Training in progress, step 187, 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 295488936
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a334fb9201f445d860c73ae2112756ab370bc5307b533de52f75d6bff0a3914c
|
3 |
size 295488936
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 150486964
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b8ee5c59809237f45abda30f70389dd874dcb6f37915ec5050f187c533a5e01
|
3 |
size 150486964
|
last-checkpoint/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:be3592b9f724213e8334a4b5adc53af9a54163ee09cbde5c7c8412412d480099
|
3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:43427779b8e6e83ef9780afcdea7472d2419de40e6f85cfedb90b60f89a2ff0f
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 17,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -487,6 +487,56 @@
|
|
487 |
"eval_samples_per_second": 35.549,
|
488 |
"eval_steps_per_second": 4.45,
|
489 |
"step": 170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
490 |
}
|
491 |
],
|
492 |
"logging_steps": 3,
|
@@ -506,7 +556,7 @@
|
|
506 |
"attributes": {}
|
507 |
}
|
508 |
},
|
509 |
-
"total_flos":
|
510 |
"train_batch_size": 8,
|
511 |
"trial_name": null,
|
512 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.10541149943630214,
|
5 |
"eval_steps": 17,
|
6 |
+
"global_step": 187,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
487 |
"eval_samples_per_second": 35.549,
|
488 |
"eval_steps_per_second": 4.45,
|
489 |
"step": 170
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.0963923337091319,
|
493 |
+
"grad_norm": 2.9714503288269043,
|
494 |
+
"learning_rate": 5.6388590278194096e-06,
|
495 |
+
"loss": 0.5568,
|
496 |
+
"step": 171
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.09808342728297632,
|
500 |
+
"grad_norm": 2.8310670852661133,
|
501 |
+
"learning_rate": 4.549673247541875e-06,
|
502 |
+
"loss": 0.6169,
|
503 |
+
"step": 174
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 0.09977452085682074,
|
507 |
+
"grad_norm": 3.3684370517730713,
|
508 |
+
"learning_rate": 3.5722980755146517e-06,
|
509 |
+
"loss": 0.4749,
|
510 |
+
"step": 177
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 0.10146561443066517,
|
514 |
+
"grad_norm": 3.3378119468688965,
|
515 |
+
"learning_rate": 2.7091379149682685e-06,
|
516 |
+
"loss": 0.5579,
|
517 |
+
"step": 180
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 0.10315670800450959,
|
521 |
+
"grad_norm": 3.996790885925293,
|
522 |
+
"learning_rate": 1.962316193157593e-06,
|
523 |
+
"loss": 0.5408,
|
524 |
+
"step": 183
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 0.10484780157835401,
|
528 |
+
"grad_norm": 3.091064453125,
|
529 |
+
"learning_rate": 1.333670137599713e-06,
|
530 |
+
"loss": 0.5679,
|
531 |
+
"step": 186
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 0.10541149943630214,
|
535 |
+
"eval_loss": 0.5559670925140381,
|
536 |
+
"eval_runtime": 84.033,
|
537 |
+
"eval_samples_per_second": 35.557,
|
538 |
+
"eval_steps_per_second": 4.451,
|
539 |
+
"step": 187
|
540 |
}
|
541 |
],
|
542 |
"logging_steps": 3,
|
|
|
556 |
"attributes": {}
|
557 |
}
|
558 |
},
|
559 |
+
"total_flos": 5.089106463817728e+16,
|
560 |
"train_batch_size": 8,
|
561 |
"trial_name": null,
|
562 |
"trial_params": null
|