Commit
•
b212e47
1
Parent(s):
4935bdb
Training in progress, step 54000
Browse files- adapter_model.bin +1 -1
- {checkpoint-50000 → checkpoint-53000/adapter_model}/README.md +0 -0
- {checkpoint-50000 → checkpoint-53000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-50000 → checkpoint-53000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-50000/adapter_model → checkpoint-54000}/README.md +0 -0
- {checkpoint-50000/adapter_model → checkpoint-54000}/adapter_config.json +0 -0
- {checkpoint-50000/adapter_model → checkpoint-54000}/adapter_model.bin +1 -1
- {checkpoint-50000 → checkpoint-54000}/optimizer.pt +1 -1
- {checkpoint-50000 → checkpoint-54000}/rng_state.pth +1 -1
- {checkpoint-50000 → checkpoint-54000}/scheduler.pt +1 -1
- {checkpoint-50000 → checkpoint-54000}/trainer_state.json +277 -5
- {checkpoint-50000 → checkpoint-54000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5739df33a1c5a62aac85a2ab78da21029f60e5e1960f7fd67f909f42e710d8f
|
| 3 |
size 500897101
|
{checkpoint-50000 → checkpoint-53000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-50000 → checkpoint-53000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-50000 → checkpoint-53000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99a57421cde685b96bae500764350fe2651e0aaccd813ec440d0624d1372eba0
|
| 3 |
size 500897101
|
{checkpoint-50000/adapter_model → checkpoint-54000}/README.md
RENAMED
|
File without changes
|
{checkpoint-50000/adapter_model → checkpoint-54000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-50000/adapter_model → checkpoint-54000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5739df33a1c5a62aac85a2ab78da21029f60e5e1960f7fd67f909f42e710d8f
|
| 3 |
size 500897101
|
{checkpoint-50000 → checkpoint-54000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:993e6ef82c425c2f345e27b366c350d42c56342038cdb59cf0f161075041fbe1
|
| 3 |
size 1001723453
|
{checkpoint-50000 → checkpoint-54000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa135a6298e01e499df0474065ae456053371e96deb5cfc76234a2e39202ca29
|
| 3 |
size 14575
|
{checkpoint-50000 → checkpoint-54000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b1d12c4db8a34c37702ab77719c13fbbb7d616d4895ab7bc2162e3f563a8277
|
| 3 |
size 627
|
{checkpoint-50000 → checkpoint-54000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -3406,11 +3406,283 @@
|
|
| 3406 |
"eval_samples_per_second": 0.441,
|
| 3407 |
"eval_steps_per_second": 0.441,
|
| 3408 |
"step": 50000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3409 |
}
|
| 3410 |
],
|
| 3411 |
"max_steps": 80463,
|
| 3412 |
"num_train_epochs": 3,
|
| 3413 |
-
"total_flos": 1.
|
| 3414 |
"trial_name": null,
|
| 3415 |
"trial_params": null
|
| 3416 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4982399046421051,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-53000",
|
| 4 |
+
"epoch": 2.0133477498974686,
|
| 5 |
+
"global_step": 54000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 3406 |
"eval_samples_per_second": 0.441,
|
| 3407 |
"eval_steps_per_second": 0.441,
|
| 3408 |
"step": 50000
|
| 3409 |
+
},
|
| 3410 |
+
{
|
| 3411 |
+
"epoch": 1.87,
|
| 3412 |
+
"learning_rate": 6.242946346261127e-05,
|
| 3413 |
+
"loss": 0.4431,
|
| 3414 |
+
"step": 50100
|
| 3415 |
+
},
|
| 3416 |
+
{
|
| 3417 |
+
"epoch": 1.87,
|
| 3418 |
+
"learning_rate": 6.20678703506642e-05,
|
| 3419 |
+
"loss": 0.4631,
|
| 3420 |
+
"step": 50200
|
| 3421 |
+
},
|
| 3422 |
+
{
|
| 3423 |
+
"epoch": 1.88,
|
| 3424 |
+
"learning_rate": 6.170685563025928e-05,
|
| 3425 |
+
"loss": 0.4655,
|
| 3426 |
+
"step": 50300
|
| 3427 |
+
},
|
| 3428 |
+
{
|
| 3429 |
+
"epoch": 1.88,
|
| 3430 |
+
"learning_rate": 6.13464248061722e-05,
|
| 3431 |
+
"loss": 0.4741,
|
| 3432 |
+
"step": 50400
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 1.88,
|
| 3436 |
+
"learning_rate": 6.098658337427517e-05,
|
| 3437 |
+
"loss": 0.4362,
|
| 3438 |
+
"step": 50500
|
| 3439 |
+
},
|
| 3440 |
+
{
|
| 3441 |
+
"epoch": 1.89,
|
| 3442 |
+
"learning_rate": 6.0627336821453426e-05,
|
| 3443 |
+
"loss": 0.4501,
|
| 3444 |
+
"step": 50600
|
| 3445 |
+
},
|
| 3446 |
+
{
|
| 3447 |
+
"epoch": 1.89,
|
| 3448 |
+
"learning_rate": 6.0268690625521364e-05,
|
| 3449 |
+
"loss": 0.4638,
|
| 3450 |
+
"step": 50700
|
| 3451 |
+
},
|
| 3452 |
+
{
|
| 3453 |
+
"epoch": 1.89,
|
| 3454 |
+
"learning_rate": 5.99106502551392e-05,
|
| 3455 |
+
"loss": 0.4499,
|
| 3456 |
+
"step": 50800
|
| 3457 |
+
},
|
| 3458 |
+
{
|
| 3459 |
+
"epoch": 1.9,
|
| 3460 |
+
"learning_rate": 5.955322116972939e-05,
|
| 3461 |
+
"loss": 0.4655,
|
| 3462 |
+
"step": 50900
|
| 3463 |
+
},
|
| 3464 |
+
{
|
| 3465 |
+
"epoch": 1.9,
|
| 3466 |
+
"learning_rate": 5.919640881939357e-05,
|
| 3467 |
+
"loss": 0.4688,
|
| 3468 |
+
"step": 51000
|
| 3469 |
+
},
|
| 3470 |
+
{
|
| 3471 |
+
"epoch": 1.9,
|
| 3472 |
+
"eval_loss": 0.50446617603302,
|
| 3473 |
+
"eval_runtime": 1222.4661,
|
| 3474 |
+
"eval_samples_per_second": 0.443,
|
| 3475 |
+
"eval_steps_per_second": 0.443,
|
| 3476 |
+
"step": 51000
|
| 3477 |
+
},
|
| 3478 |
+
{
|
| 3479 |
+
"epoch": 1.91,
|
| 3480 |
+
"learning_rate": 5.884021864482925e-05,
|
| 3481 |
+
"loss": 0.4476,
|
| 3482 |
+
"step": 51100
|
| 3483 |
+
},
|
| 3484 |
+
{
|
| 3485 |
+
"epoch": 1.91,
|
| 3486 |
+
"learning_rate": 5.8484656077247066e-05,
|
| 3487 |
+
"loss": 0.4693,
|
| 3488 |
+
"step": 51200
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 1.91,
|
| 3492 |
+
"learning_rate": 5.812972653828779e-05,
|
| 3493 |
+
"loss": 0.4592,
|
| 3494 |
+
"step": 51300
|
| 3495 |
+
},
|
| 3496 |
+
{
|
| 3497 |
+
"epoch": 1.92,
|
| 3498 |
+
"learning_rate": 5.777543543993975e-05,
|
| 3499 |
+
"loss": 0.4863,
|
| 3500 |
+
"step": 51400
|
| 3501 |
+
},
|
| 3502 |
+
{
|
| 3503 |
+
"epoch": 1.92,
|
| 3504 |
+
"learning_rate": 5.742178818445638e-05,
|
| 3505 |
+
"loss": 0.4615,
|
| 3506 |
+
"step": 51500
|
| 3507 |
+
},
|
| 3508 |
+
{
|
| 3509 |
+
"epoch": 1.92,
|
| 3510 |
+
"learning_rate": 5.706879016427364e-05,
|
| 3511 |
+
"loss": 0.4625,
|
| 3512 |
+
"step": 51600
|
| 3513 |
+
},
|
| 3514 |
+
{
|
| 3515 |
+
"epoch": 1.93,
|
| 3516 |
+
"learning_rate": 5.6716446761927885e-05,
|
| 3517 |
+
"loss": 0.4534,
|
| 3518 |
+
"step": 51700
|
| 3519 |
+
},
|
| 3520 |
+
{
|
| 3521 |
+
"epoch": 1.93,
|
| 3522 |
+
"learning_rate": 5.6364763349973995e-05,
|
| 3523 |
+
"loss": 0.4881,
|
| 3524 |
+
"step": 51800
|
| 3525 |
+
},
|
| 3526 |
+
{
|
| 3527 |
+
"epoch": 1.94,
|
| 3528 |
+
"learning_rate": 5.601374529090308e-05,
|
| 3529 |
+
"loss": 0.4474,
|
| 3530 |
+
"step": 51900
|
| 3531 |
+
},
|
| 3532 |
+
{
|
| 3533 |
+
"epoch": 1.94,
|
| 3534 |
+
"learning_rate": 5.566339793706102e-05,
|
| 3535 |
+
"loss": 0.4301,
|
| 3536 |
+
"step": 52000
|
| 3537 |
+
},
|
| 3538 |
+
{
|
| 3539 |
+
"epoch": 1.94,
|
| 3540 |
+
"eval_loss": 0.5012514591217041,
|
| 3541 |
+
"eval_runtime": 1223.7216,
|
| 3542 |
+
"eval_samples_per_second": 0.443,
|
| 3543 |
+
"eval_steps_per_second": 0.443,
|
| 3544 |
+
"step": 52000
|
| 3545 |
+
},
|
| 3546 |
+
{
|
| 3547 |
+
"epoch": 1.94,
|
| 3548 |
+
"learning_rate": 5.531372663056664e-05,
|
| 3549 |
+
"loss": 0.4523,
|
| 3550 |
+
"step": 52100
|
| 3551 |
+
},
|
| 3552 |
+
{
|
| 3553 |
+
"epoch": 1.95,
|
| 3554 |
+
"learning_rate": 5.496473670323052e-05,
|
| 3555 |
+
"loss": 0.4434,
|
| 3556 |
+
"step": 52200
|
| 3557 |
+
},
|
| 3558 |
+
{
|
| 3559 |
+
"epoch": 1.95,
|
| 3560 |
+
"learning_rate": 5.461643347647335e-05,
|
| 3561 |
+
"loss": 0.4239,
|
| 3562 |
+
"step": 52300
|
| 3563 |
+
},
|
| 3564 |
+
{
|
| 3565 |
+
"epoch": 1.95,
|
| 3566 |
+
"learning_rate": 5.4268822261245023e-05,
|
| 3567 |
+
"loss": 0.4581,
|
| 3568 |
+
"step": 52400
|
| 3569 |
+
},
|
| 3570 |
+
{
|
| 3571 |
+
"epoch": 1.96,
|
| 3572 |
+
"learning_rate": 5.392190835794369e-05,
|
| 3573 |
+
"loss": 0.4297,
|
| 3574 |
+
"step": 52500
|
| 3575 |
+
},
|
| 3576 |
+
{
|
| 3577 |
+
"epoch": 1.96,
|
| 3578 |
+
"learning_rate": 5.357569705633465e-05,
|
| 3579 |
+
"loss": 0.4431,
|
| 3580 |
+
"step": 52600
|
| 3581 |
+
},
|
| 3582 |
+
{
|
| 3583 |
+
"epoch": 1.96,
|
| 3584 |
+
"learning_rate": 5.3230193635470136e-05,
|
| 3585 |
+
"loss": 0.4406,
|
| 3586 |
+
"step": 52700
|
| 3587 |
+
},
|
| 3588 |
+
{
|
| 3589 |
+
"epoch": 1.97,
|
| 3590 |
+
"learning_rate": 5.288540336360836e-05,
|
| 3591 |
+
"loss": 0.4734,
|
| 3592 |
+
"step": 52800
|
| 3593 |
+
},
|
| 3594 |
+
{
|
| 3595 |
+
"epoch": 1.97,
|
| 3596 |
+
"learning_rate": 5.254133149813349e-05,
|
| 3597 |
+
"loss": 0.4449,
|
| 3598 |
+
"step": 52900
|
| 3599 |
+
},
|
| 3600 |
+
{
|
| 3601 |
+
"epoch": 1.98,
|
| 3602 |
+
"learning_rate": 5.2197983285475315e-05,
|
| 3603 |
+
"loss": 0.4519,
|
| 3604 |
+
"step": 53000
|
| 3605 |
+
},
|
| 3606 |
+
{
|
| 3607 |
+
"epoch": 1.98,
|
| 3608 |
+
"eval_loss": 0.4982399046421051,
|
| 3609 |
+
"eval_runtime": 1234.2642,
|
| 3610 |
+
"eval_samples_per_second": 0.439,
|
| 3611 |
+
"eval_steps_per_second": 0.439,
|
| 3612 |
+
"step": 53000
|
| 3613 |
+
},
|
| 3614 |
+
{
|
| 3615 |
+
"epoch": 1.98,
|
| 3616 |
+
"learning_rate": 5.185536396102946e-05,
|
| 3617 |
+
"loss": 0.4352,
|
| 3618 |
+
"step": 53100
|
| 3619 |
+
},
|
| 3620 |
+
{
|
| 3621 |
+
"epoch": 1.98,
|
| 3622 |
+
"learning_rate": 5.1513478749077274e-05,
|
| 3623 |
+
"loss": 0.4801,
|
| 3624 |
+
"step": 53200
|
| 3625 |
+
},
|
| 3626 |
+
{
|
| 3627 |
+
"epoch": 1.99,
|
| 3628 |
+
"learning_rate": 5.1172332862706376e-05,
|
| 3629 |
+
"loss": 0.4315,
|
| 3630 |
+
"step": 53300
|
| 3631 |
+
},
|
| 3632 |
+
{
|
| 3633 |
+
"epoch": 1.99,
|
| 3634 |
+
"learning_rate": 5.0831931503731065e-05,
|
| 3635 |
+
"loss": 0.4443,
|
| 3636 |
+
"step": 53400
|
| 3637 |
+
},
|
| 3638 |
+
{
|
| 3639 |
+
"epoch": 1.99,
|
| 3640 |
+
"learning_rate": 5.049227986261302e-05,
|
| 3641 |
+
"loss": 0.4597,
|
| 3642 |
+
"step": 53500
|
| 3643 |
+
},
|
| 3644 |
+
{
|
| 3645 |
+
"epoch": 2.0,
|
| 3646 |
+
"learning_rate": 5.0153383118382355e-05,
|
| 3647 |
+
"loss": 0.4519,
|
| 3648 |
+
"step": 53600
|
| 3649 |
+
},
|
| 3650 |
+
{
|
| 3651 |
+
"epoch": 2.0,
|
| 3652 |
+
"learning_rate": 4.9815246438558264e-05,
|
| 3653 |
+
"loss": 0.4314,
|
| 3654 |
+
"step": 53700
|
| 3655 |
+
},
|
| 3656 |
+
{
|
| 3657 |
+
"epoch": 2.01,
|
| 3658 |
+
"learning_rate": 4.9477874979070474e-05,
|
| 3659 |
+
"loss": 0.3999,
|
| 3660 |
+
"step": 53800
|
| 3661 |
+
},
|
| 3662 |
+
{
|
| 3663 |
+
"epoch": 2.01,
|
| 3664 |
+
"learning_rate": 4.914127388418062e-05,
|
| 3665 |
+
"loss": 0.3937,
|
| 3666 |
+
"step": 53900
|
| 3667 |
+
},
|
| 3668 |
+
{
|
| 3669 |
+
"epoch": 2.01,
|
| 3670 |
+
"learning_rate": 4.880544828640372e-05,
|
| 3671 |
+
"loss": 0.3597,
|
| 3672 |
+
"step": 54000
|
| 3673 |
+
},
|
| 3674 |
+
{
|
| 3675 |
+
"epoch": 2.01,
|
| 3676 |
+
"eval_loss": 0.49871668219566345,
|
| 3677 |
+
"eval_runtime": 1230.6045,
|
| 3678 |
+
"eval_samples_per_second": 0.44,
|
| 3679 |
+
"eval_steps_per_second": 0.44,
|
| 3680 |
+
"step": 54000
|
| 3681 |
}
|
| 3682 |
],
|
| 3683 |
"max_steps": 80463,
|
| 3684 |
"num_train_epochs": 3,
|
| 3685 |
+
"total_flos": 1.514015028223746e+19,
|
| 3686 |
"trial_name": null,
|
| 3687 |
"trial_params": null
|
| 3688 |
}
|
{checkpoint-50000 → checkpoint-54000}/training_args.bin
RENAMED
|
File without changes
|