"step": 3500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.32,
|
| 55 |
+
"learning_rate": 7.998315651373749e-05,
|
| 56 |
+
"loss": 2.6532,
|
| 57 |
+
"step": 4000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.36,
|
| 61 |
+
"learning_rate": 7.997800081715064e-05,
|
| 62 |
+
"loss": 2.6572,
|
| 63 |
+
"step": 4500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.4,
|
| 67 |
+
"learning_rate": 7.99721579621902e-05,
|
| 68 |
+
"loss": 2.6498,
|
| 69 |
+
"step": 5000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.43,
|
| 73 |
+
"learning_rate": 7.996562804928555e-05,
|
| 74 |
+
"loss": 2.6156,
|
| 75 |
+
"step": 5500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.47,
|
| 79 |
+
"learning_rate": 7.995841119067557e-05,
|
| 80 |
+
"loss": 2.6079,
|
| 81 |
+
"step": 6000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.51,
|
| 85 |
+
"learning_rate": 7.995050751040655e-05,
|
| 86 |
+
"loss": 2.5754,
|
| 87 |
+
"step": 6500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.55,
|
| 91 |
+
"learning_rate": 7.994191714433023e-05,
|
| 92 |
+
"loss": 2.5303,
|
| 93 |
+
"step": 7000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.59,
|
| 97 |
+
"learning_rate": 7.993264024010138e-05,
|
| 98 |
+
"loss": 2.6282,
|
| 99 |
+
"step": 7500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.63,
|
| 103 |
+
"learning_rate": 7.992267695717525e-05,
|
| 104 |
+
"loss": 2.558,
|
| 105 |
+
"step": 8000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.67,
|
| 109 |
+
"learning_rate": 7.991202746680485e-05,
|
| 110 |
+
"loss": 2.57,
|
| 111 |
+
"step": 8500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.71,
|
| 115 |
+
"learning_rate": 7.9900691952038e-05,
|
| 116 |
+
"loss": 2.6049,
|
| 117 |
+
"step": 9000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 0.75,
|
| 121 |
+
"learning_rate": 7.988867060771425e-05,
|
| 122 |
+
"loss": 2.6004,
|
| 123 |
+
"step": 9500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.79,
|
| 127 |
+
"learning_rate": 7.987596364046138e-05,
|
| 128 |
+
"loss": 2.5497,
|
| 129 |
+
"step": 10000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 0.83,
|
| 133 |
+
"learning_rate": 7.986257126869202e-05,
|
| 134 |
+
"loss": 2.5397,
|
| 135 |
+
"step": 10500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.87,
|
| 139 |
+
"learning_rate": 7.984849372259976e-05,
|
| 140 |
+
"loss": 2.5058,
|
| 141 |
+
"step": 11000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 0.91,
|
| 145 |
+
"learning_rate": 7.983373124415528e-05,
|
| 146 |
+
"loss": 2.5385,
|
| 147 |
+
"step": 11500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 0.95,
|
| 151 |
+
"learning_rate": 7.981828408710215e-05,
|
| 152 |
+
"loss": 2.523,
|
| 153 |
+
"step": 12000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.99,
|
| 157 |
+
"learning_rate": 7.980215251695248e-05,
|
| 158 |
+
"loss": 2.5043,
|
| 159 |
+
"step": 12500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 1.0,
|
| 163 |
"eval_bleu": 1.0,
|
| 164 |
"eval_brevity_penalty": 1.0,
|
| 165 |
"eval_length_ratio": 1.0,
|
| 166 |
+
"eval_loss": 2.393324613571167,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
"eval_precisions": [
|
| 168 |
1.0,
|
| 169 |
1.0,
|
|
|
|
| 171 |
1.0
|
| 172 |
],
|
| 173 |
"eval_reference_length": 1439232,
|
| 174 |
+
"eval_runtime": 1581.2805,
|
| 175 |
+
"eval_samples_per_second": 1.778,
|
| 176 |
+
"eval_steps_per_second": 0.889,
|
| 177 |
"eval_translation_length": 1439232,
|
| 178 |
+
"step": 12646
|
| 179 |
}
|
| 180 |
],
|
| 181 |
"logging_steps": 500,
|
| 182 |
+
"max_steps": 379380,
|
| 183 |
"num_train_epochs": 30,
|
| 184 |
"save_steps": 500,
|
| 185 |
+
"total_flos": 2.1666322696686797e+17,
|
| 186 |
"trial_name": null,
|
| 187 |
"trial_params": null
|
| 188 |
}
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24ee12b5b09ba5eb0b3c9a6d80467a1a4ad850b0cfd569800e5c0f0898b97979
|
| 3 |
size 4728
|