Training in progress, step 400, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9449344
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6f4af6ee9a79036c3432f7863a7ea6635bb72490198327b3753fb1ea22e3c16
|
| 3 |
size 9449344
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 18957003
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f1b6c34ca882eb6f6aeb37145c5a48effe196201acfb528d16564492d8b82a0
|
| 3 |
size 18957003
|
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:e80b99b5666058ebd2a0e81f6f0f52804008d2aa3a037ea01efc13a990d4fe7e
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -308,6 +308,306 @@
|
|
| 308 |
"rewards/margins": 10.570683479309082,
|
| 309 |
"rewards/rejected": -9.561734199523926,
|
| 310 |
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 311 |
}
|
| 312 |
],
|
| 313 |
"logging_steps": 10,
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.2385567317727747,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 400,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 308 |
"rewards/margins": 10.570683479309082,
|
| 309 |
"rewards/rejected": -9.561734199523926,
|
| 310 |
"step": 200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12524228418070674,
|
| 314 |
+
"grad_norm": 0.010515117086470127,
|
| 315 |
+
"learning_rate": 4.9861698520346155e-05,
|
| 316 |
+
"logits/chosen": -58.907005310058594,
|
| 317 |
+
"logits/rejected": -58.76160430908203,
|
| 318 |
+
"logps/chosen": -3922.297607421875,
|
| 319 |
+
"logps/rejected": -3831.158935546875,
|
| 320 |
+
"loss": 0.0106,
|
| 321 |
+
"rewards/accuracies": 1.0,
|
| 322 |
+
"rewards/chosen": 5.980830669403076,
|
| 323 |
+
"rewards/margins": 11.232372283935547,
|
| 324 |
+
"rewards/rejected": -5.251542091369629,
|
| 325 |
+
"step": 210
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.1312062024750261,
|
| 329 |
+
"grad_norm": 0.13617613911628723,
|
| 330 |
+
"learning_rate": 4.983518724298652e-05,
|
| 331 |
+
"logits/chosen": -57.034385681152344,
|
| 332 |
+
"logits/rejected": -61.353546142578125,
|
| 333 |
+
"logps/chosen": -3717.315185546875,
|
| 334 |
+
"logps/rejected": -4125.3466796875,
|
| 335 |
+
"loss": 0.0403,
|
| 336 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 337 |
+
"rewards/chosen": 11.408642768859863,
|
| 338 |
+
"rewards/margins": 12.029062271118164,
|
| 339 |
+
"rewards/rejected": -0.6204200983047485,
|
| 340 |
+
"step": 220
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.13717012076934545,
|
| 344 |
+
"grad_norm": 0.11460208892822266,
|
| 345 |
+
"learning_rate": 4.980636108439712e-05,
|
| 346 |
+
"logits/chosen": -60.95244216918945,
|
| 347 |
+
"logits/rejected": -62.050682067871094,
|
| 348 |
+
"logps/chosen": -3819.20654296875,
|
| 349 |
+
"logps/rejected": -3770.43701171875,
|
| 350 |
+
"loss": 0.2529,
|
| 351 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 352 |
+
"rewards/chosen": -2.9785029888153076,
|
| 353 |
+
"rewards/margins": 11.722539901733398,
|
| 354 |
+
"rewards/rejected": -14.701044082641602,
|
| 355 |
+
"step": 230
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.14313403906366481,
|
| 359 |
+
"grad_norm": 0.010561717674136162,
|
| 360 |
+
"learning_rate": 4.97752227314566e-05,
|
| 361 |
+
"logits/chosen": -61.96656036376953,
|
| 362 |
+
"logits/rejected": -63.59345245361328,
|
| 363 |
+
"logps/chosen": -3559.760986328125,
|
| 364 |
+
"logps/rejected": -3238.27099609375,
|
| 365 |
+
"loss": 0.0605,
|
| 366 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 367 |
+
"rewards/chosen": -3.5705113410949707,
|
| 368 |
+
"rewards/margins": 10.212821960449219,
|
| 369 |
+
"rewards/rejected": -13.783332824707031,
|
| 370 |
+
"step": 240
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.1490979573579842,
|
| 374 |
+
"grad_norm": 0.020034722983837128,
|
| 375 |
+
"learning_rate": 4.9741775086562576e-05,
|
| 376 |
+
"logits/chosen": -58.21870040893555,
|
| 377 |
+
"logits/rejected": -59.43880081176758,
|
| 378 |
+
"logps/chosen": -3953.829345703125,
|
| 379 |
+
"logps/rejected": -4362.8837890625,
|
| 380 |
+
"loss": 0.0013,
|
| 381 |
+
"rewards/accuracies": 1.0,
|
| 382 |
+
"rewards/chosen": 5.396712779998779,
|
| 383 |
+
"rewards/margins": 12.898712158203125,
|
| 384 |
+
"rewards/rejected": -7.5019989013671875,
|
| 385 |
+
"step": 250
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.15506187565230356,
|
| 389 |
+
"grad_norm": 0.0008314030710607767,
|
| 390 |
+
"learning_rate": 4.970602126736118e-05,
|
| 391 |
+
"logits/chosen": -58.50162887573242,
|
| 392 |
+
"logits/rejected": -59.48447799682617,
|
| 393 |
+
"logps/chosen": -3606.218017578125,
|
| 394 |
+
"logps/rejected": -3528.481689453125,
|
| 395 |
+
"loss": 0.0016,
|
| 396 |
+
"rewards/accuracies": 1.0,
|
| 397 |
+
"rewards/chosen": 9.148120880126953,
|
| 398 |
+
"rewards/margins": 12.437234878540039,
|
| 399 |
+
"rewards/rejected": -3.2891151905059814,
|
| 400 |
+
"step": 260
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.16102579394662292,
|
| 404 |
+
"grad_norm": 1.2535442113876343,
|
| 405 |
+
"learning_rate": 4.966796460645644e-05,
|
| 406 |
+
"logits/chosen": -59.570526123046875,
|
| 407 |
+
"logits/rejected": -63.26567459106445,
|
| 408 |
+
"logps/chosen": -3971.397705078125,
|
| 409 |
+
"logps/rejected": -4289.9853515625,
|
| 410 |
+
"loss": 0.0029,
|
| 411 |
+
"rewards/accuracies": 1.0,
|
| 412 |
+
"rewards/chosen": 4.893758296966553,
|
| 413 |
+
"rewards/margins": 15.240409851074219,
|
| 414 |
+
"rewards/rejected": -10.346652030944824,
|
| 415 |
+
"step": 270
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1669897122409423,
|
| 419 |
+
"grad_norm": 0.0032445124816149473,
|
| 420 |
+
"learning_rate": 4.962760865109964e-05,
|
| 421 |
+
"logits/chosen": -56.13077926635742,
|
| 422 |
+
"logits/rejected": -57.030982971191406,
|
| 423 |
+
"logps/chosen": -3904.32177734375,
|
| 424 |
+
"logps/rejected": -3896.450439453125,
|
| 425 |
+
"loss": 0.0019,
|
| 426 |
+
"rewards/accuracies": 1.0,
|
| 427 |
+
"rewards/chosen": 10.48493766784668,
|
| 428 |
+
"rewards/margins": 15.189654350280762,
|
| 429 |
+
"rewards/rejected": -4.704716205596924,
|
| 430 |
+
"step": 280
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.17295363053526167,
|
| 434 |
+
"grad_norm": 0.0017609696369618177,
|
| 435 |
+
"learning_rate": 4.95849571628587e-05,
|
| 436 |
+
"logits/chosen": -57.822837829589844,
|
| 437 |
+
"logits/rejected": -56.972389221191406,
|
| 438 |
+
"logps/chosen": -3716.139892578125,
|
| 439 |
+
"logps/rejected": -3481.774658203125,
|
| 440 |
+
"loss": 0.0103,
|
| 441 |
+
"rewards/accuracies": 1.0,
|
| 442 |
+
"rewards/chosen": 14.023122787475586,
|
| 443 |
+
"rewards/margins": 13.122570991516113,
|
| 444 |
+
"rewards/rejected": 0.9005520939826965,
|
| 445 |
+
"step": 290
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.17891754882958102,
|
| 449 |
+
"grad_norm": 1.900453805923462,
|
| 450 |
+
"learning_rate": 4.954001411726755e-05,
|
| 451 |
+
"logits/chosen": -57.431907653808594,
|
| 452 |
+
"logits/rejected": -60.528594970703125,
|
| 453 |
+
"logps/chosen": -4033.428955078125,
|
| 454 |
+
"logps/rejected": -4488.1083984375,
|
| 455 |
+
"loss": 0.0132,
|
| 456 |
+
"rewards/accuracies": 1.0,
|
| 457 |
+
"rewards/chosen": 9.56352710723877,
|
| 458 |
+
"rewards/margins": 13.36528491973877,
|
| 459 |
+
"rewards/rejected": -3.8017585277557373,
|
| 460 |
+
"step": 300
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.1848814671239004,
|
| 464 |
+
"grad_norm": 0.022368023172020912,
|
| 465 |
+
"learning_rate": 4.949278370345558e-05,
|
| 466 |
+
"logits/chosen": -57.648948669433594,
|
| 467 |
+
"logits/rejected": -61.04389190673828,
|
| 468 |
+
"logps/chosen": -3662.002685546875,
|
| 469 |
+
"logps/rejected": -3926.045654296875,
|
| 470 |
+
"loss": 0.2272,
|
| 471 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 472 |
+
"rewards/chosen": 6.355870246887207,
|
| 473 |
+
"rewards/margins": 15.903188705444336,
|
| 474 |
+
"rewards/rejected": -9.547318458557129,
|
| 475 |
+
"step": 310
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.19084538541821977,
|
| 479 |
+
"grad_norm": 0.0004944842658005655,
|
| 480 |
+
"learning_rate": 4.944327032375716e-05,
|
| 481 |
+
"logits/chosen": -59.660667419433594,
|
| 482 |
+
"logits/rejected": -62.95856857299805,
|
| 483 |
+
"logps/chosen": -3589.598388671875,
|
| 484 |
+
"logps/rejected": -3629.106201171875,
|
| 485 |
+
"loss": 0.0003,
|
| 486 |
+
"rewards/accuracies": 1.0,
|
| 487 |
+
"rewards/chosen": 6.476762294769287,
|
| 488 |
+
"rewards/margins": 13.896219253540039,
|
| 489 |
+
"rewards/rejected": -7.419455528259277,
|
| 490 |
+
"step": 320
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.19680930371253913,
|
| 494 |
+
"grad_norm": 0.00029502142569981515,
|
| 495 |
+
"learning_rate": 4.93914785933013e-05,
|
| 496 |
+
"logits/chosen": -55.67151641845703,
|
| 497 |
+
"logits/rejected": -57.00590896606445,
|
| 498 |
+
"logps/chosen": -3930.76025390625,
|
| 499 |
+
"logps/rejected": -3948.296142578125,
|
| 500 |
+
"loss": 0.0028,
|
| 501 |
+
"rewards/accuracies": 1.0,
|
| 502 |
+
"rewards/chosen": 11.375188827514648,
|
| 503 |
+
"rewards/margins": 14.264836311340332,
|
| 504 |
+
"rewards/rejected": -2.889647960662842,
|
| 505 |
+
"step": 330
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.20277322200685852,
|
| 509 |
+
"grad_norm": 0.003186926245689392,
|
| 510 |
+
"learning_rate": 4.9337413339581494e-05,
|
| 511 |
+
"logits/chosen": -55.91204833984375,
|
| 512 |
+
"logits/rejected": -55.10901641845703,
|
| 513 |
+
"logps/chosen": -3886.98583984375,
|
| 514 |
+
"logps/rejected": -3999.08251953125,
|
| 515 |
+
"loss": 0.0001,
|
| 516 |
+
"rewards/accuracies": 1.0,
|
| 517 |
+
"rewards/chosen": 14.179794311523438,
|
| 518 |
+
"rewards/margins": 14.276512145996094,
|
| 519 |
+
"rewards/rejected": -0.09671908617019653,
|
| 520 |
+
"step": 340
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.20873714030117788,
|
| 524 |
+
"grad_norm": 0.0017121023265644908,
|
| 525 |
+
"learning_rate": 4.928107960200573e-05,
|
| 526 |
+
"logits/chosen": -52.622962951660156,
|
| 527 |
+
"logits/rejected": -53.773536682128906,
|
| 528 |
+
"logps/chosen": -3416.02392578125,
|
| 529 |
+
"logps/rejected": -3253.180419921875,
|
| 530 |
+
"loss": 0.0008,
|
| 531 |
+
"rewards/accuracies": 1.0,
|
| 532 |
+
"rewards/chosen": 18.344406127929688,
|
| 533 |
+
"rewards/margins": 12.657001495361328,
|
| 534 |
+
"rewards/rejected": 5.687404632568359,
|
| 535 |
+
"step": 350
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.21470105859549724,
|
| 539 |
+
"grad_norm": 0.005526215769350529,
|
| 540 |
+
"learning_rate": 4.922248263142678e-05,
|
| 541 |
+
"logits/chosen": -56.997161865234375,
|
| 542 |
+
"logits/rejected": -56.7293701171875,
|
| 543 |
+
"logps/chosen": -3787.90234375,
|
| 544 |
+
"logps/rejected": -3675.44677734375,
|
| 545 |
+
"loss": 0.0075,
|
| 546 |
+
"rewards/accuracies": 1.0,
|
| 547 |
+
"rewards/chosen": 20.720577239990234,
|
| 548 |
+
"rewards/margins": 16.067211151123047,
|
| 549 |
+
"rewards/rejected": 4.653366565704346,
|
| 550 |
+
"step": 360
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.22066497688981662,
|
| 554 |
+
"grad_norm": 0.00579728651791811,
|
| 555 |
+
"learning_rate": 4.916162788965275e-05,
|
| 556 |
+
"logits/chosen": -55.556427001953125,
|
| 557 |
+
"logits/rejected": -57.451927185058594,
|
| 558 |
+
"logps/chosen": -3876.120361328125,
|
| 559 |
+
"logps/rejected": -4547.07275390625,
|
| 560 |
+
"loss": 0.058,
|
| 561 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 562 |
+
"rewards/chosen": 19.52408218383789,
|
| 563 |
+
"rewards/margins": 14.548626899719238,
|
| 564 |
+
"rewards/rejected": 4.975453853607178,
|
| 565 |
+
"step": 370
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.22662889518413598,
|
| 569 |
+
"grad_norm": 0.002147270832210779,
|
| 570 |
+
"learning_rate": 4.909852104893803e-05,
|
| 571 |
+
"logits/chosen": -55.28306198120117,
|
| 572 |
+
"logits/rejected": -57.85760498046875,
|
| 573 |
+
"logps/chosen": -3812.12646484375,
|
| 574 |
+
"logps/rejected": -3860.81787109375,
|
| 575 |
+
"loss": 0.0032,
|
| 576 |
+
"rewards/accuracies": 1.0,
|
| 577 |
+
"rewards/chosen": 14.829846382141113,
|
| 578 |
+
"rewards/margins": 15.175271987915039,
|
| 579 |
+
"rewards/rejected": -0.34542423486709595,
|
| 580 |
+
"step": 380
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 0.23259281347845534,
|
| 584 |
+
"grad_norm": 2.6284790237696143e-06,
|
| 585 |
+
"learning_rate": 4.903316799145453e-05,
|
| 586 |
+
"logits/chosen": -56.578338623046875,
|
| 587 |
+
"logits/rejected": -61.98085403442383,
|
| 588 |
+
"logps/chosen": -3731.21484375,
|
| 589 |
+
"logps/rejected": -3811.68017578125,
|
| 590 |
+
"loss": 0.0001,
|
| 591 |
+
"rewards/accuracies": 1.0,
|
| 592 |
+
"rewards/chosen": 15.379071235656738,
|
| 593 |
+
"rewards/margins": 15.874186515808105,
|
| 594 |
+
"rewards/rejected": -0.4951143264770508,
|
| 595 |
+
"step": 390
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.2385567317727747,
|
| 599 |
+
"grad_norm": 0.0006998952012509108,
|
| 600 |
+
"learning_rate": 4.896557480874345e-05,
|
| 601 |
+
"logits/chosen": -55.025390625,
|
| 602 |
+
"logits/rejected": -56.02961349487305,
|
| 603 |
+
"logps/chosen": -4079.14990234375,
|
| 604 |
+
"logps/rejected": -3936.82666015625,
|
| 605 |
+
"loss": 0.0222,
|
| 606 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 607 |
+
"rewards/chosen": 16.036401748657227,
|
| 608 |
+
"rewards/margins": 15.827476501464844,
|
| 609 |
+
"rewards/rejected": 0.20892485976219177,
|
| 610 |
+
"step": 400
|
| 611 |
}
|
| 612 |
],
|
| 613 |
"logging_steps": 10,
|