Commit
•
4935bdb
1
Parent(s):
8a92a34
Training in progress, step 53000
Browse files- adapter_model.bin +1 -1
- {checkpoint-49000 → checkpoint-52000/adapter_model}/README.md +0 -0
- {checkpoint-49000 → checkpoint-52000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-49000 → checkpoint-52000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-49000/adapter_model → checkpoint-53000}/README.md +0 -0
- {checkpoint-49000/adapter_model → checkpoint-53000}/adapter_config.json +0 -0
- {checkpoint-49000/adapter_model → checkpoint-53000}/adapter_model.bin +1 -1
- {checkpoint-49000 → checkpoint-53000}/optimizer.pt +1 -1
- {checkpoint-49000 → checkpoint-53000}/rng_state.pth +1 -1
- {checkpoint-49000 → checkpoint-53000}/scheduler.pt +1 -1
- {checkpoint-49000 → checkpoint-53000}/trainer_state.json +277 -5
- {checkpoint-49000 → checkpoint-53000}/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 sha256:99a57421cde685b96bae500764350fe2651e0aaccd813ec440d0624d1372eba0
|
| 3 |
size 500897101
|
{checkpoint-49000 → checkpoint-52000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-49000 → checkpoint-52000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-49000 → checkpoint-52000/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 sha256:30145c8f43899386a5afedd34f02e56e73fe9312e85b6dbd710652b42a9d35ee
|
| 3 |
size 500897101
|
{checkpoint-49000/adapter_model → checkpoint-53000}/README.md
RENAMED
|
File without changes
|
{checkpoint-49000/adapter_model → checkpoint-53000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-49000/adapter_model → checkpoint-53000}/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-49000 → checkpoint-53000}/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 sha256:1ad83706102ddec2d9ab08a057edef638806fd1452c631db78d7bb562a055671
|
| 3 |
size 1001723453
|
{checkpoint-49000 → checkpoint-53000}/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 sha256:6b6fb462ea132cfc986a28ed5c3f0fea0fc27fd5bbed37098c5926c8d6c4b0e8
|
| 3 |
size 14575
|
{checkpoint-49000 → checkpoint-53000}/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 sha256:56e98120f9a3c3a84f868e502ef4ed3e3a7223a4b4285cfb93038865bd6eef70
|
| 3 |
size 627
|
{checkpoint-49000 → checkpoint-53000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -3338,11 +3338,283 @@
|
|
| 3338 |
"eval_samples_per_second": 0.425,
|
| 3339 |
"eval_steps_per_second": 0.425,
|
| 3340 |
"step": 49000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3341 |
}
|
| 3342 |
],
|
| 3343 |
"max_steps": 80463,
|
| 3344 |
"num_train_epochs": 3,
|
| 3345 |
-
"total_flos": 1.
|
| 3346 |
"trial_name": null,
|
| 3347 |
"trial_params": null
|
| 3348 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4982399046421051,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-53000",
|
| 4 |
+
"epoch": 1.9760635323067746,
|
| 5 |
+
"global_step": 53000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 3338 |
"eval_samples_per_second": 0.425,
|
| 3339 |
"eval_steps_per_second": 0.425,
|
| 3340 |
"step": 49000
|
| 3341 |
+
},
|
| 3342 |
+
{
|
| 3343 |
+
"epoch": 1.83,
|
| 3344 |
+
"learning_rate": 6.607598886886645e-05,
|
| 3345 |
+
"loss": 0.4803,
|
| 3346 |
+
"step": 49100
|
| 3347 |
+
},
|
| 3348 |
+
{
|
| 3349 |
+
"epoch": 1.83,
|
| 3350 |
+
"learning_rate": 6.57089165026916e-05,
|
| 3351 |
+
"loss": 0.4703,
|
| 3352 |
+
"step": 49200
|
| 3353 |
+
},
|
| 3354 |
+
{
|
| 3355 |
+
"epoch": 1.84,
|
| 3356 |
+
"learning_rate": 6.534236700915406e-05,
|
| 3357 |
+
"loss": 0.4963,
|
| 3358 |
+
"step": 49300
|
| 3359 |
+
},
|
| 3360 |
+
{
|
| 3361 |
+
"epoch": 1.84,
|
| 3362 |
+
"learning_rate": 6.497634597742399e-05,
|
| 3363 |
+
"loss": 0.4694,
|
| 3364 |
+
"step": 49400
|
| 3365 |
+
},
|
| 3366 |
+
{
|
| 3367 |
+
"epoch": 1.85,
|
| 3368 |
+
"learning_rate": 6.46108589886135e-05,
|
| 3369 |
+
"loss": 0.4612,
|
| 3370 |
+
"step": 49500
|
| 3371 |
+
},
|
| 3372 |
+
{
|
| 3373 |
+
"epoch": 1.85,
|
| 3374 |
+
"learning_rate": 6.424591161569158e-05,
|
| 3375 |
+
"loss": 0.4361,
|
| 3376 |
+
"step": 49600
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 1.85,
|
| 3380 |
+
"learning_rate": 6.388150942339917e-05,
|
| 3381 |
+
"loss": 0.4652,
|
| 3382 |
+
"step": 49700
|
| 3383 |
+
},
|
| 3384 |
+
{
|
| 3385 |
+
"epoch": 1.86,
|
| 3386 |
+
"learning_rate": 6.35176579681642e-05,
|
| 3387 |
+
"loss": 0.4741,
|
| 3388 |
+
"step": 49800
|
| 3389 |
+
},
|
| 3390 |
+
{
|
| 3391 |
+
"epoch": 1.86,
|
| 3392 |
+
"learning_rate": 6.315436279801704e-05,
|
| 3393 |
+
"loss": 0.4447,
|
| 3394 |
+
"step": 49900
|
| 3395 |
+
},
|
| 3396 |
+
{
|
| 3397 |
+
"epoch": 1.86,
|
| 3398 |
+
"learning_rate": 6.279162945250561e-05,
|
| 3399 |
+
"loss": 0.4542,
|
| 3400 |
+
"step": 50000
|
| 3401 |
+
},
|
| 3402 |
+
{
|
| 3403 |
+
"epoch": 1.86,
|
| 3404 |
+
"eval_loss": 0.507805347442627,
|
| 3405 |
+
"eval_runtime": 1229.0064,
|
| 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 |
"max_steps": 80463,
|
| 3616 |
"num_train_epochs": 3,
|
| 3617 |
+
"total_flos": 1.4859235089879122e+19,
|
| 3618 |
"trial_name": null,
|
| 3619 |
"trial_params": null
|
| 3620 |
}
|
{checkpoint-49000 → checkpoint-53000}/training_args.bin
RENAMED
|
File without changes
|