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 204500912
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6578cecfe50e7d3a20d040c9b1cdda32c58f9cfcb54fdc5334ee95fe69c2eddf
|
| 3 |
size 204500912
|
last-checkpoint/optimizer.pt
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 sha256:9d1cf02cb795b4d866231697a267728073ceac5b926c4192928980f2ac671e58
|
| 3 |
+
size 104062923
|
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 sha256:181c5f0270cf39930062ddfa3767a2481d0c360f120b11f8e25dbf533a1cdaba
|
| 3 |
size 14645
|
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:b147ab74f32279b151eadea15baf5f3d04f7bd6e1b4f97bb054a7a56cf5a70d9
|
| 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":
|
| 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": 456.1576232910156,
|
| 309 |
"rewards/rejected": -455.028564453125,
|
| 310 |
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 311 |
}
|
| 312 |
],
|
| 313 |
"logging_steps": 10,
|
|
@@ -322,7 +622,7 @@
|
|
| 322 |
"should_evaluate": false,
|
| 323 |
"should_log": false,
|
| 324 |
"should_save": true,
|
| 325 |
-
"should_training_stop":
|
| 326 |
},
|
| 327 |
"attributes": {}
|
| 328 |
}
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0611498836822866,
|
| 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": 456.1576232910156,
|
| 309 |
"rewards/rejected": -455.028564453125,
|
| 310 |
"step": 200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5583250249252243,
|
| 314 |
+
"grad_norm": 32.07512283325195,
|
| 315 |
+
"learning_rate": 5.457142857142858e-06,
|
| 316 |
+
"logits/chosen": 7.640280723571777,
|
| 317 |
+
"logits/rejected": 8.078705787658691,
|
| 318 |
+
"logps/chosen": -507.60870361328125,
|
| 319 |
+
"logps/rejected": -1413.518798828125,
|
| 320 |
+
"loss": 0.023566444218158723,
|
| 321 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 322 |
+
"rewards/chosen": -28.288721084594727,
|
| 323 |
+
"rewards/margins": 461.63006591796875,
|
| 324 |
+
"rewards/rejected": -489.9187927246094,
|
| 325 |
+
"step": 210
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5849119308740446,
|
| 329 |
+
"grad_norm": 0.0,
|
| 330 |
+
"learning_rate": 5.171428571428571e-06,
|
| 331 |
+
"logits/chosen": 7.952935218811035,
|
| 332 |
+
"logits/rejected": 8.260855674743652,
|
| 333 |
+
"logps/chosen": -547.0781860351562,
|
| 334 |
+
"logps/rejected": -1421.848388671875,
|
| 335 |
+
"loss": 0.036107948422431944,
|
| 336 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 337 |
+
"rewards/chosen": -42.91097640991211,
|
| 338 |
+
"rewards/margins": 435.43353271484375,
|
| 339 |
+
"rewards/rejected": -478.34454345703125,
|
| 340 |
+
"step": 220
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.6114988368228648,
|
| 344 |
+
"grad_norm": 0.0,
|
| 345 |
+
"learning_rate": 4.885714285714286e-06,
|
| 346 |
+
"logits/chosen": 8.114786148071289,
|
| 347 |
+
"logits/rejected": 8.457345962524414,
|
| 348 |
+
"logps/chosen": -507.6402893066406,
|
| 349 |
+
"logps/rejected": -1458.050537109375,
|
| 350 |
+
"loss": 0.014430716633796692,
|
| 351 |
+
"rewards/accuracies": 0.981249988079071,
|
| 352 |
+
"rewards/chosen": -36.03577423095703,
|
| 353 |
+
"rewards/margins": 463.03021240234375,
|
| 354 |
+
"rewards/rejected": -499.06591796875,
|
| 355 |
+
"step": 230
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.638085742771685,
|
| 359 |
+
"grad_norm": 0.0,
|
| 360 |
+
"learning_rate": 4.600000000000001e-06,
|
| 361 |
+
"logits/chosen": 7.714877128601074,
|
| 362 |
+
"logits/rejected": 8.194944381713867,
|
| 363 |
+
"logps/chosen": -527.2571411132812,
|
| 364 |
+
"logps/rejected": -1552.201171875,
|
| 365 |
+
"loss": 0.07465354204177857,
|
| 366 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 367 |
+
"rewards/chosen": -26.272327423095703,
|
| 368 |
+
"rewards/margins": 538.8110961914062,
|
| 369 |
+
"rewards/rejected": -565.0833740234375,
|
| 370 |
+
"step": 240
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.6646726487205051,
|
| 374 |
+
"grad_norm": 0.0,
|
| 375 |
+
"learning_rate": 4.314285714285714e-06,
|
| 376 |
+
"logits/chosen": 8.05278205871582,
|
| 377 |
+
"logits/rejected": 8.397611618041992,
|
| 378 |
+
"logps/chosen": -562.1492919921875,
|
| 379 |
+
"logps/rejected": -1432.1495361328125,
|
| 380 |
+
"loss": 1.750390303248341e-16,
|
| 381 |
+
"rewards/accuracies": 1.0,
|
| 382 |
+
"rewards/chosen": -13.465299606323242,
|
| 383 |
+
"rewards/margins": 452.33758544921875,
|
| 384 |
+
"rewards/rejected": -465.80291748046875,
|
| 385 |
+
"step": 250
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.6912595546693253,
|
| 389 |
+
"grad_norm": 0.0,
|
| 390 |
+
"learning_rate": 4.028571428571429e-06,
|
| 391 |
+
"logits/chosen": 7.9839582443237305,
|
| 392 |
+
"logits/rejected": 8.325165748596191,
|
| 393 |
+
"logps/chosen": -531.9362182617188,
|
| 394 |
+
"logps/rejected": -1311.642578125,
|
| 395 |
+
"loss": 0.5010682582855225,
|
| 396 |
+
"rewards/accuracies": 0.981249988079071,
|
| 397 |
+
"rewards/chosen": -7.1212358474731445,
|
| 398 |
+
"rewards/margins": 406.3907165527344,
|
| 399 |
+
"rewards/rejected": -413.511962890625,
|
| 400 |
+
"step": 260
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.7178464606181456,
|
| 404 |
+
"grad_norm": 72.22783660888672,
|
| 405 |
+
"learning_rate": 3.742857142857143e-06,
|
| 406 |
+
"logits/chosen": 7.493854522705078,
|
| 407 |
+
"logits/rejected": 7.991011142730713,
|
| 408 |
+
"logps/chosen": -503.574462890625,
|
| 409 |
+
"logps/rejected": -1281.570068359375,
|
| 410 |
+
"loss": 0.3572836875915527,
|
| 411 |
+
"rewards/accuracies": 0.981249988079071,
|
| 412 |
+
"rewards/chosen": -10.670049667358398,
|
| 413 |
+
"rewards/margins": 382.18377685546875,
|
| 414 |
+
"rewards/rejected": -392.85382080078125,
|
| 415 |
+
"step": 270
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7444333665669658,
|
| 419 |
+
"grad_norm": 209.91946411132812,
|
| 420 |
+
"learning_rate": 3.4571428571428574e-06,
|
| 421 |
+
"logits/chosen": 7.536074161529541,
|
| 422 |
+
"logits/rejected": 7.9550018310546875,
|
| 423 |
+
"logps/chosen": -423.2442321777344,
|
| 424 |
+
"logps/rejected": -1381.945068359375,
|
| 425 |
+
"loss": 0.241387939453125,
|
| 426 |
+
"rewards/accuracies": 0.987500011920929,
|
| 427 |
+
"rewards/chosen": -2.9049324989318848,
|
| 428 |
+
"rewards/margins": 460.4957580566406,
|
| 429 |
+
"rewards/rejected": -463.40069580078125,
|
| 430 |
+
"step": 280
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.771020272515786,
|
| 434 |
+
"grad_norm": 0.0,
|
| 435 |
+
"learning_rate": 3.1714285714285714e-06,
|
| 436 |
+
"logits/chosen": 7.7217535972595215,
|
| 437 |
+
"logits/rejected": 8.095100402832031,
|
| 438 |
+
"logps/chosen": -452.80010986328125,
|
| 439 |
+
"logps/rejected": -1453.189208984375,
|
| 440 |
+
"loss": 7.635563088115305e-06,
|
| 441 |
+
"rewards/accuracies": 1.0,
|
| 442 |
+
"rewards/chosen": -13.30943489074707,
|
| 443 |
+
"rewards/margins": 486.92889404296875,
|
| 444 |
+
"rewards/rejected": -500.23834228515625,
|
| 445 |
+
"step": 290
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.7976071784646062,
|
| 449 |
+
"grad_norm": 0.0,
|
| 450 |
+
"learning_rate": 2.885714285714286e-06,
|
| 451 |
+
"logits/chosen": 7.356575012207031,
|
| 452 |
+
"logits/rejected": 7.875399589538574,
|
| 453 |
+
"logps/chosen": -446.17254638671875,
|
| 454 |
+
"logps/rejected": -1462.5728759765625,
|
| 455 |
+
"loss": 0.026995140314102172,
|
| 456 |
+
"rewards/accuracies": 0.987500011920929,
|
| 457 |
+
"rewards/chosen": -26.922204971313477,
|
| 458 |
+
"rewards/margins": 505.7860412597656,
|
| 459 |
+
"rewards/rejected": -532.708251953125,
|
| 460 |
+
"step": 300
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.8241940844134263,
|
| 464 |
+
"grad_norm": 0.0,
|
| 465 |
+
"learning_rate": 2.6e-06,
|
| 466 |
+
"logits/chosen": 7.945906162261963,
|
| 467 |
+
"logits/rejected": 8.316286087036133,
|
| 468 |
+
"logps/chosen": -521.7435302734375,
|
| 469 |
+
"logps/rejected": -1435.0,
|
| 470 |
+
"loss": 2.198499298095703,
|
| 471 |
+
"rewards/accuracies": 0.956250011920929,
|
| 472 |
+
"rewards/chosen": -26.506107330322266,
|
| 473 |
+
"rewards/margins": 458.09234619140625,
|
| 474 |
+
"rewards/rejected": -484.59844970703125,
|
| 475 |
+
"step": 310
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.8507809903622466,
|
| 479 |
+
"grad_norm": 0.0,
|
| 480 |
+
"learning_rate": 2.3142857142857145e-06,
|
| 481 |
+
"logits/chosen": 7.83941650390625,
|
| 482 |
+
"logits/rejected": 8.193696975708008,
|
| 483 |
+
"logps/chosen": -551.3535766601562,
|
| 484 |
+
"logps/rejected": -1372.135498046875,
|
| 485 |
+
"loss": 0.004357181861996651,
|
| 486 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 487 |
+
"rewards/chosen": -26.256458282470703,
|
| 488 |
+
"rewards/margins": 423.38323974609375,
|
| 489 |
+
"rewards/rejected": -449.63970947265625,
|
| 490 |
+
"step": 320
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.8773678963110668,
|
| 494 |
+
"grad_norm": 0.0,
|
| 495 |
+
"learning_rate": 2.028571428571429e-06,
|
| 496 |
+
"logits/chosen": 7.511332035064697,
|
| 497 |
+
"logits/rejected": 8.009842872619629,
|
| 498 |
+
"logps/chosen": -482.1061096191406,
|
| 499 |
+
"logps/rejected": -1416.7646484375,
|
| 500 |
+
"loss": 0.0004897473379969596,
|
| 501 |
+
"rewards/accuracies": 1.0,
|
| 502 |
+
"rewards/chosen": -24.751493453979492,
|
| 503 |
+
"rewards/margins": 472.49639892578125,
|
| 504 |
+
"rewards/rejected": -497.2478942871094,
|
| 505 |
+
"step": 330
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.903954802259887,
|
| 509 |
+
"grad_norm": 4.885124487524457e-19,
|
| 510 |
+
"learning_rate": 1.7428571428571432e-06,
|
| 511 |
+
"logits/chosen": 7.67673397064209,
|
| 512 |
+
"logits/rejected": 8.061668395996094,
|
| 513 |
+
"logps/chosen": -477.88983154296875,
|
| 514 |
+
"logps/rejected": -1371.306396484375,
|
| 515 |
+
"loss": 0.02846769690513611,
|
| 516 |
+
"rewards/accuracies": 0.987500011920929,
|
| 517 |
+
"rewards/chosen": -19.97150230407715,
|
| 518 |
+
"rewards/margins": 429.90509033203125,
|
| 519 |
+
"rewards/rejected": -449.87664794921875,
|
| 520 |
+
"step": 340
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9305417082087072,
|
| 524 |
+
"grad_norm": 0.0,
|
| 525 |
+
"learning_rate": 1.4571428571428573e-06,
|
| 526 |
+
"logits/chosen": 7.819458961486816,
|
| 527 |
+
"logits/rejected": 8.190754890441895,
|
| 528 |
+
"logps/chosen": -501.7386169433594,
|
| 529 |
+
"logps/rejected": -1499.830322265625,
|
| 530 |
+
"loss": 0.0005395257845520973,
|
| 531 |
+
"rewards/accuracies": 1.0,
|
| 532 |
+
"rewards/chosen": -13.413067817687988,
|
| 533 |
+
"rewards/margins": 516.6709594726562,
|
| 534 |
+
"rewards/rejected": -530.083984375,
|
| 535 |
+
"step": 350
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9571286141575274,
|
| 539 |
+
"grad_norm": 0.0,
|
| 540 |
+
"learning_rate": 1.1714285714285715e-06,
|
| 541 |
+
"logits/chosen": 7.6596269607543945,
|
| 542 |
+
"logits/rejected": 8.10220718383789,
|
| 543 |
+
"logps/chosen": -496.450439453125,
|
| 544 |
+
"logps/rejected": -1371.864501953125,
|
| 545 |
+
"loss": 0.00866500437259674,
|
| 546 |
+
"rewards/accuracies": 0.987500011920929,
|
| 547 |
+
"rewards/chosen": -29.596303939819336,
|
| 548 |
+
"rewards/margins": 433.4952697753906,
|
| 549 |
+
"rewards/rejected": -463.09149169921875,
|
| 550 |
+
"step": 360
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.9837155201063477,
|
| 554 |
+
"grad_norm": 1.5750356366836245e-15,
|
| 555 |
+
"learning_rate": 8.857142857142857e-07,
|
| 556 |
+
"logits/chosen": 7.884603977203369,
|
| 557 |
+
"logits/rejected": 8.13658618927002,
|
| 558 |
+
"logps/chosen": -544.7962646484375,
|
| 559 |
+
"logps/rejected": -1370.7030029296875,
|
| 560 |
+
"loss": 0.030383533239364623,
|
| 561 |
+
"rewards/accuracies": 0.987500011920929,
|
| 562 |
+
"rewards/chosen": -27.421581268310547,
|
| 563 |
+
"rewards/margins": 431.65802001953125,
|
| 564 |
+
"rewards/rejected": -459.07965087890625,
|
| 565 |
+
"step": 370
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 1.007976071784646,
|
| 569 |
+
"grad_norm": 0.0,
|
| 570 |
+
"learning_rate": 6.000000000000001e-07,
|
| 571 |
+
"logits/chosen": 7.871831893920898,
|
| 572 |
+
"logits/rejected": 8.307570457458496,
|
| 573 |
+
"logps/chosen": -519.7576904296875,
|
| 574 |
+
"logps/rejected": -1396.640869140625,
|
| 575 |
+
"loss": 0.06974923014640808,
|
| 576 |
+
"rewards/accuracies": 0.9931507110595703,
|
| 577 |
+
"rewards/chosen": -26.421640396118164,
|
| 578 |
+
"rewards/margins": 436.3954162597656,
|
| 579 |
+
"rewards/rejected": -462.8170471191406,
|
| 580 |
+
"step": 380
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 1.0345629777334662,
|
| 584 |
+
"grad_norm": 3.976285881214031e-21,
|
| 585 |
+
"learning_rate": 3.1428571428571433e-07,
|
| 586 |
+
"logits/chosen": 7.635078430175781,
|
| 587 |
+
"logits/rejected": 8.017423629760742,
|
| 588 |
+
"logps/chosen": -461.38946533203125,
|
| 589 |
+
"logps/rejected": -1499.245361328125,
|
| 590 |
+
"loss": 0.03160061240196228,
|
| 591 |
+
"rewards/accuracies": 0.987500011920929,
|
| 592 |
+
"rewards/chosen": -30.11972427368164,
|
| 593 |
+
"rewards/margins": 512.0826416015625,
|
| 594 |
+
"rewards/rejected": -542.2023315429688,
|
| 595 |
+
"step": 390
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.0611498836822866,
|
| 599 |
+
"grad_norm": 0.0,
|
| 600 |
+
"learning_rate": 2.8571428571428575e-08,
|
| 601 |
+
"logits/chosen": 7.7975664138793945,
|
| 602 |
+
"logits/rejected": 8.254159927368164,
|
| 603 |
+
"logps/chosen": -477.83282470703125,
|
| 604 |
+
"logps/rejected": -1497.345703125,
|
| 605 |
+
"loss": 0.0043321698904037476,
|
| 606 |
+
"rewards/accuracies": 0.9937499761581421,
|
| 607 |
+
"rewards/chosen": -29.0898380279541,
|
| 608 |
+
"rewards/margins": 499.4642639160156,
|
| 609 |
+
"rewards/rejected": -528.5540771484375,
|
| 610 |
+
"step": 400
|
| 611 |
}
|
| 612 |
],
|
| 613 |
"logging_steps": 10,
|
|
|
|
| 622 |
"should_evaluate": false,
|
| 623 |
"should_log": false,
|
| 624 |
"should_save": true,
|
| 625 |
+
"should_training_stop": true
|
| 626 |
},
|
| 627 |
"attributes": {}
|
| 628 |
}
|