Training in progress, step 140000
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/rng_state_4.pth +2 -2
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +2 -2
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +203 -3
- pytorch_model.bin +1 -1
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 202194449
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e1a663fa8a1428dfd0e1ea579f8f1884a2fe84a94e67275218c60462158fe8aa
|
3 |
size 202194449
|
last-checkpoint/pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 102501541
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a40275396ce53417c9c00526c28902e900eee7fe8a1176e3b167b8fa880709f1
|
3 |
size 102501541
|
last-checkpoint/rng_state_0.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ff6473d36a3b0d61e638f6f7adf0e3c4e3a4c90311741bef45eedc8752b4abbb
|
3 |
size 14503
|
last-checkpoint/rng_state_1.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14439
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:abd20d5dfc80e3bb5cba79ef18d635e6b9d3fba3685d49236ba9bbfcbd245f3d
|
3 |
size 14439
|
last-checkpoint/rng_state_2.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2963be3c0faa28379747d9429ff572f2788db89f6185c9c53832cb9290c1c849
|
3 |
size 14503
|
last-checkpoint/rng_state_3.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14439
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bc4f1c2a017f1ba8bd941830cb07d47e6bde89cb5e80ca0663de3f779aa3e16b
|
3 |
size 14439
|
last-checkpoint/rng_state_4.pth
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 sha256:52188af98e11936af8bc69f5366b88b44cd6c73d9b1b37682a5e46e4706f5e3f
|
3 |
+
size 14439
|
last-checkpoint/rng_state_5.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:299331297e38ed6c53d0026997de39e06ade131719f6e2f227655bc716b72e7d
|
3 |
size 14503
|
last-checkpoint/rng_state_6.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d145aa3067eb9b9e66af4f261f18929e80b865f364d9ea9075acacb8581bccd8
|
3 |
size 14503
|
last-checkpoint/rng_state_7.pth
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:e8dc2be3757aa011e41ae06df8c828d05d218a32a27ad90767a1b48fa565a909
|
3 |
+
size 14439
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f9522f343e5b1c37fe7e600f7b39d619e850350d8948dc240940a440582a0eb9
|
3 |
size 623
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -2606,11 +2606,211 @@
|
|
2606 |
"eval_samples_per_second": 1938.903,
|
2607 |
"eval_steps_per_second": 31.022,
|
2608 |
"step": 130000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2609 |
}
|
2610 |
],
|
2611 |
"max_steps": 500000,
|
2612 |
"num_train_epochs": 16,
|
2613 |
-
"total_flos": 4.
|
2614 |
"trial_name": null,
|
2615 |
"trial_params": null
|
2616 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 4.289347100094979,
|
5 |
+
"global_step": 140000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
2606 |
"eval_samples_per_second": 1938.903,
|
2607 |
"eval_steps_per_second": 31.022,
|
2608 |
"step": 130000
|
2609 |
+
},
|
2610 |
+
{
|
2611 |
+
"epoch": 4.0,
|
2612 |
+
"learning_rate": 0.0002661106834948409,
|
2613 |
+
"loss": 0.3626,
|
2614 |
+
"step": 130500
|
2615 |
+
},
|
2616 |
+
{
|
2617 |
+
"epoch": 4.01,
|
2618 |
+
"learning_rate": 0.0002658019908268041,
|
2619 |
+
"loss": 0.3625,
|
2620 |
+
"step": 131000
|
2621 |
+
},
|
2622 |
+
{
|
2623 |
+
"epoch": 4.01,
|
2624 |
+
"eval_loss": 0.7918245792388916,
|
2625 |
+
"eval_runtime": 0.521,
|
2626 |
+
"eval_samples_per_second": 1919.333,
|
2627 |
+
"eval_steps_per_second": 30.709,
|
2628 |
+
"step": 131000
|
2629 |
+
},
|
2630 |
+
{
|
2631 |
+
"epoch": 4.03,
|
2632 |
+
"learning_rate": 0.00026549208644594766,
|
2633 |
+
"loss": 0.3622,
|
2634 |
+
"step": 131500
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 4.04,
|
2638 |
+
"learning_rate": 0.00026518097374133627,
|
2639 |
+
"loss": 0.3622,
|
2640 |
+
"step": 132000
|
2641 |
+
},
|
2642 |
+
{
|
2643 |
+
"epoch": 4.04,
|
2644 |
+
"eval_loss": 0.7886719107627869,
|
2645 |
+
"eval_runtime": 0.5166,
|
2646 |
+
"eval_samples_per_second": 1935.893,
|
2647 |
+
"eval_steps_per_second": 30.974,
|
2648 |
+
"step": 132000
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 4.06,
|
2652 |
+
"learning_rate": 0.00026486865611524853,
|
2653 |
+
"loss": 0.3622,
|
2654 |
+
"step": 132500
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 4.07,
|
2658 |
+
"learning_rate": 0.00026455513698314003,
|
2659 |
+
"loss": 0.3616,
|
2660 |
+
"step": 133000
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 4.07,
|
2664 |
+
"eval_loss": 0.7906731963157654,
|
2665 |
+
"eval_runtime": 0.5202,
|
2666 |
+
"eval_samples_per_second": 1922.502,
|
2667 |
+
"eval_steps_per_second": 30.76,
|
2668 |
+
"step": 133000
|
2669 |
+
},
|
2670 |
+
{
|
2671 |
+
"epoch": 4.09,
|
2672 |
+
"learning_rate": 0.0002642404197736058,
|
2673 |
+
"loss": 0.3618,
|
2674 |
+
"step": 133500
|
2675 |
+
},
|
2676 |
+
{
|
2677 |
+
"epoch": 4.11,
|
2678 |
+
"learning_rate": 0.0002639245079283428,
|
2679 |
+
"loss": 0.3646,
|
2680 |
+
"step": 134000
|
2681 |
+
},
|
2682 |
+
{
|
2683 |
+
"epoch": 4.11,
|
2684 |
+
"eval_loss": 0.7901037931442261,
|
2685 |
+
"eval_runtime": 0.5304,
|
2686 |
+
"eval_samples_per_second": 1885.342,
|
2687 |
+
"eval_steps_per_second": 30.165,
|
2688 |
+
"step": 134000
|
2689 |
+
},
|
2690 |
+
{
|
2691 |
+
"epoch": 4.12,
|
2692 |
+
"learning_rate": 0.00026360740490211234,
|
2693 |
+
"loss": 0.3612,
|
2694 |
+
"step": 134500
|
2695 |
+
},
|
2696 |
+
{
|
2697 |
+
"epoch": 4.14,
|
2698 |
+
"learning_rate": 0.0002632891141627023,
|
2699 |
+
"loss": 0.3604,
|
2700 |
+
"step": 135000
|
2701 |
+
},
|
2702 |
+
{
|
2703 |
+
"epoch": 4.14,
|
2704 |
+
"eval_loss": 0.7916610836982727,
|
2705 |
+
"eval_runtime": 0.5131,
|
2706 |
+
"eval_samples_per_second": 1948.858,
|
2707 |
+
"eval_steps_per_second": 31.182,
|
2708 |
+
"step": 135000
|
2709 |
+
},
|
2710 |
+
{
|
2711 |
+
"epoch": 4.15,
|
2712 |
+
"learning_rate": 0.00026296963919088923,
|
2713 |
+
"loss": 0.3603,
|
2714 |
+
"step": 135500
|
2715 |
+
},
|
2716 |
+
{
|
2717 |
+
"epoch": 4.17,
|
2718 |
+
"learning_rate": 0.00026264898348040024,
|
2719 |
+
"loss": 0.3604,
|
2720 |
+
"step": 136000
|
2721 |
+
},
|
2722 |
+
{
|
2723 |
+
"epoch": 4.17,
|
2724 |
+
"eval_loss": 0.7712200284004211,
|
2725 |
+
"eval_runtime": 0.519,
|
2726 |
+
"eval_samples_per_second": 1926.697,
|
2727 |
+
"eval_steps_per_second": 30.827,
|
2728 |
+
"step": 136000
|
2729 |
+
},
|
2730 |
+
{
|
2731 |
+
"epoch": 4.18,
|
2732 |
+
"learning_rate": 0.0002623271505378748,
|
2733 |
+
"loss": 0.3608,
|
2734 |
+
"step": 136500
|
2735 |
+
},
|
2736 |
+
{
|
2737 |
+
"epoch": 4.2,
|
2738 |
+
"learning_rate": 0.00026200414388282637,
|
2739 |
+
"loss": 0.361,
|
2740 |
+
"step": 137000
|
2741 |
+
},
|
2742 |
+
{
|
2743 |
+
"epoch": 4.2,
|
2744 |
+
"eval_loss": 0.7850333452224731,
|
2745 |
+
"eval_runtime": 0.5415,
|
2746 |
+
"eval_samples_per_second": 1846.815,
|
2747 |
+
"eval_steps_per_second": 29.549,
|
2748 |
+
"step": 137000
|
2749 |
+
},
|
2750 |
+
{
|
2751 |
+
"epoch": 4.21,
|
2752 |
+
"learning_rate": 0.00026167996704760406,
|
2753 |
+
"loss": 0.3603,
|
2754 |
+
"step": 137500
|
2755 |
+
},
|
2756 |
+
{
|
2757 |
+
"epoch": 4.23,
|
2758 |
+
"learning_rate": 0.00026135462357735375,
|
2759 |
+
"loss": 0.3604,
|
2760 |
+
"step": 138000
|
2761 |
+
},
|
2762 |
+
{
|
2763 |
+
"epoch": 4.23,
|
2764 |
+
"eval_loss": 0.7824327349662781,
|
2765 |
+
"eval_runtime": 0.5307,
|
2766 |
+
"eval_samples_per_second": 1884.218,
|
2767 |
+
"eval_steps_per_second": 30.147,
|
2768 |
+
"step": 138000
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 4.24,
|
2772 |
+
"learning_rate": 0.0002610281170299795,
|
2773 |
+
"loss": 0.3602,
|
2774 |
+
"step": 138500
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 4.26,
|
2778 |
+
"learning_rate": 0.00026070045097610465,
|
2779 |
+
"loss": 0.36,
|
2780 |
+
"step": 139000
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 4.26,
|
2784 |
+
"eval_loss": 0.787578821182251,
|
2785 |
+
"eval_runtime": 0.5306,
|
2786 |
+
"eval_samples_per_second": 1884.712,
|
2787 |
+
"eval_steps_per_second": 30.155,
|
2788 |
+
"step": 139000
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 4.27,
|
2792 |
+
"learning_rate": 0.0002603716289990326,
|
2793 |
+
"loss": 0.3601,
|
2794 |
+
"step": 139500
|
2795 |
+
},
|
2796 |
+
{
|
2797 |
+
"epoch": 4.29,
|
2798 |
+
"learning_rate": 0.00026004165469470787,
|
2799 |
+
"loss": 0.3597,
|
2800 |
+
"step": 140000
|
2801 |
+
},
|
2802 |
+
{
|
2803 |
+
"epoch": 4.29,
|
2804 |
+
"eval_loss": 0.7803857326507568,
|
2805 |
+
"eval_runtime": 0.5242,
|
2806 |
+
"eval_samples_per_second": 1907.849,
|
2807 |
+
"eval_steps_per_second": 30.526,
|
2808 |
+
"step": 140000
|
2809 |
}
|
2810 |
],
|
2811 |
"max_steps": 500000,
|
2812 |
"num_train_epochs": 16,
|
2813 |
+
"total_flos": 4.4727879910350966e+21,
|
2814 |
"trial_name": null,
|
2815 |
"trial_params": null
|
2816 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 102501541
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a40275396ce53417c9c00526c28902e900eee7fe8a1176e3b167b8fa880709f1
|
3 |
size 102501541
|