Training in progress, step 437, checkpoint
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/rng_state.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +1390 -4
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 304998596
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94dab5bfb47f564438af988f3574bb0f0e9507107f7995229ba7c0b407533a9b
|
| 3 |
size 304998596
|
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 sha256:05483b02c0f77daae352ffb8f45d8d17bcaf90caa568067a2ed2383fb7d92545
|
| 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 sha256:43cdf166fd391100ddd836aff220772a70fdca8f5f3ab782a162cdf93eef1cc4
|
| 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": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1680,6 +1680,1392 @@
|
|
| 1680 |
"learning_rate": 6.357763004631104e-05,
|
| 1681 |
"loss": 0.0,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +3080,12 @@
|
|
| 1694 |
"should_evaluate": false,
|
| 1695 |
"should_log": false,
|
| 1696 |
"should_save": true,
|
| 1697 |
-
"should_training_stop":
|
| 1698 |
},
|
| 1699 |
"attributes": {}
|
| 1700 |
}
|
| 1701 |
},
|
| 1702 |
-
"total_flos":
|
| 1703 |
"train_batch_size": 8,
|
| 1704 |
"trial_name": null,
|
| 1705 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9988571428571429,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 437,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 6.357763004631104e-05,
|
| 1681 |
"loss": 0.0,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.5485714285714286,
|
| 1686 |
+
"grad_norm": NaN,
|
| 1687 |
+
"learning_rate": 6.31284497017055e-05,
|
| 1688 |
+
"loss": 0.0,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.5508571428571428,
|
| 1693 |
+
"grad_norm": NaN,
|
| 1694 |
+
"learning_rate": 6.26781284501043e-05,
|
| 1695 |
+
"loss": 0.0,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.5531428571428572,
|
| 1700 |
+
"grad_norm": NaN,
|
| 1701 |
+
"learning_rate": 6.2226705425958e-05,
|
| 1702 |
+
"loss": 0.0,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.5554285714285714,
|
| 1707 |
+
"grad_norm": NaN,
|
| 1708 |
+
"learning_rate": 6.177421985946499e-05,
|
| 1709 |
+
"loss": 0.0,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.5577142857142857,
|
| 1714 |
+
"grad_norm": NaN,
|
| 1715 |
+
"learning_rate": 6.13207110731622e-05,
|
| 1716 |
+
"loss": 0.0,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.56,
|
| 1721 |
+
"grad_norm": NaN,
|
| 1722 |
+
"learning_rate": 6.086621847850788e-05,
|
| 1723 |
+
"loss": 0.0,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.5622857142857143,
|
| 1728 |
+
"grad_norm": NaN,
|
| 1729 |
+
"learning_rate": 6.0410781572456486e-05,
|
| 1730 |
+
"loss": 0.0,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.5645714285714286,
|
| 1735 |
+
"grad_norm": NaN,
|
| 1736 |
+
"learning_rate": 5.995443993402647e-05,
|
| 1737 |
+
"loss": 0.0,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5668571428571428,
|
| 1742 |
+
"grad_norm": NaN,
|
| 1743 |
+
"learning_rate": 5.949723322086053e-05,
|
| 1744 |
+
"loss": 0.0,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5691428571428572,
|
| 1749 |
+
"grad_norm": NaN,
|
| 1750 |
+
"learning_rate": 5.9039201165779315e-05,
|
| 1751 |
+
"loss": 0.0,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5714285714285714,
|
| 1756 |
+
"grad_norm": NaN,
|
| 1757 |
+
"learning_rate": 5.85803835733285e-05,
|
| 1758 |
+
"loss": 0.0,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5737142857142857,
|
| 1763 |
+
"grad_norm": NaN,
|
| 1764 |
+
"learning_rate": 5.812082031631966e-05,
|
| 1765 |
+
"loss": 0.0,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.576,
|
| 1770 |
+
"grad_norm": NaN,
|
| 1771 |
+
"learning_rate": 5.766055133236513e-05,
|
| 1772 |
+
"loss": 0.0,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5782857142857143,
|
| 1777 |
+
"grad_norm": NaN,
|
| 1778 |
+
"learning_rate": 5.719961662040733e-05,
|
| 1779 |
+
"loss": 0.0,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5805714285714285,
|
| 1784 |
+
"grad_norm": NaN,
|
| 1785 |
+
"learning_rate": 5.673805623724272e-05,
|
| 1786 |
+
"loss": 0.0,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.5828571428571429,
|
| 1791 |
+
"grad_norm": NaN,
|
| 1792 |
+
"learning_rate": 5.627591029404071e-05,
|
| 1793 |
+
"loss": 0.0,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.5851428571428572,
|
| 1798 |
+
"grad_norm": NaN,
|
| 1799 |
+
"learning_rate": 5.581321895285787e-05,
|
| 1800 |
+
"loss": 0.0,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.5874285714285714,
|
| 1805 |
+
"grad_norm": NaN,
|
| 1806 |
+
"learning_rate": 5.535002242314772e-05,
|
| 1807 |
+
"loss": 0.0,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.5897142857142857,
|
| 1812 |
+
"grad_norm": NaN,
|
| 1813 |
+
"learning_rate": 5.488636095826636e-05,
|
| 1814 |
+
"loss": 0.0,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.592,
|
| 1819 |
+
"grad_norm": NaN,
|
| 1820 |
+
"learning_rate": 5.442227485197435e-05,
|
| 1821 |
+
"loss": 0.0,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.5942857142857143,
|
| 1826 |
+
"grad_norm": NaN,
|
| 1827 |
+
"learning_rate": 5.395780443493508e-05,
|
| 1828 |
+
"loss": 0.0,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5965714285714285,
|
| 1833 |
+
"grad_norm": NaN,
|
| 1834 |
+
"learning_rate": 5.3492990071209806e-05,
|
| 1835 |
+
"loss": 0.0,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.5988571428571429,
|
| 1840 |
+
"grad_norm": NaN,
|
| 1841 |
+
"learning_rate": 5.3027872154749915e-05,
|
| 1842 |
+
"loss": 0.0,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6011428571428571,
|
| 1847 |
+
"grad_norm": NaN,
|
| 1848 |
+
"learning_rate": 5.256249110588659e-05,
|
| 1849 |
+
"loss": 0.0,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6034285714285714,
|
| 1854 |
+
"grad_norm": NaN,
|
| 1855 |
+
"learning_rate": 5.2096887367818105e-05,
|
| 1856 |
+
"loss": 0.0,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6057142857142858,
|
| 1861 |
+
"grad_norm": NaN,
|
| 1862 |
+
"learning_rate": 5.1631101403095184e-05,
|
| 1863 |
+
"loss": 0.0,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.608,
|
| 1868 |
+
"grad_norm": NaN,
|
| 1869 |
+
"learning_rate": 5.116517369010466e-05,
|
| 1870 |
+
"loss": 0.0,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6102857142857143,
|
| 1875 |
+
"grad_norm": NaN,
|
| 1876 |
+
"learning_rate": 5.069914471955178e-05,
|
| 1877 |
+
"loss": 0.0,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6125714285714285,
|
| 1882 |
+
"grad_norm": NaN,
|
| 1883 |
+
"learning_rate": 5.023305499094144e-05,
|
| 1884 |
+
"loss": 0.0,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6148571428571429,
|
| 1889 |
+
"grad_norm": NaN,
|
| 1890 |
+
"learning_rate": 4.976694500905857e-05,
|
| 1891 |
+
"loss": 0.0,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6171428571428571,
|
| 1896 |
+
"grad_norm": NaN,
|
| 1897 |
+
"learning_rate": 4.930085528044823e-05,
|
| 1898 |
+
"loss": 0.0,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6194285714285714,
|
| 1903 |
+
"grad_norm": NaN,
|
| 1904 |
+
"learning_rate": 4.883482630989535e-05,
|
| 1905 |
+
"loss": 0.0,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6217142857142857,
|
| 1910 |
+
"grad_norm": NaN,
|
| 1911 |
+
"learning_rate": 4.8368898596904834e-05,
|
| 1912 |
+
"loss": 0.0,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.624,
|
| 1917 |
+
"grad_norm": NaN,
|
| 1918 |
+
"learning_rate": 4.790311263218191e-05,
|
| 1919 |
+
"loss": 0.0,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.6262857142857143,
|
| 1924 |
+
"grad_norm": NaN,
|
| 1925 |
+
"learning_rate": 4.743750889411342e-05,
|
| 1926 |
+
"loss": 0.0,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6285714285714286,
|
| 1931 |
+
"grad_norm": NaN,
|
| 1932 |
+
"learning_rate": 4.697212784525008e-05,
|
| 1933 |
+
"loss": 0.0,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6308571428571429,
|
| 1938 |
+
"grad_norm": NaN,
|
| 1939 |
+
"learning_rate": 4.65070099287902e-05,
|
| 1940 |
+
"loss": 0.0,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6331428571428571,
|
| 1945 |
+
"grad_norm": NaN,
|
| 1946 |
+
"learning_rate": 4.604219556506492e-05,
|
| 1947 |
+
"loss": 0.0,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.6354285714285715,
|
| 1952 |
+
"grad_norm": NaN,
|
| 1953 |
+
"learning_rate": 4.5577725148025646e-05,
|
| 1954 |
+
"loss": 0.0,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.6377142857142857,
|
| 1959 |
+
"grad_norm": NaN,
|
| 1960 |
+
"learning_rate": 4.511363904173366e-05,
|
| 1961 |
+
"loss": 0.0,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.64,
|
| 1966 |
+
"grad_norm": NaN,
|
| 1967 |
+
"learning_rate": 4.46499775768523e-05,
|
| 1968 |
+
"loss": 0.0,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.6422857142857142,
|
| 1973 |
+
"grad_norm": NaN,
|
| 1974 |
+
"learning_rate": 4.418678104714214e-05,
|
| 1975 |
+
"loss": 0.0,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.6445714285714286,
|
| 1980 |
+
"grad_norm": NaN,
|
| 1981 |
+
"learning_rate": 4.3724089705959305e-05,
|
| 1982 |
+
"loss": 0.0,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.6468571428571429,
|
| 1987 |
+
"grad_norm": NaN,
|
| 1988 |
+
"learning_rate": 4.3261943762757287e-05,
|
| 1989 |
+
"loss": 0.0,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.6491428571428571,
|
| 1994 |
+
"grad_norm": NaN,
|
| 1995 |
+
"learning_rate": 4.280038337959268e-05,
|
| 1996 |
+
"loss": 0.0,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.6514285714285715,
|
| 2001 |
+
"grad_norm": NaN,
|
| 2002 |
+
"learning_rate": 4.233944866763489e-05,
|
| 2003 |
+
"loss": 0.0,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.6537142857142857,
|
| 2008 |
+
"grad_norm": NaN,
|
| 2009 |
+
"learning_rate": 4.187917968368036e-05,
|
| 2010 |
+
"loss": 0.0,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.656,
|
| 2015 |
+
"grad_norm": NaN,
|
| 2016 |
+
"learning_rate": 4.141961642667152e-05,
|
| 2017 |
+
"loss": 0.0,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.6582857142857143,
|
| 2022 |
+
"grad_norm": NaN,
|
| 2023 |
+
"learning_rate": 4.0960798834220704e-05,
|
| 2024 |
+
"loss": 0.0,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.6605714285714286,
|
| 2029 |
+
"grad_norm": NaN,
|
| 2030 |
+
"learning_rate": 4.0502766779139484e-05,
|
| 2031 |
+
"loss": 0.0,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6628571428571428,
|
| 2036 |
+
"grad_norm": NaN,
|
| 2037 |
+
"learning_rate": 4.004556006597353e-05,
|
| 2038 |
+
"loss": 0.0,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.6651428571428571,
|
| 2043 |
+
"grad_norm": NaN,
|
| 2044 |
+
"learning_rate": 3.958921842754351e-05,
|
| 2045 |
+
"loss": 0.0,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.6674285714285715,
|
| 2050 |
+
"grad_norm": NaN,
|
| 2051 |
+
"learning_rate": 3.913378152149214e-05,
|
| 2052 |
+
"loss": 0.0,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.6697142857142857,
|
| 2057 |
+
"grad_norm": NaN,
|
| 2058 |
+
"learning_rate": 3.8679288926837804e-05,
|
| 2059 |
+
"loss": 0.0,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.672,
|
| 2064 |
+
"grad_norm": NaN,
|
| 2065 |
+
"learning_rate": 3.8225780140535025e-05,
|
| 2066 |
+
"loss": 0.0,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.6742857142857143,
|
| 2071 |
+
"grad_norm": NaN,
|
| 2072 |
+
"learning_rate": 3.777329457404202e-05,
|
| 2073 |
+
"loss": 0.0,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.6765714285714286,
|
| 2078 |
+
"grad_norm": NaN,
|
| 2079 |
+
"learning_rate": 3.7321871549895714e-05,
|
| 2080 |
+
"loss": 0.0,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.6788571428571428,
|
| 2085 |
+
"grad_norm": NaN,
|
| 2086 |
+
"learning_rate": 3.68715502982945e-05,
|
| 2087 |
+
"loss": 0.0,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.6811428571428572,
|
| 2092 |
+
"grad_norm": NaN,
|
| 2093 |
+
"learning_rate": 3.642236995368897e-05,
|
| 2094 |
+
"loss": 0.0,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.6834285714285714,
|
| 2099 |
+
"grad_norm": NaN,
|
| 2100 |
+
"learning_rate": 3.597436955138102e-05,
|
| 2101 |
+
"loss": 0.0,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.6857142857142857,
|
| 2106 |
+
"grad_norm": NaN,
|
| 2107 |
+
"learning_rate": 3.5527588024131544e-05,
|
| 2108 |
+
"loss": 0.0,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.688,
|
| 2113 |
+
"grad_norm": NaN,
|
| 2114 |
+
"learning_rate": 3.5082064198777e-05,
|
| 2115 |
+
"loss": 0.0,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.6902857142857143,
|
| 2120 |
+
"grad_norm": NaN,
|
| 2121 |
+
"learning_rate": 3.463783679285535e-05,
|
| 2122 |
+
"loss": 0.0,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.6925714285714286,
|
| 2127 |
+
"grad_norm": NaN,
|
| 2128 |
+
"learning_rate": 3.419494441124121e-05,
|
| 2129 |
+
"loss": 0.0,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.6948571428571428,
|
| 2134 |
+
"grad_norm": NaN,
|
| 2135 |
+
"learning_rate": 3.375342554279111e-05,
|
| 2136 |
+
"loss": 0.0,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.6971428571428572,
|
| 2141 |
+
"grad_norm": NaN,
|
| 2142 |
+
"learning_rate": 3.3313318556998526e-05,
|
| 2143 |
+
"loss": 0.0,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.6994285714285714,
|
| 2148 |
+
"grad_norm": NaN,
|
| 2149 |
+
"learning_rate": 3.287466170065959e-05,
|
| 2150 |
+
"loss": 0.0,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7017142857142857,
|
| 2155 |
+
"grad_norm": NaN,
|
| 2156 |
+
"learning_rate": 3.243749309454922e-05,
|
| 2157 |
+
"loss": 0.0,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.704,
|
| 2162 |
+
"grad_norm": NaN,
|
| 2163 |
+
"learning_rate": 3.200185073010831e-05,
|
| 2164 |
+
"loss": 0.0,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7062857142857143,
|
| 2169 |
+
"grad_norm": NaN,
|
| 2170 |
+
"learning_rate": 3.1567772466142156e-05,
|
| 2171 |
+
"loss": 0.0,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7085714285714285,
|
| 2176 |
+
"grad_norm": NaN,
|
| 2177 |
+
"learning_rate": 3.113529602553042e-05,
|
| 2178 |
+
"loss": 0.0,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.7108571428571429,
|
| 2183 |
+
"grad_norm": NaN,
|
| 2184 |
+
"learning_rate": 3.070445899194885e-05,
|
| 2185 |
+
"loss": 0.0,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7131428571428572,
|
| 2190 |
+
"grad_norm": NaN,
|
| 2191 |
+
"learning_rate": 3.02752988066031e-05,
|
| 2192 |
+
"loss": 0.0,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.7154285714285714,
|
| 2197 |
+
"grad_norm": NaN,
|
| 2198 |
+
"learning_rate": 2.984785276497507e-05,
|
| 2199 |
+
"loss": 0.0,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7177142857142857,
|
| 2204 |
+
"grad_norm": NaN,
|
| 2205 |
+
"learning_rate": 2.9422158013581658e-05,
|
| 2206 |
+
"loss": 0.0,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.72,
|
| 2211 |
+
"grad_norm": NaN,
|
| 2212 |
+
"learning_rate": 2.899825154674674e-05,
|
| 2213 |
+
"loss": 0.0,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.7222857142857143,
|
| 2218 |
+
"grad_norm": NaN,
|
| 2219 |
+
"learning_rate": 2.8576170203386143e-05,
|
| 2220 |
+
"loss": 0.0,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.7245714285714285,
|
| 2225 |
+
"grad_norm": NaN,
|
| 2226 |
+
"learning_rate": 2.8155950663806235e-05,
|
| 2227 |
+
"loss": 0.0,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.7268571428571429,
|
| 2232 |
+
"grad_norm": NaN,
|
| 2233 |
+
"learning_rate": 2.773762944651632e-05,
|
| 2234 |
+
"loss": 0.0,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.7291428571428571,
|
| 2239 |
+
"grad_norm": NaN,
|
| 2240 |
+
"learning_rate": 2.7321242905055013e-05,
|
| 2241 |
+
"loss": 0.0,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.7314285714285714,
|
| 2246 |
+
"grad_norm": NaN,
|
| 2247 |
+
"learning_rate": 2.690682722483102e-05,
|
| 2248 |
+
"loss": 0.0,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.7337142857142858,
|
| 2253 |
+
"grad_norm": NaN,
|
| 2254 |
+
"learning_rate": 2.6494418419978482e-05,
|
| 2255 |
+
"loss": 0.0,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.736,
|
| 2260 |
+
"grad_norm": NaN,
|
| 2261 |
+
"learning_rate": 2.6084052330227238e-05,
|
| 2262 |
+
"loss": 0.0,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.7382857142857143,
|
| 2267 |
+
"grad_norm": NaN,
|
| 2268 |
+
"learning_rate": 2.5675764617788234e-05,
|
| 2269 |
+
"loss": 0.0,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7405714285714285,
|
| 2274 |
+
"grad_norm": NaN,
|
| 2275 |
+
"learning_rate": 2.526959076425434e-05,
|
| 2276 |
+
"loss": 0.0,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.7428571428571429,
|
| 2281 |
+
"grad_norm": NaN,
|
| 2282 |
+
"learning_rate": 2.4865566067516893e-05,
|
| 2283 |
+
"loss": 0.0,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.7451428571428571,
|
| 2288 |
+
"grad_norm": NaN,
|
| 2289 |
+
"learning_rate": 2.4463725638698183e-05,
|
| 2290 |
+
"loss": 0.0,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.7474285714285714,
|
| 2295 |
+
"grad_norm": NaN,
|
| 2296 |
+
"learning_rate": 2.406410439910017e-05,
|
| 2297 |
+
"loss": 0.0,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.7497142857142857,
|
| 2302 |
+
"grad_norm": NaN,
|
| 2303 |
+
"learning_rate": 2.3666737077169726e-05,
|
| 2304 |
+
"loss": 0.0,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.752,
|
| 2309 |
+
"grad_norm": NaN,
|
| 2310 |
+
"learning_rate": 2.327165820548059e-05,
|
| 2311 |
+
"loss": 0.0,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.7542857142857143,
|
| 2316 |
+
"grad_norm": NaN,
|
| 2317 |
+
"learning_rate": 2.287890211773238e-05,
|
| 2318 |
+
"loss": 0.0,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7565714285714286,
|
| 2323 |
+
"grad_norm": NaN,
|
| 2324 |
+
"learning_rate": 2.2488502945766894e-05,
|
| 2325 |
+
"loss": 0.0,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.7588571428571429,
|
| 2330 |
+
"grad_norm": NaN,
|
| 2331 |
+
"learning_rate": 2.2100494616601893e-05,
|
| 2332 |
+
"loss": 0.0,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7611428571428571,
|
| 2337 |
+
"grad_norm": NaN,
|
| 2338 |
+
"learning_rate": 2.171491084948278e-05,
|
| 2339 |
+
"loss": 0.0,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7634285714285715,
|
| 2344 |
+
"grad_norm": NaN,
|
| 2345 |
+
"learning_rate": 2.1331785152952243e-05,
|
| 2346 |
+
"loss": 0.0,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.7657142857142857,
|
| 2351 |
+
"grad_norm": NaN,
|
| 2352 |
+
"learning_rate": 2.0951150821938275e-05,
|
| 2353 |
+
"loss": 0.0,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.768,
|
| 2358 |
+
"grad_norm": NaN,
|
| 2359 |
+
"learning_rate": 2.0573040934860717e-05,
|
| 2360 |
+
"loss": 0.0,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.7702857142857142,
|
| 2365 |
+
"grad_norm": NaN,
|
| 2366 |
+
"learning_rate": 2.0197488350756616e-05,
|
| 2367 |
+
"loss": 0.0,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.7725714285714286,
|
| 2372 |
+
"grad_norm": NaN,
|
| 2373 |
+
"learning_rate": 1.98245257064247e-05,
|
| 2374 |
+
"loss": 0.0,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.7748571428571429,
|
| 2379 |
+
"grad_norm": NaN,
|
| 2380 |
+
"learning_rate": 1.945418541358911e-05,
|
| 2381 |
+
"loss": 0.0,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.7771428571428571,
|
| 2386 |
+
"grad_norm": NaN,
|
| 2387 |
+
"learning_rate": 1.9086499656082686e-05,
|
| 2388 |
+
"loss": 0.0,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.7794285714285715,
|
| 2393 |
+
"grad_norm": NaN,
|
| 2394 |
+
"learning_rate": 1.872150038705015e-05,
|
| 2395 |
+
"loss": 0.0,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.7817142857142857,
|
| 2400 |
+
"grad_norm": NaN,
|
| 2401 |
+
"learning_rate": 1.835921932617119e-05,
|
| 2402 |
+
"loss": 0.0,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.784,
|
| 2407 |
+
"grad_norm": NaN,
|
| 2408 |
+
"learning_rate": 1.7999687956903953e-05,
|
| 2409 |
+
"loss": 0.0,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.7862857142857143,
|
| 2414 |
+
"grad_norm": NaN,
|
| 2415 |
+
"learning_rate": 1.7642937523749038e-05,
|
| 2416 |
+
"loss": 0.0,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.7885714285714286,
|
| 2421 |
+
"grad_norm": NaN,
|
| 2422 |
+
"learning_rate": 1.7288999029534176e-05,
|
| 2423 |
+
"loss": 0.0,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.7908571428571428,
|
| 2428 |
+
"grad_norm": NaN,
|
| 2429 |
+
"learning_rate": 1.6937903232720077e-05,
|
| 2430 |
+
"loss": 0.0,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.7931428571428571,
|
| 2435 |
+
"grad_norm": NaN,
|
| 2436 |
+
"learning_rate": 1.6589680644727344e-05,
|
| 2437 |
+
"loss": 0.0,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7954285714285714,
|
| 2442 |
+
"grad_norm": NaN,
|
| 2443 |
+
"learning_rate": 1.6244361527284952e-05,
|
| 2444 |
+
"loss": 0.0,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.7977142857142857,
|
| 2449 |
+
"grad_norm": NaN,
|
| 2450 |
+
"learning_rate": 1.5901975889800386e-05,
|
| 2451 |
+
"loss": 0.0,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8,
|
| 2456 |
+
"grad_norm": NaN,
|
| 2457 |
+
"learning_rate": 1.5562553486751743e-05,
|
| 2458 |
+
"loss": 0.0,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.8022857142857143,
|
| 2463 |
+
"grad_norm": NaN,
|
| 2464 |
+
"learning_rate": 1.5226123815101951e-05,
|
| 2465 |
+
"loss": 0.0,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8045714285714286,
|
| 2470 |
+
"grad_norm": NaN,
|
| 2471 |
+
"learning_rate": 1.4892716111735378e-05,
|
| 2472 |
+
"loss": 0.0,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8068571428571428,
|
| 2477 |
+
"grad_norm": NaN,
|
| 2478 |
+
"learning_rate": 1.4562359350917054e-05,
|
| 2479 |
+
"loss": 0.0,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8091428571428572,
|
| 2484 |
+
"grad_norm": NaN,
|
| 2485 |
+
"learning_rate": 1.423508224177474e-05,
|
| 2486 |
+
"loss": 0.0,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8114285714285714,
|
| 2491 |
+
"grad_norm": NaN,
|
| 2492 |
+
"learning_rate": 1.3910913225803945e-05,
|
| 2493 |
+
"loss": 0.0,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8137142857142857,
|
| 2498 |
+
"grad_norm": NaN,
|
| 2499 |
+
"learning_rate": 1.35898804743963e-05,
|
| 2500 |
+
"loss": 0.0,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.816,
|
| 2505 |
+
"grad_norm": NaN,
|
| 2506 |
+
"learning_rate": 1.3272011886391366e-05,
|
| 2507 |
+
"loss": 0.0,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8182857142857143,
|
| 2512 |
+
"grad_norm": NaN,
|
| 2513 |
+
"learning_rate": 1.2957335085652129e-05,
|
| 2514 |
+
"loss": 0.0,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.8205714285714286,
|
| 2519 |
+
"grad_norm": NaN,
|
| 2520 |
+
"learning_rate": 1.2645877418664392e-05,
|
| 2521 |
+
"loss": 0.0,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.8228571428571428,
|
| 2526 |
+
"grad_norm": NaN,
|
| 2527 |
+
"learning_rate": 1.2337665952160266e-05,
|
| 2528 |
+
"loss": 0.0,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.8251428571428572,
|
| 2533 |
+
"grad_norm": NaN,
|
| 2534 |
+
"learning_rate": 1.203272747076598e-05,
|
| 2535 |
+
"loss": 0.0,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.8274285714285714,
|
| 2540 |
+
"grad_norm": NaN,
|
| 2541 |
+
"learning_rate": 1.1731088474674234e-05,
|
| 2542 |
+
"loss": 0.0,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8297142857142857,
|
| 2547 |
+
"grad_norm": NaN,
|
| 2548 |
+
"learning_rate": 1.1432775177341165e-05,
|
| 2549 |
+
"loss": 0.0,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.832,
|
| 2554 |
+
"grad_norm": NaN,
|
| 2555 |
+
"learning_rate": 1.11378135032084e-05,
|
| 2556 |
+
"loss": 0.0,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8342857142857143,
|
| 2561 |
+
"grad_norm": NaN,
|
| 2562 |
+
"learning_rate": 1.0846229085450099e-05,
|
| 2563 |
+
"loss": 0.0,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.8365714285714285,
|
| 2568 |
+
"grad_norm": NaN,
|
| 2569 |
+
"learning_rate": 1.0558047263745297e-05,
|
| 2570 |
+
"loss": 0.0,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.8388571428571429,
|
| 2575 |
+
"grad_norm": NaN,
|
| 2576 |
+
"learning_rate": 1.0273293082075913e-05,
|
| 2577 |
+
"loss": 0.0,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8411428571428572,
|
| 2582 |
+
"grad_norm": NaN,
|
| 2583 |
+
"learning_rate": 9.991991286550207e-06,
|
| 2584 |
+
"loss": 0.0,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8434285714285714,
|
| 2589 |
+
"grad_norm": NaN,
|
| 2590 |
+
"learning_rate": 9.71416632325235e-06,
|
| 2591 |
+
"loss": 0.0,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8457142857142858,
|
| 2596 |
+
"grad_norm": NaN,
|
| 2597 |
+
"learning_rate": 9.439842336117954e-06,
|
| 2598 |
+
"loss": 0.0,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.848,
|
| 2603 |
+
"grad_norm": NaN,
|
| 2604 |
+
"learning_rate": 9.169043164835867e-06,
|
| 2605 |
+
"loss": 0.0,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8502857142857143,
|
| 2610 |
+
"grad_norm": NaN,
|
| 2611 |
+
"learning_rate": 8.901792342776437e-06,
|
| 2612 |
+
"loss": 0.0,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8525714285714285,
|
| 2617 |
+
"grad_norm": NaN,
|
| 2618 |
+
"learning_rate": 8.638113094946381e-06,
|
| 2619 |
+
"loss": 0.0,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8548571428571429,
|
| 2624 |
+
"grad_norm": NaN,
|
| 2625 |
+
"learning_rate": 8.37802833597045e-06,
|
| 2626 |
+
"loss": 0.0,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8571428571428571,
|
| 2631 |
+
"grad_norm": NaN,
|
| 2632 |
+
"learning_rate": 8.121560668100064e-06,
|
| 2633 |
+
"loss": 0.0,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.8594285714285714,
|
| 2638 |
+
"grad_norm": NaN,
|
| 2639 |
+
"learning_rate": 7.868732379249121e-06,
|
| 2640 |
+
"loss": 0.0,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8617142857142858,
|
| 2645 |
+
"grad_norm": NaN,
|
| 2646 |
+
"learning_rate": 7.619565441057075e-06,
|
| 2647 |
+
"loss": 0.0,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.864,
|
| 2652 |
+
"grad_norm": NaN,
|
| 2653 |
+
"learning_rate": 7.37408150697953e-06,
|
| 2654 |
+
"loss": 0.0,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8662857142857143,
|
| 2659 |
+
"grad_norm": NaN,
|
| 2660 |
+
"learning_rate": 7.132301910406503e-06,
|
| 2661 |
+
"loss": 0.0,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.8685714285714285,
|
| 2666 |
+
"grad_norm": NaN,
|
| 2667 |
+
"learning_rate": 6.894247662808456e-06,
|
| 2668 |
+
"loss": 0.0,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.8708571428571429,
|
| 2673 |
+
"grad_norm": NaN,
|
| 2674 |
+
"learning_rate": 6.659939451910341e-06,
|
| 2675 |
+
"loss": 0.0,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.8731428571428571,
|
| 2680 |
+
"grad_norm": NaN,
|
| 2681 |
+
"learning_rate": 6.429397639893758e-06,
|
| 2682 |
+
"loss": 0.0,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.8754285714285714,
|
| 2687 |
+
"grad_norm": NaN,
|
| 2688 |
+
"learning_rate": 6.202642261627411e-06,
|
| 2689 |
+
"loss": 0.0,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.8777142857142857,
|
| 2694 |
+
"grad_norm": NaN,
|
| 2695 |
+
"learning_rate": 5.979693022926025e-06,
|
| 2696 |
+
"loss": 0.0,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.88,
|
| 2701 |
+
"grad_norm": NaN,
|
| 2702 |
+
"learning_rate": 5.7605692988378255e-06,
|
| 2703 |
+
"loss": 0.0,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.8822857142857143,
|
| 2708 |
+
"grad_norm": NaN,
|
| 2709 |
+
"learning_rate": 5.5452901319607894e-06,
|
| 2710 |
+
"loss": 0.0,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.8845714285714286,
|
| 2715 |
+
"grad_norm": NaN,
|
| 2716 |
+
"learning_rate": 5.333874230787772e-06,
|
| 2717 |
+
"loss": 0.0,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.8868571428571429,
|
| 2722 |
+
"grad_norm": NaN,
|
| 2723 |
+
"learning_rate": 5.126339968080696e-06,
|
| 2724 |
+
"loss": 0.0,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.8891428571428571,
|
| 2729 |
+
"grad_norm": NaN,
|
| 2730 |
+
"learning_rate": 4.922705379273862e-06,
|
| 2731 |
+
"loss": 0.0,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.8914285714285715,
|
| 2736 |
+
"grad_norm": NaN,
|
| 2737 |
+
"learning_rate": 4.722988160906638e-06,
|
| 2738 |
+
"loss": 0.0,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.8937142857142857,
|
| 2743 |
+
"grad_norm": NaN,
|
| 2744 |
+
"learning_rate": 4.527205669085549e-06,
|
| 2745 |
+
"loss": 0.0,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.896,
|
| 2750 |
+
"grad_norm": NaN,
|
| 2751 |
+
"learning_rate": 4.335374917975981e-06,
|
| 2752 |
+
"loss": 0.0,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.8982857142857142,
|
| 2757 |
+
"grad_norm": NaN,
|
| 2758 |
+
"learning_rate": 4.147512578323614e-06,
|
| 2759 |
+
"loss": 0.0,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.9005714285714286,
|
| 2764 |
+
"grad_norm": NaN,
|
| 2765 |
+
"learning_rate": 3.963634976005642e-06,
|
| 2766 |
+
"loss": 0.0,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.9028571428571428,
|
| 2771 |
+
"grad_norm": NaN,
|
| 2772 |
+
"learning_rate": 3.783758090611983e-06,
|
| 2773 |
+
"loss": 0.0,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.9051428571428571,
|
| 2778 |
+
"grad_norm": NaN,
|
| 2779 |
+
"learning_rate": 3.6078975540566716e-06,
|
| 2780 |
+
"loss": 0.0,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.9074285714285715,
|
| 2785 |
+
"grad_norm": NaN,
|
| 2786 |
+
"learning_rate": 3.436068649219326e-06,
|
| 2787 |
+
"loss": 0.0,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.9097142857142857,
|
| 2792 |
+
"grad_norm": NaN,
|
| 2793 |
+
"learning_rate": 3.268286308617041e-06,
|
| 2794 |
+
"loss": 0.0,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.912,
|
| 2799 |
+
"grad_norm": NaN,
|
| 2800 |
+
"learning_rate": 3.1045651131066887e-06,
|
| 2801 |
+
"loss": 0.0,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.9142857142857143,
|
| 2806 |
+
"grad_norm": NaN,
|
| 2807 |
+
"learning_rate": 2.94491929061782e-06,
|
| 2808 |
+
"loss": 0.0,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.9165714285714286,
|
| 2813 |
+
"grad_norm": NaN,
|
| 2814 |
+
"learning_rate": 2.7893627149161716e-06,
|
| 2815 |
+
"loss": 0.0,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.9188571428571428,
|
| 2820 |
+
"grad_norm": NaN,
|
| 2821 |
+
"learning_rate": 2.6379089043980067e-06,
|
| 2822 |
+
"loss": 0.0,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.9211428571428572,
|
| 2827 |
+
"grad_norm": NaN,
|
| 2828 |
+
"learning_rate": 2.4905710209153223e-06,
|
| 2829 |
+
"loss": 0.0,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.9234285714285714,
|
| 2834 |
+
"grad_norm": NaN,
|
| 2835 |
+
"learning_rate": 2.3473618686320474e-06,
|
| 2836 |
+
"loss": 0.0,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.9257142857142857,
|
| 2841 |
+
"grad_norm": NaN,
|
| 2842 |
+
"learning_rate": 2.208293892911284e-06,
|
| 2843 |
+
"loss": 0.0,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.928,
|
| 2848 |
+
"grad_norm": NaN,
|
| 2849 |
+
"learning_rate": 2.07337917923382e-06,
|
| 2850 |
+
"loss": 0.0,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.9302857142857143,
|
| 2855 |
+
"grad_norm": NaN,
|
| 2856 |
+
"learning_rate": 1.9426294521477873e-06,
|
| 2857 |
+
"loss": 0.0,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9325714285714286,
|
| 2862 |
+
"grad_norm": NaN,
|
| 2863 |
+
"learning_rate": 1.8160560742498222e-06,
|
| 2864 |
+
"loss": 0.0,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9348571428571428,
|
| 2869 |
+
"grad_norm": NaN,
|
| 2870 |
+
"learning_rate": 1.6936700451975817e-06,
|
| 2871 |
+
"loss": 0.0,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9371428571428572,
|
| 2876 |
+
"grad_norm": NaN,
|
| 2877 |
+
"learning_rate": 1.5754820007538474e-06,
|
| 2878 |
+
"loss": 0.0,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9394285714285714,
|
| 2883 |
+
"grad_norm": NaN,
|
| 2884 |
+
"learning_rate": 1.4615022118622367e-06,
|
| 2885 |
+
"loss": 0.0,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.9417142857142857,
|
| 2890 |
+
"grad_norm": NaN,
|
| 2891 |
+
"learning_rate": 1.3517405837546403e-06,
|
| 2892 |
+
"loss": 0.0,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.944,
|
| 2897 |
+
"grad_norm": NaN,
|
| 2898 |
+
"learning_rate": 1.2462066550903817e-06,
|
| 2899 |
+
"loss": 0.0,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9462857142857143,
|
| 2904 |
+
"grad_norm": NaN,
|
| 2905 |
+
"learning_rate": 1.1449095971273305e-06,
|
| 2906 |
+
"loss": 0.0,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9485714285714286,
|
| 2911 |
+
"grad_norm": NaN,
|
| 2912 |
+
"learning_rate": 1.0478582129248515e-06,
|
| 2913 |
+
"loss": 0.0,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9508571428571428,
|
| 2918 |
+
"grad_norm": NaN,
|
| 2919 |
+
"learning_rate": 9.550609365787888e-07,
|
| 2920 |
+
"loss": 0.0,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.9531428571428572,
|
| 2925 |
+
"grad_norm": NaN,
|
| 2926 |
+
"learning_rate": 8.66525832488535e-07,
|
| 2927 |
+
"loss": 0.0,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9554285714285714,
|
| 2932 |
+
"grad_norm": NaN,
|
| 2933 |
+
"learning_rate": 7.822605946561923e-07,
|
| 2934 |
+
"loss": 0.0,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9577142857142857,
|
| 2939 |
+
"grad_norm": NaN,
|
| 2940 |
+
"learning_rate": 7.022725460179457e-07,
|
| 2941 |
+
"loss": 0.0,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.96,
|
| 2946 |
+
"grad_norm": NaN,
|
| 2947 |
+
"learning_rate": 6.265686378076729e-07,
|
| 2948 |
+
"loss": 0.0,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9622857142857143,
|
| 2953 |
+
"grad_norm": NaN,
|
| 2954 |
+
"learning_rate": 5.551554489528432e-07,
|
| 2955 |
+
"loss": 0.0,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.9645714285714285,
|
| 2960 |
+
"grad_norm": NaN,
|
| 2961 |
+
"learning_rate": 4.880391855028088e-07,
|
| 2962 |
+
"loss": 0.0,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.9668571428571429,
|
| 2967 |
+
"grad_norm": NaN,
|
| 2968 |
+
"learning_rate": 4.252256800894694e-07,
|
| 2969 |
+
"loss": 0.0,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.9691428571428572,
|
| 2974 |
+
"grad_norm": NaN,
|
| 2975 |
+
"learning_rate": 3.667203914203943e-07,
|
| 2976 |
+
"loss": 0.0,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.9714285714285714,
|
| 2981 |
+
"grad_norm": NaN,
|
| 2982 |
+
"learning_rate": 3.1252840380444073e-07,
|
| 2983 |
+
"loss": 0.0,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.9737142857142858,
|
| 2988 |
+
"grad_norm": NaN,
|
| 2989 |
+
"learning_rate": 2.626544267099129e-07,
|
| 2990 |
+
"loss": 0.0,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.976,
|
| 2995 |
+
"grad_norm": NaN,
|
| 2996 |
+
"learning_rate": 2.1710279435530057e-07,
|
| 2997 |
+
"loss": 0.0,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.9782857142857143,
|
| 3002 |
+
"grad_norm": NaN,
|
| 3003 |
+
"learning_rate": 1.7587746533260784e-07,
|
| 3004 |
+
"loss": 0.0,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.9805714285714285,
|
| 3009 |
+
"grad_norm": NaN,
|
| 3010 |
+
"learning_rate": 1.3898202226333423e-07,
|
| 3011 |
+
"loss": 0.0,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.9828571428571429,
|
| 3016 |
+
"grad_norm": NaN,
|
| 3017 |
+
"learning_rate": 1.0641967148716236e-07,
|
| 3018 |
+
"loss": 0.0,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.9851428571428571,
|
| 3023 |
+
"grad_norm": NaN,
|
| 3024 |
+
"learning_rate": 7.8193242783281e-08,
|
| 3025 |
+
"loss": 0.0,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.9874285714285714,
|
| 3030 |
+
"grad_norm": NaN,
|
| 3031 |
+
"learning_rate": 5.430518912448168e-08,
|
| 3032 |
+
"loss": 0.0,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.9897142857142858,
|
| 3037 |
+
"grad_norm": NaN,
|
| 3038 |
+
"learning_rate": 3.475758646400151e-08,
|
| 3039 |
+
"loss": 0.0,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.992,
|
| 3044 |
+
"grad_norm": NaN,
|
| 3045 |
+
"learning_rate": 1.9552133555084117e-08,
|
| 3046 |
+
"loss": 0.0,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.9942857142857143,
|
| 3051 |
+
"grad_norm": NaN,
|
| 3052 |
+
"learning_rate": 8.690151803386616e-09,
|
| 3053 |
+
"loss": 0.0,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.9965714285714286,
|
| 3058 |
+
"grad_norm": NaN,
|
| 3059 |
+
"learning_rate": 2.1725851521103846e-09,
|
| 3060 |
+
"loss": 0.0,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.9988571428571429,
|
| 3065 |
+
"grad_norm": NaN,
|
| 3066 |
+
"learning_rate": 0.0,
|
| 3067 |
+
"loss": 0.0,
|
| 3068 |
+
"step": 437
|
| 3069 |
}
|
| 3070 |
],
|
| 3071 |
"logging_steps": 1,
|
|
|
|
| 3080 |
"should_evaluate": false,
|
| 3081 |
"should_log": false,
|
| 3082 |
"should_save": true,
|
| 3083 |
+
"should_training_stop": true
|
| 3084 |
},
|
| 3085 |
"attributes": {}
|
| 3086 |
}
|
| 3087 |
},
|
| 3088 |
+
"total_flos": 5599012097359872.0,
|
| 3089 |
"train_batch_size": 8,
|
| 3090 |
"trial_name": null,
|
| 3091 |
"trial_params": null
|