Training in progress, step 3000, checkpoint
Browse files- last-checkpoint/model.safetensors +1 -1
- last-checkpoint/optimizer.pt +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 +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +711 -3
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 791869518
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:266509c575dfdf72a458e7b2d84b76257d04155a692ce14c2d7de35a15a19280
|
| 3 |
size 791869518
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2375752250
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83d9948277ff1a200f25a50447020e053cceb593a430ae25d078c5789fa9e9b9
|
| 3 |
size 2375752250
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb18ac8d6db3307b1c242f7cb069fc8b8dab957434ddfcafcac997cfd6a43abf
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bdab708057b5f34a402d9a2b4443f5f93a8e8ee2ddb66d955f0a15ad394ecc5
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:599882a30c163a5a2a000c4e74b320ecc4a55aa1b079882fd66aa3d2559d19e7
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:567c3b482c209c2778fc017e39a38642c488edda20673ef29f571ef7177ad81e
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f9ffe9a916e778423aaed4ec842923c9ccfdd3d7a4fbad10dc6a3bfc278fb8e
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7ede8a81aa3c780fb9c3cb57537752a782c4aed1dcecb7aafd6ca5a7ea90252
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b1c5c0c0afa907d332467e631e6cee80ba476689aa0caa77689ca273d83b3e4
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73025ac422abb13303ee974109cf39f6f848de7f7013e828d04aa4e2ec0e6757
|
| 3 |
size 15984
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1000
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a82dd56e5d1fd0f536eca0ab3c8df16cc4dea4fcf65ca478171fc5290a211afc
|
| 3 |
size 1000
|
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": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -3547,6 +3547,714 @@
|
|
| 3547 |
"eval_samples_per_second": 610.776,
|
| 3548 |
"eval_steps_per_second": 38.174,
|
| 3549 |
"step": 2500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3550 |
}
|
| 3551 |
],
|
| 3552 |
"logging_steps": 5,
|
|
@@ -3566,7 +4274,7 @@
|
|
| 3566 |
"attributes": {}
|
| 3567 |
}
|
| 3568 |
},
|
| 3569 |
-
"total_flos": 1.
|
| 3570 |
"train_batch_size": 4,
|
| 3571 |
"trial_name": null,
|
| 3572 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.6935263648382133,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 3000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 3547 |
"eval_samples_per_second": 610.776,
|
| 3548 |
"eval_steps_per_second": 38.174,
|
| 3549 |
"step": 2500
|
| 3550 |
+
},
|
| 3551 |
+
{
|
| 3552 |
+
"epoch": 0.5790945146399081,
|
| 3553 |
+
"grad_norm": 172.125,
|
| 3554 |
+
"learning_rate": 4.676258992805755e-06,
|
| 3555 |
+
"loss": 69.669,
|
| 3556 |
+
"step": 2505
|
| 3557 |
+
},
|
| 3558 |
+
{
|
| 3559 |
+
"epoch": 0.5802503919146385,
|
| 3560 |
+
"grad_norm": 157.25,
|
| 3561 |
+
"learning_rate": 4.663412127440905e-06,
|
| 3562 |
+
"loss": 70.5345,
|
| 3563 |
+
"step": 2510
|
| 3564 |
+
},
|
| 3565 |
+
{
|
| 3566 |
+
"epoch": 0.5814062691893688,
|
| 3567 |
+
"grad_norm": 152.375,
|
| 3568 |
+
"learning_rate": 4.650565262076054e-06,
|
| 3569 |
+
"loss": 68.8704,
|
| 3570 |
+
"step": 2515
|
| 3571 |
+
},
|
| 3572 |
+
{
|
| 3573 |
+
"epoch": 0.5825621464640992,
|
| 3574 |
+
"grad_norm": 182.375,
|
| 3575 |
+
"learning_rate": 4.637718396711203e-06,
|
| 3576 |
+
"loss": 68.8443,
|
| 3577 |
+
"step": 2520
|
| 3578 |
+
},
|
| 3579 |
+
{
|
| 3580 |
+
"epoch": 0.5837180237388295,
|
| 3581 |
+
"grad_norm": 174.125,
|
| 3582 |
+
"learning_rate": 4.624871531346352e-06,
|
| 3583 |
+
"loss": 69.324,
|
| 3584 |
+
"step": 2525
|
| 3585 |
+
},
|
| 3586 |
+
{
|
| 3587 |
+
"epoch": 0.5848739010135598,
|
| 3588 |
+
"grad_norm": 170.375,
|
| 3589 |
+
"learning_rate": 4.612024665981501e-06,
|
| 3590 |
+
"loss": 69.2429,
|
| 3591 |
+
"step": 2530
|
| 3592 |
+
},
|
| 3593 |
+
{
|
| 3594 |
+
"epoch": 0.5860297782882903,
|
| 3595 |
+
"grad_norm": 162.875,
|
| 3596 |
+
"learning_rate": 4.59917780061665e-06,
|
| 3597 |
+
"loss": 68.8715,
|
| 3598 |
+
"step": 2535
|
| 3599 |
+
},
|
| 3600 |
+
{
|
| 3601 |
+
"epoch": 0.5871856555630206,
|
| 3602 |
+
"grad_norm": 154.875,
|
| 3603 |
+
"learning_rate": 4.586330935251799e-06,
|
| 3604 |
+
"loss": 70.5938,
|
| 3605 |
+
"step": 2540
|
| 3606 |
+
},
|
| 3607 |
+
{
|
| 3608 |
+
"epoch": 0.5883415328377509,
|
| 3609 |
+
"grad_norm": 164.125,
|
| 3610 |
+
"learning_rate": 4.5734840698869486e-06,
|
| 3611 |
+
"loss": 69.1381,
|
| 3612 |
+
"step": 2545
|
| 3613 |
+
},
|
| 3614 |
+
{
|
| 3615 |
+
"epoch": 0.5894974101124814,
|
| 3616 |
+
"grad_norm": 168.5,
|
| 3617 |
+
"learning_rate": 4.560637204522097e-06,
|
| 3618 |
+
"loss": 68.9723,
|
| 3619 |
+
"step": 2550
|
| 3620 |
+
},
|
| 3621 |
+
{
|
| 3622 |
+
"epoch": 0.5906532873872117,
|
| 3623 |
+
"grad_norm": 157.125,
|
| 3624 |
+
"learning_rate": 4.547790339157246e-06,
|
| 3625 |
+
"loss": 68.8563,
|
| 3626 |
+
"step": 2555
|
| 3627 |
+
},
|
| 3628 |
+
{
|
| 3629 |
+
"epoch": 0.591809164661942,
|
| 3630 |
+
"grad_norm": 170.125,
|
| 3631 |
+
"learning_rate": 4.534943473792395e-06,
|
| 3632 |
+
"loss": 69.5333,
|
| 3633 |
+
"step": 2560
|
| 3634 |
+
},
|
| 3635 |
+
{
|
| 3636 |
+
"epoch": 0.5929650419366723,
|
| 3637 |
+
"grad_norm": 168.75,
|
| 3638 |
+
"learning_rate": 4.522096608427544e-06,
|
| 3639 |
+
"loss": 68.3743,
|
| 3640 |
+
"step": 2565
|
| 3641 |
+
},
|
| 3642 |
+
{
|
| 3643 |
+
"epoch": 0.5941209192114028,
|
| 3644 |
+
"grad_norm": 161.5,
|
| 3645 |
+
"learning_rate": 4.509249743062693e-06,
|
| 3646 |
+
"loss": 68.4732,
|
| 3647 |
+
"step": 2570
|
| 3648 |
+
},
|
| 3649 |
+
{
|
| 3650 |
+
"epoch": 0.5952767964861331,
|
| 3651 |
+
"grad_norm": 177.375,
|
| 3652 |
+
"learning_rate": 4.496402877697842e-06,
|
| 3653 |
+
"loss": 69.9702,
|
| 3654 |
+
"step": 2575
|
| 3655 |
+
},
|
| 3656 |
+
{
|
| 3657 |
+
"epoch": 0.5964326737608634,
|
| 3658 |
+
"grad_norm": 160.5,
|
| 3659 |
+
"learning_rate": 4.483556012332991e-06,
|
| 3660 |
+
"loss": 69.5753,
|
| 3661 |
+
"step": 2580
|
| 3662 |
+
},
|
| 3663 |
+
{
|
| 3664 |
+
"epoch": 0.5975885510355938,
|
| 3665 |
+
"grad_norm": 170.625,
|
| 3666 |
+
"learning_rate": 4.47070914696814e-06,
|
| 3667 |
+
"loss": 69.2494,
|
| 3668 |
+
"step": 2585
|
| 3669 |
+
},
|
| 3670 |
+
{
|
| 3671 |
+
"epoch": 0.5987444283103242,
|
| 3672 |
+
"grad_norm": 160.5,
|
| 3673 |
+
"learning_rate": 4.457862281603289e-06,
|
| 3674 |
+
"loss": 69.5917,
|
| 3675 |
+
"step": 2590
|
| 3676 |
+
},
|
| 3677 |
+
{
|
| 3678 |
+
"epoch": 0.5999003055850545,
|
| 3679 |
+
"grad_norm": 157.875,
|
| 3680 |
+
"learning_rate": 4.445015416238438e-06,
|
| 3681 |
+
"loss": 69.0619,
|
| 3682 |
+
"step": 2595
|
| 3683 |
+
},
|
| 3684 |
+
{
|
| 3685 |
+
"epoch": 0.6010561828597849,
|
| 3686 |
+
"grad_norm": 150.875,
|
| 3687 |
+
"learning_rate": 4.432168550873588e-06,
|
| 3688 |
+
"loss": 69.3003,
|
| 3689 |
+
"step": 2600
|
| 3690 |
+
},
|
| 3691 |
+
{
|
| 3692 |
+
"epoch": 0.6022120601345152,
|
| 3693 |
+
"grad_norm": 163.125,
|
| 3694 |
+
"learning_rate": 4.4193216855087364e-06,
|
| 3695 |
+
"loss": 68.8558,
|
| 3696 |
+
"step": 2605
|
| 3697 |
+
},
|
| 3698 |
+
{
|
| 3699 |
+
"epoch": 0.6033679374092455,
|
| 3700 |
+
"grad_norm": 168.375,
|
| 3701 |
+
"learning_rate": 4.406474820143885e-06,
|
| 3702 |
+
"loss": 70.8878,
|
| 3703 |
+
"step": 2610
|
| 3704 |
+
},
|
| 3705 |
+
{
|
| 3706 |
+
"epoch": 0.604523814683976,
|
| 3707 |
+
"grad_norm": 157.25,
|
| 3708 |
+
"learning_rate": 4.393627954779034e-06,
|
| 3709 |
+
"loss": 69.8354,
|
| 3710 |
+
"step": 2615
|
| 3711 |
+
},
|
| 3712 |
+
{
|
| 3713 |
+
"epoch": 0.6056796919587063,
|
| 3714 |
+
"grad_norm": 154.625,
|
| 3715 |
+
"learning_rate": 4.380781089414183e-06,
|
| 3716 |
+
"loss": 69.2094,
|
| 3717 |
+
"step": 2620
|
| 3718 |
+
},
|
| 3719 |
+
{
|
| 3720 |
+
"epoch": 0.6068355692334366,
|
| 3721 |
+
"grad_norm": 178.875,
|
| 3722 |
+
"learning_rate": 4.367934224049332e-06,
|
| 3723 |
+
"loss": 69.4816,
|
| 3724 |
+
"step": 2625
|
| 3725 |
+
},
|
| 3726 |
+
{
|
| 3727 |
+
"epoch": 0.6079914465081669,
|
| 3728 |
+
"grad_norm": 148.5,
|
| 3729 |
+
"learning_rate": 4.355087358684481e-06,
|
| 3730 |
+
"loss": 69.5445,
|
| 3731 |
+
"step": 2630
|
| 3732 |
+
},
|
| 3733 |
+
{
|
| 3734 |
+
"epoch": 0.6091473237828974,
|
| 3735 |
+
"grad_norm": 189.375,
|
| 3736 |
+
"learning_rate": 4.34224049331963e-06,
|
| 3737 |
+
"loss": 69.7445,
|
| 3738 |
+
"step": 2635
|
| 3739 |
+
},
|
| 3740 |
+
{
|
| 3741 |
+
"epoch": 0.6103032010576277,
|
| 3742 |
+
"grad_norm": 163.375,
|
| 3743 |
+
"learning_rate": 4.329393627954779e-06,
|
| 3744 |
+
"loss": 70.1707,
|
| 3745 |
+
"step": 2640
|
| 3746 |
+
},
|
| 3747 |
+
{
|
| 3748 |
+
"epoch": 0.611459078332358,
|
| 3749 |
+
"grad_norm": 159.625,
|
| 3750 |
+
"learning_rate": 4.316546762589928e-06,
|
| 3751 |
+
"loss": 70.575,
|
| 3752 |
+
"step": 2645
|
| 3753 |
+
},
|
| 3754 |
+
{
|
| 3755 |
+
"epoch": 0.6126149556070885,
|
| 3756 |
+
"grad_norm": 163.875,
|
| 3757 |
+
"learning_rate": 4.303699897225078e-06,
|
| 3758 |
+
"loss": 68.2103,
|
| 3759 |
+
"step": 2650
|
| 3760 |
+
},
|
| 3761 |
+
{
|
| 3762 |
+
"epoch": 0.6137708328818188,
|
| 3763 |
+
"grad_norm": 173.25,
|
| 3764 |
+
"learning_rate": 4.290853031860227e-06,
|
| 3765 |
+
"loss": 68.2952,
|
| 3766 |
+
"step": 2655
|
| 3767 |
+
},
|
| 3768 |
+
{
|
| 3769 |
+
"epoch": 0.6149267101565491,
|
| 3770 |
+
"grad_norm": 173.75,
|
| 3771 |
+
"learning_rate": 4.2780061664953754e-06,
|
| 3772 |
+
"loss": 68.7904,
|
| 3773 |
+
"step": 2660
|
| 3774 |
+
},
|
| 3775 |
+
{
|
| 3776 |
+
"epoch": 0.6160825874312795,
|
| 3777 |
+
"grad_norm": 163.875,
|
| 3778 |
+
"learning_rate": 4.265159301130524e-06,
|
| 3779 |
+
"loss": 70.0719,
|
| 3780 |
+
"step": 2665
|
| 3781 |
+
},
|
| 3782 |
+
{
|
| 3783 |
+
"epoch": 0.6172384647060098,
|
| 3784 |
+
"grad_norm": 180.625,
|
| 3785 |
+
"learning_rate": 4.252312435765674e-06,
|
| 3786 |
+
"loss": 68.3693,
|
| 3787 |
+
"step": 2670
|
| 3788 |
+
},
|
| 3789 |
+
{
|
| 3790 |
+
"epoch": 0.6183943419807402,
|
| 3791 |
+
"grad_norm": 154.5,
|
| 3792 |
+
"learning_rate": 4.239465570400823e-06,
|
| 3793 |
+
"loss": 68.8033,
|
| 3794 |
+
"step": 2675
|
| 3795 |
+
},
|
| 3796 |
+
{
|
| 3797 |
+
"epoch": 0.6195502192554706,
|
| 3798 |
+
"grad_norm": 160.25,
|
| 3799 |
+
"learning_rate": 4.226618705035972e-06,
|
| 3800 |
+
"loss": 67.7923,
|
| 3801 |
+
"step": 2680
|
| 3802 |
+
},
|
| 3803 |
+
{
|
| 3804 |
+
"epoch": 0.6207060965302009,
|
| 3805 |
+
"grad_norm": 159.5,
|
| 3806 |
+
"learning_rate": 4.2137718396711205e-06,
|
| 3807 |
+
"loss": 69.2746,
|
| 3808 |
+
"step": 2685
|
| 3809 |
+
},
|
| 3810 |
+
{
|
| 3811 |
+
"epoch": 0.6218619738049312,
|
| 3812 |
+
"grad_norm": 164.0,
|
| 3813 |
+
"learning_rate": 4.200924974306269e-06,
|
| 3814 |
+
"loss": 68.537,
|
| 3815 |
+
"step": 2690
|
| 3816 |
+
},
|
| 3817 |
+
{
|
| 3818 |
+
"epoch": 0.6230178510796616,
|
| 3819 |
+
"grad_norm": 156.25,
|
| 3820 |
+
"learning_rate": 4.188078108941418e-06,
|
| 3821 |
+
"loss": 68.6553,
|
| 3822 |
+
"step": 2695
|
| 3823 |
+
},
|
| 3824 |
+
{
|
| 3825 |
+
"epoch": 0.624173728354392,
|
| 3826 |
+
"grad_norm": 158.625,
|
| 3827 |
+
"learning_rate": 4.175231243576567e-06,
|
| 3828 |
+
"loss": 68.8911,
|
| 3829 |
+
"step": 2700
|
| 3830 |
+
},
|
| 3831 |
+
{
|
| 3832 |
+
"epoch": 0.6253296056291223,
|
| 3833 |
+
"grad_norm": 166.25,
|
| 3834 |
+
"learning_rate": 4.162384378211717e-06,
|
| 3835 |
+
"loss": 69.7975,
|
| 3836 |
+
"step": 2705
|
| 3837 |
+
},
|
| 3838 |
+
{
|
| 3839 |
+
"epoch": 0.6264854829038526,
|
| 3840 |
+
"grad_norm": 160.25,
|
| 3841 |
+
"learning_rate": 4.149537512846866e-06,
|
| 3842 |
+
"loss": 69.9966,
|
| 3843 |
+
"step": 2710
|
| 3844 |
+
},
|
| 3845 |
+
{
|
| 3846 |
+
"epoch": 0.6276413601785831,
|
| 3847 |
+
"grad_norm": 167.0,
|
| 3848 |
+
"learning_rate": 4.1366906474820145e-06,
|
| 3849 |
+
"loss": 70.2918,
|
| 3850 |
+
"step": 2715
|
| 3851 |
+
},
|
| 3852 |
+
{
|
| 3853 |
+
"epoch": 0.6287972374533134,
|
| 3854 |
+
"grad_norm": 156.5,
|
| 3855 |
+
"learning_rate": 4.123843782117164e-06,
|
| 3856 |
+
"loss": 67.3517,
|
| 3857 |
+
"step": 2720
|
| 3858 |
+
},
|
| 3859 |
+
{
|
| 3860 |
+
"epoch": 0.6299531147280437,
|
| 3861 |
+
"grad_norm": 171.25,
|
| 3862 |
+
"learning_rate": 4.110996916752313e-06,
|
| 3863 |
+
"loss": 68.5519,
|
| 3864 |
+
"step": 2725
|
| 3865 |
+
},
|
| 3866 |
+
{
|
| 3867 |
+
"epoch": 0.6311089920027741,
|
| 3868 |
+
"grad_norm": 178.75,
|
| 3869 |
+
"learning_rate": 4.098150051387462e-06,
|
| 3870 |
+
"loss": 69.3377,
|
| 3871 |
+
"step": 2730
|
| 3872 |
+
},
|
| 3873 |
+
{
|
| 3874 |
+
"epoch": 0.6322648692775045,
|
| 3875 |
+
"grad_norm": 167.875,
|
| 3876 |
+
"learning_rate": 4.085303186022611e-06,
|
| 3877 |
+
"loss": 68.7498,
|
| 3878 |
+
"step": 2735
|
| 3879 |
+
},
|
| 3880 |
+
{
|
| 3881 |
+
"epoch": 0.6334207465522348,
|
| 3882 |
+
"grad_norm": 164.875,
|
| 3883 |
+
"learning_rate": 4.07245632065776e-06,
|
| 3884 |
+
"loss": 68.4089,
|
| 3885 |
+
"step": 2740
|
| 3886 |
+
},
|
| 3887 |
+
{
|
| 3888 |
+
"epoch": 0.6345766238269652,
|
| 3889 |
+
"grad_norm": 171.5,
|
| 3890 |
+
"learning_rate": 4.059609455292909e-06,
|
| 3891 |
+
"loss": 68.4604,
|
| 3892 |
+
"step": 2745
|
| 3893 |
+
},
|
| 3894 |
+
{
|
| 3895 |
+
"epoch": 0.6357325011016955,
|
| 3896 |
+
"grad_norm": 169.5,
|
| 3897 |
+
"learning_rate": 4.046762589928058e-06,
|
| 3898 |
+
"loss": 69.7823,
|
| 3899 |
+
"step": 2750
|
| 3900 |
+
},
|
| 3901 |
+
{
|
| 3902 |
+
"epoch": 0.6368883783764259,
|
| 3903 |
+
"grad_norm": 167.125,
|
| 3904 |
+
"learning_rate": 4.033915724563207e-06,
|
| 3905 |
+
"loss": 68.461,
|
| 3906 |
+
"step": 2755
|
| 3907 |
+
},
|
| 3908 |
+
{
|
| 3909 |
+
"epoch": 0.6380442556511562,
|
| 3910 |
+
"grad_norm": 157.75,
|
| 3911 |
+
"learning_rate": 4.021068859198356e-06,
|
| 3912 |
+
"loss": 69.8009,
|
| 3913 |
+
"step": 2760
|
| 3914 |
+
},
|
| 3915 |
+
{
|
| 3916 |
+
"epoch": 0.6392001329258866,
|
| 3917 |
+
"grad_norm": 153.25,
|
| 3918 |
+
"learning_rate": 4.008221993833505e-06,
|
| 3919 |
+
"loss": 69.0233,
|
| 3920 |
+
"step": 2765
|
| 3921 |
+
},
|
| 3922 |
+
{
|
| 3923 |
+
"epoch": 0.6403560102006169,
|
| 3924 |
+
"grad_norm": 153.375,
|
| 3925 |
+
"learning_rate": 3.9953751284686535e-06,
|
| 3926 |
+
"loss": 69.8202,
|
| 3927 |
+
"step": 2770
|
| 3928 |
+
},
|
| 3929 |
+
{
|
| 3930 |
+
"epoch": 0.6415118874753473,
|
| 3931 |
+
"grad_norm": 152.25,
|
| 3932 |
+
"learning_rate": 3.982528263103803e-06,
|
| 3933 |
+
"loss": 70.0482,
|
| 3934 |
+
"step": 2775
|
| 3935 |
+
},
|
| 3936 |
+
{
|
| 3937 |
+
"epoch": 0.6426677647500777,
|
| 3938 |
+
"grad_norm": 151.0,
|
| 3939 |
+
"learning_rate": 3.969681397738952e-06,
|
| 3940 |
+
"loss": 68.4574,
|
| 3941 |
+
"step": 2780
|
| 3942 |
+
},
|
| 3943 |
+
{
|
| 3944 |
+
"epoch": 0.643823642024808,
|
| 3945 |
+
"grad_norm": 153.75,
|
| 3946 |
+
"learning_rate": 3.956834532374101e-06,
|
| 3947 |
+
"loss": 68.9972,
|
| 3948 |
+
"step": 2785
|
| 3949 |
+
},
|
| 3950 |
+
{
|
| 3951 |
+
"epoch": 0.6449795192995383,
|
| 3952 |
+
"grad_norm": 155.0,
|
| 3953 |
+
"learning_rate": 3.9439876670092506e-06,
|
| 3954 |
+
"loss": 69.0547,
|
| 3955 |
+
"step": 2790
|
| 3956 |
+
},
|
| 3957 |
+
{
|
| 3958 |
+
"epoch": 0.6461353965742688,
|
| 3959 |
+
"grad_norm": 165.75,
|
| 3960 |
+
"learning_rate": 3.931140801644399e-06,
|
| 3961 |
+
"loss": 69.5258,
|
| 3962 |
+
"step": 2795
|
| 3963 |
+
},
|
| 3964 |
+
{
|
| 3965 |
+
"epoch": 0.6472912738489991,
|
| 3966 |
+
"grad_norm": 148.625,
|
| 3967 |
+
"learning_rate": 3.918293936279548e-06,
|
| 3968 |
+
"loss": 68.9722,
|
| 3969 |
+
"step": 2800
|
| 3970 |
+
},
|
| 3971 |
+
{
|
| 3972 |
+
"epoch": 0.6484471511237294,
|
| 3973 |
+
"grad_norm": 146.125,
|
| 3974 |
+
"learning_rate": 3.905447070914697e-06,
|
| 3975 |
+
"loss": 69.1489,
|
| 3976 |
+
"step": 2805
|
| 3977 |
+
},
|
| 3978 |
+
{
|
| 3979 |
+
"epoch": 0.6496030283984598,
|
| 3980 |
+
"grad_norm": 167.625,
|
| 3981 |
+
"learning_rate": 3.892600205549846e-06,
|
| 3982 |
+
"loss": 67.9903,
|
| 3983 |
+
"step": 2810
|
| 3984 |
+
},
|
| 3985 |
+
{
|
| 3986 |
+
"epoch": 0.6507589056731902,
|
| 3987 |
+
"grad_norm": 154.0,
|
| 3988 |
+
"learning_rate": 3.879753340184995e-06,
|
| 3989 |
+
"loss": 69.16,
|
| 3990 |
+
"step": 2815
|
| 3991 |
+
},
|
| 3992 |
+
{
|
| 3993 |
+
"epoch": 0.6519147829479205,
|
| 3994 |
+
"grad_norm": 157.25,
|
| 3995 |
+
"learning_rate": 3.866906474820144e-06,
|
| 3996 |
+
"loss": 70.0419,
|
| 3997 |
+
"step": 2820
|
| 3998 |
+
},
|
| 3999 |
+
{
|
| 4000 |
+
"epoch": 0.6530706602226508,
|
| 4001 |
+
"grad_norm": 156.625,
|
| 4002 |
+
"learning_rate": 3.854059609455293e-06,
|
| 4003 |
+
"loss": 69.2703,
|
| 4004 |
+
"step": 2825
|
| 4005 |
+
},
|
| 4006 |
+
{
|
| 4007 |
+
"epoch": 0.6542265374973812,
|
| 4008 |
+
"grad_norm": 153.875,
|
| 4009 |
+
"learning_rate": 3.841212744090442e-06,
|
| 4010 |
+
"loss": 69.3849,
|
| 4011 |
+
"step": 2830
|
| 4012 |
+
},
|
| 4013 |
+
{
|
| 4014 |
+
"epoch": 0.6553824147721116,
|
| 4015 |
+
"grad_norm": 161.125,
|
| 4016 |
+
"learning_rate": 3.828365878725591e-06,
|
| 4017 |
+
"loss": 69.6766,
|
| 4018 |
+
"step": 2835
|
| 4019 |
+
},
|
| 4020 |
+
{
|
| 4021 |
+
"epoch": 0.6565382920468419,
|
| 4022 |
+
"grad_norm": 147.875,
|
| 4023 |
+
"learning_rate": 3.81551901336074e-06,
|
| 4024 |
+
"loss": 67.732,
|
| 4025 |
+
"step": 2840
|
| 4026 |
+
},
|
| 4027 |
+
{
|
| 4028 |
+
"epoch": 0.6576941693215723,
|
| 4029 |
+
"grad_norm": 168.0,
|
| 4030 |
+
"learning_rate": 3.8026721479958896e-06,
|
| 4031 |
+
"loss": 69.0334,
|
| 4032 |
+
"step": 2845
|
| 4033 |
+
},
|
| 4034 |
+
{
|
| 4035 |
+
"epoch": 0.6588500465963026,
|
| 4036 |
+
"grad_norm": 153.75,
|
| 4037 |
+
"learning_rate": 3.7898252826310384e-06,
|
| 4038 |
+
"loss": 68.3081,
|
| 4039 |
+
"step": 2850
|
| 4040 |
+
},
|
| 4041 |
+
{
|
| 4042 |
+
"epoch": 0.660005923871033,
|
| 4043 |
+
"grad_norm": 184.25,
|
| 4044 |
+
"learning_rate": 3.7769784172661873e-06,
|
| 4045 |
+
"loss": 69.0602,
|
| 4046 |
+
"step": 2855
|
| 4047 |
+
},
|
| 4048 |
+
{
|
| 4049 |
+
"epoch": 0.6611618011457634,
|
| 4050 |
+
"grad_norm": 189.75,
|
| 4051 |
+
"learning_rate": 3.7641315519013365e-06,
|
| 4052 |
+
"loss": 67.2991,
|
| 4053 |
+
"step": 2860
|
| 4054 |
+
},
|
| 4055 |
+
{
|
| 4056 |
+
"epoch": 0.6623176784204937,
|
| 4057 |
+
"grad_norm": 169.0,
|
| 4058 |
+
"learning_rate": 3.7512846865364854e-06,
|
| 4059 |
+
"loss": 70.8647,
|
| 4060 |
+
"step": 2865
|
| 4061 |
+
},
|
| 4062 |
+
{
|
| 4063 |
+
"epoch": 0.663473555695224,
|
| 4064 |
+
"grad_norm": 157.75,
|
| 4065 |
+
"learning_rate": 3.7384378211716342e-06,
|
| 4066 |
+
"loss": 68.4063,
|
| 4067 |
+
"step": 2870
|
| 4068 |
+
},
|
| 4069 |
+
{
|
| 4070 |
+
"epoch": 0.6646294329699545,
|
| 4071 |
+
"grad_norm": 176.375,
|
| 4072 |
+
"learning_rate": 3.725590955806783e-06,
|
| 4073 |
+
"loss": 68.3025,
|
| 4074 |
+
"step": 2875
|
| 4075 |
+
},
|
| 4076 |
+
{
|
| 4077 |
+
"epoch": 0.6657853102446848,
|
| 4078 |
+
"grad_norm": 180.0,
|
| 4079 |
+
"learning_rate": 3.7127440904419328e-06,
|
| 4080 |
+
"loss": 68.6778,
|
| 4081 |
+
"step": 2880
|
| 4082 |
+
},
|
| 4083 |
+
{
|
| 4084 |
+
"epoch": 0.6669411875194151,
|
| 4085 |
+
"grad_norm": 168.5,
|
| 4086 |
+
"learning_rate": 3.6998972250770816e-06,
|
| 4087 |
+
"loss": 67.2759,
|
| 4088 |
+
"step": 2885
|
| 4089 |
+
},
|
| 4090 |
+
{
|
| 4091 |
+
"epoch": 0.6680970647941454,
|
| 4092 |
+
"grad_norm": 164.25,
|
| 4093 |
+
"learning_rate": 3.6870503597122305e-06,
|
| 4094 |
+
"loss": 69.538,
|
| 4095 |
+
"step": 2890
|
| 4096 |
+
},
|
| 4097 |
+
{
|
| 4098 |
+
"epoch": 0.6692529420688759,
|
| 4099 |
+
"grad_norm": 157.625,
|
| 4100 |
+
"learning_rate": 3.6742034943473797e-06,
|
| 4101 |
+
"loss": 68.8391,
|
| 4102 |
+
"step": 2895
|
| 4103 |
+
},
|
| 4104 |
+
{
|
| 4105 |
+
"epoch": 0.6704088193436062,
|
| 4106 |
+
"grad_norm": 171.625,
|
| 4107 |
+
"learning_rate": 3.6613566289825286e-06,
|
| 4108 |
+
"loss": 68.3289,
|
| 4109 |
+
"step": 2900
|
| 4110 |
+
},
|
| 4111 |
+
{
|
| 4112 |
+
"epoch": 0.6715646966183365,
|
| 4113 |
+
"grad_norm": 161.375,
|
| 4114 |
+
"learning_rate": 3.6485097636176774e-06,
|
| 4115 |
+
"loss": 67.6343,
|
| 4116 |
+
"step": 2905
|
| 4117 |
+
},
|
| 4118 |
+
{
|
| 4119 |
+
"epoch": 0.6727205738930669,
|
| 4120 |
+
"grad_norm": 159.0,
|
| 4121 |
+
"learning_rate": 3.6356628982528263e-06,
|
| 4122 |
+
"loss": 67.9106,
|
| 4123 |
+
"step": 2910
|
| 4124 |
+
},
|
| 4125 |
+
{
|
| 4126 |
+
"epoch": 0.6738764511677973,
|
| 4127 |
+
"grad_norm": 148.625,
|
| 4128 |
+
"learning_rate": 3.622816032887976e-06,
|
| 4129 |
+
"loss": 68.7488,
|
| 4130 |
+
"step": 2915
|
| 4131 |
+
},
|
| 4132 |
+
{
|
| 4133 |
+
"epoch": 0.6750323284425276,
|
| 4134 |
+
"grad_norm": 172.0,
|
| 4135 |
+
"learning_rate": 3.609969167523125e-06,
|
| 4136 |
+
"loss": 68.8937,
|
| 4137 |
+
"step": 2920
|
| 4138 |
+
},
|
| 4139 |
+
{
|
| 4140 |
+
"epoch": 0.676188205717258,
|
| 4141 |
+
"grad_norm": 169.875,
|
| 4142 |
+
"learning_rate": 3.5971223021582737e-06,
|
| 4143 |
+
"loss": 68.0938,
|
| 4144 |
+
"step": 2925
|
| 4145 |
+
},
|
| 4146 |
+
{
|
| 4147 |
+
"epoch": 0.6773440829919883,
|
| 4148 |
+
"grad_norm": 161.25,
|
| 4149 |
+
"learning_rate": 3.584275436793423e-06,
|
| 4150 |
+
"loss": 68.6723,
|
| 4151 |
+
"step": 2930
|
| 4152 |
+
},
|
| 4153 |
+
{
|
| 4154 |
+
"epoch": 0.6784999602667187,
|
| 4155 |
+
"grad_norm": 159.0,
|
| 4156 |
+
"learning_rate": 3.5714285714285718e-06,
|
| 4157 |
+
"loss": 68.2037,
|
| 4158 |
+
"step": 2935
|
| 4159 |
+
},
|
| 4160 |
+
{
|
| 4161 |
+
"epoch": 0.679655837541449,
|
| 4162 |
+
"grad_norm": 165.875,
|
| 4163 |
+
"learning_rate": 3.5585817060637206e-06,
|
| 4164 |
+
"loss": 68.3861,
|
| 4165 |
+
"step": 2940
|
| 4166 |
+
},
|
| 4167 |
+
{
|
| 4168 |
+
"epoch": 0.6808117148161794,
|
| 4169 |
+
"grad_norm": 160.625,
|
| 4170 |
+
"learning_rate": 3.5457348406988695e-06,
|
| 4171 |
+
"loss": 67.6351,
|
| 4172 |
+
"step": 2945
|
| 4173 |
+
},
|
| 4174 |
+
{
|
| 4175 |
+
"epoch": 0.6819675920909097,
|
| 4176 |
+
"grad_norm": 172.625,
|
| 4177 |
+
"learning_rate": 3.5328879753340187e-06,
|
| 4178 |
+
"loss": 68.8567,
|
| 4179 |
+
"step": 2950
|
| 4180 |
+
},
|
| 4181 |
+
{
|
| 4182 |
+
"epoch": 0.6831234693656401,
|
| 4183 |
+
"grad_norm": 185.125,
|
| 4184 |
+
"learning_rate": 3.520041109969168e-06,
|
| 4185 |
+
"loss": 66.5128,
|
| 4186 |
+
"step": 2955
|
| 4187 |
+
},
|
| 4188 |
+
{
|
| 4189 |
+
"epoch": 0.6842793466403705,
|
| 4190 |
+
"grad_norm": 157.75,
|
| 4191 |
+
"learning_rate": 3.507194244604317e-06,
|
| 4192 |
+
"loss": 68.0714,
|
| 4193 |
+
"step": 2960
|
| 4194 |
+
},
|
| 4195 |
+
{
|
| 4196 |
+
"epoch": 0.6854352239151008,
|
| 4197 |
+
"grad_norm": 151.375,
|
| 4198 |
+
"learning_rate": 3.494347379239466e-06,
|
| 4199 |
+
"loss": 67.9339,
|
| 4200 |
+
"step": 2965
|
| 4201 |
+
},
|
| 4202 |
+
{
|
| 4203 |
+
"epoch": 0.6865911011898311,
|
| 4204 |
+
"grad_norm": 164.25,
|
| 4205 |
+
"learning_rate": 3.481500513874615e-06,
|
| 4206 |
+
"loss": 68.5463,
|
| 4207 |
+
"step": 2970
|
| 4208 |
+
},
|
| 4209 |
+
{
|
| 4210 |
+
"epoch": 0.6877469784645616,
|
| 4211 |
+
"grad_norm": 138.0,
|
| 4212 |
+
"learning_rate": 3.468653648509764e-06,
|
| 4213 |
+
"loss": 69.1578,
|
| 4214 |
+
"step": 2975
|
| 4215 |
+
},
|
| 4216 |
+
{
|
| 4217 |
+
"epoch": 0.6889028557392919,
|
| 4218 |
+
"grad_norm": 200.75,
|
| 4219 |
+
"learning_rate": 3.4558067831449127e-06,
|
| 4220 |
+
"loss": 68.674,
|
| 4221 |
+
"step": 2980
|
| 4222 |
+
},
|
| 4223 |
+
{
|
| 4224 |
+
"epoch": 0.6900587330140222,
|
| 4225 |
+
"grad_norm": 149.75,
|
| 4226 |
+
"learning_rate": 3.442959917780062e-06,
|
| 4227 |
+
"loss": 68.8552,
|
| 4228 |
+
"step": 2985
|
| 4229 |
+
},
|
| 4230 |
+
{
|
| 4231 |
+
"epoch": 0.6912146102887526,
|
| 4232 |
+
"grad_norm": 165.125,
|
| 4233 |
+
"learning_rate": 3.430113052415211e-06,
|
| 4234 |
+
"loss": 68.541,
|
| 4235 |
+
"step": 2990
|
| 4236 |
+
},
|
| 4237 |
+
{
|
| 4238 |
+
"epoch": 0.692370487563483,
|
| 4239 |
+
"grad_norm": 175.625,
|
| 4240 |
+
"learning_rate": 3.4172661870503596e-06,
|
| 4241 |
+
"loss": 68.8684,
|
| 4242 |
+
"step": 2995
|
| 4243 |
+
},
|
| 4244 |
+
{
|
| 4245 |
+
"epoch": 0.6935263648382133,
|
| 4246 |
+
"grad_norm": 165.5,
|
| 4247 |
+
"learning_rate": 3.4044193216855093e-06,
|
| 4248 |
+
"loss": 68.3212,
|
| 4249 |
+
"step": 3000
|
| 4250 |
+
},
|
| 4251 |
+
{
|
| 4252 |
+
"epoch": 0.6935263648382133,
|
| 4253 |
+
"eval_loss": NaN,
|
| 4254 |
+
"eval_runtime": 380.967,
|
| 4255 |
+
"eval_samples_per_second": 611.953,
|
| 4256 |
+
"eval_steps_per_second": 38.247,
|
| 4257 |
+
"step": 3000
|
| 4258 |
}
|
| 4259 |
],
|
| 4260 |
"logging_steps": 5,
|
|
|
|
| 4274 |
"attributes": {}
|
| 4275 |
}
|
| 4276 |
},
|
| 4277 |
+
"total_flos": 1.2997073978366362e+19,
|
| 4278 |
"train_batch_size": 4,
|
| 4279 |
"trial_name": null,
|
| 4280 |
"trial_params": null
|