Commit
•
8e479b2
1
Parent(s):
78ce51e
Training in progress, step 40000
Browse files- adapter_model.bin +1 -1
- {checkpoint-36000 → checkpoint-39000/adapter_model}/README.md +0 -0
- {checkpoint-36000 → checkpoint-39000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-36000 → checkpoint-39000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-36000/adapter_model → checkpoint-40000}/README.md +0 -0
- {checkpoint-36000/adapter_model → checkpoint-40000}/adapter_config.json +0 -0
- {checkpoint-36000/adapter_model → checkpoint-40000}/adapter_model.bin +1 -1
- {checkpoint-36000 → checkpoint-40000}/optimizer.pt +1 -1
- {checkpoint-36000 → checkpoint-40000}/rng_state.pth +1 -1
- {checkpoint-36000 → checkpoint-40000}/scheduler.pt +1 -1
- {checkpoint-36000 → checkpoint-40000}/trainer_state.json +277 -5
- {checkpoint-36000 → checkpoint-40000}/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 sha256:8b58822fe1e8262161f44bc61bc0ad0a9e7afd47e4892189dfa46e0a57e6dee1
|
| 3 |
size 500897101
|
{checkpoint-36000 → checkpoint-39000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-36000 → checkpoint-39000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-36000 → checkpoint-39000/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 sha256:4ac02c8b60ff2af2043c8bbd7012fcc6335dd99c02ab7c2e9f0bff405da9b9af
|
| 3 |
size 500897101
|
{checkpoint-36000/adapter_model → checkpoint-40000}/README.md
RENAMED
|
File without changes
|
{checkpoint-36000/adapter_model → checkpoint-40000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-36000/adapter_model → checkpoint-40000}/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 sha256:8b58822fe1e8262161f44bc61bc0ad0a9e7afd47e4892189dfa46e0a57e6dee1
|
| 3 |
size 500897101
|
{checkpoint-36000 → checkpoint-40000}/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:dd8245362ab2a8b70d0dc9675794736e143bd184a4de380caa12617c8833da7b
|
| 3 |
size 1001723453
|
{checkpoint-36000 → checkpoint-40000}/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 sha256:0c27ba48931454fb9cea28943695a625f9f93123a1cb450835a0e8e0f21bcf26
|
| 3 |
size 14575
|
{checkpoint-36000 → checkpoint-40000}/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:c82428b9badf946a212907d36a6d1a7db1b15e77a0dbc7be8d578794535936cc
|
| 3 |
size 627
|
{checkpoint-36000 → checkpoint-40000}/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,
|
|
@@ -2454,11 +2454,283 @@
|
|
| 2454 |
"eval_samples_per_second": 0.427,
|
| 2455 |
"eval_steps_per_second": 0.427,
|
| 2456 |
"step": 36000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2457 |
}
|
| 2458 |
],
|
| 2459 |
"max_steps": 80463,
|
| 2460 |
"num_train_epochs": 3,
|
| 2461 |
-
"total_flos": 1.
|
| 2462 |
"trial_name": null,
|
| 2463 |
"trial_params": null
|
| 2464 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5438262820243835,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-40000",
|
| 4 |
+
"epoch": 1.4913687036277543,
|
| 5 |
+
"global_step": 40000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 2454 |
"eval_samples_per_second": 0.427,
|
| 2455 |
"eval_steps_per_second": 0.427,
|
| 2456 |
"step": 36000
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 1.35,
|
| 2460 |
+
"learning_rate": 0.00011608238759007561,
|
| 2461 |
+
"loss": 0.5268,
|
| 2462 |
+
"step": 36100
|
| 2463 |
+
},
|
| 2464 |
+
{
|
| 2465 |
+
"epoch": 1.35,
|
| 2466 |
+
"learning_rate": 0.00011569686094752101,
|
| 2467 |
+
"loss": 0.5179,
|
| 2468 |
+
"step": 36200
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 1.35,
|
| 2472 |
+
"learning_rate": 0.00011531109495822545,
|
| 2473 |
+
"loss": 0.5236,
|
| 2474 |
+
"step": 36300
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.36,
|
| 2478 |
+
"learning_rate": 0.00011492509550437339,
|
| 2479 |
+
"loss": 0.5197,
|
| 2480 |
+
"step": 36400
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 1.36,
|
| 2484 |
+
"learning_rate": 0.0001145388684717092,
|
| 2485 |
+
"loss": 0.5109,
|
| 2486 |
+
"step": 36500
|
| 2487 |
+
},
|
| 2488 |
+
{
|
| 2489 |
+
"epoch": 1.36,
|
| 2490 |
+
"learning_rate": 0.00011415241974944744,
|
| 2491 |
+
"loss": 0.5126,
|
| 2492 |
+
"step": 36600
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 1.37,
|
| 2496 |
+
"learning_rate": 0.00011376575523018296,
|
| 2497 |
+
"loss": 0.501,
|
| 2498 |
+
"step": 36700
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 1.37,
|
| 2502 |
+
"learning_rate": 0.00011337888080980115,
|
| 2503 |
+
"loss": 0.4888,
|
| 2504 |
+
"step": 36800
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 1.38,
|
| 2508 |
+
"learning_rate": 0.00011299180238738789,
|
| 2509 |
+
"loss": 0.5324,
|
| 2510 |
+
"step": 36900
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 1.38,
|
| 2514 |
+
"learning_rate": 0.00011260452586513981,
|
| 2515 |
+
"loss": 0.5053,
|
| 2516 |
+
"step": 37000
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.38,
|
| 2520 |
+
"eval_loss": 0.5555862188339233,
|
| 2521 |
+
"eval_runtime": 1324.5375,
|
| 2522 |
+
"eval_samples_per_second": 0.409,
|
| 2523 |
+
"eval_steps_per_second": 0.409,
|
| 2524 |
+
"step": 37000
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 1.38,
|
| 2528 |
+
"learning_rate": 0.00011221705714827408,
|
| 2529 |
+
"loss": 0.4994,
|
| 2530 |
+
"step": 37100
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 1.39,
|
| 2534 |
+
"learning_rate": 0.00011182940214493858,
|
| 2535 |
+
"loss": 0.4826,
|
| 2536 |
+
"step": 37200
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 1.39,
|
| 2540 |
+
"learning_rate": 0.0001114415667661215,
|
| 2541 |
+
"loss": 0.5173,
|
| 2542 |
+
"step": 37300
|
| 2543 |
+
},
|
| 2544 |
+
{
|
| 2545 |
+
"epoch": 1.39,
|
| 2546 |
+
"learning_rate": 0.00011105355692556165,
|
| 2547 |
+
"loss": 0.4992,
|
| 2548 |
+
"step": 37400
|
| 2549 |
+
},
|
| 2550 |
+
{
|
| 2551 |
+
"epoch": 1.4,
|
| 2552 |
+
"learning_rate": 0.00011066537853965788,
|
| 2553 |
+
"loss": 0.5155,
|
| 2554 |
+
"step": 37500
|
| 2555 |
+
},
|
| 2556 |
+
{
|
| 2557 |
+
"epoch": 1.4,
|
| 2558 |
+
"learning_rate": 0.00011027703752737913,
|
| 2559 |
+
"loss": 0.5118,
|
| 2560 |
+
"step": 37600
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 1.41,
|
| 2564 |
+
"learning_rate": 0.00010988853981017393,
|
| 2565 |
+
"loss": 0.5078,
|
| 2566 |
+
"step": 37700
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 1.41,
|
| 2570 |
+
"learning_rate": 0.00010949989131188043,
|
| 2571 |
+
"loss": 0.506,
|
| 2572 |
+
"step": 37800
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 1.41,
|
| 2576 |
+
"learning_rate": 0.00010911109795863581,
|
| 2577 |
+
"loss": 0.5074,
|
| 2578 |
+
"step": 37900
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 1.42,
|
| 2582 |
+
"learning_rate": 0.00010872216567878599,
|
| 2583 |
+
"loss": 0.4837,
|
| 2584 |
+
"step": 38000
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 1.42,
|
| 2588 |
+
"eval_loss": 0.5529844164848328,
|
| 2589 |
+
"eval_runtime": 1281.6978,
|
| 2590 |
+
"eval_samples_per_second": 0.423,
|
| 2591 |
+
"eval_steps_per_second": 0.423,
|
| 2592 |
+
"step": 38000
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 1.42,
|
| 2596 |
+
"learning_rate": 0.00010833310040279531,
|
| 2597 |
+
"loss": 0.5161,
|
| 2598 |
+
"step": 38100
|
| 2599 |
+
},
|
| 2600 |
+
{
|
| 2601 |
+
"epoch": 1.42,
|
| 2602 |
+
"learning_rate": 0.00010794390806315602,
|
| 2603 |
+
"loss": 0.4929,
|
| 2604 |
+
"step": 38200
|
| 2605 |
+
},
|
| 2606 |
+
{
|
| 2607 |
+
"epoch": 1.43,
|
| 2608 |
+
"learning_rate": 0.0001075545945942978,
|
| 2609 |
+
"loss": 0.4819,
|
| 2610 |
+
"step": 38300
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 1.43,
|
| 2614 |
+
"learning_rate": 0.00010716516593249742,
|
| 2615 |
+
"loss": 0.4594,
|
| 2616 |
+
"step": 38400
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 1.44,
|
| 2620 |
+
"learning_rate": 0.00010677562801578798,
|
| 2621 |
+
"loss": 0.4802,
|
| 2622 |
+
"step": 38500
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 1.44,
|
| 2626 |
+
"learning_rate": 0.00010638598678386864,
|
| 2627 |
+
"loss": 0.4793,
|
| 2628 |
+
"step": 38600
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.44,
|
| 2632 |
+
"learning_rate": 0.00010599624817801383,
|
| 2633 |
+
"loss": 0.4912,
|
| 2634 |
+
"step": 38700
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 1.45,
|
| 2638 |
+
"learning_rate": 0.0001056064181409828,
|
| 2639 |
+
"loss": 0.519,
|
| 2640 |
+
"step": 38800
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 1.45,
|
| 2644 |
+
"learning_rate": 0.00010521650261692886,
|
| 2645 |
+
"loss": 0.4993,
|
| 2646 |
+
"step": 38900
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 1.45,
|
| 2650 |
+
"learning_rate": 0.00010482650755130898,
|
| 2651 |
+
"loss": 0.4688,
|
| 2652 |
+
"step": 39000
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 1.45,
|
| 2656 |
+
"eval_loss": 0.5486682057380676,
|
| 2657 |
+
"eval_runtime": 1265.4476,
|
| 2658 |
+
"eval_samples_per_second": 0.428,
|
| 2659 |
+
"eval_steps_per_second": 0.428,
|
| 2660 |
+
"step": 39000
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 1.46,
|
| 2664 |
+
"learning_rate": 0.00010443643889079282,
|
| 2665 |
+
"loss": 0.4901,
|
| 2666 |
+
"step": 39100
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 1.46,
|
| 2670 |
+
"learning_rate": 0.00010404630258317236,
|
| 2671 |
+
"loss": 0.4783,
|
| 2672 |
+
"step": 39200
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 1.47,
|
| 2676 |
+
"learning_rate": 0.00010365610457727095,
|
| 2677 |
+
"loss": 0.5026,
|
| 2678 |
+
"step": 39300
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 1.47,
|
| 2682 |
+
"learning_rate": 0.00010326585082285279,
|
| 2683 |
+
"loss": 0.4602,
|
| 2684 |
+
"step": 39400
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.47,
|
| 2688 |
+
"learning_rate": 0.00010287554727053215,
|
| 2689 |
+
"loss": 0.5299,
|
| 2690 |
+
"step": 39500
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 1.48,
|
| 2694 |
+
"learning_rate": 0.00010248519987168252,
|
| 2695 |
+
"loss": 0.5135,
|
| 2696 |
+
"step": 39600
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 1.48,
|
| 2700 |
+
"learning_rate": 0.00010209481457834616,
|
| 2701 |
+
"loss": 0.4792,
|
| 2702 |
+
"step": 39700
|
| 2703 |
+
},
|
| 2704 |
+
{
|
| 2705 |
+
"epoch": 1.48,
|
| 2706 |
+
"learning_rate": 0.0001017043973431429,
|
| 2707 |
+
"loss": 0.5004,
|
| 2708 |
+
"step": 39800
|
| 2709 |
+
},
|
| 2710 |
+
{
|
| 2711 |
+
"epoch": 1.49,
|
| 2712 |
+
"learning_rate": 0.00010131395411917979,
|
| 2713 |
+
"loss": 0.5013,
|
| 2714 |
+
"step": 39900
|
| 2715 |
+
},
|
| 2716 |
+
{
|
| 2717 |
+
"epoch": 1.49,
|
| 2718 |
+
"learning_rate": 0.00010092349085996011,
|
| 2719 |
+
"loss": 0.501,
|
| 2720 |
+
"step": 40000
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 1.49,
|
| 2724 |
+
"eval_loss": 0.5438262820243835,
|
| 2725 |
+
"eval_runtime": 1262.0322,
|
| 2726 |
+
"eval_samples_per_second": 0.429,
|
| 2727 |
+
"eval_steps_per_second": 0.429,
|
| 2728 |
+
"step": 40000
|
| 2729 |
}
|
| 2730 |
],
|
| 2731 |
"max_steps": 80463,
|
| 2732 |
"num_train_epochs": 3,
|
| 2733 |
+
"total_flos": 1.1217846349451428e+19,
|
| 2734 |
"trial_name": null,
|
| 2735 |
"trial_params": null
|
| 2736 |
}
|
{checkpoint-36000 → checkpoint-40000}/training_args.bin
RENAMED
|
File without changes
|