End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec05_23-48-52_8fb6626b3f6f/events.out.tfevents.1733447829.8fb6626b3f6f.752.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +2952 -0
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 1.
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.3695652173913043
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 1.3992
|
| 37 |
+
- Accuracy: 0.3696
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 117.33333333333333,
|
| 3 |
+
"eval_accuracy": 0.3695652173913043,
|
| 4 |
+
"eval_loss": 1.3992067575454712,
|
| 5 |
+
"eval_runtime": 1.0044,
|
| 6 |
+
"eval_samples_per_second": 45.799,
|
| 7 |
+
"eval_steps_per_second": 2.987,
|
| 8 |
+
"total_flos": 5.466852859010089e+18,
|
| 9 |
+
"train_loss": 1.1631801536588957,
|
| 10 |
+
"train_runtime": 5282.7516,
|
| 11 |
+
"train_samples_per_second": 32.529,
|
| 12 |
+
"train_steps_per_second": 0.5
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 117.33333333333333,
|
| 3 |
+
"eval_accuracy": 0.3695652173913043,
|
| 4 |
+
"eval_loss": 1.3992067575454712,
|
| 5 |
+
"eval_runtime": 1.0044,
|
| 6 |
+
"eval_samples_per_second": 45.799,
|
| 7 |
+
"eval_steps_per_second": 2.987
|
| 8 |
+
}
|
runs/Dec05_23-48-52_8fb6626b3f6f/events.out.tfevents.1733447829.8fb6626b3f6f.752.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2922ef2a37c1f5d10aee2bbc62b7c71e977da9cb38dfdb9eb2bb41557343d2ea
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 117.33333333333333,
|
| 3 |
+
"total_flos": 5.466852859010089e+18,
|
| 4 |
+
"train_loss": 1.1631801536588957,
|
| 5 |
+
"train_runtime": 5282.7516,
|
| 6 |
+
"train_samples_per_second": 32.529,
|
| 7 |
+
"train_steps_per_second": 0.5
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2952 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.3695652173913043,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-da3-colab/checkpoint-1485",
|
| 4 |
+
"epoch": 117.33333333333333,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2640,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.4444444444444444,
|
| 13 |
+
"grad_norm": 3.825187921524048,
|
| 14 |
+
"learning_rate": 0.0009962121212121213,
|
| 15 |
+
"loss": 1.5683,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.8888888888888888,
|
| 20 |
+
"grad_norm": 5.65834903717041,
|
| 21 |
+
"learning_rate": 0.0009924242424242424,
|
| 22 |
+
"loss": 1.3523,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.9777777777777777,
|
| 27 |
+
"eval_accuracy": 0.32608695652173914,
|
| 28 |
+
"eval_loss": 1.4024076461791992,
|
| 29 |
+
"eval_runtime": 1.0691,
|
| 30 |
+
"eval_samples_per_second": 43.028,
|
| 31 |
+
"eval_steps_per_second": 2.806,
|
| 32 |
+
"step": 22
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.3333333333333333,
|
| 36 |
+
"grad_norm": 10.152283668518066,
|
| 37 |
+
"learning_rate": 0.0009886363636363636,
|
| 38 |
+
"loss": 1.2778,
|
| 39 |
+
"step": 30
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.7777777777777777,
|
| 43 |
+
"grad_norm": 3.099470376968384,
|
| 44 |
+
"learning_rate": 0.000984848484848485,
|
| 45 |
+
"loss": 1.3805,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_accuracy": 0.2608695652173913,
|
| 51 |
+
"eval_loss": 1.3775243759155273,
|
| 52 |
+
"eval_runtime": 0.9313,
|
| 53 |
+
"eval_samples_per_second": 49.394,
|
| 54 |
+
"eval_steps_per_second": 3.221,
|
| 55 |
+
"step": 45
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.2222222222222223,
|
| 59 |
+
"grad_norm": 4.7406511306762695,
|
| 60 |
+
"learning_rate": 0.0009810606060606062,
|
| 61 |
+
"loss": 1.3708,
|
| 62 |
+
"step": 50
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.6666666666666665,
|
| 66 |
+
"grad_norm": 3.888740301132202,
|
| 67 |
+
"learning_rate": 0.0009772727272727272,
|
| 68 |
+
"loss": 1.3221,
|
| 69 |
+
"step": 60
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.977777777777778,
|
| 73 |
+
"eval_accuracy": 0.30434782608695654,
|
| 74 |
+
"eval_loss": 1.4418785572052002,
|
| 75 |
+
"eval_runtime": 0.8857,
|
| 76 |
+
"eval_samples_per_second": 51.938,
|
| 77 |
+
"eval_steps_per_second": 3.387,
|
| 78 |
+
"step": 67
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.111111111111111,
|
| 82 |
+
"grad_norm": 2.308217763900757,
|
| 83 |
+
"learning_rate": 0.0009734848484848485,
|
| 84 |
+
"loss": 1.2355,
|
| 85 |
+
"step": 70
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.5555555555555554,
|
| 89 |
+
"grad_norm": 5.7985053062438965,
|
| 90 |
+
"learning_rate": 0.0009696969696969698,
|
| 91 |
+
"loss": 1.2409,
|
| 92 |
+
"step": 80
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.0,
|
| 96 |
+
"grad_norm": 110.87601470947266,
|
| 97 |
+
"learning_rate": 0.000965909090909091,
|
| 98 |
+
"loss": 1.297,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.0,
|
| 103 |
+
"eval_accuracy": 0.32608695652173914,
|
| 104 |
+
"eval_loss": 1.3581539392471313,
|
| 105 |
+
"eval_runtime": 0.8843,
|
| 106 |
+
"eval_samples_per_second": 52.019,
|
| 107 |
+
"eval_steps_per_second": 3.393,
|
| 108 |
+
"step": 90
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.444444444444445,
|
| 112 |
+
"grad_norm": 4.244718551635742,
|
| 113 |
+
"learning_rate": 0.0009621212121212122,
|
| 114 |
+
"loss": 1.3056,
|
| 115 |
+
"step": 100
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 4.888888888888889,
|
| 119 |
+
"grad_norm": 4.412840843200684,
|
| 120 |
+
"learning_rate": 0.0009583333333333334,
|
| 121 |
+
"loss": 1.353,
|
| 122 |
+
"step": 110
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.977777777777778,
|
| 126 |
+
"eval_accuracy": 0.34782608695652173,
|
| 127 |
+
"eval_loss": 1.3405784368515015,
|
| 128 |
+
"eval_runtime": 0.8733,
|
| 129 |
+
"eval_samples_per_second": 52.672,
|
| 130 |
+
"eval_steps_per_second": 3.435,
|
| 131 |
+
"step": 112
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.333333333333333,
|
| 135 |
+
"grad_norm": 2.0132997035980225,
|
| 136 |
+
"learning_rate": 0.0009545454545454546,
|
| 137 |
+
"loss": 1.3115,
|
| 138 |
+
"step": 120
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 5.777777777777778,
|
| 142 |
+
"grad_norm": 3.7579846382141113,
|
| 143 |
+
"learning_rate": 0.0009507575757575758,
|
| 144 |
+
"loss": 1.2627,
|
| 145 |
+
"step": 130
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 6.0,
|
| 149 |
+
"eval_accuracy": 0.15217391304347827,
|
| 150 |
+
"eval_loss": 1.3823662996292114,
|
| 151 |
+
"eval_runtime": 0.8973,
|
| 152 |
+
"eval_samples_per_second": 51.264,
|
| 153 |
+
"eval_steps_per_second": 3.343,
|
| 154 |
+
"step": 135
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.222222222222222,
|
| 158 |
+
"grad_norm": 3.731177806854248,
|
| 159 |
+
"learning_rate": 0.000946969696969697,
|
| 160 |
+
"loss": 1.2545,
|
| 161 |
+
"step": 140
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.666666666666667,
|
| 165 |
+
"grad_norm": 3.8037898540496826,
|
| 166 |
+
"learning_rate": 0.0009431818181818183,
|
| 167 |
+
"loss": 1.3006,
|
| 168 |
+
"step": 150
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.977777777777778,
|
| 172 |
+
"eval_accuracy": 0.15217391304347827,
|
| 173 |
+
"eval_loss": 1.4008022546768188,
|
| 174 |
+
"eval_runtime": 1.1693,
|
| 175 |
+
"eval_samples_per_second": 39.34,
|
| 176 |
+
"eval_steps_per_second": 2.566,
|
| 177 |
+
"step": 157
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.111111111111111,
|
| 181 |
+
"grad_norm": 2.627021074295044,
|
| 182 |
+
"learning_rate": 0.0009393939393939394,
|
| 183 |
+
"loss": 1.2558,
|
| 184 |
+
"step": 160
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.555555555555555,
|
| 188 |
+
"grad_norm": 2.487823247909546,
|
| 189 |
+
"learning_rate": 0.0009356060606060606,
|
| 190 |
+
"loss": 1.216,
|
| 191 |
+
"step": 170
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 8.0,
|
| 195 |
+
"grad_norm": 2.525559902191162,
|
| 196 |
+
"learning_rate": 0.0009318181818181818,
|
| 197 |
+
"loss": 1.2438,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 8.0,
|
| 202 |
+
"eval_accuracy": 0.32608695652173914,
|
| 203 |
+
"eval_loss": 1.3769112825393677,
|
| 204 |
+
"eval_runtime": 0.9024,
|
| 205 |
+
"eval_samples_per_second": 50.973,
|
| 206 |
+
"eval_steps_per_second": 3.324,
|
| 207 |
+
"step": 180
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 8.444444444444445,
|
| 211 |
+
"grad_norm": 9.093722343444824,
|
| 212 |
+
"learning_rate": 0.000928030303030303,
|
| 213 |
+
"loss": 1.2023,
|
| 214 |
+
"step": 190
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 8.88888888888889,
|
| 218 |
+
"grad_norm": 7.364065170288086,
|
| 219 |
+
"learning_rate": 0.0009242424242424242,
|
| 220 |
+
"loss": 1.222,
|
| 221 |
+
"step": 200
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 8.977777777777778,
|
| 225 |
+
"eval_accuracy": 0.30434782608695654,
|
| 226 |
+
"eval_loss": 1.421162486076355,
|
| 227 |
+
"eval_runtime": 0.8938,
|
| 228 |
+
"eval_samples_per_second": 51.468,
|
| 229 |
+
"eval_steps_per_second": 3.357,
|
| 230 |
+
"step": 202
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.333333333333334,
|
| 234 |
+
"grad_norm": 3.667402744293213,
|
| 235 |
+
"learning_rate": 0.0009204545454545455,
|
| 236 |
+
"loss": 1.2186,
|
| 237 |
+
"step": 210
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 9.777777777777779,
|
| 241 |
+
"grad_norm": 10.466683387756348,
|
| 242 |
+
"learning_rate": 0.0009166666666666666,
|
| 243 |
+
"loss": 1.2221,
|
| 244 |
+
"step": 220
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 10.0,
|
| 248 |
+
"eval_accuracy": 0.2391304347826087,
|
| 249 |
+
"eval_loss": 1.4223273992538452,
|
| 250 |
+
"eval_runtime": 0.8601,
|
| 251 |
+
"eval_samples_per_second": 53.48,
|
| 252 |
+
"eval_steps_per_second": 3.488,
|
| 253 |
+
"step": 225
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 10.222222222222221,
|
| 257 |
+
"grad_norm": 13.964035987854004,
|
| 258 |
+
"learning_rate": 0.0009128787878787878,
|
| 259 |
+
"loss": 1.2394,
|
| 260 |
+
"step": 230
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 10.666666666666666,
|
| 264 |
+
"grad_norm": 4.716994762420654,
|
| 265 |
+
"learning_rate": 0.0009090909090909091,
|
| 266 |
+
"loss": 1.2262,
|
| 267 |
+
"step": 240
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 10.977777777777778,
|
| 271 |
+
"eval_accuracy": 0.2608695652173913,
|
| 272 |
+
"eval_loss": 1.4154268503189087,
|
| 273 |
+
"eval_runtime": 1.2346,
|
| 274 |
+
"eval_samples_per_second": 37.258,
|
| 275 |
+
"eval_steps_per_second": 2.43,
|
| 276 |
+
"step": 247
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 11.11111111111111,
|
| 280 |
+
"grad_norm": 11.025979995727539,
|
| 281 |
+
"learning_rate": 0.0009053030303030303,
|
| 282 |
+
"loss": 1.2196,
|
| 283 |
+
"step": 250
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 11.555555555555555,
|
| 287 |
+
"grad_norm": 22.928346633911133,
|
| 288 |
+
"learning_rate": 0.0009015151515151515,
|
| 289 |
+
"loss": 1.2131,
|
| 290 |
+
"step": 260
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 12.0,
|
| 294 |
+
"grad_norm": 7.0689263343811035,
|
| 295 |
+
"learning_rate": 0.0008977272727272727,
|
| 296 |
+
"loss": 1.2381,
|
| 297 |
+
"step": 270
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 12.0,
|
| 301 |
+
"eval_accuracy": 0.2391304347826087,
|
| 302 |
+
"eval_loss": 1.3327127695083618,
|
| 303 |
+
"eval_runtime": 0.884,
|
| 304 |
+
"eval_samples_per_second": 52.039,
|
| 305 |
+
"eval_steps_per_second": 3.394,
|
| 306 |
+
"step": 270
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 12.444444444444445,
|
| 310 |
+
"grad_norm": 3.511734962463379,
|
| 311 |
+
"learning_rate": 0.000893939393939394,
|
| 312 |
+
"loss": 1.1634,
|
| 313 |
+
"step": 280
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 12.88888888888889,
|
| 317 |
+
"grad_norm": 33.742347717285156,
|
| 318 |
+
"learning_rate": 0.0008901515151515151,
|
| 319 |
+
"loss": 1.227,
|
| 320 |
+
"step": 290
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 12.977777777777778,
|
| 324 |
+
"eval_accuracy": 0.2826086956521739,
|
| 325 |
+
"eval_loss": 1.288680076599121,
|
| 326 |
+
"eval_runtime": 0.8658,
|
| 327 |
+
"eval_samples_per_second": 53.132,
|
| 328 |
+
"eval_steps_per_second": 3.465,
|
| 329 |
+
"step": 292
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.333333333333334,
|
| 333 |
+
"grad_norm": 6.175626277923584,
|
| 334 |
+
"learning_rate": 0.0008863636363636364,
|
| 335 |
+
"loss": 1.2082,
|
| 336 |
+
"step": 300
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 13.777777777777779,
|
| 340 |
+
"grad_norm": 11.057406425476074,
|
| 341 |
+
"learning_rate": 0.0008825757575757576,
|
| 342 |
+
"loss": 1.2158,
|
| 343 |
+
"step": 310
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 14.0,
|
| 347 |
+
"eval_accuracy": 0.2608695652173913,
|
| 348 |
+
"eval_loss": 1.3465280532836914,
|
| 349 |
+
"eval_runtime": 0.872,
|
| 350 |
+
"eval_samples_per_second": 52.749,
|
| 351 |
+
"eval_steps_per_second": 3.44,
|
| 352 |
+
"step": 315
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 14.222222222222221,
|
| 356 |
+
"grad_norm": 37.48932647705078,
|
| 357 |
+
"learning_rate": 0.0008787878787878789,
|
| 358 |
+
"loss": 1.2026,
|
| 359 |
+
"step": 320
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 14.666666666666666,
|
| 363 |
+
"grad_norm": 4.4914960861206055,
|
| 364 |
+
"learning_rate": 0.000875,
|
| 365 |
+
"loss": 1.2174,
|
| 366 |
+
"step": 330
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 14.977777777777778,
|
| 370 |
+
"eval_accuracy": 0.30434782608695654,
|
| 371 |
+
"eval_loss": 1.34762704372406,
|
| 372 |
+
"eval_runtime": 1.1749,
|
| 373 |
+
"eval_samples_per_second": 39.152,
|
| 374 |
+
"eval_steps_per_second": 2.553,
|
| 375 |
+
"step": 337
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.11111111111111,
|
| 379 |
+
"grad_norm": 3.3884503841400146,
|
| 380 |
+
"learning_rate": 0.0008712121212121212,
|
| 381 |
+
"loss": 1.1947,
|
| 382 |
+
"step": 340
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 15.555555555555555,
|
| 386 |
+
"grad_norm": 3.722648859024048,
|
| 387 |
+
"learning_rate": 0.0008674242424242425,
|
| 388 |
+
"loss": 1.2054,
|
| 389 |
+
"step": 350
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 16.0,
|
| 393 |
+
"grad_norm": 15.185981750488281,
|
| 394 |
+
"learning_rate": 0.0008636363636363636,
|
| 395 |
+
"loss": 1.1767,
|
| 396 |
+
"step": 360
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 16.0,
|
| 400 |
+
"eval_accuracy": 0.1956521739130435,
|
| 401 |
+
"eval_loss": 1.4023534059524536,
|
| 402 |
+
"eval_runtime": 0.8914,
|
| 403 |
+
"eval_samples_per_second": 51.601,
|
| 404 |
+
"eval_steps_per_second": 3.365,
|
| 405 |
+
"step": 360
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 16.444444444444443,
|
| 409 |
+
"grad_norm": 4.847630500793457,
|
| 410 |
+
"learning_rate": 0.0008598484848484849,
|
| 411 |
+
"loss": 1.1721,
|
| 412 |
+
"step": 370
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 16.88888888888889,
|
| 416 |
+
"grad_norm": 3.731781482696533,
|
| 417 |
+
"learning_rate": 0.0008560606060606061,
|
| 418 |
+
"loss": 1.2067,
|
| 419 |
+
"step": 380
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 16.977777777777778,
|
| 423 |
+
"eval_accuracy": 0.17391304347826086,
|
| 424 |
+
"eval_loss": 1.3664109706878662,
|
| 425 |
+
"eval_runtime": 0.8715,
|
| 426 |
+
"eval_samples_per_second": 52.783,
|
| 427 |
+
"eval_steps_per_second": 3.442,
|
| 428 |
+
"step": 382
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 17.333333333333332,
|
| 432 |
+
"grad_norm": 3.2619216442108154,
|
| 433 |
+
"learning_rate": 0.0008522727272727273,
|
| 434 |
+
"loss": 1.1866,
|
| 435 |
+
"step": 390
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 17.77777777777778,
|
| 439 |
+
"grad_norm": 11.475706100463867,
|
| 440 |
+
"learning_rate": 0.0008484848484848485,
|
| 441 |
+
"loss": 1.2303,
|
| 442 |
+
"step": 400
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 18.0,
|
| 446 |
+
"eval_accuracy": 0.2826086956521739,
|
| 447 |
+
"eval_loss": 1.4259557723999023,
|
| 448 |
+
"eval_runtime": 0.8714,
|
| 449 |
+
"eval_samples_per_second": 52.786,
|
| 450 |
+
"eval_steps_per_second": 3.443,
|
| 451 |
+
"step": 405
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 18.22222222222222,
|
| 455 |
+
"grad_norm": 18.98031997680664,
|
| 456 |
+
"learning_rate": 0.0008446969696969698,
|
| 457 |
+
"loss": 1.1939,
|
| 458 |
+
"step": 410
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 18.666666666666668,
|
| 462 |
+
"grad_norm": 2.759476900100708,
|
| 463 |
+
"learning_rate": 0.000840909090909091,
|
| 464 |
+
"loss": 1.222,
|
| 465 |
+
"step": 420
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 18.977777777777778,
|
| 469 |
+
"eval_accuracy": 0.17391304347826086,
|
| 470 |
+
"eval_loss": 1.480705976486206,
|
| 471 |
+
"eval_runtime": 1.1937,
|
| 472 |
+
"eval_samples_per_second": 38.535,
|
| 473 |
+
"eval_steps_per_second": 2.513,
|
| 474 |
+
"step": 427
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 19.11111111111111,
|
| 478 |
+
"grad_norm": 4.408325672149658,
|
| 479 |
+
"learning_rate": 0.0008371212121212122,
|
| 480 |
+
"loss": 1.1725,
|
| 481 |
+
"step": 430
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 19.555555555555557,
|
| 485 |
+
"grad_norm": 2.795828104019165,
|
| 486 |
+
"learning_rate": 0.0008333333333333334,
|
| 487 |
+
"loss": 1.1577,
|
| 488 |
+
"step": 440
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 20.0,
|
| 492 |
+
"grad_norm": 8.652217864990234,
|
| 493 |
+
"learning_rate": 0.0008295454545454546,
|
| 494 |
+
"loss": 1.2026,
|
| 495 |
+
"step": 450
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 20.0,
|
| 499 |
+
"eval_accuracy": 0.17391304347826086,
|
| 500 |
+
"eval_loss": 1.3851475715637207,
|
| 501 |
+
"eval_runtime": 0.8768,
|
| 502 |
+
"eval_samples_per_second": 52.465,
|
| 503 |
+
"eval_steps_per_second": 3.422,
|
| 504 |
+
"step": 450
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 20.444444444444443,
|
| 508 |
+
"grad_norm": 4.27825927734375,
|
| 509 |
+
"learning_rate": 0.0008257575757575758,
|
| 510 |
+
"loss": 1.1934,
|
| 511 |
+
"step": 460
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 20.88888888888889,
|
| 515 |
+
"grad_norm": 3.146332263946533,
|
| 516 |
+
"learning_rate": 0.000821969696969697,
|
| 517 |
+
"loss": 1.2185,
|
| 518 |
+
"step": 470
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 20.977777777777778,
|
| 522 |
+
"eval_accuracy": 0.2608695652173913,
|
| 523 |
+
"eval_loss": 1.3213552236557007,
|
| 524 |
+
"eval_runtime": 0.889,
|
| 525 |
+
"eval_samples_per_second": 51.741,
|
| 526 |
+
"eval_steps_per_second": 3.374,
|
| 527 |
+
"step": 472
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 21.333333333333332,
|
| 531 |
+
"grad_norm": 6.067198753356934,
|
| 532 |
+
"learning_rate": 0.0008181818181818183,
|
| 533 |
+
"loss": 1.2416,
|
| 534 |
+
"step": 480
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 21.77777777777778,
|
| 538 |
+
"grad_norm": 1.9041274785995483,
|
| 539 |
+
"learning_rate": 0.0008143939393939394,
|
| 540 |
+
"loss": 1.2773,
|
| 541 |
+
"step": 490
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 22.0,
|
| 545 |
+
"eval_accuracy": 0.1956521739130435,
|
| 546 |
+
"eval_loss": 1.4403716325759888,
|
| 547 |
+
"eval_runtime": 0.9305,
|
| 548 |
+
"eval_samples_per_second": 49.436,
|
| 549 |
+
"eval_steps_per_second": 3.224,
|
| 550 |
+
"step": 495
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 22.22222222222222,
|
| 554 |
+
"grad_norm": 3.5751218795776367,
|
| 555 |
+
"learning_rate": 0.0008106060606060606,
|
| 556 |
+
"loss": 1.2544,
|
| 557 |
+
"step": 500
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 22.666666666666668,
|
| 561 |
+
"grad_norm": 6.662022590637207,
|
| 562 |
+
"learning_rate": 0.0008068181818181818,
|
| 563 |
+
"loss": 1.227,
|
| 564 |
+
"step": 510
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 22.977777777777778,
|
| 568 |
+
"eval_accuracy": 0.2391304347826087,
|
| 569 |
+
"eval_loss": 1.453503966331482,
|
| 570 |
+
"eval_runtime": 1.2139,
|
| 571 |
+
"eval_samples_per_second": 37.894,
|
| 572 |
+
"eval_steps_per_second": 2.471,
|
| 573 |
+
"step": 517
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 23.11111111111111,
|
| 577 |
+
"grad_norm": 2.344249725341797,
|
| 578 |
+
"learning_rate": 0.000803030303030303,
|
| 579 |
+
"loss": 1.1912,
|
| 580 |
+
"step": 520
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 23.555555555555557,
|
| 584 |
+
"grad_norm": 5.093134880065918,
|
| 585 |
+
"learning_rate": 0.0007992424242424242,
|
| 586 |
+
"loss": 1.211,
|
| 587 |
+
"step": 530
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 24.0,
|
| 591 |
+
"grad_norm": 3.9864282608032227,
|
| 592 |
+
"learning_rate": 0.0007954545454545455,
|
| 593 |
+
"loss": 1.2032,
|
| 594 |
+
"step": 540
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 24.0,
|
| 598 |
+
"eval_accuracy": 0.30434782608695654,
|
| 599 |
+
"eval_loss": 1.3966683149337769,
|
| 600 |
+
"eval_runtime": 0.8732,
|
| 601 |
+
"eval_samples_per_second": 52.682,
|
| 602 |
+
"eval_steps_per_second": 3.436,
|
| 603 |
+
"step": 540
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 24.444444444444443,
|
| 607 |
+
"grad_norm": 5.3367438316345215,
|
| 608 |
+
"learning_rate": 0.0007916666666666666,
|
| 609 |
+
"loss": 1.2051,
|
| 610 |
+
"step": 550
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 24.88888888888889,
|
| 614 |
+
"grad_norm": 3.129544973373413,
|
| 615 |
+
"learning_rate": 0.0007878787878787878,
|
| 616 |
+
"loss": 1.2223,
|
| 617 |
+
"step": 560
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 24.977777777777778,
|
| 621 |
+
"eval_accuracy": 0.32608695652173914,
|
| 622 |
+
"eval_loss": 1.408994436264038,
|
| 623 |
+
"eval_runtime": 0.8542,
|
| 624 |
+
"eval_samples_per_second": 53.849,
|
| 625 |
+
"eval_steps_per_second": 3.512,
|
| 626 |
+
"step": 562
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 25.333333333333332,
|
| 630 |
+
"grad_norm": 2.691143751144409,
|
| 631 |
+
"learning_rate": 0.0007840909090909091,
|
| 632 |
+
"loss": 1.2041,
|
| 633 |
+
"step": 570
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 25.77777777777778,
|
| 637 |
+
"grad_norm": 9.64910888671875,
|
| 638 |
+
"learning_rate": 0.0007803030303030303,
|
| 639 |
+
"loss": 1.2527,
|
| 640 |
+
"step": 580
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 26.0,
|
| 644 |
+
"eval_accuracy": 0.2608695652173913,
|
| 645 |
+
"eval_loss": 1.4858431816101074,
|
| 646 |
+
"eval_runtime": 0.9084,
|
| 647 |
+
"eval_samples_per_second": 50.639,
|
| 648 |
+
"eval_steps_per_second": 3.303,
|
| 649 |
+
"step": 585
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 26.22222222222222,
|
| 653 |
+
"grad_norm": 2.8862998485565186,
|
| 654 |
+
"learning_rate": 0.0007765151515151515,
|
| 655 |
+
"loss": 1.1968,
|
| 656 |
+
"step": 590
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 26.666666666666668,
|
| 660 |
+
"grad_norm": 2.620485544204712,
|
| 661 |
+
"learning_rate": 0.0007727272727272727,
|
| 662 |
+
"loss": 1.2203,
|
| 663 |
+
"step": 600
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 26.977777777777778,
|
| 667 |
+
"eval_accuracy": 0.17391304347826086,
|
| 668 |
+
"eval_loss": 1.4366178512573242,
|
| 669 |
+
"eval_runtime": 0.888,
|
| 670 |
+
"eval_samples_per_second": 51.8,
|
| 671 |
+
"eval_steps_per_second": 3.378,
|
| 672 |
+
"step": 607
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 27.11111111111111,
|
| 676 |
+
"grad_norm": 1.672736406326294,
|
| 677 |
+
"learning_rate": 0.000768939393939394,
|
| 678 |
+
"loss": 1.182,
|
| 679 |
+
"step": 610
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 27.555555555555557,
|
| 683 |
+
"grad_norm": 4.047964096069336,
|
| 684 |
+
"learning_rate": 0.0007651515151515151,
|
| 685 |
+
"loss": 1.1824,
|
| 686 |
+
"step": 620
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 28.0,
|
| 690 |
+
"grad_norm": 7.370471477508545,
|
| 691 |
+
"learning_rate": 0.0007613636363636364,
|
| 692 |
+
"loss": 1.1993,
|
| 693 |
+
"step": 630
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 28.0,
|
| 697 |
+
"eval_accuracy": 0.2608695652173913,
|
| 698 |
+
"eval_loss": 1.4055887460708618,
|
| 699 |
+
"eval_runtime": 0.8703,
|
| 700 |
+
"eval_samples_per_second": 52.856,
|
| 701 |
+
"eval_steps_per_second": 3.447,
|
| 702 |
+
"step": 630
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 28.444444444444443,
|
| 706 |
+
"grad_norm": 2.3139352798461914,
|
| 707 |
+
"learning_rate": 0.0007575757575757576,
|
| 708 |
+
"loss": 1.1908,
|
| 709 |
+
"step": 640
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 28.88888888888889,
|
| 713 |
+
"grad_norm": 2.4274730682373047,
|
| 714 |
+
"learning_rate": 0.0007537878787878788,
|
| 715 |
+
"loss": 1.2014,
|
| 716 |
+
"step": 650
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 28.977777777777778,
|
| 720 |
+
"eval_accuracy": 0.30434782608695654,
|
| 721 |
+
"eval_loss": 1.3755403757095337,
|
| 722 |
+
"eval_runtime": 0.8706,
|
| 723 |
+
"eval_samples_per_second": 52.837,
|
| 724 |
+
"eval_steps_per_second": 3.446,
|
| 725 |
+
"step": 652
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 29.333333333333332,
|
| 729 |
+
"grad_norm": 2.012920618057251,
|
| 730 |
+
"learning_rate": 0.00075,
|
| 731 |
+
"loss": 1.2134,
|
| 732 |
+
"step": 660
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 29.77777777777778,
|
| 736 |
+
"grad_norm": 3.179375171661377,
|
| 737 |
+
"learning_rate": 0.0007462121212121212,
|
| 738 |
+
"loss": 1.2027,
|
| 739 |
+
"step": 670
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 30.0,
|
| 743 |
+
"eval_accuracy": 0.2608695652173913,
|
| 744 |
+
"eval_loss": 1.457945466041565,
|
| 745 |
+
"eval_runtime": 1.1362,
|
| 746 |
+
"eval_samples_per_second": 40.485,
|
| 747 |
+
"eval_steps_per_second": 2.64,
|
| 748 |
+
"step": 675
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 30.22222222222222,
|
| 752 |
+
"grad_norm": 3.6860008239746094,
|
| 753 |
+
"learning_rate": 0.0007424242424242425,
|
| 754 |
+
"loss": 1.2086,
|
| 755 |
+
"step": 680
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 30.666666666666668,
|
| 759 |
+
"grad_norm": 1.5310014486312866,
|
| 760 |
+
"learning_rate": 0.0007386363636363636,
|
| 761 |
+
"loss": 1.1961,
|
| 762 |
+
"step": 690
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 30.977777777777778,
|
| 766 |
+
"eval_accuracy": 0.2608695652173913,
|
| 767 |
+
"eval_loss": 1.4524133205413818,
|
| 768 |
+
"eval_runtime": 0.9718,
|
| 769 |
+
"eval_samples_per_second": 47.335,
|
| 770 |
+
"eval_steps_per_second": 3.087,
|
| 771 |
+
"step": 697
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 31.11111111111111,
|
| 775 |
+
"grad_norm": 24.489850997924805,
|
| 776 |
+
"learning_rate": 0.0007348484848484849,
|
| 777 |
+
"loss": 1.2059,
|
| 778 |
+
"step": 700
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 31.555555555555557,
|
| 782 |
+
"grad_norm": 2.539658308029175,
|
| 783 |
+
"learning_rate": 0.0007310606060606061,
|
| 784 |
+
"loss": 1.1874,
|
| 785 |
+
"step": 710
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 32.0,
|
| 789 |
+
"grad_norm": 2.7552974224090576,
|
| 790 |
+
"learning_rate": 0.0007272727272727273,
|
| 791 |
+
"loss": 1.1939,
|
| 792 |
+
"step": 720
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 32.0,
|
| 796 |
+
"eval_accuracy": 0.2391304347826087,
|
| 797 |
+
"eval_loss": 1.448819875717163,
|
| 798 |
+
"eval_runtime": 0.8862,
|
| 799 |
+
"eval_samples_per_second": 51.905,
|
| 800 |
+
"eval_steps_per_second": 3.385,
|
| 801 |
+
"step": 720
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 32.44444444444444,
|
| 805 |
+
"grad_norm": 3.2986652851104736,
|
| 806 |
+
"learning_rate": 0.0007234848484848485,
|
| 807 |
+
"loss": 1.2196,
|
| 808 |
+
"step": 730
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 32.888888888888886,
|
| 812 |
+
"grad_norm": 2.5736711025238037,
|
| 813 |
+
"learning_rate": 0.0007196969696969698,
|
| 814 |
+
"loss": 1.1889,
|
| 815 |
+
"step": 740
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 32.977777777777774,
|
| 819 |
+
"eval_accuracy": 0.15217391304347827,
|
| 820 |
+
"eval_loss": 1.456831693649292,
|
| 821 |
+
"eval_runtime": 0.8853,
|
| 822 |
+
"eval_samples_per_second": 51.958,
|
| 823 |
+
"eval_steps_per_second": 3.389,
|
| 824 |
+
"step": 742
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 33.333333333333336,
|
| 828 |
+
"grad_norm": 5.722994327545166,
|
| 829 |
+
"learning_rate": 0.0007159090909090909,
|
| 830 |
+
"loss": 1.178,
|
| 831 |
+
"step": 750
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 33.77777777777778,
|
| 835 |
+
"grad_norm": 1.9777653217315674,
|
| 836 |
+
"learning_rate": 0.0007121212121212122,
|
| 837 |
+
"loss": 1.1871,
|
| 838 |
+
"step": 760
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 34.0,
|
| 842 |
+
"eval_accuracy": 0.32608695652173914,
|
| 843 |
+
"eval_loss": 1.3814184665679932,
|
| 844 |
+
"eval_runtime": 0.8797,
|
| 845 |
+
"eval_samples_per_second": 52.292,
|
| 846 |
+
"eval_steps_per_second": 3.41,
|
| 847 |
+
"step": 765
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 34.22222222222222,
|
| 851 |
+
"grad_norm": 2.5516679286956787,
|
| 852 |
+
"learning_rate": 0.0007083333333333334,
|
| 853 |
+
"loss": 1.2329,
|
| 854 |
+
"step": 770
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 34.666666666666664,
|
| 858 |
+
"grad_norm": 3.3046047687530518,
|
| 859 |
+
"learning_rate": 0.0007045454545454546,
|
| 860 |
+
"loss": 1.1778,
|
| 861 |
+
"step": 780
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 34.977777777777774,
|
| 865 |
+
"eval_accuracy": 0.13043478260869565,
|
| 866 |
+
"eval_loss": 1.44027578830719,
|
| 867 |
+
"eval_runtime": 1.196,
|
| 868 |
+
"eval_samples_per_second": 38.461,
|
| 869 |
+
"eval_steps_per_second": 2.508,
|
| 870 |
+
"step": 787
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 35.111111111111114,
|
| 874 |
+
"grad_norm": 1.8858623504638672,
|
| 875 |
+
"learning_rate": 0.0007007575757575758,
|
| 876 |
+
"loss": 1.2006,
|
| 877 |
+
"step": 790
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 35.55555555555556,
|
| 881 |
+
"grad_norm": 5.07556676864624,
|
| 882 |
+
"learning_rate": 0.000696969696969697,
|
| 883 |
+
"loss": 1.2925,
|
| 884 |
+
"step": 800
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 36.0,
|
| 888 |
+
"grad_norm": 5.501034259796143,
|
| 889 |
+
"learning_rate": 0.0006931818181818183,
|
| 890 |
+
"loss": 1.2404,
|
| 891 |
+
"step": 810
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 36.0,
|
| 895 |
+
"eval_accuracy": 0.1956521739130435,
|
| 896 |
+
"eval_loss": 1.4436728954315186,
|
| 897 |
+
"eval_runtime": 0.9177,
|
| 898 |
+
"eval_samples_per_second": 50.125,
|
| 899 |
+
"eval_steps_per_second": 3.269,
|
| 900 |
+
"step": 810
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 36.44444444444444,
|
| 904 |
+
"grad_norm": 2.563979387283325,
|
| 905 |
+
"learning_rate": 0.0006893939393939394,
|
| 906 |
+
"loss": 1.222,
|
| 907 |
+
"step": 820
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 36.888888888888886,
|
| 911 |
+
"grad_norm": 2.562971591949463,
|
| 912 |
+
"learning_rate": 0.0006856060606060606,
|
| 913 |
+
"loss": 1.197,
|
| 914 |
+
"step": 830
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 36.977777777777774,
|
| 918 |
+
"eval_accuracy": 0.21739130434782608,
|
| 919 |
+
"eval_loss": 1.476518988609314,
|
| 920 |
+
"eval_runtime": 0.8797,
|
| 921 |
+
"eval_samples_per_second": 52.292,
|
| 922 |
+
"eval_steps_per_second": 3.41,
|
| 923 |
+
"step": 832
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 37.333333333333336,
|
| 927 |
+
"grad_norm": 4.266129016876221,
|
| 928 |
+
"learning_rate": 0.0006818181818181818,
|
| 929 |
+
"loss": 1.1733,
|
| 930 |
+
"step": 840
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 37.77777777777778,
|
| 934 |
+
"grad_norm": 2.0012102127075195,
|
| 935 |
+
"learning_rate": 0.000678030303030303,
|
| 936 |
+
"loss": 1.2161,
|
| 937 |
+
"step": 850
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 38.0,
|
| 941 |
+
"eval_accuracy": 0.2391304347826087,
|
| 942 |
+
"eval_loss": 1.3720359802246094,
|
| 943 |
+
"eval_runtime": 0.8896,
|
| 944 |
+
"eval_samples_per_second": 51.708,
|
| 945 |
+
"eval_steps_per_second": 3.372,
|
| 946 |
+
"step": 855
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 38.22222222222222,
|
| 950 |
+
"grad_norm": 5.82732629776001,
|
| 951 |
+
"learning_rate": 0.0006742424242424242,
|
| 952 |
+
"loss": 1.194,
|
| 953 |
+
"step": 860
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 38.666666666666664,
|
| 957 |
+
"grad_norm": 3.146446466445923,
|
| 958 |
+
"learning_rate": 0.0006704545454545455,
|
| 959 |
+
"loss": 1.221,
|
| 960 |
+
"step": 870
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 38.977777777777774,
|
| 964 |
+
"eval_accuracy": 0.34782608695652173,
|
| 965 |
+
"eval_loss": 1.3749516010284424,
|
| 966 |
+
"eval_runtime": 1.1844,
|
| 967 |
+
"eval_samples_per_second": 38.839,
|
| 968 |
+
"eval_steps_per_second": 2.533,
|
| 969 |
+
"step": 877
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 39.111111111111114,
|
| 973 |
+
"grad_norm": 2.4704267978668213,
|
| 974 |
+
"learning_rate": 0.0006666666666666666,
|
| 975 |
+
"loss": 1.2082,
|
| 976 |
+
"step": 880
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 39.55555555555556,
|
| 980 |
+
"grad_norm": 3.377894878387451,
|
| 981 |
+
"learning_rate": 0.0006628787878787878,
|
| 982 |
+
"loss": 1.2203,
|
| 983 |
+
"step": 890
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 40.0,
|
| 987 |
+
"grad_norm": 3.2576770782470703,
|
| 988 |
+
"learning_rate": 0.0006590909090909091,
|
| 989 |
+
"loss": 1.229,
|
| 990 |
+
"step": 900
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 40.0,
|
| 994 |
+
"eval_accuracy": 0.2391304347826087,
|
| 995 |
+
"eval_loss": 1.3404773473739624,
|
| 996 |
+
"eval_runtime": 0.876,
|
| 997 |
+
"eval_samples_per_second": 52.511,
|
| 998 |
+
"eval_steps_per_second": 3.425,
|
| 999 |
+
"step": 900
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 40.44444444444444,
|
| 1003 |
+
"grad_norm": 2.0540192127227783,
|
| 1004 |
+
"learning_rate": 0.0006553030303030303,
|
| 1005 |
+
"loss": 1.2222,
|
| 1006 |
+
"step": 910
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 40.888888888888886,
|
| 1010 |
+
"grad_norm": 2.336094617843628,
|
| 1011 |
+
"learning_rate": 0.0006515151515151515,
|
| 1012 |
+
"loss": 1.2046,
|
| 1013 |
+
"step": 920
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 40.977777777777774,
|
| 1017 |
+
"eval_accuracy": 0.2608695652173913,
|
| 1018 |
+
"eval_loss": 1.4231040477752686,
|
| 1019 |
+
"eval_runtime": 0.899,
|
| 1020 |
+
"eval_samples_per_second": 51.168,
|
| 1021 |
+
"eval_steps_per_second": 3.337,
|
| 1022 |
+
"step": 922
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 41.333333333333336,
|
| 1026 |
+
"grad_norm": 3.401913642883301,
|
| 1027 |
+
"learning_rate": 0.0006477272727272727,
|
| 1028 |
+
"loss": 1.2028,
|
| 1029 |
+
"step": 930
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 41.77777777777778,
|
| 1033 |
+
"grad_norm": 1.3715537786483765,
|
| 1034 |
+
"learning_rate": 0.000643939393939394,
|
| 1035 |
+
"loss": 1.2077,
|
| 1036 |
+
"step": 940
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 42.0,
|
| 1040 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1041 |
+
"eval_loss": 1.4383732080459595,
|
| 1042 |
+
"eval_runtime": 1.0424,
|
| 1043 |
+
"eval_samples_per_second": 44.128,
|
| 1044 |
+
"eval_steps_per_second": 2.878,
|
| 1045 |
+
"step": 945
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 42.22222222222222,
|
| 1049 |
+
"grad_norm": 3.548417806625366,
|
| 1050 |
+
"learning_rate": 0.0006401515151515151,
|
| 1051 |
+
"loss": 1.2126,
|
| 1052 |
+
"step": 950
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 42.666666666666664,
|
| 1056 |
+
"grad_norm": 15.91876220703125,
|
| 1057 |
+
"learning_rate": 0.0006363636363636364,
|
| 1058 |
+
"loss": 1.1865,
|
| 1059 |
+
"step": 960
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 42.977777777777774,
|
| 1063 |
+
"eval_accuracy": 0.2608695652173913,
|
| 1064 |
+
"eval_loss": 1.4346174001693726,
|
| 1065 |
+
"eval_runtime": 0.9529,
|
| 1066 |
+
"eval_samples_per_second": 48.275,
|
| 1067 |
+
"eval_steps_per_second": 3.148,
|
| 1068 |
+
"step": 967
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 43.111111111111114,
|
| 1072 |
+
"grad_norm": 3.9566547870635986,
|
| 1073 |
+
"learning_rate": 0.0006325757575757576,
|
| 1074 |
+
"loss": 1.175,
|
| 1075 |
+
"step": 970
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 43.55555555555556,
|
| 1079 |
+
"grad_norm": 2.891098976135254,
|
| 1080 |
+
"learning_rate": 0.0006287878787878788,
|
| 1081 |
+
"loss": 1.2044,
|
| 1082 |
+
"step": 980
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 44.0,
|
| 1086 |
+
"grad_norm": 1.5360569953918457,
|
| 1087 |
+
"learning_rate": 0.000625,
|
| 1088 |
+
"loss": 1.1882,
|
| 1089 |
+
"step": 990
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 44.0,
|
| 1093 |
+
"eval_accuracy": 0.2826086956521739,
|
| 1094 |
+
"eval_loss": 1.367881417274475,
|
| 1095 |
+
"eval_runtime": 0.8863,
|
| 1096 |
+
"eval_samples_per_second": 51.903,
|
| 1097 |
+
"eval_steps_per_second": 3.385,
|
| 1098 |
+
"step": 990
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 44.44444444444444,
|
| 1102 |
+
"grad_norm": 2.5285489559173584,
|
| 1103 |
+
"learning_rate": 0.0006212121212121212,
|
| 1104 |
+
"loss": 1.1957,
|
| 1105 |
+
"step": 1000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 44.888888888888886,
|
| 1109 |
+
"grad_norm": 4.244635105133057,
|
| 1110 |
+
"learning_rate": 0.0006174242424242425,
|
| 1111 |
+
"loss": 1.2528,
|
| 1112 |
+
"step": 1010
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 44.977777777777774,
|
| 1116 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1117 |
+
"eval_loss": 1.3451467752456665,
|
| 1118 |
+
"eval_runtime": 0.8705,
|
| 1119 |
+
"eval_samples_per_second": 52.845,
|
| 1120 |
+
"eval_steps_per_second": 3.446,
|
| 1121 |
+
"step": 1012
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 45.333333333333336,
|
| 1125 |
+
"grad_norm": 1.46609365940094,
|
| 1126 |
+
"learning_rate": 0.0006136363636363636,
|
| 1127 |
+
"loss": 1.2534,
|
| 1128 |
+
"step": 1020
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 45.77777777777778,
|
| 1132 |
+
"grad_norm": 9.328937530517578,
|
| 1133 |
+
"learning_rate": 0.0006098484848484849,
|
| 1134 |
+
"loss": 1.1836,
|
| 1135 |
+
"step": 1030
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 46.0,
|
| 1139 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1140 |
+
"eval_loss": 1.4912604093551636,
|
| 1141 |
+
"eval_runtime": 0.8907,
|
| 1142 |
+
"eval_samples_per_second": 51.643,
|
| 1143 |
+
"eval_steps_per_second": 3.368,
|
| 1144 |
+
"step": 1035
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 46.22222222222222,
|
| 1148 |
+
"grad_norm": 2.901005744934082,
|
| 1149 |
+
"learning_rate": 0.0006060606060606061,
|
| 1150 |
+
"loss": 1.2226,
|
| 1151 |
+
"step": 1040
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 46.666666666666664,
|
| 1155 |
+
"grad_norm": 4.646663188934326,
|
| 1156 |
+
"learning_rate": 0.0006022727272727273,
|
| 1157 |
+
"loss": 1.2009,
|
| 1158 |
+
"step": 1050
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 46.977777777777774,
|
| 1162 |
+
"eval_accuracy": 0.32608695652173914,
|
| 1163 |
+
"eval_loss": 1.4841315746307373,
|
| 1164 |
+
"eval_runtime": 1.1901,
|
| 1165 |
+
"eval_samples_per_second": 38.652,
|
| 1166 |
+
"eval_steps_per_second": 2.521,
|
| 1167 |
+
"step": 1057
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 47.111111111111114,
|
| 1171 |
+
"grad_norm": 2.45768141746521,
|
| 1172 |
+
"learning_rate": 0.0005984848484848485,
|
| 1173 |
+
"loss": 1.219,
|
| 1174 |
+
"step": 1060
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 47.55555555555556,
|
| 1178 |
+
"grad_norm": 5.484715938568115,
|
| 1179 |
+
"learning_rate": 0.0005946969696969698,
|
| 1180 |
+
"loss": 1.2043,
|
| 1181 |
+
"step": 1070
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 48.0,
|
| 1185 |
+
"grad_norm": 62.28512954711914,
|
| 1186 |
+
"learning_rate": 0.0005909090909090909,
|
| 1187 |
+
"loss": 1.203,
|
| 1188 |
+
"step": 1080
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 48.0,
|
| 1192 |
+
"eval_accuracy": 0.30434782608695654,
|
| 1193 |
+
"eval_loss": 1.4326200485229492,
|
| 1194 |
+
"eval_runtime": 0.8718,
|
| 1195 |
+
"eval_samples_per_second": 52.767,
|
| 1196 |
+
"eval_steps_per_second": 3.441,
|
| 1197 |
+
"step": 1080
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 48.44444444444444,
|
| 1201 |
+
"grad_norm": 2.838622808456421,
|
| 1202 |
+
"learning_rate": 0.0005871212121212122,
|
| 1203 |
+
"loss": 1.2071,
|
| 1204 |
+
"step": 1090
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 48.888888888888886,
|
| 1208 |
+
"grad_norm": 4.207474708557129,
|
| 1209 |
+
"learning_rate": 0.0005833333333333334,
|
| 1210 |
+
"loss": 1.1679,
|
| 1211 |
+
"step": 1100
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 48.977777777777774,
|
| 1215 |
+
"eval_accuracy": 0.30434782608695654,
|
| 1216 |
+
"eval_loss": 1.3934518098831177,
|
| 1217 |
+
"eval_runtime": 0.857,
|
| 1218 |
+
"eval_samples_per_second": 53.673,
|
| 1219 |
+
"eval_steps_per_second": 3.5,
|
| 1220 |
+
"step": 1102
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 49.333333333333336,
|
| 1224 |
+
"grad_norm": 2.942615270614624,
|
| 1225 |
+
"learning_rate": 0.0005795454545454545,
|
| 1226 |
+
"loss": 1.1719,
|
| 1227 |
+
"step": 1110
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 49.77777777777778,
|
| 1231 |
+
"grad_norm": 54.94511032104492,
|
| 1232 |
+
"learning_rate": 0.0005757575757575758,
|
| 1233 |
+
"loss": 1.179,
|
| 1234 |
+
"step": 1120
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 50.0,
|
| 1238 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1239 |
+
"eval_loss": 1.4185277223587036,
|
| 1240 |
+
"eval_runtime": 0.8854,
|
| 1241 |
+
"eval_samples_per_second": 51.953,
|
| 1242 |
+
"eval_steps_per_second": 3.388,
|
| 1243 |
+
"step": 1125
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 50.22222222222222,
|
| 1247 |
+
"grad_norm": 11.695096969604492,
|
| 1248 |
+
"learning_rate": 0.000571969696969697,
|
| 1249 |
+
"loss": 1.1624,
|
| 1250 |
+
"step": 1130
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 50.666666666666664,
|
| 1254 |
+
"grad_norm": 1.823878288269043,
|
| 1255 |
+
"learning_rate": 0.0005681818181818183,
|
| 1256 |
+
"loss": 1.1687,
|
| 1257 |
+
"step": 1140
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 50.977777777777774,
|
| 1261 |
+
"eval_accuracy": 0.2826086956521739,
|
| 1262 |
+
"eval_loss": 1.3686347007751465,
|
| 1263 |
+
"eval_runtime": 1.2456,
|
| 1264 |
+
"eval_samples_per_second": 36.929,
|
| 1265 |
+
"eval_steps_per_second": 2.408,
|
| 1266 |
+
"step": 1147
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 51.111111111111114,
|
| 1270 |
+
"grad_norm": 17.766942977905273,
|
| 1271 |
+
"learning_rate": 0.0005643939393939394,
|
| 1272 |
+
"loss": 1.17,
|
| 1273 |
+
"step": 1150
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 51.55555555555556,
|
| 1277 |
+
"grad_norm": 4.572807788848877,
|
| 1278 |
+
"learning_rate": 0.0005606060606060606,
|
| 1279 |
+
"loss": 1.1363,
|
| 1280 |
+
"step": 1160
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 52.0,
|
| 1284 |
+
"grad_norm": 7.582568168640137,
|
| 1285 |
+
"learning_rate": 0.0005568181818181818,
|
| 1286 |
+
"loss": 1.1779,
|
| 1287 |
+
"step": 1170
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 52.0,
|
| 1291 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1292 |
+
"eval_loss": 1.4319127798080444,
|
| 1293 |
+
"eval_runtime": 0.895,
|
| 1294 |
+
"eval_samples_per_second": 51.396,
|
| 1295 |
+
"eval_steps_per_second": 3.352,
|
| 1296 |
+
"step": 1170
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 52.44444444444444,
|
| 1300 |
+
"grad_norm": 2.6896812915802,
|
| 1301 |
+
"learning_rate": 0.000553030303030303,
|
| 1302 |
+
"loss": 1.1698,
|
| 1303 |
+
"step": 1180
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 52.888888888888886,
|
| 1307 |
+
"grad_norm": 3.666240692138672,
|
| 1308 |
+
"learning_rate": 0.0005492424242424242,
|
| 1309 |
+
"loss": 1.1566,
|
| 1310 |
+
"step": 1190
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 52.977777777777774,
|
| 1314 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1315 |
+
"eval_loss": 1.3800519704818726,
|
| 1316 |
+
"eval_runtime": 0.9305,
|
| 1317 |
+
"eval_samples_per_second": 49.434,
|
| 1318 |
+
"eval_steps_per_second": 3.224,
|
| 1319 |
+
"step": 1192
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 53.333333333333336,
|
| 1323 |
+
"grad_norm": 4.159182071685791,
|
| 1324 |
+
"learning_rate": 0.0005454545454545455,
|
| 1325 |
+
"loss": 1.1785,
|
| 1326 |
+
"step": 1200
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 53.77777777777778,
|
| 1330 |
+
"grad_norm": 5.972134590148926,
|
| 1331 |
+
"learning_rate": 0.0005416666666666666,
|
| 1332 |
+
"loss": 1.192,
|
| 1333 |
+
"step": 1210
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 54.0,
|
| 1337 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1338 |
+
"eval_loss": 1.3745651245117188,
|
| 1339 |
+
"eval_runtime": 1.1774,
|
| 1340 |
+
"eval_samples_per_second": 39.07,
|
| 1341 |
+
"eval_steps_per_second": 2.548,
|
| 1342 |
+
"step": 1215
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 54.22222222222222,
|
| 1346 |
+
"grad_norm": 5.2233428955078125,
|
| 1347 |
+
"learning_rate": 0.0005378787878787878,
|
| 1348 |
+
"loss": 1.1768,
|
| 1349 |
+
"step": 1220
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 54.666666666666664,
|
| 1353 |
+
"grad_norm": 5.244997501373291,
|
| 1354 |
+
"learning_rate": 0.0005340909090909091,
|
| 1355 |
+
"loss": 1.1803,
|
| 1356 |
+
"step": 1230
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 54.977777777777774,
|
| 1360 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1361 |
+
"eval_loss": 1.4016964435577393,
|
| 1362 |
+
"eval_runtime": 0.8875,
|
| 1363 |
+
"eval_samples_per_second": 51.83,
|
| 1364 |
+
"eval_steps_per_second": 3.38,
|
| 1365 |
+
"step": 1237
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 55.111111111111114,
|
| 1369 |
+
"grad_norm": 4.16229772567749,
|
| 1370 |
+
"learning_rate": 0.0005303030303030302,
|
| 1371 |
+
"loss": 1.1548,
|
| 1372 |
+
"step": 1240
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 55.55555555555556,
|
| 1376 |
+
"grad_norm": 3.8485047817230225,
|
| 1377 |
+
"learning_rate": 0.0005265151515151515,
|
| 1378 |
+
"loss": 1.1629,
|
| 1379 |
+
"step": 1250
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 56.0,
|
| 1383 |
+
"grad_norm": 3.398857593536377,
|
| 1384 |
+
"learning_rate": 0.0005227272727272727,
|
| 1385 |
+
"loss": 1.194,
|
| 1386 |
+
"step": 1260
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 56.0,
|
| 1390 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1391 |
+
"eval_loss": 1.4288326501846313,
|
| 1392 |
+
"eval_runtime": 0.8835,
|
| 1393 |
+
"eval_samples_per_second": 52.063,
|
| 1394 |
+
"eval_steps_per_second": 3.395,
|
| 1395 |
+
"step": 1260
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 56.44444444444444,
|
| 1399 |
+
"grad_norm": 2.9012982845306396,
|
| 1400 |
+
"learning_rate": 0.000518939393939394,
|
| 1401 |
+
"loss": 1.1283,
|
| 1402 |
+
"step": 1270
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 56.888888888888886,
|
| 1406 |
+
"grad_norm": 2.652462959289551,
|
| 1407 |
+
"learning_rate": 0.0005151515151515151,
|
| 1408 |
+
"loss": 1.1486,
|
| 1409 |
+
"step": 1280
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 56.977777777777774,
|
| 1413 |
+
"eval_accuracy": 0.30434782608695654,
|
| 1414 |
+
"eval_loss": 1.392043113708496,
|
| 1415 |
+
"eval_runtime": 1.1872,
|
| 1416 |
+
"eval_samples_per_second": 38.747,
|
| 1417 |
+
"eval_steps_per_second": 2.527,
|
| 1418 |
+
"step": 1282
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 57.333333333333336,
|
| 1422 |
+
"grad_norm": 16.2806453704834,
|
| 1423 |
+
"learning_rate": 0.0005113636363636364,
|
| 1424 |
+
"loss": 1.154,
|
| 1425 |
+
"step": 1290
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 57.77777777777778,
|
| 1429 |
+
"grad_norm": 2.9445173740386963,
|
| 1430 |
+
"learning_rate": 0.0005075757575757576,
|
| 1431 |
+
"loss": 1.1429,
|
| 1432 |
+
"step": 1300
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 58.0,
|
| 1436 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1437 |
+
"eval_loss": 1.461561918258667,
|
| 1438 |
+
"eval_runtime": 0.8733,
|
| 1439 |
+
"eval_samples_per_second": 52.675,
|
| 1440 |
+
"eval_steps_per_second": 3.435,
|
| 1441 |
+
"step": 1305
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 58.22222222222222,
|
| 1445 |
+
"grad_norm": 4.916953086853027,
|
| 1446 |
+
"learning_rate": 0.0005037878787878788,
|
| 1447 |
+
"loss": 1.1694,
|
| 1448 |
+
"step": 1310
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 58.666666666666664,
|
| 1452 |
+
"grad_norm": 5.632236957550049,
|
| 1453 |
+
"learning_rate": 0.0005,
|
| 1454 |
+
"loss": 1.1655,
|
| 1455 |
+
"step": 1320
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 58.977777777777774,
|
| 1459 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1460 |
+
"eval_loss": 1.4119428396224976,
|
| 1461 |
+
"eval_runtime": 0.8592,
|
| 1462 |
+
"eval_samples_per_second": 53.535,
|
| 1463 |
+
"eval_steps_per_second": 3.491,
|
| 1464 |
+
"step": 1327
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 59.111111111111114,
|
| 1468 |
+
"grad_norm": 10.658681869506836,
|
| 1469 |
+
"learning_rate": 0.0004962121212121212,
|
| 1470 |
+
"loss": 1.148,
|
| 1471 |
+
"step": 1330
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 59.55555555555556,
|
| 1475 |
+
"grad_norm": 4.212299823760986,
|
| 1476 |
+
"learning_rate": 0.0004924242424242425,
|
| 1477 |
+
"loss": 1.1508,
|
| 1478 |
+
"step": 1340
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 60.0,
|
| 1482 |
+
"grad_norm": 4.59652853012085,
|
| 1483 |
+
"learning_rate": 0.0004886363636363636,
|
| 1484 |
+
"loss": 1.1697,
|
| 1485 |
+
"step": 1350
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 60.0,
|
| 1489 |
+
"eval_accuracy": 0.2608695652173913,
|
| 1490 |
+
"eval_loss": 1.3811644315719604,
|
| 1491 |
+
"eval_runtime": 0.9127,
|
| 1492 |
+
"eval_samples_per_second": 50.4,
|
| 1493 |
+
"eval_steps_per_second": 3.287,
|
| 1494 |
+
"step": 1350
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 60.44444444444444,
|
| 1498 |
+
"grad_norm": 3.656782865524292,
|
| 1499 |
+
"learning_rate": 0.0004848484848484849,
|
| 1500 |
+
"loss": 1.1312,
|
| 1501 |
+
"step": 1360
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 60.888888888888886,
|
| 1505 |
+
"grad_norm": 9.660019874572754,
|
| 1506 |
+
"learning_rate": 0.0004810606060606061,
|
| 1507 |
+
"loss": 1.1898,
|
| 1508 |
+
"step": 1370
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 60.977777777777774,
|
| 1512 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1513 |
+
"eval_loss": 1.4008588790893555,
|
| 1514 |
+
"eval_runtime": 1.1831,
|
| 1515 |
+
"eval_samples_per_second": 38.883,
|
| 1516 |
+
"eval_steps_per_second": 2.536,
|
| 1517 |
+
"step": 1372
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 61.333333333333336,
|
| 1521 |
+
"grad_norm": 5.424683570861816,
|
| 1522 |
+
"learning_rate": 0.0004772727272727273,
|
| 1523 |
+
"loss": 1.2188,
|
| 1524 |
+
"step": 1380
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 61.77777777777778,
|
| 1528 |
+
"grad_norm": 3.420642375946045,
|
| 1529 |
+
"learning_rate": 0.0004734848484848485,
|
| 1530 |
+
"loss": 1.1882,
|
| 1531 |
+
"step": 1390
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 62.0,
|
| 1535 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1536 |
+
"eval_loss": 1.422127604484558,
|
| 1537 |
+
"eval_runtime": 1.0461,
|
| 1538 |
+
"eval_samples_per_second": 43.972,
|
| 1539 |
+
"eval_steps_per_second": 2.868,
|
| 1540 |
+
"step": 1395
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 62.22222222222222,
|
| 1544 |
+
"grad_norm": 37.69420623779297,
|
| 1545 |
+
"learning_rate": 0.0004696969696969697,
|
| 1546 |
+
"loss": 1.1428,
|
| 1547 |
+
"step": 1400
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 62.666666666666664,
|
| 1551 |
+
"grad_norm": 6.073638439178467,
|
| 1552 |
+
"learning_rate": 0.0004659090909090909,
|
| 1553 |
+
"loss": 1.134,
|
| 1554 |
+
"step": 1410
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 62.977777777777774,
|
| 1558 |
+
"eval_accuracy": 0.2608695652173913,
|
| 1559 |
+
"eval_loss": 1.618972897529602,
|
| 1560 |
+
"eval_runtime": 0.888,
|
| 1561 |
+
"eval_samples_per_second": 51.801,
|
| 1562 |
+
"eval_steps_per_second": 3.378,
|
| 1563 |
+
"step": 1417
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 63.111111111111114,
|
| 1567 |
+
"grad_norm": 14.145001411437988,
|
| 1568 |
+
"learning_rate": 0.0004621212121212121,
|
| 1569 |
+
"loss": 1.177,
|
| 1570 |
+
"step": 1420
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 63.55555555555556,
|
| 1574 |
+
"grad_norm": 3.415473222732544,
|
| 1575 |
+
"learning_rate": 0.0004583333333333333,
|
| 1576 |
+
"loss": 1.1739,
|
| 1577 |
+
"step": 1430
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 64.0,
|
| 1581 |
+
"grad_norm": 9.372400283813477,
|
| 1582 |
+
"learning_rate": 0.00045454545454545455,
|
| 1583 |
+
"loss": 1.1748,
|
| 1584 |
+
"step": 1440
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 64.0,
|
| 1588 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1589 |
+
"eval_loss": 1.4336298704147339,
|
| 1590 |
+
"eval_runtime": 0.9015,
|
| 1591 |
+
"eval_samples_per_second": 51.025,
|
| 1592 |
+
"eval_steps_per_second": 3.328,
|
| 1593 |
+
"step": 1440
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 64.44444444444444,
|
| 1597 |
+
"grad_norm": 3.4770920276641846,
|
| 1598 |
+
"learning_rate": 0.00045075757575757577,
|
| 1599 |
+
"loss": 1.1419,
|
| 1600 |
+
"step": 1450
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 64.88888888888889,
|
| 1604 |
+
"grad_norm": 17.83055877685547,
|
| 1605 |
+
"learning_rate": 0.000446969696969697,
|
| 1606 |
+
"loss": 1.1439,
|
| 1607 |
+
"step": 1460
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 64.97777777777777,
|
| 1611 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1612 |
+
"eval_loss": 1.3744150400161743,
|
| 1613 |
+
"eval_runtime": 1.1623,
|
| 1614 |
+
"eval_samples_per_second": 39.576,
|
| 1615 |
+
"eval_steps_per_second": 2.581,
|
| 1616 |
+
"step": 1462
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 65.33333333333333,
|
| 1620 |
+
"grad_norm": 5.167716026306152,
|
| 1621 |
+
"learning_rate": 0.0004431818181818182,
|
| 1622 |
+
"loss": 1.1155,
|
| 1623 |
+
"step": 1470
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 65.77777777777777,
|
| 1627 |
+
"grad_norm": 2.334927558898926,
|
| 1628 |
+
"learning_rate": 0.0004393939393939394,
|
| 1629 |
+
"loss": 1.1585,
|
| 1630 |
+
"step": 1480
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 66.0,
|
| 1634 |
+
"eval_accuracy": 0.3695652173913043,
|
| 1635 |
+
"eval_loss": 1.3992067575454712,
|
| 1636 |
+
"eval_runtime": 0.8747,
|
| 1637 |
+
"eval_samples_per_second": 52.591,
|
| 1638 |
+
"eval_steps_per_second": 3.43,
|
| 1639 |
+
"step": 1485
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 66.22222222222223,
|
| 1643 |
+
"grad_norm": 4.668221473693848,
|
| 1644 |
+
"learning_rate": 0.0004356060606060606,
|
| 1645 |
+
"loss": 1.136,
|
| 1646 |
+
"step": 1490
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 66.66666666666667,
|
| 1650 |
+
"grad_norm": 2.9979135990142822,
|
| 1651 |
+
"learning_rate": 0.0004318181818181818,
|
| 1652 |
+
"loss": 1.1344,
|
| 1653 |
+
"step": 1500
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 66.97777777777777,
|
| 1657 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1658 |
+
"eval_loss": 1.3951774835586548,
|
| 1659 |
+
"eval_runtime": 0.8935,
|
| 1660 |
+
"eval_samples_per_second": 51.481,
|
| 1661 |
+
"eval_steps_per_second": 3.357,
|
| 1662 |
+
"step": 1507
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 67.11111111111111,
|
| 1666 |
+
"grad_norm": 2.9879891872406006,
|
| 1667 |
+
"learning_rate": 0.00042803030303030303,
|
| 1668 |
+
"loss": 1.1615,
|
| 1669 |
+
"step": 1510
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 67.55555555555556,
|
| 1673 |
+
"grad_norm": 3.322258710861206,
|
| 1674 |
+
"learning_rate": 0.00042424242424242425,
|
| 1675 |
+
"loss": 1.1635,
|
| 1676 |
+
"step": 1520
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 68.0,
|
| 1680 |
+
"grad_norm": 4.408764362335205,
|
| 1681 |
+
"learning_rate": 0.0004204545454545455,
|
| 1682 |
+
"loss": 1.1374,
|
| 1683 |
+
"step": 1530
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 68.0,
|
| 1687 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1688 |
+
"eval_loss": 1.3666102886199951,
|
| 1689 |
+
"eval_runtime": 0.8532,
|
| 1690 |
+
"eval_samples_per_second": 53.917,
|
| 1691 |
+
"eval_steps_per_second": 3.516,
|
| 1692 |
+
"step": 1530
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 68.44444444444444,
|
| 1696 |
+
"grad_norm": 18.494497299194336,
|
| 1697 |
+
"learning_rate": 0.0004166666666666667,
|
| 1698 |
+
"loss": 1.126,
|
| 1699 |
+
"step": 1540
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 68.88888888888889,
|
| 1703 |
+
"grad_norm": 6.762816905975342,
|
| 1704 |
+
"learning_rate": 0.0004128787878787879,
|
| 1705 |
+
"loss": 1.1252,
|
| 1706 |
+
"step": 1550
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 68.97777777777777,
|
| 1710 |
+
"eval_accuracy": 0.2826086956521739,
|
| 1711 |
+
"eval_loss": 1.3704602718353271,
|
| 1712 |
+
"eval_runtime": 1.2029,
|
| 1713 |
+
"eval_samples_per_second": 38.24,
|
| 1714 |
+
"eval_steps_per_second": 2.494,
|
| 1715 |
+
"step": 1552
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 69.33333333333333,
|
| 1719 |
+
"grad_norm": 4.585610389709473,
|
| 1720 |
+
"learning_rate": 0.00040909090909090913,
|
| 1721 |
+
"loss": 1.1272,
|
| 1722 |
+
"step": 1560
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 69.77777777777777,
|
| 1726 |
+
"grad_norm": 10.724448204040527,
|
| 1727 |
+
"learning_rate": 0.0004053030303030303,
|
| 1728 |
+
"loss": 1.1339,
|
| 1729 |
+
"step": 1570
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 70.0,
|
| 1733 |
+
"eval_accuracy": 0.2826086956521739,
|
| 1734 |
+
"eval_loss": 1.3982820510864258,
|
| 1735 |
+
"eval_runtime": 1.0724,
|
| 1736 |
+
"eval_samples_per_second": 42.893,
|
| 1737 |
+
"eval_steps_per_second": 2.797,
|
| 1738 |
+
"step": 1575
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 70.22222222222223,
|
| 1742 |
+
"grad_norm": 5.506129741668701,
|
| 1743 |
+
"learning_rate": 0.0004015151515151515,
|
| 1744 |
+
"loss": 1.1491,
|
| 1745 |
+
"step": 1580
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 70.66666666666667,
|
| 1749 |
+
"grad_norm": 17.69223976135254,
|
| 1750 |
+
"learning_rate": 0.00039772727272727274,
|
| 1751 |
+
"loss": 1.1344,
|
| 1752 |
+
"step": 1590
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 70.97777777777777,
|
| 1756 |
+
"eval_accuracy": 0.30434782608695654,
|
| 1757 |
+
"eval_loss": 1.3792437314987183,
|
| 1758 |
+
"eval_runtime": 0.8913,
|
| 1759 |
+
"eval_samples_per_second": 51.609,
|
| 1760 |
+
"eval_steps_per_second": 3.366,
|
| 1761 |
+
"step": 1597
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 71.11111111111111,
|
| 1765 |
+
"grad_norm": 10.686148643493652,
|
| 1766 |
+
"learning_rate": 0.0003939393939393939,
|
| 1767 |
+
"loss": 1.1495,
|
| 1768 |
+
"step": 1600
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 71.55555555555556,
|
| 1772 |
+
"grad_norm": 2.353846549987793,
|
| 1773 |
+
"learning_rate": 0.0003901515151515151,
|
| 1774 |
+
"loss": 1.1566,
|
| 1775 |
+
"step": 1610
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 72.0,
|
| 1779 |
+
"grad_norm": 3.250394821166992,
|
| 1780 |
+
"learning_rate": 0.00038636363636363635,
|
| 1781 |
+
"loss": 1.1343,
|
| 1782 |
+
"step": 1620
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 72.0,
|
| 1786 |
+
"eval_accuracy": 0.2826086956521739,
|
| 1787 |
+
"eval_loss": 1.4466689825057983,
|
| 1788 |
+
"eval_runtime": 0.891,
|
| 1789 |
+
"eval_samples_per_second": 51.629,
|
| 1790 |
+
"eval_steps_per_second": 3.367,
|
| 1791 |
+
"step": 1620
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 72.44444444444444,
|
| 1795 |
+
"grad_norm": 6.152634143829346,
|
| 1796 |
+
"learning_rate": 0.00038257575757575757,
|
| 1797 |
+
"loss": 1.1417,
|
| 1798 |
+
"step": 1630
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 72.88888888888889,
|
| 1802 |
+
"grad_norm": 93.5090560913086,
|
| 1803 |
+
"learning_rate": 0.0003787878787878788,
|
| 1804 |
+
"loss": 1.1555,
|
| 1805 |
+
"step": 1640
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 72.97777777777777,
|
| 1809 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1810 |
+
"eval_loss": 1.4822701215744019,
|
| 1811 |
+
"eval_runtime": 1.1509,
|
| 1812 |
+
"eval_samples_per_second": 39.968,
|
| 1813 |
+
"eval_steps_per_second": 2.607,
|
| 1814 |
+
"step": 1642
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 73.33333333333333,
|
| 1818 |
+
"grad_norm": 90.8385238647461,
|
| 1819 |
+
"learning_rate": 0.000375,
|
| 1820 |
+
"loss": 1.1227,
|
| 1821 |
+
"step": 1650
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 73.77777777777777,
|
| 1825 |
+
"grad_norm": 3.00873064994812,
|
| 1826 |
+
"learning_rate": 0.00037121212121212123,
|
| 1827 |
+
"loss": 1.1329,
|
| 1828 |
+
"step": 1660
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 74.0,
|
| 1832 |
+
"eval_accuracy": 0.15217391304347827,
|
| 1833 |
+
"eval_loss": 1.5136324167251587,
|
| 1834 |
+
"eval_runtime": 1.0615,
|
| 1835 |
+
"eval_samples_per_second": 43.334,
|
| 1836 |
+
"eval_steps_per_second": 2.826,
|
| 1837 |
+
"step": 1665
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 74.22222222222223,
|
| 1841 |
+
"grad_norm": 9.816000938415527,
|
| 1842 |
+
"learning_rate": 0.00036742424242424245,
|
| 1843 |
+
"loss": 1.1719,
|
| 1844 |
+
"step": 1670
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 74.66666666666667,
|
| 1848 |
+
"grad_norm": 2.1428558826446533,
|
| 1849 |
+
"learning_rate": 0.00036363636363636367,
|
| 1850 |
+
"loss": 1.1513,
|
| 1851 |
+
"step": 1680
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 74.97777777777777,
|
| 1855 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1856 |
+
"eval_loss": 1.479099988937378,
|
| 1857 |
+
"eval_runtime": 0.8852,
|
| 1858 |
+
"eval_samples_per_second": 51.967,
|
| 1859 |
+
"eval_steps_per_second": 3.389,
|
| 1860 |
+
"step": 1687
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 75.11111111111111,
|
| 1864 |
+
"grad_norm": 4.258810997009277,
|
| 1865 |
+
"learning_rate": 0.0003598484848484849,
|
| 1866 |
+
"loss": 1.1449,
|
| 1867 |
+
"step": 1690
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"epoch": 75.55555555555556,
|
| 1871 |
+
"grad_norm": 3.7126126289367676,
|
| 1872 |
+
"learning_rate": 0.0003560606060606061,
|
| 1873 |
+
"loss": 1.1289,
|
| 1874 |
+
"step": 1700
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 76.0,
|
| 1878 |
+
"grad_norm": 4.913658142089844,
|
| 1879 |
+
"learning_rate": 0.0003522727272727273,
|
| 1880 |
+
"loss": 1.1278,
|
| 1881 |
+
"step": 1710
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 76.0,
|
| 1885 |
+
"eval_accuracy": 0.2608695652173913,
|
| 1886 |
+
"eval_loss": 1.4527482986450195,
|
| 1887 |
+
"eval_runtime": 0.8695,
|
| 1888 |
+
"eval_samples_per_second": 52.906,
|
| 1889 |
+
"eval_steps_per_second": 3.45,
|
| 1890 |
+
"step": 1710
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 76.44444444444444,
|
| 1894 |
+
"grad_norm": 5.169142723083496,
|
| 1895 |
+
"learning_rate": 0.0003484848484848485,
|
| 1896 |
+
"loss": 1.1212,
|
| 1897 |
+
"step": 1720
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 76.88888888888889,
|
| 1901 |
+
"grad_norm": 6.800394058227539,
|
| 1902 |
+
"learning_rate": 0.0003446969696969697,
|
| 1903 |
+
"loss": 1.0956,
|
| 1904 |
+
"step": 1730
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 76.97777777777777,
|
| 1908 |
+
"eval_accuracy": 0.2391304347826087,
|
| 1909 |
+
"eval_loss": 1.4839533567428589,
|
| 1910 |
+
"eval_runtime": 0.9059,
|
| 1911 |
+
"eval_samples_per_second": 50.779,
|
| 1912 |
+
"eval_steps_per_second": 3.312,
|
| 1913 |
+
"step": 1732
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 77.33333333333333,
|
| 1917 |
+
"grad_norm": 4.102919578552246,
|
| 1918 |
+
"learning_rate": 0.0003409090909090909,
|
| 1919 |
+
"loss": 1.1466,
|
| 1920 |
+
"step": 1740
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 77.77777777777777,
|
| 1924 |
+
"grad_norm": 4.560825824737549,
|
| 1925 |
+
"learning_rate": 0.0003371212121212121,
|
| 1926 |
+
"loss": 1.1131,
|
| 1927 |
+
"step": 1750
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 78.0,
|
| 1931 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1932 |
+
"eval_loss": 1.4900346994400024,
|
| 1933 |
+
"eval_runtime": 1.1932,
|
| 1934 |
+
"eval_samples_per_second": 38.552,
|
| 1935 |
+
"eval_steps_per_second": 2.514,
|
| 1936 |
+
"step": 1755
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 78.22222222222223,
|
| 1940 |
+
"grad_norm": 1.923434853553772,
|
| 1941 |
+
"learning_rate": 0.0003333333333333333,
|
| 1942 |
+
"loss": 1.1285,
|
| 1943 |
+
"step": 1760
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 78.66666666666667,
|
| 1947 |
+
"grad_norm": 4.646895885467529,
|
| 1948 |
+
"learning_rate": 0.00032954545454545454,
|
| 1949 |
+
"loss": 1.1376,
|
| 1950 |
+
"step": 1770
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 78.97777777777777,
|
| 1954 |
+
"eval_accuracy": 0.21739130434782608,
|
| 1955 |
+
"eval_loss": 1.5395020246505737,
|
| 1956 |
+
"eval_runtime": 0.8752,
|
| 1957 |
+
"eval_samples_per_second": 52.557,
|
| 1958 |
+
"eval_steps_per_second": 3.428,
|
| 1959 |
+
"step": 1777
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 79.11111111111111,
|
| 1963 |
+
"grad_norm": 8.433492660522461,
|
| 1964 |
+
"learning_rate": 0.00032575757575757576,
|
| 1965 |
+
"loss": 1.1072,
|
| 1966 |
+
"step": 1780
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 79.55555555555556,
|
| 1970 |
+
"grad_norm": 5.669383525848389,
|
| 1971 |
+
"learning_rate": 0.000321969696969697,
|
| 1972 |
+
"loss": 1.1135,
|
| 1973 |
+
"step": 1790
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 80.0,
|
| 1977 |
+
"grad_norm": 18.24361801147461,
|
| 1978 |
+
"learning_rate": 0.0003181818181818182,
|
| 1979 |
+
"loss": 1.0883,
|
| 1980 |
+
"step": 1800
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 80.0,
|
| 1984 |
+
"eval_accuracy": 0.1956521739130435,
|
| 1985 |
+
"eval_loss": 1.5037870407104492,
|
| 1986 |
+
"eval_runtime": 0.8647,
|
| 1987 |
+
"eval_samples_per_second": 53.198,
|
| 1988 |
+
"eval_steps_per_second": 3.469,
|
| 1989 |
+
"step": 1800
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 80.44444444444444,
|
| 1993 |
+
"grad_norm": 8.68807315826416,
|
| 1994 |
+
"learning_rate": 0.0003143939393939394,
|
| 1995 |
+
"loss": 1.0899,
|
| 1996 |
+
"step": 1810
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 80.88888888888889,
|
| 2000 |
+
"grad_norm": 39.473899841308594,
|
| 2001 |
+
"learning_rate": 0.0003106060606060606,
|
| 2002 |
+
"loss": 1.1017,
|
| 2003 |
+
"step": 1820
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 80.97777777777777,
|
| 2007 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2008 |
+
"eval_loss": 1.5392367839813232,
|
| 2009 |
+
"eval_runtime": 0.8702,
|
| 2010 |
+
"eval_samples_per_second": 52.864,
|
| 2011 |
+
"eval_steps_per_second": 3.448,
|
| 2012 |
+
"step": 1822
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 81.33333333333333,
|
| 2016 |
+
"grad_norm": 3.8075191974639893,
|
| 2017 |
+
"learning_rate": 0.0003068181818181818,
|
| 2018 |
+
"loss": 1.0607,
|
| 2019 |
+
"step": 1830
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 81.77777777777777,
|
| 2023 |
+
"grad_norm": 4.356723308563232,
|
| 2024 |
+
"learning_rate": 0.00030303030303030303,
|
| 2025 |
+
"loss": 1.1608,
|
| 2026 |
+
"step": 1840
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 82.0,
|
| 2030 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2031 |
+
"eval_loss": 1.4875361919403076,
|
| 2032 |
+
"eval_runtime": 1.2071,
|
| 2033 |
+
"eval_samples_per_second": 38.106,
|
| 2034 |
+
"eval_steps_per_second": 2.485,
|
| 2035 |
+
"step": 1845
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 82.22222222222223,
|
| 2039 |
+
"grad_norm": 4.517791748046875,
|
| 2040 |
+
"learning_rate": 0.00029924242424242425,
|
| 2041 |
+
"loss": 1.1144,
|
| 2042 |
+
"step": 1850
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 82.66666666666667,
|
| 2046 |
+
"grad_norm": 4.995427131652832,
|
| 2047 |
+
"learning_rate": 0.00029545454545454547,
|
| 2048 |
+
"loss": 1.1308,
|
| 2049 |
+
"step": 1860
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 82.97777777777777,
|
| 2053 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2054 |
+
"eval_loss": 1.5079646110534668,
|
| 2055 |
+
"eval_runtime": 0.8614,
|
| 2056 |
+
"eval_samples_per_second": 53.403,
|
| 2057 |
+
"eval_steps_per_second": 3.483,
|
| 2058 |
+
"step": 1867
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 83.11111111111111,
|
| 2062 |
+
"grad_norm": 4.479354381561279,
|
| 2063 |
+
"learning_rate": 0.0002916666666666667,
|
| 2064 |
+
"loss": 1.0821,
|
| 2065 |
+
"step": 1870
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 83.55555555555556,
|
| 2069 |
+
"grad_norm": 4.680623531341553,
|
| 2070 |
+
"learning_rate": 0.0002878787878787879,
|
| 2071 |
+
"loss": 1.0904,
|
| 2072 |
+
"step": 1880
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 84.0,
|
| 2076 |
+
"grad_norm": 11.278671264648438,
|
| 2077 |
+
"learning_rate": 0.00028409090909090913,
|
| 2078 |
+
"loss": 1.1382,
|
| 2079 |
+
"step": 1890
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 84.0,
|
| 2083 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2084 |
+
"eval_loss": 1.4835433959960938,
|
| 2085 |
+
"eval_runtime": 0.8759,
|
| 2086 |
+
"eval_samples_per_second": 52.52,
|
| 2087 |
+
"eval_steps_per_second": 3.425,
|
| 2088 |
+
"step": 1890
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 84.44444444444444,
|
| 2092 |
+
"grad_norm": 4.505593776702881,
|
| 2093 |
+
"learning_rate": 0.0002803030303030303,
|
| 2094 |
+
"loss": 1.0869,
|
| 2095 |
+
"step": 1900
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 84.88888888888889,
|
| 2099 |
+
"grad_norm": 3.064387083053589,
|
| 2100 |
+
"learning_rate": 0.0002765151515151515,
|
| 2101 |
+
"loss": 1.1195,
|
| 2102 |
+
"step": 1910
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 84.97777777777777,
|
| 2106 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2107 |
+
"eval_loss": 1.4076049327850342,
|
| 2108 |
+
"eval_runtime": 0.865,
|
| 2109 |
+
"eval_samples_per_second": 53.179,
|
| 2110 |
+
"eval_steps_per_second": 3.468,
|
| 2111 |
+
"step": 1912
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 85.33333333333333,
|
| 2115 |
+
"grad_norm": 4.745396137237549,
|
| 2116 |
+
"learning_rate": 0.00027272727272727274,
|
| 2117 |
+
"loss": 1.1153,
|
| 2118 |
+
"step": 1920
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 85.77777777777777,
|
| 2122 |
+
"grad_norm": 14.576072692871094,
|
| 2123 |
+
"learning_rate": 0.0002689393939393939,
|
| 2124 |
+
"loss": 1.1149,
|
| 2125 |
+
"step": 1930
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 86.0,
|
| 2129 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2130 |
+
"eval_loss": 1.4840431213378906,
|
| 2131 |
+
"eval_runtime": 1.1314,
|
| 2132 |
+
"eval_samples_per_second": 40.656,
|
| 2133 |
+
"eval_steps_per_second": 2.651,
|
| 2134 |
+
"step": 1935
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 86.22222222222223,
|
| 2138 |
+
"grad_norm": 4.840237617492676,
|
| 2139 |
+
"learning_rate": 0.0002651515151515151,
|
| 2140 |
+
"loss": 1.1314,
|
| 2141 |
+
"step": 1940
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 86.66666666666667,
|
| 2145 |
+
"grad_norm": 7.4581756591796875,
|
| 2146 |
+
"learning_rate": 0.00026136363636363634,
|
| 2147 |
+
"loss": 1.1344,
|
| 2148 |
+
"step": 1950
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 86.97777777777777,
|
| 2152 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2153 |
+
"eval_loss": 1.473250150680542,
|
| 2154 |
+
"eval_runtime": 1.086,
|
| 2155 |
+
"eval_samples_per_second": 42.357,
|
| 2156 |
+
"eval_steps_per_second": 2.762,
|
| 2157 |
+
"step": 1957
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 87.11111111111111,
|
| 2161 |
+
"grad_norm": 4.788184642791748,
|
| 2162 |
+
"learning_rate": 0.00025757575757575756,
|
| 2163 |
+
"loss": 1.0985,
|
| 2164 |
+
"step": 1960
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 87.55555555555556,
|
| 2168 |
+
"grad_norm": 7.187996864318848,
|
| 2169 |
+
"learning_rate": 0.0002537878787878788,
|
| 2170 |
+
"loss": 1.1018,
|
| 2171 |
+
"step": 1970
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 88.0,
|
| 2175 |
+
"grad_norm": 6.393486976623535,
|
| 2176 |
+
"learning_rate": 0.00025,
|
| 2177 |
+
"loss": 1.1268,
|
| 2178 |
+
"step": 1980
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 88.0,
|
| 2182 |
+
"eval_accuracy": 0.2391304347826087,
|
| 2183 |
+
"eval_loss": 1.4446380138397217,
|
| 2184 |
+
"eval_runtime": 0.877,
|
| 2185 |
+
"eval_samples_per_second": 52.451,
|
| 2186 |
+
"eval_steps_per_second": 3.421,
|
| 2187 |
+
"step": 1980
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 88.44444444444444,
|
| 2191 |
+
"grad_norm": 19.996049880981445,
|
| 2192 |
+
"learning_rate": 0.0002462121212121212,
|
| 2193 |
+
"loss": 1.115,
|
| 2194 |
+
"step": 1990
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 88.88888888888889,
|
| 2198 |
+
"grad_norm": 11.436882019042969,
|
| 2199 |
+
"learning_rate": 0.00024242424242424245,
|
| 2200 |
+
"loss": 1.1267,
|
| 2201 |
+
"step": 2000
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 88.97777777777777,
|
| 2205 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2206 |
+
"eval_loss": 1.4359674453735352,
|
| 2207 |
+
"eval_runtime": 0.8812,
|
| 2208 |
+
"eval_samples_per_second": 52.199,
|
| 2209 |
+
"eval_steps_per_second": 3.404,
|
| 2210 |
+
"step": 2002
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 89.33333333333333,
|
| 2214 |
+
"grad_norm": 7.150528430938721,
|
| 2215 |
+
"learning_rate": 0.00023863636363636364,
|
| 2216 |
+
"loss": 1.1335,
|
| 2217 |
+
"step": 2010
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 89.77777777777777,
|
| 2221 |
+
"grad_norm": 142.4888153076172,
|
| 2222 |
+
"learning_rate": 0.00023484848484848486,
|
| 2223 |
+
"loss": 1.1034,
|
| 2224 |
+
"step": 2020
|
| 2225 |
+
},
|
| 2226 |
+
{
|
| 2227 |
+
"epoch": 90.0,
|
| 2228 |
+
"eval_accuracy": 0.15217391304347827,
|
| 2229 |
+
"eval_loss": 1.4328769445419312,
|
| 2230 |
+
"eval_runtime": 0.8938,
|
| 2231 |
+
"eval_samples_per_second": 51.465,
|
| 2232 |
+
"eval_steps_per_second": 3.356,
|
| 2233 |
+
"step": 2025
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"epoch": 90.22222222222223,
|
| 2237 |
+
"grad_norm": 13.855992317199707,
|
| 2238 |
+
"learning_rate": 0.00023106060606060605,
|
| 2239 |
+
"loss": 1.0987,
|
| 2240 |
+
"step": 2030
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 90.66666666666667,
|
| 2244 |
+
"grad_norm": 4.523609638214111,
|
| 2245 |
+
"learning_rate": 0.00022727272727272727,
|
| 2246 |
+
"loss": 1.1113,
|
| 2247 |
+
"step": 2040
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"epoch": 90.97777777777777,
|
| 2251 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2252 |
+
"eval_loss": 1.4670028686523438,
|
| 2253 |
+
"eval_runtime": 1.1907,
|
| 2254 |
+
"eval_samples_per_second": 38.631,
|
| 2255 |
+
"eval_steps_per_second": 2.519,
|
| 2256 |
+
"step": 2047
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 91.11111111111111,
|
| 2260 |
+
"grad_norm": 8.413890838623047,
|
| 2261 |
+
"learning_rate": 0.0002234848484848485,
|
| 2262 |
+
"loss": 1.0848,
|
| 2263 |
+
"step": 2050
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 91.55555555555556,
|
| 2267 |
+
"grad_norm": 2.8552653789520264,
|
| 2268 |
+
"learning_rate": 0.0002196969696969697,
|
| 2269 |
+
"loss": 1.0788,
|
| 2270 |
+
"step": 2060
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 92.0,
|
| 2274 |
+
"grad_norm": 22.520193099975586,
|
| 2275 |
+
"learning_rate": 0.0002159090909090909,
|
| 2276 |
+
"loss": 1.0957,
|
| 2277 |
+
"step": 2070
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 92.0,
|
| 2281 |
+
"eval_accuracy": 0.2391304347826087,
|
| 2282 |
+
"eval_loss": 1.4802157878875732,
|
| 2283 |
+
"eval_runtime": 0.8728,
|
| 2284 |
+
"eval_samples_per_second": 52.706,
|
| 2285 |
+
"eval_steps_per_second": 3.437,
|
| 2286 |
+
"step": 2070
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 92.44444444444444,
|
| 2290 |
+
"grad_norm": 17.830493927001953,
|
| 2291 |
+
"learning_rate": 0.00021212121212121213,
|
| 2292 |
+
"loss": 1.122,
|
| 2293 |
+
"step": 2080
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 92.88888888888889,
|
| 2297 |
+
"grad_norm": 4.072429656982422,
|
| 2298 |
+
"learning_rate": 0.00020833333333333335,
|
| 2299 |
+
"loss": 1.1227,
|
| 2300 |
+
"step": 2090
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 92.97777777777777,
|
| 2304 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2305 |
+
"eval_loss": 1.4715131521224976,
|
| 2306 |
+
"eval_runtime": 0.8786,
|
| 2307 |
+
"eval_samples_per_second": 52.357,
|
| 2308 |
+
"eval_steps_per_second": 3.415,
|
| 2309 |
+
"step": 2092
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 93.33333333333333,
|
| 2313 |
+
"grad_norm": 4.416510581970215,
|
| 2314 |
+
"learning_rate": 0.00020454545454545457,
|
| 2315 |
+
"loss": 1.0755,
|
| 2316 |
+
"step": 2100
|
| 2317 |
+
},
|
| 2318 |
+
{
|
| 2319 |
+
"epoch": 93.77777777777777,
|
| 2320 |
+
"grad_norm": 19.229928970336914,
|
| 2321 |
+
"learning_rate": 0.00020075757575757576,
|
| 2322 |
+
"loss": 1.1083,
|
| 2323 |
+
"step": 2110
|
| 2324 |
+
},
|
| 2325 |
+
{
|
| 2326 |
+
"epoch": 94.0,
|
| 2327 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2328 |
+
"eval_loss": 1.4812626838684082,
|
| 2329 |
+
"eval_runtime": 0.8656,
|
| 2330 |
+
"eval_samples_per_second": 53.14,
|
| 2331 |
+
"eval_steps_per_second": 3.466,
|
| 2332 |
+
"step": 2115
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 94.22222222222223,
|
| 2336 |
+
"grad_norm": 4.718553066253662,
|
| 2337 |
+
"learning_rate": 0.00019696969696969695,
|
| 2338 |
+
"loss": 1.0757,
|
| 2339 |
+
"step": 2120
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 94.66666666666667,
|
| 2343 |
+
"grad_norm": 5.6447601318359375,
|
| 2344 |
+
"learning_rate": 0.00019318181818181817,
|
| 2345 |
+
"loss": 1.0583,
|
| 2346 |
+
"step": 2130
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 94.97777777777777,
|
| 2350 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2351 |
+
"eval_loss": 1.520257830619812,
|
| 2352 |
+
"eval_runtime": 1.1597,
|
| 2353 |
+
"eval_samples_per_second": 39.665,
|
| 2354 |
+
"eval_steps_per_second": 2.587,
|
| 2355 |
+
"step": 2137
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 95.11111111111111,
|
| 2359 |
+
"grad_norm": 5.209888935089111,
|
| 2360 |
+
"learning_rate": 0.0001893939393939394,
|
| 2361 |
+
"loss": 1.098,
|
| 2362 |
+
"step": 2140
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 95.55555555555556,
|
| 2366 |
+
"grad_norm": 7.493114471435547,
|
| 2367 |
+
"learning_rate": 0.00018560606060606061,
|
| 2368 |
+
"loss": 1.0796,
|
| 2369 |
+
"step": 2150
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 96.0,
|
| 2373 |
+
"grad_norm": 11.232746124267578,
|
| 2374 |
+
"learning_rate": 0.00018181818181818183,
|
| 2375 |
+
"loss": 1.093,
|
| 2376 |
+
"step": 2160
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 96.0,
|
| 2380 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2381 |
+
"eval_loss": 1.5394465923309326,
|
| 2382 |
+
"eval_runtime": 0.8817,
|
| 2383 |
+
"eval_samples_per_second": 52.171,
|
| 2384 |
+
"eval_steps_per_second": 3.402,
|
| 2385 |
+
"step": 2160
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 96.44444444444444,
|
| 2389 |
+
"grad_norm": 9.968954086303711,
|
| 2390 |
+
"learning_rate": 0.00017803030303030305,
|
| 2391 |
+
"loss": 1.0958,
|
| 2392 |
+
"step": 2170
|
| 2393 |
+
},
|
| 2394 |
+
{
|
| 2395 |
+
"epoch": 96.88888888888889,
|
| 2396 |
+
"grad_norm": 15.155268669128418,
|
| 2397 |
+
"learning_rate": 0.00017424242424242425,
|
| 2398 |
+
"loss": 1.0809,
|
| 2399 |
+
"step": 2180
|
| 2400 |
+
},
|
| 2401 |
+
{
|
| 2402 |
+
"epoch": 96.97777777777777,
|
| 2403 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2404 |
+
"eval_loss": 1.4620193243026733,
|
| 2405 |
+
"eval_runtime": 0.8656,
|
| 2406 |
+
"eval_samples_per_second": 53.142,
|
| 2407 |
+
"eval_steps_per_second": 3.466,
|
| 2408 |
+
"step": 2182
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 97.33333333333333,
|
| 2412 |
+
"grad_norm": 10.245019912719727,
|
| 2413 |
+
"learning_rate": 0.00017045454545454544,
|
| 2414 |
+
"loss": 1.0822,
|
| 2415 |
+
"step": 2190
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 97.77777777777777,
|
| 2419 |
+
"grad_norm": 56.624778747558594,
|
| 2420 |
+
"learning_rate": 0.00016666666666666666,
|
| 2421 |
+
"loss": 1.0888,
|
| 2422 |
+
"step": 2200
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 98.0,
|
| 2426 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2427 |
+
"eval_loss": 1.4407347440719604,
|
| 2428 |
+
"eval_runtime": 0.8737,
|
| 2429 |
+
"eval_samples_per_second": 52.651,
|
| 2430 |
+
"eval_steps_per_second": 3.434,
|
| 2431 |
+
"step": 2205
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 98.22222222222223,
|
| 2435 |
+
"grad_norm": 16.125377655029297,
|
| 2436 |
+
"learning_rate": 0.00016287878787878788,
|
| 2437 |
+
"loss": 1.0803,
|
| 2438 |
+
"step": 2210
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 98.66666666666667,
|
| 2442 |
+
"grad_norm": 4.7502546310424805,
|
| 2443 |
+
"learning_rate": 0.0001590909090909091,
|
| 2444 |
+
"loss": 1.1292,
|
| 2445 |
+
"step": 2220
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 98.97777777777777,
|
| 2449 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2450 |
+
"eval_loss": 1.4577943086624146,
|
| 2451 |
+
"eval_runtime": 1.2602,
|
| 2452 |
+
"eval_samples_per_second": 36.503,
|
| 2453 |
+
"eval_steps_per_second": 2.381,
|
| 2454 |
+
"step": 2227
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 99.11111111111111,
|
| 2458 |
+
"grad_norm": 12.011200904846191,
|
| 2459 |
+
"learning_rate": 0.0001553030303030303,
|
| 2460 |
+
"loss": 1.076,
|
| 2461 |
+
"step": 2230
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 99.55555555555556,
|
| 2465 |
+
"grad_norm": 10.061297416687012,
|
| 2466 |
+
"learning_rate": 0.00015151515151515152,
|
| 2467 |
+
"loss": 1.0789,
|
| 2468 |
+
"step": 2240
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 100.0,
|
| 2472 |
+
"grad_norm": 7.683816432952881,
|
| 2473 |
+
"learning_rate": 0.00014772727272727274,
|
| 2474 |
+
"loss": 1.0754,
|
| 2475 |
+
"step": 2250
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 100.0,
|
| 2479 |
+
"eval_accuracy": 0.17391304347826086,
|
| 2480 |
+
"eval_loss": 1.5031030178070068,
|
| 2481 |
+
"eval_runtime": 0.8791,
|
| 2482 |
+
"eval_samples_per_second": 52.328,
|
| 2483 |
+
"eval_steps_per_second": 3.413,
|
| 2484 |
+
"step": 2250
|
| 2485 |
+
},
|
| 2486 |
+
{
|
| 2487 |
+
"epoch": 100.44444444444444,
|
| 2488 |
+
"grad_norm": 10.789087295532227,
|
| 2489 |
+
"learning_rate": 0.00014393939393939396,
|
| 2490 |
+
"loss": 1.0662,
|
| 2491 |
+
"step": 2260
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 100.88888888888889,
|
| 2495 |
+
"grad_norm": 14.93548583984375,
|
| 2496 |
+
"learning_rate": 0.00014015151515151515,
|
| 2497 |
+
"loss": 1.0817,
|
| 2498 |
+
"step": 2270
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 100.97777777777777,
|
| 2502 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2503 |
+
"eval_loss": 1.4460813999176025,
|
| 2504 |
+
"eval_runtime": 0.8764,
|
| 2505 |
+
"eval_samples_per_second": 52.489,
|
| 2506 |
+
"eval_steps_per_second": 3.423,
|
| 2507 |
+
"step": 2272
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 101.33333333333333,
|
| 2511 |
+
"grad_norm": 7.292104244232178,
|
| 2512 |
+
"learning_rate": 0.00013636363636363637,
|
| 2513 |
+
"loss": 1.1087,
|
| 2514 |
+
"step": 2280
|
| 2515 |
+
},
|
| 2516 |
+
{
|
| 2517 |
+
"epoch": 101.77777777777777,
|
| 2518 |
+
"grad_norm": 21.396413803100586,
|
| 2519 |
+
"learning_rate": 0.00013257575757575756,
|
| 2520 |
+
"loss": 1.0671,
|
| 2521 |
+
"step": 2290
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 102.0,
|
| 2525 |
+
"eval_accuracy": 0.2391304347826087,
|
| 2526 |
+
"eval_loss": 1.4722661972045898,
|
| 2527 |
+
"eval_runtime": 0.8883,
|
| 2528 |
+
"eval_samples_per_second": 51.786,
|
| 2529 |
+
"eval_steps_per_second": 3.377,
|
| 2530 |
+
"step": 2295
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 102.22222222222223,
|
| 2534 |
+
"grad_norm": 7.56445837020874,
|
| 2535 |
+
"learning_rate": 0.00012878787878787878,
|
| 2536 |
+
"loss": 1.0837,
|
| 2537 |
+
"step": 2300
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 102.66666666666667,
|
| 2541 |
+
"grad_norm": 12.246611595153809,
|
| 2542 |
+
"learning_rate": 0.000125,
|
| 2543 |
+
"loss": 1.0815,
|
| 2544 |
+
"step": 2310
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 102.97777777777777,
|
| 2548 |
+
"eval_accuracy": 0.1956521739130435,
|
| 2549 |
+
"eval_loss": 1.4988662004470825,
|
| 2550 |
+
"eval_runtime": 1.1771,
|
| 2551 |
+
"eval_samples_per_second": 39.079,
|
| 2552 |
+
"eval_steps_per_second": 2.549,
|
| 2553 |
+
"step": 2317
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 103.11111111111111,
|
| 2557 |
+
"grad_norm": 10.69598388671875,
|
| 2558 |
+
"learning_rate": 0.00012121212121212122,
|
| 2559 |
+
"loss": 1.0852,
|
| 2560 |
+
"step": 2320
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 103.55555555555556,
|
| 2564 |
+
"grad_norm": 9.647980690002441,
|
| 2565 |
+
"learning_rate": 0.00011742424242424243,
|
| 2566 |
+
"loss": 1.076,
|
| 2567 |
+
"step": 2330
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 104.0,
|
| 2571 |
+
"grad_norm": 38.92795944213867,
|
| 2572 |
+
"learning_rate": 0.00011363636363636364,
|
| 2573 |
+
"loss": 1.0967,
|
| 2574 |
+
"step": 2340
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 104.0,
|
| 2578 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2579 |
+
"eval_loss": 1.465432047843933,
|
| 2580 |
+
"eval_runtime": 0.8692,
|
| 2581 |
+
"eval_samples_per_second": 52.922,
|
| 2582 |
+
"eval_steps_per_second": 3.451,
|
| 2583 |
+
"step": 2340
|
| 2584 |
+
},
|
| 2585 |
+
{
|
| 2586 |
+
"epoch": 104.44444444444444,
|
| 2587 |
+
"grad_norm": 9.53869915008545,
|
| 2588 |
+
"learning_rate": 0.00010984848484848486,
|
| 2589 |
+
"loss": 1.0838,
|
| 2590 |
+
"step": 2350
|
| 2591 |
+
},
|
| 2592 |
+
{
|
| 2593 |
+
"epoch": 104.88888888888889,
|
| 2594 |
+
"grad_norm": 5.269750118255615,
|
| 2595 |
+
"learning_rate": 0.00010606060606060606,
|
| 2596 |
+
"loss": 1.091,
|
| 2597 |
+
"step": 2360
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 104.97777777777777,
|
| 2601 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2602 |
+
"eval_loss": 1.4559190273284912,
|
| 2603 |
+
"eval_runtime": 0.8756,
|
| 2604 |
+
"eval_samples_per_second": 52.535,
|
| 2605 |
+
"eval_steps_per_second": 3.426,
|
| 2606 |
+
"step": 2362
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 105.33333333333333,
|
| 2610 |
+
"grad_norm": 12.788183212280273,
|
| 2611 |
+
"learning_rate": 0.00010227272727272728,
|
| 2612 |
+
"loss": 1.1085,
|
| 2613 |
+
"step": 2370
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 105.77777777777777,
|
| 2617 |
+
"grad_norm": 7.0092058181762695,
|
| 2618 |
+
"learning_rate": 9.848484848484848e-05,
|
| 2619 |
+
"loss": 1.0895,
|
| 2620 |
+
"step": 2380
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 106.0,
|
| 2624 |
+
"eval_accuracy": 0.2826086956521739,
|
| 2625 |
+
"eval_loss": 1.4221450090408325,
|
| 2626 |
+
"eval_runtime": 1.1631,
|
| 2627 |
+
"eval_samples_per_second": 39.549,
|
| 2628 |
+
"eval_steps_per_second": 2.579,
|
| 2629 |
+
"step": 2385
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 106.22222222222223,
|
| 2633 |
+
"grad_norm": 12.933279037475586,
|
| 2634 |
+
"learning_rate": 9.46969696969697e-05,
|
| 2635 |
+
"loss": 1.1548,
|
| 2636 |
+
"step": 2390
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 106.66666666666667,
|
| 2640 |
+
"grad_norm": 4.65310001373291,
|
| 2641 |
+
"learning_rate": 9.090909090909092e-05,
|
| 2642 |
+
"loss": 1.0847,
|
| 2643 |
+
"step": 2400
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 106.97777777777777,
|
| 2647 |
+
"eval_accuracy": 0.2826086956521739,
|
| 2648 |
+
"eval_loss": 1.4292521476745605,
|
| 2649 |
+
"eval_runtime": 0.8992,
|
| 2650 |
+
"eval_samples_per_second": 51.156,
|
| 2651 |
+
"eval_steps_per_second": 3.336,
|
| 2652 |
+
"step": 2407
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 107.11111111111111,
|
| 2656 |
+
"grad_norm": 7.354497909545898,
|
| 2657 |
+
"learning_rate": 8.712121212121212e-05,
|
| 2658 |
+
"loss": 1.0996,
|
| 2659 |
+
"step": 2410
|
| 2660 |
+
},
|
| 2661 |
+
{
|
| 2662 |
+
"epoch": 107.55555555555556,
|
| 2663 |
+
"grad_norm": 5.702237129211426,
|
| 2664 |
+
"learning_rate": 8.333333333333333e-05,
|
| 2665 |
+
"loss": 1.0883,
|
| 2666 |
+
"step": 2420
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 108.0,
|
| 2670 |
+
"grad_norm": 4.873330116271973,
|
| 2671 |
+
"learning_rate": 7.954545454545455e-05,
|
| 2672 |
+
"loss": 1.102,
|
| 2673 |
+
"step": 2430
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 108.0,
|
| 2677 |
+
"eval_accuracy": 0.2391304347826087,
|
| 2678 |
+
"eval_loss": 1.4582384824752808,
|
| 2679 |
+
"eval_runtime": 0.9053,
|
| 2680 |
+
"eval_samples_per_second": 50.81,
|
| 2681 |
+
"eval_steps_per_second": 3.314,
|
| 2682 |
+
"step": 2430
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 108.44444444444444,
|
| 2686 |
+
"grad_norm": 78.408447265625,
|
| 2687 |
+
"learning_rate": 7.575757575757576e-05,
|
| 2688 |
+
"loss": 1.1048,
|
| 2689 |
+
"step": 2440
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 108.88888888888889,
|
| 2693 |
+
"grad_norm": 6.651626110076904,
|
| 2694 |
+
"learning_rate": 7.196969696969698e-05,
|
| 2695 |
+
"loss": 1.0404,
|
| 2696 |
+
"step": 2450
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 108.97777777777777,
|
| 2700 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2701 |
+
"eval_loss": 1.4655812978744507,
|
| 2702 |
+
"eval_runtime": 0.8722,
|
| 2703 |
+
"eval_samples_per_second": 52.741,
|
| 2704 |
+
"eval_steps_per_second": 3.44,
|
| 2705 |
+
"step": 2452
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 109.33333333333333,
|
| 2709 |
+
"grad_norm": 10.666740417480469,
|
| 2710 |
+
"learning_rate": 6.818181818181818e-05,
|
| 2711 |
+
"loss": 1.0799,
|
| 2712 |
+
"step": 2460
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 109.77777777777777,
|
| 2716 |
+
"grad_norm": 6.093382358551025,
|
| 2717 |
+
"learning_rate": 6.439393939393939e-05,
|
| 2718 |
+
"loss": 1.0488,
|
| 2719 |
+
"step": 2470
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 110.0,
|
| 2723 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2724 |
+
"eval_loss": 1.489004373550415,
|
| 2725 |
+
"eval_runtime": 1.0926,
|
| 2726 |
+
"eval_samples_per_second": 42.101,
|
| 2727 |
+
"eval_steps_per_second": 2.746,
|
| 2728 |
+
"step": 2475
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 110.22222222222223,
|
| 2732 |
+
"grad_norm": 5.012928009033203,
|
| 2733 |
+
"learning_rate": 6.060606060606061e-05,
|
| 2734 |
+
"loss": 1.1006,
|
| 2735 |
+
"step": 2480
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 110.66666666666667,
|
| 2739 |
+
"grad_norm": 8.174079895019531,
|
| 2740 |
+
"learning_rate": 5.681818181818182e-05,
|
| 2741 |
+
"loss": 1.0966,
|
| 2742 |
+
"step": 2490
|
| 2743 |
+
},
|
| 2744 |
+
{
|
| 2745 |
+
"epoch": 110.97777777777777,
|
| 2746 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2747 |
+
"eval_loss": 1.4631787538528442,
|
| 2748 |
+
"eval_runtime": 0.8917,
|
| 2749 |
+
"eval_samples_per_second": 51.586,
|
| 2750 |
+
"eval_steps_per_second": 3.364,
|
| 2751 |
+
"step": 2497
|
| 2752 |
+
},
|
| 2753 |
+
{
|
| 2754 |
+
"epoch": 111.11111111111111,
|
| 2755 |
+
"grad_norm": 4.650646686553955,
|
| 2756 |
+
"learning_rate": 5.303030303030303e-05,
|
| 2757 |
+
"loss": 1.0629,
|
| 2758 |
+
"step": 2500
|
| 2759 |
+
},
|
| 2760 |
+
{
|
| 2761 |
+
"epoch": 111.55555555555556,
|
| 2762 |
+
"grad_norm": 5.132219314575195,
|
| 2763 |
+
"learning_rate": 4.924242424242424e-05,
|
| 2764 |
+
"loss": 1.0835,
|
| 2765 |
+
"step": 2510
|
| 2766 |
+
},
|
| 2767 |
+
{
|
| 2768 |
+
"epoch": 112.0,
|
| 2769 |
+
"grad_norm": 8.792770385742188,
|
| 2770 |
+
"learning_rate": 4.545454545454546e-05,
|
| 2771 |
+
"loss": 1.0901,
|
| 2772 |
+
"step": 2520
|
| 2773 |
+
},
|
| 2774 |
+
{
|
| 2775 |
+
"epoch": 112.0,
|
| 2776 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2777 |
+
"eval_loss": 1.4494850635528564,
|
| 2778 |
+
"eval_runtime": 0.872,
|
| 2779 |
+
"eval_samples_per_second": 52.75,
|
| 2780 |
+
"eval_steps_per_second": 3.44,
|
| 2781 |
+
"step": 2520
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 112.44444444444444,
|
| 2785 |
+
"grad_norm": 6.029526710510254,
|
| 2786 |
+
"learning_rate": 4.1666666666666665e-05,
|
| 2787 |
+
"loss": 1.0764,
|
| 2788 |
+
"step": 2530
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 112.88888888888889,
|
| 2792 |
+
"grad_norm": 5.1245927810668945,
|
| 2793 |
+
"learning_rate": 3.787878787878788e-05,
|
| 2794 |
+
"loss": 1.1008,
|
| 2795 |
+
"step": 2540
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 112.97777777777777,
|
| 2799 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2800 |
+
"eval_loss": 1.4332908391952515,
|
| 2801 |
+
"eval_runtime": 0.9004,
|
| 2802 |
+
"eval_samples_per_second": 51.09,
|
| 2803 |
+
"eval_steps_per_second": 3.332,
|
| 2804 |
+
"step": 2542
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 113.33333333333333,
|
| 2808 |
+
"grad_norm": 11.329251289367676,
|
| 2809 |
+
"learning_rate": 3.409090909090909e-05,
|
| 2810 |
+
"loss": 1.0763,
|
| 2811 |
+
"step": 2550
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 113.77777777777777,
|
| 2815 |
+
"grad_norm": 3.865112066268921,
|
| 2816 |
+
"learning_rate": 3.0303030303030306e-05,
|
| 2817 |
+
"loss": 1.0884,
|
| 2818 |
+
"step": 2560
|
| 2819 |
+
},
|
| 2820 |
+
{
|
| 2821 |
+
"epoch": 114.0,
|
| 2822 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2823 |
+
"eval_loss": 1.4406064748764038,
|
| 2824 |
+
"eval_runtime": 0.9189,
|
| 2825 |
+
"eval_samples_per_second": 50.057,
|
| 2826 |
+
"eval_steps_per_second": 3.265,
|
| 2827 |
+
"step": 2565
|
| 2828 |
+
},
|
| 2829 |
+
{
|
| 2830 |
+
"epoch": 114.22222222222223,
|
| 2831 |
+
"grad_norm": 7.575568675994873,
|
| 2832 |
+
"learning_rate": 2.6515151515151516e-05,
|
| 2833 |
+
"loss": 1.088,
|
| 2834 |
+
"step": 2570
|
| 2835 |
+
},
|
| 2836 |
+
{
|
| 2837 |
+
"epoch": 114.66666666666667,
|
| 2838 |
+
"grad_norm": 7.7806620597839355,
|
| 2839 |
+
"learning_rate": 2.272727272727273e-05,
|
| 2840 |
+
"loss": 1.0889,
|
| 2841 |
+
"step": 2580
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 114.97777777777777,
|
| 2845 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2846 |
+
"eval_loss": 1.447421908378601,
|
| 2847 |
+
"eval_runtime": 0.9635,
|
| 2848 |
+
"eval_samples_per_second": 47.742,
|
| 2849 |
+
"eval_steps_per_second": 3.114,
|
| 2850 |
+
"step": 2587
|
| 2851 |
+
},
|
| 2852 |
+
{
|
| 2853 |
+
"epoch": 115.11111111111111,
|
| 2854 |
+
"grad_norm": 20.770198822021484,
|
| 2855 |
+
"learning_rate": 1.893939393939394e-05,
|
| 2856 |
+
"loss": 1.0754,
|
| 2857 |
+
"step": 2590
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 115.55555555555556,
|
| 2861 |
+
"grad_norm": 5.939628601074219,
|
| 2862 |
+
"learning_rate": 1.5151515151515153e-05,
|
| 2863 |
+
"loss": 1.07,
|
| 2864 |
+
"step": 2600
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 116.0,
|
| 2868 |
+
"grad_norm": 35.29814147949219,
|
| 2869 |
+
"learning_rate": 1.1363636363636365e-05,
|
| 2870 |
+
"loss": 1.0729,
|
| 2871 |
+
"step": 2610
|
| 2872 |
+
},
|
| 2873 |
+
{
|
| 2874 |
+
"epoch": 116.0,
|
| 2875 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2876 |
+
"eval_loss": 1.4561296701431274,
|
| 2877 |
+
"eval_runtime": 0.899,
|
| 2878 |
+
"eval_samples_per_second": 51.169,
|
| 2879 |
+
"eval_steps_per_second": 3.337,
|
| 2880 |
+
"step": 2610
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 116.44444444444444,
|
| 2884 |
+
"grad_norm": 24.017641067504883,
|
| 2885 |
+
"learning_rate": 7.5757575757575764e-06,
|
| 2886 |
+
"loss": 1.0589,
|
| 2887 |
+
"step": 2620
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 116.88888888888889,
|
| 2891 |
+
"grad_norm": 14.974956512451172,
|
| 2892 |
+
"learning_rate": 3.7878787878787882e-06,
|
| 2893 |
+
"loss": 1.0671,
|
| 2894 |
+
"step": 2630
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 116.97777777777777,
|
| 2898 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2899 |
+
"eval_loss": 1.4538123607635498,
|
| 2900 |
+
"eval_runtime": 0.8903,
|
| 2901 |
+
"eval_samples_per_second": 51.667,
|
| 2902 |
+
"eval_steps_per_second": 3.37,
|
| 2903 |
+
"step": 2632
|
| 2904 |
+
},
|
| 2905 |
+
{
|
| 2906 |
+
"epoch": 117.33333333333333,
|
| 2907 |
+
"grad_norm": 8.030067443847656,
|
| 2908 |
+
"learning_rate": 0.0,
|
| 2909 |
+
"loss": 1.0937,
|
| 2910 |
+
"step": 2640
|
| 2911 |
+
},
|
| 2912 |
+
{
|
| 2913 |
+
"epoch": 117.33333333333333,
|
| 2914 |
+
"eval_accuracy": 0.21739130434782608,
|
| 2915 |
+
"eval_loss": 1.453188180923462,
|
| 2916 |
+
"eval_runtime": 1.73,
|
| 2917 |
+
"eval_samples_per_second": 26.589,
|
| 2918 |
+
"eval_steps_per_second": 1.734,
|
| 2919 |
+
"step": 2640
|
| 2920 |
+
},
|
| 2921 |
+
{
|
| 2922 |
+
"epoch": 117.33333333333333,
|
| 2923 |
+
"step": 2640,
|
| 2924 |
+
"total_flos": 5.466852859010089e+18,
|
| 2925 |
+
"train_loss": 1.1631801536588957,
|
| 2926 |
+
"train_runtime": 5282.7516,
|
| 2927 |
+
"train_samples_per_second": 32.529,
|
| 2928 |
+
"train_steps_per_second": 0.5
|
| 2929 |
+
}
|
| 2930 |
+
],
|
| 2931 |
+
"logging_steps": 10,
|
| 2932 |
+
"max_steps": 2640,
|
| 2933 |
+
"num_input_tokens_seen": 0,
|
| 2934 |
+
"num_train_epochs": 120,
|
| 2935 |
+
"save_steps": 500,
|
| 2936 |
+
"stateful_callbacks": {
|
| 2937 |
+
"TrainerControl": {
|
| 2938 |
+
"args": {
|
| 2939 |
+
"should_epoch_stop": false,
|
| 2940 |
+
"should_evaluate": false,
|
| 2941 |
+
"should_log": false,
|
| 2942 |
+
"should_save": true,
|
| 2943 |
+
"should_training_stop": true
|
| 2944 |
+
},
|
| 2945 |
+
"attributes": {}
|
| 2946 |
+
}
|
| 2947 |
+
},
|
| 2948 |
+
"total_flos": 5.466852859010089e+18,
|
| 2949 |
+
"train_batch_size": 16,
|
| 2950 |
+
"trial_name": null,
|
| 2951 |
+
"trial_params": null
|
| 2952 |
+
}
|