Training in progress, step 1000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1583480280
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:872dafd4f7a04e803d7c539219fc84f71fe6a69c1babbc798b79455dc4e1cc48
|
| 3 |
size 1583480280
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3166958572
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9ad0354ec70f7cee4872b10ddc2b723a017a3711c3e9c5a25eba5b7823716aa
|
| 3 |
size 3166958572
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ef1ee026e99fb5f8e5a5e72bb4657b4cedd8651ee5b752f5c314d3a89fd9ad7
|
| 3 |
size 14645
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68aec417c91400a5fbe9c98d7447dabd74ed3b0812272a5f21d640985e919bad
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 1.
|
| 4 |
-
"best_model_checkpoint": "hieptt/vietnamese-correction-finetuning/checkpoint-
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 1000,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -82,959 +82,11 @@
|
|
| 82 |
{
|
| 83 |
"epoch": 0.02686222365487415,
|
| 84 |
"eval_loss": 1.550229787826538,
|
| 85 |
-
"eval_runtime":
|
| 86 |
"eval_sacrebleu": 96.21455916515954,
|
| 87 |
-
"eval_samples_per_second":
|
| 88 |
-
"eval_steps_per_second": 0.563,
|
| 89 |
-
"step": 1000
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"epoch": 0.029548446020361566,
|
| 93 |
-
"grad_norm": 0.43504711985588074,
|
| 94 |
-
"learning_rate": 4.999949639906304e-05,
|
| 95 |
-
"loss": 1.5642,
|
| 96 |
-
"step": 1100
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 0.03223466838584898,
|
| 100 |
-
"grad_norm": 0.5235740542411804,
|
| 101 |
-
"learning_rate": 4.999796521812822e-05,
|
| 102 |
-
"loss": 1.5535,
|
| 103 |
-
"step": 1200
|
| 104 |
-
},
|
| 105 |
-
{
|
| 106 |
-
"epoch": 0.0349208907513364,
|
| 107 |
-
"grad_norm": 0.40589195489883423,
|
| 108 |
-
"learning_rate": 4.999540646875361e-05,
|
| 109 |
-
"loss": 1.5474,
|
| 110 |
-
"step": 1300
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 0.03760711311682381,
|
| 114 |
-
"grad_norm": 0.47619858384132385,
|
| 115 |
-
"learning_rate": 4.9991820256119385e-05,
|
| 116 |
-
"loss": 1.5467,
|
| 117 |
-
"step": 1400
|
| 118 |
-
},
|
| 119 |
-
{
|
| 120 |
-
"epoch": 0.040293335482311225,
|
| 121 |
-
"grad_norm": 0.5160552263259888,
|
| 122 |
-
"learning_rate": 4.9987206727640703e-05,
|
| 123 |
-
"loss": 1.5404,
|
| 124 |
-
"step": 1500
|
| 125 |
-
},
|
| 126 |
-
{
|
| 127 |
-
"epoch": 0.04297955784779864,
|
| 128 |
-
"grad_norm": 0.7639428973197937,
|
| 129 |
-
"learning_rate": 4.998156607296163e-05,
|
| 130 |
-
"loss": 1.5391,
|
| 131 |
-
"step": 1600
|
| 132 |
-
},
|
| 133 |
-
{
|
| 134 |
-
"epoch": 0.04566578021328606,
|
| 135 |
-
"grad_norm": Infinity,
|
| 136 |
-
"learning_rate": 4.997489852394741e-05,
|
| 137 |
-
"loss": 1.5411,
|
| 138 |
-
"step": 1700
|
| 139 |
-
},
|
| 140 |
-
{
|
| 141 |
-
"epoch": 0.04835200257877347,
|
| 142 |
-
"grad_norm": 11.356963157653809,
|
| 143 |
-
"learning_rate": 4.996720435467485e-05,
|
| 144 |
-
"loss": 3.6358,
|
| 145 |
-
"step": 1800
|
| 146 |
-
},
|
| 147 |
-
{
|
| 148 |
-
"epoch": 0.051038224944260885,
|
| 149 |
-
"grad_norm": 0.5683807730674744,
|
| 150 |
-
"learning_rate": 4.995848388142112e-05,
|
| 151 |
-
"loss": 5.6818,
|
| 152 |
-
"step": 1900
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch": 0.0537244473097483,
|
| 156 |
-
"grad_norm": 0.47999536991119385,
|
| 157 |
-
"learning_rate": 4.994873746265073e-05,
|
| 158 |
-
"loss": 1.8588,
|
| 159 |
-
"step": 2000
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 0.0537244473097483,
|
| 163 |
-
"eval_loss": 1.5401405096054077,
|
| 164 |
-
"eval_runtime": 3621.9751,
|
| 165 |
-
"eval_sacrebleu": 96.05509046827605,
|
| 166 |
-
"eval_samples_per_second": 73.089,
|
| 167 |
-
"eval_steps_per_second": 0.571,
|
| 168 |
-
"step": 2000
|
| 169 |
-
},
|
| 170 |
-
{
|
| 171 |
-
"epoch": 0.05641066967523572,
|
| 172 |
-
"grad_norm": 0.4521150290966034,
|
| 173 |
-
"learning_rate": 4.993796549900076e-05,
|
| 174 |
-
"loss": 1.5427,
|
| 175 |
-
"step": 2100
|
| 176 |
-
},
|
| 177 |
-
{
|
| 178 |
-
"epoch": 0.05909689204072313,
|
| 179 |
-
"grad_norm": 0.4296003580093384,
|
| 180 |
-
"learning_rate": 4.992616843326446e-05,
|
| 181 |
-
"loss": 1.5634,
|
| 182 |
-
"step": 2200
|
| 183 |
-
},
|
| 184 |
-
{
|
| 185 |
-
"epoch": 0.061783114406210544,
|
| 186 |
-
"grad_norm": 0.6067550182342529,
|
| 187 |
-
"learning_rate": 4.991334675037299e-05,
|
| 188 |
-
"loss": 1.534,
|
| 189 |
-
"step": 2300
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 0.06446933677169796,
|
| 193 |
-
"grad_norm": 0.5774173140525818,
|
| 194 |
-
"learning_rate": 4.989950097737552e-05,
|
| 195 |
-
"loss": 1.542,
|
| 196 |
-
"step": 2400
|
| 197 |
-
},
|
| 198 |
-
{
|
| 199 |
-
"epoch": 0.06715555913718538,
|
| 200 |
-
"grad_norm": 0.40837734937667847,
|
| 201 |
-
"learning_rate": 4.988463168341755e-05,
|
| 202 |
-
"loss": 1.5252,
|
| 203 |
-
"step": 2500
|
| 204 |
-
},
|
| 205 |
-
{
|
| 206 |
-
"epoch": 0.0698417815026728,
|
| 207 |
-
"grad_norm": 0.5584208965301514,
|
| 208 |
-
"learning_rate": 4.986873947971751e-05,
|
| 209 |
-
"loss": 1.5311,
|
| 210 |
-
"step": 2600
|
| 211 |
-
},
|
| 212 |
-
{
|
| 213 |
-
"epoch": 0.0725280038681602,
|
| 214 |
-
"grad_norm": 0.430813193321228,
|
| 215 |
-
"learning_rate": 4.9851825019541644e-05,
|
| 216 |
-
"loss": 1.5247,
|
| 217 |
-
"step": 2700
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"epoch": 0.07521422623364762,
|
| 221 |
-
"grad_norm": 0.3637406826019287,
|
| 222 |
-
"learning_rate": 4.9833888998177165e-05,
|
| 223 |
-
"loss": 1.5221,
|
| 224 |
-
"step": 2800
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 0.07790044859913503,
|
| 228 |
-
"grad_norm": 0.530950129032135,
|
| 229 |
-
"learning_rate": 4.981493215290366e-05,
|
| 230 |
-
"loss": 1.5233,
|
| 231 |
-
"step": 2900
|
| 232 |
-
},
|
| 233 |
-
{
|
| 234 |
-
"epoch": 0.08058667096462245,
|
| 235 |
-
"grad_norm": 1.674155831336975,
|
| 236 |
-
"learning_rate": 4.979495526296279e-05,
|
| 237 |
-
"loss": 1.5204,
|
| 238 |
-
"step": 3000
|
| 239 |
-
},
|
| 240 |
-
{
|
| 241 |
-
"epoch": 0.08058667096462245,
|
| 242 |
-
"eval_loss": 1.5201951265335083,
|
| 243 |
-
"eval_runtime": 3188.7059,
|
| 244 |
-
"eval_sacrebleu": 96.36598044790144,
|
| 245 |
-
"eval_samples_per_second": 83.02,
|
| 246 |
-
"eval_steps_per_second": 0.649,
|
| 247 |
-
"step": 3000
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 0.08327289333010987,
|
| 251 |
-
"grad_norm": 0.39490917325019836,
|
| 252 |
-
"learning_rate": 4.977395914952627e-05,
|
| 253 |
-
"loss": 1.5193,
|
| 254 |
-
"step": 3100
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.08595911569559728,
|
| 258 |
-
"grad_norm": 0.37447476387023926,
|
| 259 |
-
"learning_rate": 4.975194467566206e-05,
|
| 260 |
-
"loss": 1.5185,
|
| 261 |
-
"step": 3200
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 0.0886453380610847,
|
| 265 |
-
"grad_norm": 0.4466964900493622,
|
| 266 |
-
"learning_rate": 4.9728912746298975e-05,
|
| 267 |
-
"loss": 1.5143,
|
| 268 |
-
"step": 3300
|
| 269 |
-
},
|
| 270 |
-
{
|
| 271 |
-
"epoch": 0.09133156042657212,
|
| 272 |
-
"grad_norm": 0.45723655819892883,
|
| 273 |
-
"learning_rate": 4.9704864308189415e-05,
|
| 274 |
-
"loss": 1.5119,
|
| 275 |
-
"step": 3400
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 0.09401778279205952,
|
| 279 |
-
"grad_norm": 0.3492431938648224,
|
| 280 |
-
"learning_rate": 4.967980034987048e-05,
|
| 281 |
-
"loss": 1.5118,
|
| 282 |
-
"step": 3500
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 0.09670400515754694,
|
| 286 |
-
"grad_norm": 0.6189600825309753,
|
| 287 |
-
"learning_rate": 4.965372190162333e-05,
|
| 288 |
-
"loss": 1.5079,
|
| 289 |
-
"step": 3600
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 0.09939022752303436,
|
| 293 |
-
"grad_norm": 0.4467921555042267,
|
| 294 |
-
"learning_rate": 4.962663003543083e-05,
|
| 295 |
-
"loss": 1.5091,
|
| 296 |
-
"step": 3700
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 0.10207644988852177,
|
| 300 |
-
"grad_norm": 0.4007367491722107,
|
| 301 |
-
"learning_rate": 4.959852586493349e-05,
|
| 302 |
-
"loss": 1.5111,
|
| 303 |
-
"step": 3800
|
| 304 |
-
},
|
| 305 |
-
{
|
| 306 |
-
"epoch": 0.10476267225400919,
|
| 307 |
-
"grad_norm": 0.4409460127353668,
|
| 308 |
-
"learning_rate": 4.9569410545383665e-05,
|
| 309 |
-
"loss": 1.5099,
|
| 310 |
-
"step": 3900
|
| 311 |
-
},
|
| 312 |
-
{
|
| 313 |
-
"epoch": 0.1074488946194966,
|
| 314 |
-
"grad_norm": 0.3737949728965759,
|
| 315 |
-
"learning_rate": 4.953928527359812e-05,
|
| 316 |
-
"loss": 1.5107,
|
| 317 |
-
"step": 4000
|
| 318 |
-
},
|
| 319 |
-
{
|
| 320 |
-
"epoch": 0.1074488946194966,
|
| 321 |
-
"eval_loss": 1.5068457126617432,
|
| 322 |
-
"eval_runtime": 3197.023,
|
| 323 |
-
"eval_sacrebleu": 96.60403832626737,
|
| 324 |
-
"eval_samples_per_second": 82.804,
|
| 325 |
-
"eval_steps_per_second": 0.647,
|
| 326 |
-
"step": 4000
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.11013511698498402,
|
| 330 |
-
"grad_norm": 0.3667586147785187,
|
| 331 |
-
"learning_rate": 4.95081512879088e-05,
|
| 332 |
-
"loss": 1.505,
|
| 333 |
-
"step": 4100
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 0.11282133935047144,
|
| 337 |
-
"grad_norm": 0.3441518545150757,
|
| 338 |
-
"learning_rate": 4.947600986811188e-05,
|
| 339 |
-
"loss": 1.5082,
|
| 340 |
-
"step": 4200
|
| 341 |
-
},
|
| 342 |
-
{
|
| 343 |
-
"epoch": 0.11550756171595884,
|
| 344 |
-
"grad_norm": 0.3704121708869934,
|
| 345 |
-
"learning_rate": 4.9442862335415266e-05,
|
| 346 |
-
"loss": 1.5026,
|
| 347 |
-
"step": 4300
|
| 348 |
-
},
|
| 349 |
-
{
|
| 350 |
-
"epoch": 0.11819378408144626,
|
| 351 |
-
"grad_norm": 0.3782528042793274,
|
| 352 |
-
"learning_rate": 4.940871005238418e-05,
|
| 353 |
-
"loss": 1.5041,
|
| 354 |
-
"step": 4400
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 0.12088000644693368,
|
| 358 |
-
"grad_norm": 0.3923707902431488,
|
| 359 |
-
"learning_rate": 4.937355442288523e-05,
|
| 360 |
-
"loss": 1.5008,
|
| 361 |
-
"step": 4500
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 0.12356622881242109,
|
| 365 |
-
"grad_norm": 0.38463589549064636,
|
| 366 |
-
"learning_rate": 4.933739689202862e-05,
|
| 367 |
-
"loss": 1.5035,
|
| 368 |
-
"step": 4600
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 0.1262524511779085,
|
| 372 |
-
"grad_norm": 0.44041115045547485,
|
| 373 |
-
"learning_rate": 4.9300238946108856e-05,
|
| 374 |
-
"loss": 1.5005,
|
| 375 |
-
"step": 4700
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 0.12893867354339592,
|
| 379 |
-
"grad_norm": 0.41506507992744446,
|
| 380 |
-
"learning_rate": 4.926208211254354e-05,
|
| 381 |
-
"loss": 1.5014,
|
| 382 |
-
"step": 4800
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 0.13162489590888335,
|
| 386 |
-
"grad_norm": 3.7683701515197754,
|
| 387 |
-
"learning_rate": 4.922292795981065e-05,
|
| 388 |
-
"loss": 3.7267,
|
| 389 |
-
"step": 4900
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 0.13431111827437076,
|
| 393 |
-
"grad_norm": 3.7164273262023926,
|
| 394 |
-
"learning_rate": 4.9182778097384055e-05,
|
| 395 |
-
"loss": 5.7642,
|
| 396 |
-
"step": 5000
|
| 397 |
-
},
|
| 398 |
-
{
|
| 399 |
-
"epoch": 0.13431111827437076,
|
| 400 |
-
"eval_loss": 4.501608371734619,
|
| 401 |
-
"eval_runtime": 2795.6042,
|
| 402 |
-
"eval_sacrebleu": 17.695398660139727,
|
| 403 |
-
"eval_samples_per_second": 94.694,
|
| 404 |
-
"eval_steps_per_second": 0.74,
|
| 405 |
-
"step": 5000
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 0.13699734063985816,
|
| 409 |
-
"grad_norm": 7.479543209075928,
|
| 410 |
-
"learning_rate": 4.914163417566734e-05,
|
| 411 |
-
"loss": 2.1475,
|
| 412 |
-
"step": 5100
|
| 413 |
-
},
|
| 414 |
-
{
|
| 415 |
-
"epoch": 0.1396835630053456,
|
| 416 |
-
"grad_norm": 0.3754720091819763,
|
| 417 |
-
"learning_rate": 4.909949788592598e-05,
|
| 418 |
-
"loss": 2.107,
|
| 419 |
-
"step": 5200
|
| 420 |
-
},
|
| 421 |
-
{
|
| 422 |
-
"epoch": 0.142369785370833,
|
| 423 |
-
"grad_norm": 0.42401301860809326,
|
| 424 |
-
"learning_rate": 4.9056370960217804e-05,
|
| 425 |
-
"loss": 1.5074,
|
| 426 |
-
"step": 5300
|
| 427 |
-
},
|
| 428 |
-
{
|
| 429 |
-
"epoch": 0.1450560077363204,
|
| 430 |
-
"grad_norm": 0.261708527803421,
|
| 431 |
-
"learning_rate": 4.9012255171321814e-05,
|
| 432 |
-
"loss": 1.5061,
|
| 433 |
-
"step": 5400
|
| 434 |
-
},
|
| 435 |
-
{
|
| 436 |
-
"epoch": 0.14774223010180781,
|
| 437 |
-
"grad_norm": 0.35923367738723755,
|
| 438 |
-
"learning_rate": 4.89671523326653e-05,
|
| 439 |
-
"loss": 1.5086,
|
| 440 |
-
"step": 5500
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 0.15042845246729525,
|
| 444 |
-
"grad_norm": 0.3558649718761444,
|
| 445 |
-
"learning_rate": 4.892106429824931e-05,
|
| 446 |
-
"loss": 1.5026,
|
| 447 |
-
"step": 5600
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 0.15311467483278265,
|
| 451 |
-
"grad_norm": 0.3130311369895935,
|
| 452 |
-
"learning_rate": 4.8873992962572413e-05,
|
| 453 |
-
"loss": 1.5008,
|
| 454 |
-
"step": 5700
|
| 455 |
-
},
|
| 456 |
-
{
|
| 457 |
-
"epoch": 0.15580089719827006,
|
| 458 |
-
"grad_norm": 0.40723615884780884,
|
| 459 |
-
"learning_rate": 4.882594026055286e-05,
|
| 460 |
-
"loss": 1.497,
|
| 461 |
-
"step": 5800
|
| 462 |
-
},
|
| 463 |
-
{
|
| 464 |
-
"epoch": 0.1584871195637575,
|
| 465 |
-
"grad_norm": 0.3690826892852783,
|
| 466 |
-
"learning_rate": 4.877690816744903e-05,
|
| 467 |
-
"loss": 1.493,
|
| 468 |
-
"step": 5900
|
| 469 |
-
},
|
| 470 |
-
{
|
| 471 |
-
"epoch": 0.1611733419292449,
|
| 472 |
-
"grad_norm": 0.4789445400238037,
|
| 473 |
-
"learning_rate": 4.87268986987782e-05,
|
| 474 |
-
"loss": 1.4936,
|
| 475 |
-
"step": 6000
|
| 476 |
-
},
|
| 477 |
-
{
|
| 478 |
-
"epoch": 0.1611733419292449,
|
| 479 |
-
"eval_loss": 1.4912784099578857,
|
| 480 |
-
"eval_runtime": 3279.389,
|
| 481 |
-
"eval_sacrebleu": 96.84419951407857,
|
| 482 |
-
"eval_samples_per_second": 80.724,
|
| 483 |
-
"eval_steps_per_second": 0.631,
|
| 484 |
-
"step": 6000
|
| 485 |
-
},
|
| 486 |
-
{
|
| 487 |
-
"epoch": 0.1638595642947323,
|
| 488 |
-
"grad_norm": 0.7140536308288574,
|
| 489 |
-
"learning_rate": 4.8675913910233784e-05,
|
| 490 |
-
"loss": 1.4966,
|
| 491 |
-
"step": 6100
|
| 492 |
-
},
|
| 493 |
-
{
|
| 494 |
-
"epoch": 0.16654578666021974,
|
| 495 |
-
"grad_norm": 0.348763108253479,
|
| 496 |
-
"learning_rate": 4.8623955897600724e-05,
|
| 497 |
-
"loss": 1.4944,
|
| 498 |
-
"step": 6200
|
| 499 |
-
},
|
| 500 |
-
{
|
| 501 |
-
"epoch": 0.16923200902570715,
|
| 502 |
-
"grad_norm": 0.33172258734703064,
|
| 503 |
-
"learning_rate": 4.8571026796669416e-05,
|
| 504 |
-
"loss": 1.4922,
|
| 505 |
-
"step": 6300
|
| 506 |
-
},
|
| 507 |
-
{
|
| 508 |
-
"epoch": 0.17191823139119455,
|
| 509 |
-
"grad_norm": 0.42076775431632996,
|
| 510 |
-
"learning_rate": 4.85171287831479e-05,
|
| 511 |
-
"loss": 1.4935,
|
| 512 |
-
"step": 6400
|
| 513 |
-
},
|
| 514 |
-
{
|
| 515 |
-
"epoch": 0.174604453756682,
|
| 516 |
-
"grad_norm": 0.4683443009853363,
|
| 517 |
-
"learning_rate": 4.846226407257241e-05,
|
| 518 |
-
"loss": 1.4956,
|
| 519 |
-
"step": 6500
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 0.1772906761221694,
|
| 523 |
-
"grad_norm": 0.45604732632637024,
|
| 524 |
-
"learning_rate": 4.840643492021628e-05,
|
| 525 |
-
"loss": 1.4916,
|
| 526 |
-
"step": 6600
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 0.1799768984876568,
|
| 530 |
-
"grad_norm": 0.4385334849357605,
|
| 531 |
-
"learning_rate": 4.834964362099733e-05,
|
| 532 |
-
"loss": 1.489,
|
| 533 |
-
"step": 6700
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 0.18266312085314423,
|
| 537 |
-
"grad_norm": 0.48662105202674866,
|
| 538 |
-
"learning_rate": 4.829189250938341e-05,
|
| 539 |
-
"loss": 1.4893,
|
| 540 |
-
"step": 6800
|
| 541 |
-
},
|
| 542 |
-
{
|
| 543 |
-
"epoch": 0.18534934321863164,
|
| 544 |
-
"grad_norm": 0.4047809839248657,
|
| 545 |
-
"learning_rate": 4.8233183959296544e-05,
|
| 546 |
-
"loss": 1.489,
|
| 547 |
-
"step": 6900
|
| 548 |
-
},
|
| 549 |
-
{
|
| 550 |
-
"epoch": 0.18803556558411905,
|
| 551 |
-
"grad_norm": 0.31221145391464233,
|
| 552 |
-
"learning_rate": 4.817352038401526e-05,
|
| 553 |
-
"loss": 1.4888,
|
| 554 |
-
"step": 7000
|
| 555 |
-
},
|
| 556 |
-
{
|
| 557 |
-
"epoch": 0.18803556558411905,
|
| 558 |
-
"eval_loss": 1.48853600025177,
|
| 559 |
-
"eval_runtime": 3164.6016,
|
| 560 |
-
"eval_sacrebleu": 96.86234024863172,
|
| 561 |
-
"eval_samples_per_second": 83.652,
|
| 562 |
-
"eval_steps_per_second": 0.654,
|
| 563 |
-
"step": 7000
|
| 564 |
-
},
|
| 565 |
-
{
|
| 566 |
-
"epoch": 0.19072178794960648,
|
| 567 |
-
"grad_norm": 0.33513781428337097,
|
| 568 |
-
"learning_rate": 4.8112904236075464e-05,
|
| 569 |
-
"loss": 1.4882,
|
| 570 |
-
"step": 7100
|
| 571 |
-
},
|
| 572 |
-
{
|
| 573 |
-
"epoch": 0.1934080103150939,
|
| 574 |
-
"grad_norm": 0.3263476490974426,
|
| 575 |
-
"learning_rate": 4.8051338007169574e-05,
|
| 576 |
-
"loss": 1.4931,
|
| 577 |
-
"step": 7200
|
| 578 |
-
},
|
| 579 |
-
{
|
| 580 |
-
"epoch": 0.1960942326805813,
|
| 581 |
-
"grad_norm": 0.40779784321784973,
|
| 582 |
-
"learning_rate": 4.798882422804413e-05,
|
| 583 |
-
"loss": 1.4894,
|
| 584 |
-
"step": 7300
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 0.19878045504606873,
|
| 588 |
-
"grad_norm": 0.3289216458797455,
|
| 589 |
-
"learning_rate": 4.792536546839573e-05,
|
| 590 |
-
"loss": 1.4872,
|
| 591 |
-
"step": 7400
|
| 592 |
-
},
|
| 593 |
-
{
|
| 594 |
-
"epoch": 0.20146667741155613,
|
| 595 |
-
"grad_norm": 0.3560020625591278,
|
| 596 |
-
"learning_rate": 4.786096433676545e-05,
|
| 597 |
-
"loss": 1.4879,
|
| 598 |
-
"step": 7500
|
| 599 |
-
},
|
| 600 |
-
{
|
| 601 |
-
"epoch": 0.20415289977704354,
|
| 602 |
-
"grad_norm": 0.34368178248405457,
|
| 603 |
-
"learning_rate": 4.779562348043155e-05,
|
| 604 |
-
"loss": 1.4888,
|
| 605 |
-
"step": 7600
|
| 606 |
-
},
|
| 607 |
-
{
|
| 608 |
-
"epoch": 0.20683912214253095,
|
| 609 |
-
"grad_norm": 0.7768594622612,
|
| 610 |
-
"learning_rate": 4.7729345585300716e-05,
|
| 611 |
-
"loss": 1.4891,
|
| 612 |
-
"step": 7700
|
| 613 |
-
},
|
| 614 |
-
{
|
| 615 |
-
"epoch": 0.20952534450801838,
|
| 616 |
-
"grad_norm": 0.32350119948387146,
|
| 617 |
-
"learning_rate": 4.7662133375797635e-05,
|
| 618 |
-
"loss": 1.4846,
|
| 619 |
-
"step": 7800
|
| 620 |
-
},
|
| 621 |
-
{
|
| 622 |
-
"epoch": 0.21221156687350579,
|
| 623 |
-
"grad_norm": 0.3208902180194855,
|
| 624 |
-
"learning_rate": 4.7593989614752974e-05,
|
| 625 |
-
"loss": 1.4889,
|
| 626 |
-
"step": 7900
|
| 627 |
-
},
|
| 628 |
-
{
|
| 629 |
-
"epoch": 0.2148977892389932,
|
| 630 |
-
"grad_norm": 0.2309061586856842,
|
| 631 |
-
"learning_rate": 4.7524917103289863e-05,
|
| 632 |
-
"loss": 1.4831,
|
| 633 |
-
"step": 8000
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"epoch": 0.2148977892389932,
|
| 637 |
-
"eval_loss": 1.4867380857467651,
|
| 638 |
-
"eval_runtime": 3183.5875,
|
| 639 |
-
"eval_sacrebleu": 96.88666750889345,
|
| 640 |
-
"eval_samples_per_second": 83.153,
|
| 641 |
"eval_steps_per_second": 0.65,
|
| 642 |
-
"step":
|
| 643 |
-
},
|
| 644 |
-
{
|
| 645 |
-
"epoch": 0.21758401160448063,
|
| 646 |
-
"grad_norm": 0.33346688747406006,
|
| 647 |
-
"learning_rate": 4.74549186807087e-05,
|
| 648 |
-
"loss": 1.4812,
|
| 649 |
-
"step": 8100
|
| 650 |
-
},
|
| 651 |
-
{
|
| 652 |
-
"epoch": 0.22027023396996803,
|
| 653 |
-
"grad_norm": 0.5275943279266357,
|
| 654 |
-
"learning_rate": 4.738399722437047e-05,
|
| 655 |
-
"loss": 1.4874,
|
| 656 |
-
"step": 8200
|
| 657 |
-
},
|
| 658 |
-
{
|
| 659 |
-
"epoch": 0.22295645633545544,
|
| 660 |
-
"grad_norm": 0.4352237284183502,
|
| 661 |
-
"learning_rate": 4.731215564957847e-05,
|
| 662 |
-
"loss": 1.4912,
|
| 663 |
-
"step": 8300
|
| 664 |
-
},
|
| 665 |
-
{
|
| 666 |
-
"epoch": 0.22564267870094287,
|
| 667 |
-
"grad_norm": 0.3379707336425781,
|
| 668 |
-
"learning_rate": 4.723939690945846e-05,
|
| 669 |
-
"loss": 1.4852,
|
| 670 |
-
"step": 8400
|
| 671 |
-
},
|
| 672 |
-
{
|
| 673 |
-
"epoch": 0.22832890106643028,
|
| 674 |
-
"grad_norm": 0.40677231550216675,
|
| 675 |
-
"learning_rate": 4.7165723994837246e-05,
|
| 676 |
-
"loss": 1.4857,
|
| 677 |
-
"step": 8500
|
| 678 |
-
},
|
| 679 |
-
{
|
| 680 |
-
"epoch": 0.23101512343191768,
|
| 681 |
-
"grad_norm": 0.4737798273563385,
|
| 682 |
-
"learning_rate": 4.709113993411981e-05,
|
| 683 |
-
"loss": 1.4863,
|
| 684 |
-
"step": 8600
|
| 685 |
-
},
|
| 686 |
-
{
|
| 687 |
-
"epoch": 0.23370134579740512,
|
| 688 |
-
"grad_norm": 0.33614835143089294,
|
| 689 |
-
"learning_rate": 4.701564779316476e-05,
|
| 690 |
-
"loss": 1.485,
|
| 691 |
-
"step": 8700
|
| 692 |
-
},
|
| 693 |
-
{
|
| 694 |
-
"epoch": 0.23638756816289253,
|
| 695 |
-
"grad_norm": 0.4807875454425812,
|
| 696 |
-
"learning_rate": 4.693925067515834e-05,
|
| 697 |
-
"loss": 1.4875,
|
| 698 |
-
"step": 8800
|
| 699 |
-
},
|
| 700 |
-
{
|
| 701 |
-
"epoch": 0.23907379052837993,
|
| 702 |
-
"grad_norm": 0.3510216772556305,
|
| 703 |
-
"learning_rate": 4.686195172048682e-05,
|
| 704 |
-
"loss": 1.4839,
|
| 705 |
-
"step": 8900
|
| 706 |
-
},
|
| 707 |
-
{
|
| 708 |
-
"epoch": 0.24176001289386737,
|
| 709 |
-
"grad_norm": 0.24355977773666382,
|
| 710 |
-
"learning_rate": 4.678375410660748e-05,
|
| 711 |
-
"loss": 1.483,
|
| 712 |
-
"step": 9000
|
| 713 |
-
},
|
| 714 |
-
{
|
| 715 |
-
"epoch": 0.24176001289386737,
|
| 716 |
-
"eval_loss": 1.4836901426315308,
|
| 717 |
-
"eval_runtime": 3169.9641,
|
| 718 |
-
"eval_sacrebleu": 96.9796698855444,
|
| 719 |
-
"eval_samples_per_second": 83.511,
|
| 720 |
-
"eval_steps_per_second": 0.653,
|
| 721 |
-
"step": 9000
|
| 722 |
-
},
|
| 723 |
-
{
|
| 724 |
-
"epoch": 0.24444623525935477,
|
| 725 |
-
"grad_norm": 0.26222726702690125,
|
| 726 |
-
"learning_rate": 4.6704661047917955e-05,
|
| 727 |
-
"loss": 1.4855,
|
| 728 |
-
"step": 9100
|
| 729 |
-
},
|
| 730 |
-
{
|
| 731 |
-
"epoch": 0.24713245762484218,
|
| 732 |
-
"grad_norm": 0.3217809796333313,
|
| 733 |
-
"learning_rate": 4.6624675795624084e-05,
|
| 734 |
-
"loss": 1.4823,
|
| 735 |
-
"step": 9200
|
| 736 |
-
},
|
| 737 |
-
{
|
| 738 |
-
"epoch": 0.2498186799903296,
|
| 739 |
-
"grad_norm": 0.31736332178115845,
|
| 740 |
-
"learning_rate": 4.6543801637606324e-05,
|
| 741 |
-
"loss": 1.4791,
|
| 742 |
-
"step": 9300
|
| 743 |
-
},
|
| 744 |
-
{
|
| 745 |
-
"epoch": 0.252504902355817,
|
| 746 |
-
"grad_norm": 0.42374491691589355,
|
| 747 |
-
"learning_rate": 4.646204189828453e-05,
|
| 748 |
-
"loss": 1.4793,
|
| 749 |
-
"step": 9400
|
| 750 |
-
},
|
| 751 |
-
{
|
| 752 |
-
"epoch": 0.25519112472130445,
|
| 753 |
-
"grad_norm": 0.3236493468284607,
|
| 754 |
-
"learning_rate": 4.637939993848137e-05,
|
| 755 |
-
"loss": 1.4815,
|
| 756 |
-
"step": 9500
|
| 757 |
-
},
|
| 758 |
-
{
|
| 759 |
-
"epoch": 0.25787734708679183,
|
| 760 |
-
"grad_norm": 0.3557586371898651,
|
| 761 |
-
"learning_rate": 4.6295879155284125e-05,
|
| 762 |
-
"loss": 1.4822,
|
| 763 |
-
"step": 9600
|
| 764 |
-
},
|
| 765 |
-
{
|
| 766 |
-
"epoch": 0.26056356945227926,
|
| 767 |
-
"grad_norm": 0.2667744755744934,
|
| 768 |
-
"learning_rate": 4.621148298190506e-05,
|
| 769 |
-
"loss": 1.4804,
|
| 770 |
-
"step": 9700
|
| 771 |
-
},
|
| 772 |
-
{
|
| 773 |
-
"epoch": 0.2632497918177667,
|
| 774 |
-
"grad_norm": 0.2917531132698059,
|
| 775 |
-
"learning_rate": 4.61262148875403e-05,
|
| 776 |
-
"loss": 1.4782,
|
| 777 |
-
"step": 9800
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 0.2659360141832541,
|
| 781 |
-
"grad_norm": 0.32322603464126587,
|
| 782 |
-
"learning_rate": 4.604007837722725e-05,
|
| 783 |
-
"loss": 1.4815,
|
| 784 |
-
"step": 9900
|
| 785 |
-
},
|
| 786 |
-
{
|
| 787 |
-
"epoch": 0.2686222365487415,
|
| 788 |
-
"grad_norm": 0.3380342125892639,
|
| 789 |
-
"learning_rate": 4.595307699170045e-05,
|
| 790 |
-
"loss": 1.4816,
|
| 791 |
-
"step": 10000
|
| 792 |
-
},
|
| 793 |
-
{
|
| 794 |
-
"epoch": 0.2686222365487415,
|
| 795 |
-
"eval_loss": 1.4826488494873047,
|
| 796 |
-
"eval_runtime": 3178.5973,
|
| 797 |
-
"eval_sacrebleu": 97.05288129988465,
|
| 798 |
-
"eval_samples_per_second": 83.284,
|
| 799 |
-
"eval_steps_per_second": 0.651,
|
| 800 |
-
"step": 10000
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 0.27130845891422894,
|
| 804 |
-
"grad_norm": 0.34705039858818054,
|
| 805 |
-
"learning_rate": 4.586521430724612e-05,
|
| 806 |
-
"loss": 1.4829,
|
| 807 |
-
"step": 10100
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 0.2739946812797163,
|
| 811 |
-
"grad_norm": 0.3544181287288666,
|
| 812 |
-
"learning_rate": 4.5776493935555065e-05,
|
| 813 |
-
"loss": 1.481,
|
| 814 |
-
"step": 10200
|
| 815 |
-
},
|
| 816 |
-
{
|
| 817 |
-
"epoch": 0.27668090364520376,
|
| 818 |
-
"grad_norm": 0.3179134726524353,
|
| 819 |
-
"learning_rate": 4.568691952357428e-05,
|
| 820 |
-
"loss": 1.4783,
|
| 821 |
-
"step": 10300
|
| 822 |
-
},
|
| 823 |
-
{
|
| 824 |
-
"epoch": 0.2793671260106912,
|
| 825 |
-
"grad_norm": 0.32810860872268677,
|
| 826 |
-
"learning_rate": 4.5596494753357005e-05,
|
| 827 |
-
"loss": 1.4777,
|
| 828 |
-
"step": 10400
|
| 829 |
-
},
|
| 830 |
-
{
|
| 831 |
-
"epoch": 0.28205334837617857,
|
| 832 |
-
"grad_norm": 0.36927270889282227,
|
| 833 |
-
"learning_rate": 4.550522334191138e-05,
|
| 834 |
-
"loss": 1.4776,
|
| 835 |
-
"step": 10500
|
| 836 |
-
},
|
| 837 |
-
{
|
| 838 |
-
"epoch": 0.284739570741666,
|
| 839 |
-
"grad_norm": 0.2897135615348816,
|
| 840 |
-
"learning_rate": 4.541310904104762e-05,
|
| 841 |
-
"loss": 1.4768,
|
| 842 |
-
"step": 10600
|
| 843 |
-
},
|
| 844 |
-
{
|
| 845 |
-
"epoch": 0.28742579310715344,
|
| 846 |
-
"grad_norm": 0.31894442439079285,
|
| 847 |
-
"learning_rate": 4.532015563722387e-05,
|
| 848 |
-
"loss": 1.4753,
|
| 849 |
-
"step": 10700
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 0.2901120154726408,
|
| 853 |
-
"grad_norm": 0.3248063921928406,
|
| 854 |
-
"learning_rate": 4.522636695139049e-05,
|
| 855 |
-
"loss": 1.4755,
|
| 856 |
-
"step": 10800
|
| 857 |
-
},
|
| 858 |
-
{
|
| 859 |
-
"epoch": 0.29279823783812825,
|
| 860 |
-
"grad_norm": 0.28601497411727905,
|
| 861 |
-
"learning_rate": 4.513174683883299e-05,
|
| 862 |
-
"loss": 1.4789,
|
| 863 |
-
"step": 10900
|
| 864 |
-
},
|
| 865 |
-
{
|
| 866 |
-
"epoch": 0.29548446020361563,
|
| 867 |
-
"grad_norm": 0.263621985912323,
|
| 868 |
-
"learning_rate": 4.50362991890136e-05,
|
| 869 |
-
"loss": 1.4772,
|
| 870 |
-
"step": 11000
|
| 871 |
-
},
|
| 872 |
-
{
|
| 873 |
-
"epoch": 0.29548446020361563,
|
| 874 |
-
"eval_loss": 1.483279824256897,
|
| 875 |
-
"eval_runtime": 3185.3454,
|
| 876 |
-
"eval_sacrebleu": 97.12672190258174,
|
| 877 |
-
"eval_samples_per_second": 83.107,
|
| 878 |
-
"eval_steps_per_second": 0.65,
|
| 879 |
-
"step": 11000
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 0.29817068256910306,
|
| 883 |
-
"grad_norm": 0.32398489117622375,
|
| 884 |
-
"learning_rate": 4.494002792541137e-05,
|
| 885 |
-
"loss": 1.475,
|
| 886 |
-
"step": 11100
|
| 887 |
-
},
|
| 888 |
-
{
|
| 889 |
-
"epoch": 0.3008569049345905,
|
| 890 |
-
"grad_norm": 0.28194740414619446,
|
| 891 |
-
"learning_rate": 4.484293700536088e-05,
|
| 892 |
-
"loss": 1.478,
|
| 893 |
-
"step": 11200
|
| 894 |
-
},
|
| 895 |
-
{
|
| 896 |
-
"epoch": 0.3035431273000779,
|
| 897 |
-
"grad_norm": 0.25159361958503723,
|
| 898 |
-
"learning_rate": 4.4745030419889555e-05,
|
| 899 |
-
"loss": 1.4733,
|
| 900 |
-
"step": 11300
|
| 901 |
-
},
|
| 902 |
-
{
|
| 903 |
-
"epoch": 0.3062293496655653,
|
| 904 |
-
"grad_norm": 0.33246248960494995,
|
| 905 |
-
"learning_rate": 4.4646312193553675e-05,
|
| 906 |
-
"loss": 1.4786,
|
| 907 |
-
"step": 11400
|
| 908 |
-
},
|
| 909 |
-
{
|
| 910 |
-
"epoch": 0.30891557203105274,
|
| 911 |
-
"grad_norm": 0.2811319828033447,
|
| 912 |
-
"learning_rate": 4.4546786384272855e-05,
|
| 913 |
-
"loss": 1.4739,
|
| 914 |
-
"step": 11500
|
| 915 |
-
},
|
| 916 |
-
{
|
| 917 |
-
"epoch": 0.3116017943965401,
|
| 918 |
-
"grad_norm": 0.3391018807888031,
|
| 919 |
-
"learning_rate": 4.44464570831633e-05,
|
| 920 |
-
"loss": 1.4758,
|
| 921 |
-
"step": 11600
|
| 922 |
-
},
|
| 923 |
-
{
|
| 924 |
-
"epoch": 0.31428801676202756,
|
| 925 |
-
"grad_norm": 0.3480006158351898,
|
| 926 |
-
"learning_rate": 4.434532841436962e-05,
|
| 927 |
-
"loss": 1.4739,
|
| 928 |
-
"step": 11700
|
| 929 |
-
},
|
| 930 |
-
{
|
| 931 |
-
"epoch": 0.316974239127515,
|
| 932 |
-
"grad_norm": 0.32334116101264954,
|
| 933 |
-
"learning_rate": 4.42434045348953e-05,
|
| 934 |
-
"loss": 1.4757,
|
| 935 |
-
"step": 11800
|
| 936 |
-
},
|
| 937 |
-
{
|
| 938 |
-
"epoch": 0.31966046149300237,
|
| 939 |
-
"grad_norm": 0.35196009278297424,
|
| 940 |
-
"learning_rate": 4.41406896344318e-05,
|
| 941 |
-
"loss": 1.4759,
|
| 942 |
-
"step": 11900
|
| 943 |
-
},
|
| 944 |
-
{
|
| 945 |
-
"epoch": 0.3223466838584898,
|
| 946 |
-
"grad_norm": 0.31370338797569275,
|
| 947 |
-
"learning_rate": 4.4037187935186394e-05,
|
| 948 |
-
"loss": 1.4735,
|
| 949 |
-
"step": 12000
|
| 950 |
-
},
|
| 951 |
-
{
|
| 952 |
-
"epoch": 0.3223466838584898,
|
| 953 |
-
"eval_loss": 1.4803506135940552,
|
| 954 |
-
"eval_runtime": 3210.9887,
|
| 955 |
-
"eval_sacrebleu": 97.12202597988144,
|
| 956 |
-
"eval_samples_per_second": 82.444,
|
| 957 |
-
"eval_steps_per_second": 0.644,
|
| 958 |
-
"step": 12000
|
| 959 |
-
},
|
| 960 |
-
{
|
| 961 |
-
"epoch": 0.32503290622397724,
|
| 962 |
-
"grad_norm": 0.36732399463653564,
|
| 963 |
-
"learning_rate": 4.3932903691708534e-05,
|
| 964 |
-
"loss": 1.4759,
|
| 965 |
-
"step": 12100
|
| 966 |
-
},
|
| 967 |
-
{
|
| 968 |
-
"epoch": 0.3277191285894646,
|
| 969 |
-
"grad_norm": 0.23831887543201447,
|
| 970 |
-
"learning_rate": 4.382784119071502e-05,
|
| 971 |
-
"loss": 1.4719,
|
| 972 |
-
"step": 12200
|
| 973 |
-
},
|
| 974 |
-
{
|
| 975 |
-
"epoch": 0.33040535095495205,
|
| 976 |
-
"grad_norm": 0.3706836700439453,
|
| 977 |
-
"learning_rate": 4.372200475091376e-05,
|
| 978 |
-
"loss": 1.4739,
|
| 979 |
-
"step": 12300
|
| 980 |
-
},
|
| 981 |
-
{
|
| 982 |
-
"epoch": 0.3330915733204395,
|
| 983 |
-
"grad_norm": 0.3490886688232422,
|
| 984 |
-
"learning_rate": 4.3615398722826276e-05,
|
| 985 |
-
"loss": 1.474,
|
| 986 |
-
"step": 12400
|
| 987 |
-
},
|
| 988 |
-
{
|
| 989 |
-
"epoch": 0.33577779568592686,
|
| 990 |
-
"grad_norm": 1.0285382270812988,
|
| 991 |
-
"learning_rate": 4.350802748860882e-05,
|
| 992 |
-
"loss": 1.4734,
|
| 993 |
-
"step": 12500
|
| 994 |
-
},
|
| 995 |
-
{
|
| 996 |
-
"epoch": 0.3384640180514143,
|
| 997 |
-
"grad_norm": 0.26471439003944397,
|
| 998 |
-
"learning_rate": 4.339989546187229e-05,
|
| 999 |
-
"loss": 1.4745,
|
| 1000 |
-
"step": 12600
|
| 1001 |
-
},
|
| 1002 |
-
{
|
| 1003 |
-
"epoch": 0.34115024041690173,
|
| 1004 |
-
"grad_norm": 0.30338284373283386,
|
| 1005 |
-
"learning_rate": 4.329100708750079e-05,
|
| 1006 |
-
"loss": 1.4727,
|
| 1007 |
-
"step": 12700
|
| 1008 |
-
},
|
| 1009 |
-
{
|
| 1010 |
-
"epoch": 0.3438364627823891,
|
| 1011 |
-
"grad_norm": 0.290783166885376,
|
| 1012 |
-
"learning_rate": 4.3181366841468896e-05,
|
| 1013 |
-
"loss": 1.4722,
|
| 1014 |
-
"step": 12800
|
| 1015 |
-
},
|
| 1016 |
-
{
|
| 1017 |
-
"epoch": 0.34652268514787654,
|
| 1018 |
-
"grad_norm": 0.2663359045982361,
|
| 1019 |
-
"learning_rate": 4.3070979230657694e-05,
|
| 1020 |
-
"loss": 1.4729,
|
| 1021 |
-
"step": 12900
|
| 1022 |
-
},
|
| 1023 |
-
{
|
| 1024 |
-
"epoch": 0.349208907513364,
|
| 1025 |
-
"grad_norm": 1.0900615453720093,
|
| 1026 |
-
"learning_rate": 4.29598487926695e-05,
|
| 1027 |
-
"loss": 1.4707,
|
| 1028 |
-
"step": 13000
|
| 1029 |
-
},
|
| 1030 |
-
{
|
| 1031 |
-
"epoch": 0.349208907513364,
|
| 1032 |
-
"eval_loss": 1.4730943441390991,
|
| 1033 |
-
"eval_runtime": 3189.0879,
|
| 1034 |
-
"eval_sacrebleu": 97.19666890806242,
|
| 1035 |
-
"eval_samples_per_second": 83.01,
|
| 1036 |
-
"eval_steps_per_second": 0.649,
|
| 1037 |
-
"step": 13000
|
| 1038 |
}
|
| 1039 |
],
|
| 1040 |
"logging_steps": 100,
|
|
@@ -1063,7 +115,7 @@
|
|
| 1063 |
"attributes": {}
|
| 1064 |
}
|
| 1065 |
},
|
| 1066 |
-
"total_flos":
|
| 1067 |
"train_batch_size": 64,
|
| 1068 |
"trial_name": null,
|
| 1069 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 1.550229787826538,
|
| 4 |
+
"best_model_checkpoint": "hieptt/vietnamese-correction-finetuning/checkpoint-1000",
|
| 5 |
+
"epoch": 0.02686222365487415,
|
| 6 |
"eval_steps": 1000,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 82 |
{
|
| 83 |
"epoch": 0.02686222365487415,
|
| 84 |
"eval_loss": 1.550229787826538,
|
| 85 |
+
"eval_runtime": 3183.759,
|
| 86 |
"eval_sacrebleu": 96.21455916515954,
|
| 87 |
+
"eval_samples_per_second": 83.149,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
"eval_steps_per_second": 0.65,
|
| 89 |
+
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
}
|
| 91 |
],
|
| 92 |
"logging_steps": 100,
|
|
|
|
| 115 |
"attributes": {}
|
| 116 |
}
|
| 117 |
},
|
| 118 |
+
"total_flos": 1.6604548281925632e+16,
|
| 119 |
"train_batch_size": 64,
|
| 120 |
"trial_name": null,
|
| 121 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6033
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:522844a580fb08e9801add7b55973eec21ec1422f4a0aab9509ee801930f6aa1
|
| 3 |
size 6033
|