Commit
•
5c75957
1
Parent(s):
3c91b52
End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May20_19-46-34_b94dbfcd7693/events.out.tfevents.1716235649.b94dbfcd7693.1289.30 +3 -0
- train_results.json +8 -0
- trainer_state.json +1079 -0
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.9772727272727273
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.1430
|
| 36 |
+
- Accuracy: 0.9773
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_accuracy": 0.9772727272727273,
|
| 4 |
+
"eval_loss": 0.14295423030853271,
|
| 5 |
+
"eval_runtime": 0.7189,
|
| 6 |
+
"eval_samples_per_second": 61.203,
|
| 7 |
+
"eval_steps_per_second": 2.782,
|
| 8 |
+
"total_flos": 1.8822505078960128e+18,
|
| 9 |
+
"train_loss": 0.21107831209897995,
|
| 10 |
+
"train_runtime": 1239.3905,
|
| 11 |
+
"train_samples_per_second": 19.606,
|
| 12 |
+
"train_steps_per_second": 0.161
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_accuracy": 0.9772727272727273,
|
| 4 |
+
"eval_loss": 0.14295423030853271,
|
| 5 |
+
"eval_runtime": 0.7189,
|
| 6 |
+
"eval_samples_per_second": 61.203,
|
| 7 |
+
"eval_steps_per_second": 2.782
|
| 8 |
+
}
|
runs/May20_19-46-34_b94dbfcd7693/events.out.tfevents.1716235649.b94dbfcd7693.1289.30
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb913ba71e799c933ee4b9995b8cd09eebcca8c824b10fd8de25f690e02477b9
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"total_flos": 1.8822505078960128e+18,
|
| 4 |
+
"train_loss": 0.21107831209897995,
|
| 5 |
+
"train_runtime": 1239.3905,
|
| 6 |
+
"train_samples_per_second": 19.606,
|
| 7 |
+
"train_steps_per_second": 0.161
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1079 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9772727272727273,
|
| 3 |
+
"best_model_checkpoint": "beit-base-patch16-224-85-fold1/checkpoint-112",
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.5454545454545454,
|
| 14 |
+
"eval_loss": 0.7308238744735718,
|
| 15 |
+
"eval_runtime": 0.7278,
|
| 16 |
+
"eval_samples_per_second": 60.457,
|
| 17 |
+
"eval_steps_per_second": 2.748,
|
| 18 |
+
"step": 2
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 2.0,
|
| 22 |
+
"eval_accuracy": 0.7045454545454546,
|
| 23 |
+
"eval_loss": 0.7927291393280029,
|
| 24 |
+
"eval_runtime": 0.7032,
|
| 25 |
+
"eval_samples_per_second": 62.571,
|
| 26 |
+
"eval_steps_per_second": 2.844,
|
| 27 |
+
"step": 4
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 3.0,
|
| 31 |
+
"eval_accuracy": 0.7045454545454546,
|
| 32 |
+
"eval_loss": 0.9672332406044006,
|
| 33 |
+
"eval_runtime": 0.6786,
|
| 34 |
+
"eval_samples_per_second": 64.835,
|
| 35 |
+
"eval_steps_per_second": 2.947,
|
| 36 |
+
"step": 6
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 4.0,
|
| 40 |
+
"eval_accuracy": 0.7045454545454546,
|
| 41 |
+
"eval_loss": 0.6257070302963257,
|
| 42 |
+
"eval_runtime": 0.6667,
|
| 43 |
+
"eval_samples_per_second": 65.994,
|
| 44 |
+
"eval_steps_per_second": 3.0,
|
| 45 |
+
"step": 8
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.0,
|
| 49 |
+
"grad_norm": 11.562158584594727,
|
| 50 |
+
"learning_rate": 2.5e-05,
|
| 51 |
+
"loss": 0.6404,
|
| 52 |
+
"step": 10
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 5.0,
|
| 56 |
+
"eval_accuracy": 0.7954545454545454,
|
| 57 |
+
"eval_loss": 0.4645519554615021,
|
| 58 |
+
"eval_runtime": 0.6715,
|
| 59 |
+
"eval_samples_per_second": 65.525,
|
| 60 |
+
"eval_steps_per_second": 2.978,
|
| 61 |
+
"step": 10
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 6.0,
|
| 65 |
+
"eval_accuracy": 0.7045454545454546,
|
| 66 |
+
"eval_loss": 0.5648249983787537,
|
| 67 |
+
"eval_runtime": 0.6749,
|
| 68 |
+
"eval_samples_per_second": 65.19,
|
| 69 |
+
"eval_steps_per_second": 2.963,
|
| 70 |
+
"step": 12
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 7.0,
|
| 74 |
+
"eval_accuracy": 0.7727272727272727,
|
| 75 |
+
"eval_loss": 0.4388914406299591,
|
| 76 |
+
"eval_runtime": 0.6837,
|
| 77 |
+
"eval_samples_per_second": 64.359,
|
| 78 |
+
"eval_steps_per_second": 2.925,
|
| 79 |
+
"step": 14
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 8.0,
|
| 83 |
+
"eval_accuracy": 0.75,
|
| 84 |
+
"eval_loss": 0.4522952139377594,
|
| 85 |
+
"eval_runtime": 0.7027,
|
| 86 |
+
"eval_samples_per_second": 62.616,
|
| 87 |
+
"eval_steps_per_second": 2.846,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 9.0,
|
| 92 |
+
"eval_accuracy": 0.75,
|
| 93 |
+
"eval_loss": 0.46983444690704346,
|
| 94 |
+
"eval_runtime": 0.6818,
|
| 95 |
+
"eval_samples_per_second": 64.535,
|
| 96 |
+
"eval_steps_per_second": 2.933,
|
| 97 |
+
"step": 18
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 10.0,
|
| 101 |
+
"grad_norm": 8.737435340881348,
|
| 102 |
+
"learning_rate": 5e-05,
|
| 103 |
+
"loss": 0.455,
|
| 104 |
+
"step": 20
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 10.0,
|
| 108 |
+
"eval_accuracy": 0.8409090909090909,
|
| 109 |
+
"eval_loss": 0.3706974983215332,
|
| 110 |
+
"eval_runtime": 0.6808,
|
| 111 |
+
"eval_samples_per_second": 64.631,
|
| 112 |
+
"eval_steps_per_second": 2.938,
|
| 113 |
+
"step": 20
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 11.0,
|
| 117 |
+
"eval_accuracy": 0.8181818181818182,
|
| 118 |
+
"eval_loss": 0.35938122868537903,
|
| 119 |
+
"eval_runtime": 0.6781,
|
| 120 |
+
"eval_samples_per_second": 64.883,
|
| 121 |
+
"eval_steps_per_second": 2.949,
|
| 122 |
+
"step": 22
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 12.0,
|
| 126 |
+
"eval_accuracy": 0.7272727272727273,
|
| 127 |
+
"eval_loss": 0.6136276721954346,
|
| 128 |
+
"eval_runtime": 0.6807,
|
| 129 |
+
"eval_samples_per_second": 64.643,
|
| 130 |
+
"eval_steps_per_second": 2.938,
|
| 131 |
+
"step": 24
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 13.0,
|
| 135 |
+
"eval_accuracy": 0.8863636363636364,
|
| 136 |
+
"eval_loss": 0.30222317576408386,
|
| 137 |
+
"eval_runtime": 0.6745,
|
| 138 |
+
"eval_samples_per_second": 65.234,
|
| 139 |
+
"eval_steps_per_second": 2.965,
|
| 140 |
+
"step": 26
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 14.0,
|
| 144 |
+
"eval_accuracy": 0.8409090909090909,
|
| 145 |
+
"eval_loss": 0.29185205698013306,
|
| 146 |
+
"eval_runtime": 0.6861,
|
| 147 |
+
"eval_samples_per_second": 64.127,
|
| 148 |
+
"eval_steps_per_second": 2.915,
|
| 149 |
+
"step": 28
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 15.0,
|
| 153 |
+
"grad_norm": 8.116207122802734,
|
| 154 |
+
"learning_rate": 4.722222222222222e-05,
|
| 155 |
+
"loss": 0.3981,
|
| 156 |
+
"step": 30
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 15.0,
|
| 160 |
+
"eval_accuracy": 0.8181818181818182,
|
| 161 |
+
"eval_loss": 0.3611550033092499,
|
| 162 |
+
"eval_runtime": 0.6792,
|
| 163 |
+
"eval_samples_per_second": 64.785,
|
| 164 |
+
"eval_steps_per_second": 2.945,
|
| 165 |
+
"step": 30
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 16.0,
|
| 169 |
+
"eval_accuracy": 0.8863636363636364,
|
| 170 |
+
"eval_loss": 0.24922798573970795,
|
| 171 |
+
"eval_runtime": 0.6808,
|
| 172 |
+
"eval_samples_per_second": 64.63,
|
| 173 |
+
"eval_steps_per_second": 2.938,
|
| 174 |
+
"step": 32
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 17.0,
|
| 178 |
+
"eval_accuracy": 0.9090909090909091,
|
| 179 |
+
"eval_loss": 0.24598444998264313,
|
| 180 |
+
"eval_runtime": 0.6781,
|
| 181 |
+
"eval_samples_per_second": 64.888,
|
| 182 |
+
"eval_steps_per_second": 2.949,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 18.0,
|
| 187 |
+
"eval_accuracy": 0.8636363636363636,
|
| 188 |
+
"eval_loss": 0.2930772006511688,
|
| 189 |
+
"eval_runtime": 0.6891,
|
| 190 |
+
"eval_samples_per_second": 63.847,
|
| 191 |
+
"eval_steps_per_second": 2.902,
|
| 192 |
+
"step": 36
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 19.0,
|
| 196 |
+
"eval_accuracy": 0.9090909090909091,
|
| 197 |
+
"eval_loss": 0.18223300576210022,
|
| 198 |
+
"eval_runtime": 0.6759,
|
| 199 |
+
"eval_samples_per_second": 65.101,
|
| 200 |
+
"eval_steps_per_second": 2.959,
|
| 201 |
+
"step": 38
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 20.0,
|
| 205 |
+
"grad_norm": 15.090621948242188,
|
| 206 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 207 |
+
"loss": 0.3257,
|
| 208 |
+
"step": 40
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 20.0,
|
| 212 |
+
"eval_accuracy": 0.9090909090909091,
|
| 213 |
+
"eval_loss": 0.20600570738315582,
|
| 214 |
+
"eval_runtime": 0.6756,
|
| 215 |
+
"eval_samples_per_second": 65.128,
|
| 216 |
+
"eval_steps_per_second": 2.96,
|
| 217 |
+
"step": 40
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 21.0,
|
| 221 |
+
"eval_accuracy": 0.8863636363636364,
|
| 222 |
+
"eval_loss": 0.21951809525489807,
|
| 223 |
+
"eval_runtime": 0.6733,
|
| 224 |
+
"eval_samples_per_second": 65.351,
|
| 225 |
+
"eval_steps_per_second": 2.971,
|
| 226 |
+
"step": 42
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 22.0,
|
| 230 |
+
"eval_accuracy": 0.9090909090909091,
|
| 231 |
+
"eval_loss": 0.2624146640300751,
|
| 232 |
+
"eval_runtime": 0.6747,
|
| 233 |
+
"eval_samples_per_second": 65.219,
|
| 234 |
+
"eval_steps_per_second": 2.964,
|
| 235 |
+
"step": 44
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 23.0,
|
| 239 |
+
"eval_accuracy": 0.9090909090909091,
|
| 240 |
+
"eval_loss": 0.2383754700422287,
|
| 241 |
+
"eval_runtime": 0.6742,
|
| 242 |
+
"eval_samples_per_second": 65.258,
|
| 243 |
+
"eval_steps_per_second": 2.966,
|
| 244 |
+
"step": 46
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 24.0,
|
| 248 |
+
"eval_accuracy": 0.9318181818181818,
|
| 249 |
+
"eval_loss": 0.1767176240682602,
|
| 250 |
+
"eval_runtime": 0.6822,
|
| 251 |
+
"eval_samples_per_second": 64.498,
|
| 252 |
+
"eval_steps_per_second": 2.932,
|
| 253 |
+
"step": 48
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 25.0,
|
| 257 |
+
"grad_norm": 8.298179626464844,
|
| 258 |
+
"learning_rate": 4.166666666666667e-05,
|
| 259 |
+
"loss": 0.2553,
|
| 260 |
+
"step": 50
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 25.0,
|
| 264 |
+
"eval_accuracy": 0.9318181818181818,
|
| 265 |
+
"eval_loss": 0.20396804809570312,
|
| 266 |
+
"eval_runtime": 0.6827,
|
| 267 |
+
"eval_samples_per_second": 64.45,
|
| 268 |
+
"eval_steps_per_second": 2.93,
|
| 269 |
+
"step": 50
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 26.0,
|
| 273 |
+
"eval_accuracy": 0.9090909090909091,
|
| 274 |
+
"eval_loss": 0.1980859339237213,
|
| 275 |
+
"eval_runtime": 0.6738,
|
| 276 |
+
"eval_samples_per_second": 65.297,
|
| 277 |
+
"eval_steps_per_second": 2.968,
|
| 278 |
+
"step": 52
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 27.0,
|
| 282 |
+
"eval_accuracy": 0.9318181818181818,
|
| 283 |
+
"eval_loss": 0.18346600234508514,
|
| 284 |
+
"eval_runtime": 0.6738,
|
| 285 |
+
"eval_samples_per_second": 65.303,
|
| 286 |
+
"eval_steps_per_second": 2.968,
|
| 287 |
+
"step": 54
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 28.0,
|
| 291 |
+
"eval_accuracy": 0.9318181818181818,
|
| 292 |
+
"eval_loss": 0.1820111870765686,
|
| 293 |
+
"eval_runtime": 0.6748,
|
| 294 |
+
"eval_samples_per_second": 65.209,
|
| 295 |
+
"eval_steps_per_second": 2.964,
|
| 296 |
+
"step": 56
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 29.0,
|
| 300 |
+
"eval_accuracy": 0.9545454545454546,
|
| 301 |
+
"eval_loss": 0.1465969681739807,
|
| 302 |
+
"eval_runtime": 0.6762,
|
| 303 |
+
"eval_samples_per_second": 65.067,
|
| 304 |
+
"eval_steps_per_second": 2.958,
|
| 305 |
+
"step": 58
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 30.0,
|
| 309 |
+
"grad_norm": 6.1258087158203125,
|
| 310 |
+
"learning_rate": 3.888888888888889e-05,
|
| 311 |
+
"loss": 0.2083,
|
| 312 |
+
"step": 60
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 30.0,
|
| 316 |
+
"eval_accuracy": 0.9318181818181818,
|
| 317 |
+
"eval_loss": 0.16679802536964417,
|
| 318 |
+
"eval_runtime": 0.6722,
|
| 319 |
+
"eval_samples_per_second": 65.452,
|
| 320 |
+
"eval_steps_per_second": 2.975,
|
| 321 |
+
"step": 60
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 31.0,
|
| 325 |
+
"eval_accuracy": 0.9318181818181818,
|
| 326 |
+
"eval_loss": 0.2228713482618332,
|
| 327 |
+
"eval_runtime": 0.6833,
|
| 328 |
+
"eval_samples_per_second": 64.389,
|
| 329 |
+
"eval_steps_per_second": 2.927,
|
| 330 |
+
"step": 62
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 32.0,
|
| 334 |
+
"eval_accuracy": 0.9545454545454546,
|
| 335 |
+
"eval_loss": 0.17827552556991577,
|
| 336 |
+
"eval_runtime": 0.6819,
|
| 337 |
+
"eval_samples_per_second": 64.521,
|
| 338 |
+
"eval_steps_per_second": 2.933,
|
| 339 |
+
"step": 64
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 33.0,
|
| 343 |
+
"eval_accuracy": 0.8863636363636364,
|
| 344 |
+
"eval_loss": 0.19438548386096954,
|
| 345 |
+
"eval_runtime": 0.6804,
|
| 346 |
+
"eval_samples_per_second": 64.672,
|
| 347 |
+
"eval_steps_per_second": 2.94,
|
| 348 |
+
"step": 66
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 34.0,
|
| 352 |
+
"eval_accuracy": 0.9090909090909091,
|
| 353 |
+
"eval_loss": 0.30252915620803833,
|
| 354 |
+
"eval_runtime": 0.6694,
|
| 355 |
+
"eval_samples_per_second": 65.735,
|
| 356 |
+
"eval_steps_per_second": 2.988,
|
| 357 |
+
"step": 68
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 35.0,
|
| 361 |
+
"grad_norm": 10.506503105163574,
|
| 362 |
+
"learning_rate": 3.611111111111111e-05,
|
| 363 |
+
"loss": 0.2353,
|
| 364 |
+
"step": 70
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 35.0,
|
| 368 |
+
"eval_accuracy": 0.8409090909090909,
|
| 369 |
+
"eval_loss": 0.4457201659679413,
|
| 370 |
+
"eval_runtime": 0.6705,
|
| 371 |
+
"eval_samples_per_second": 65.624,
|
| 372 |
+
"eval_steps_per_second": 2.983,
|
| 373 |
+
"step": 70
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 36.0,
|
| 377 |
+
"eval_accuracy": 0.9318181818181818,
|
| 378 |
+
"eval_loss": 0.27586841583251953,
|
| 379 |
+
"eval_runtime": 0.7004,
|
| 380 |
+
"eval_samples_per_second": 62.824,
|
| 381 |
+
"eval_steps_per_second": 2.856,
|
| 382 |
+
"step": 72
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 37.0,
|
| 386 |
+
"eval_accuracy": 0.9318181818181818,
|
| 387 |
+
"eval_loss": 0.2178523689508438,
|
| 388 |
+
"eval_runtime": 0.705,
|
| 389 |
+
"eval_samples_per_second": 62.41,
|
| 390 |
+
"eval_steps_per_second": 2.837,
|
| 391 |
+
"step": 74
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 38.0,
|
| 395 |
+
"eval_accuracy": 0.9090909090909091,
|
| 396 |
+
"eval_loss": 0.39105215668678284,
|
| 397 |
+
"eval_runtime": 0.7066,
|
| 398 |
+
"eval_samples_per_second": 62.271,
|
| 399 |
+
"eval_steps_per_second": 2.831,
|
| 400 |
+
"step": 76
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 39.0,
|
| 404 |
+
"eval_accuracy": 0.8409090909090909,
|
| 405 |
+
"eval_loss": 0.5785248279571533,
|
| 406 |
+
"eval_runtime": 0.7084,
|
| 407 |
+
"eval_samples_per_second": 62.115,
|
| 408 |
+
"eval_steps_per_second": 2.823,
|
| 409 |
+
"step": 78
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 40.0,
|
| 413 |
+
"grad_norm": 3.042407751083374,
|
| 414 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 415 |
+
"loss": 0.1782,
|
| 416 |
+
"step": 80
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 40.0,
|
| 420 |
+
"eval_accuracy": 0.9318181818181818,
|
| 421 |
+
"eval_loss": 0.2338746190071106,
|
| 422 |
+
"eval_runtime": 0.6767,
|
| 423 |
+
"eval_samples_per_second": 65.017,
|
| 424 |
+
"eval_steps_per_second": 2.955,
|
| 425 |
+
"step": 80
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 41.0,
|
| 429 |
+
"eval_accuracy": 0.9090909090909091,
|
| 430 |
+
"eval_loss": 0.23022708296775818,
|
| 431 |
+
"eval_runtime": 0.6612,
|
| 432 |
+
"eval_samples_per_second": 66.541,
|
| 433 |
+
"eval_steps_per_second": 3.025,
|
| 434 |
+
"step": 82
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 42.0,
|
| 438 |
+
"eval_accuracy": 0.8863636363636364,
|
| 439 |
+
"eval_loss": 0.39668968319892883,
|
| 440 |
+
"eval_runtime": 0.6712,
|
| 441 |
+
"eval_samples_per_second": 65.555,
|
| 442 |
+
"eval_steps_per_second": 2.98,
|
| 443 |
+
"step": 84
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 43.0,
|
| 447 |
+
"eval_accuracy": 0.8636363636363636,
|
| 448 |
+
"eval_loss": 0.4447011649608612,
|
| 449 |
+
"eval_runtime": 0.6775,
|
| 450 |
+
"eval_samples_per_second": 64.94,
|
| 451 |
+
"eval_steps_per_second": 2.952,
|
| 452 |
+
"step": 86
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 44.0,
|
| 456 |
+
"eval_accuracy": 0.9090909090909091,
|
| 457 |
+
"eval_loss": 0.20197413861751556,
|
| 458 |
+
"eval_runtime": 0.6933,
|
| 459 |
+
"eval_samples_per_second": 63.464,
|
| 460 |
+
"eval_steps_per_second": 2.885,
|
| 461 |
+
"step": 88
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 45.0,
|
| 465 |
+
"grad_norm": 6.790353775024414,
|
| 466 |
+
"learning_rate": 3.055555555555556e-05,
|
| 467 |
+
"loss": 0.2059,
|
| 468 |
+
"step": 90
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 45.0,
|
| 472 |
+
"eval_accuracy": 0.9318181818181818,
|
| 473 |
+
"eval_loss": 0.191122904419899,
|
| 474 |
+
"eval_runtime": 0.6988,
|
| 475 |
+
"eval_samples_per_second": 62.967,
|
| 476 |
+
"eval_steps_per_second": 2.862,
|
| 477 |
+
"step": 90
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 46.0,
|
| 481 |
+
"eval_accuracy": 0.9090909090909091,
|
| 482 |
+
"eval_loss": 0.26089540123939514,
|
| 483 |
+
"eval_runtime": 0.6777,
|
| 484 |
+
"eval_samples_per_second": 64.925,
|
| 485 |
+
"eval_steps_per_second": 2.951,
|
| 486 |
+
"step": 92
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 47.0,
|
| 490 |
+
"eval_accuracy": 0.9090909090909091,
|
| 491 |
+
"eval_loss": 0.2924693822860718,
|
| 492 |
+
"eval_runtime": 0.6842,
|
| 493 |
+
"eval_samples_per_second": 64.304,
|
| 494 |
+
"eval_steps_per_second": 2.923,
|
| 495 |
+
"step": 94
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 48.0,
|
| 499 |
+
"eval_accuracy": 0.9318181818181818,
|
| 500 |
+
"eval_loss": 0.2078903764486313,
|
| 501 |
+
"eval_runtime": 0.693,
|
| 502 |
+
"eval_samples_per_second": 63.488,
|
| 503 |
+
"eval_steps_per_second": 2.886,
|
| 504 |
+
"step": 96
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 49.0,
|
| 508 |
+
"eval_accuracy": 0.9318181818181818,
|
| 509 |
+
"eval_loss": 0.18529368937015533,
|
| 510 |
+
"eval_runtime": 0.6921,
|
| 511 |
+
"eval_samples_per_second": 63.572,
|
| 512 |
+
"eval_steps_per_second": 2.89,
|
| 513 |
+
"step": 98
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 50.0,
|
| 517 |
+
"grad_norm": 4.374385833740234,
|
| 518 |
+
"learning_rate": 2.777777777777778e-05,
|
| 519 |
+
"loss": 0.1706,
|
| 520 |
+
"step": 100
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 50.0,
|
| 524 |
+
"eval_accuracy": 0.9318181818181818,
|
| 525 |
+
"eval_loss": 0.2860322594642639,
|
| 526 |
+
"eval_runtime": 0.697,
|
| 527 |
+
"eval_samples_per_second": 63.13,
|
| 528 |
+
"eval_steps_per_second": 2.87,
|
| 529 |
+
"step": 100
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 51.0,
|
| 533 |
+
"eval_accuracy": 0.8636363636363636,
|
| 534 |
+
"eval_loss": 0.3734706938266754,
|
| 535 |
+
"eval_runtime": 0.6878,
|
| 536 |
+
"eval_samples_per_second": 63.971,
|
| 537 |
+
"eval_steps_per_second": 2.908,
|
| 538 |
+
"step": 102
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 52.0,
|
| 542 |
+
"eval_accuracy": 0.9318181818181818,
|
| 543 |
+
"eval_loss": 0.19677944481372833,
|
| 544 |
+
"eval_runtime": 0.6875,
|
| 545 |
+
"eval_samples_per_second": 63.996,
|
| 546 |
+
"eval_steps_per_second": 2.909,
|
| 547 |
+
"step": 104
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 53.0,
|
| 551 |
+
"eval_accuracy": 0.9318181818181818,
|
| 552 |
+
"eval_loss": 0.17221951484680176,
|
| 553 |
+
"eval_runtime": 0.679,
|
| 554 |
+
"eval_samples_per_second": 64.798,
|
| 555 |
+
"eval_steps_per_second": 2.945,
|
| 556 |
+
"step": 106
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 54.0,
|
| 560 |
+
"eval_accuracy": 0.8636363636363636,
|
| 561 |
+
"eval_loss": 0.312281996011734,
|
| 562 |
+
"eval_runtime": 0.6832,
|
| 563 |
+
"eval_samples_per_second": 64.399,
|
| 564 |
+
"eval_steps_per_second": 2.927,
|
| 565 |
+
"step": 108
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 55.0,
|
| 569 |
+
"grad_norm": 3.807600259780884,
|
| 570 |
+
"learning_rate": 2.5e-05,
|
| 571 |
+
"loss": 0.1429,
|
| 572 |
+
"step": 110
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 55.0,
|
| 576 |
+
"eval_accuracy": 0.8863636363636364,
|
| 577 |
+
"eval_loss": 0.32969653606414795,
|
| 578 |
+
"eval_runtime": 0.6845,
|
| 579 |
+
"eval_samples_per_second": 64.284,
|
| 580 |
+
"eval_steps_per_second": 2.922,
|
| 581 |
+
"step": 110
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 56.0,
|
| 585 |
+
"eval_accuracy": 0.9772727272727273,
|
| 586 |
+
"eval_loss": 0.14295423030853271,
|
| 587 |
+
"eval_runtime": 0.6744,
|
| 588 |
+
"eval_samples_per_second": 65.244,
|
| 589 |
+
"eval_steps_per_second": 2.966,
|
| 590 |
+
"step": 112
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 57.0,
|
| 594 |
+
"eval_accuracy": 0.9772727272727273,
|
| 595 |
+
"eval_loss": 0.11342933028936386,
|
| 596 |
+
"eval_runtime": 0.6865,
|
| 597 |
+
"eval_samples_per_second": 64.09,
|
| 598 |
+
"eval_steps_per_second": 2.913,
|
| 599 |
+
"step": 114
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 58.0,
|
| 603 |
+
"eval_accuracy": 0.9090909090909091,
|
| 604 |
+
"eval_loss": 0.231190487742424,
|
| 605 |
+
"eval_runtime": 0.6772,
|
| 606 |
+
"eval_samples_per_second": 64.97,
|
| 607 |
+
"eval_steps_per_second": 2.953,
|
| 608 |
+
"step": 116
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 59.0,
|
| 612 |
+
"eval_accuracy": 0.9090909090909091,
|
| 613 |
+
"eval_loss": 0.28261473774909973,
|
| 614 |
+
"eval_runtime": 0.6857,
|
| 615 |
+
"eval_samples_per_second": 64.163,
|
| 616 |
+
"eval_steps_per_second": 2.917,
|
| 617 |
+
"step": 118
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 60.0,
|
| 621 |
+
"grad_norm": 3.425337314605713,
|
| 622 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 623 |
+
"loss": 0.1325,
|
| 624 |
+
"step": 120
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 60.0,
|
| 628 |
+
"eval_accuracy": 0.9090909090909091,
|
| 629 |
+
"eval_loss": 0.2417471557855606,
|
| 630 |
+
"eval_runtime": 0.6829,
|
| 631 |
+
"eval_samples_per_second": 64.43,
|
| 632 |
+
"eval_steps_per_second": 2.929,
|
| 633 |
+
"step": 120
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 61.0,
|
| 637 |
+
"eval_accuracy": 0.9318181818181818,
|
| 638 |
+
"eval_loss": 0.139329731464386,
|
| 639 |
+
"eval_runtime": 0.6819,
|
| 640 |
+
"eval_samples_per_second": 64.521,
|
| 641 |
+
"eval_steps_per_second": 2.933,
|
| 642 |
+
"step": 122
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 62.0,
|
| 646 |
+
"eval_accuracy": 0.9318181818181818,
|
| 647 |
+
"eval_loss": 0.21779733896255493,
|
| 648 |
+
"eval_runtime": 0.6874,
|
| 649 |
+
"eval_samples_per_second": 64.008,
|
| 650 |
+
"eval_steps_per_second": 2.909,
|
| 651 |
+
"step": 124
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 63.0,
|
| 655 |
+
"eval_accuracy": 0.9090909090909091,
|
| 656 |
+
"eval_loss": 0.3990946114063263,
|
| 657 |
+
"eval_runtime": 0.6762,
|
| 658 |
+
"eval_samples_per_second": 65.068,
|
| 659 |
+
"eval_steps_per_second": 2.958,
|
| 660 |
+
"step": 126
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 64.0,
|
| 664 |
+
"eval_accuracy": 0.9090909090909091,
|
| 665 |
+
"eval_loss": 0.33248183131217957,
|
| 666 |
+
"eval_runtime": 0.6851,
|
| 667 |
+
"eval_samples_per_second": 64.222,
|
| 668 |
+
"eval_steps_per_second": 2.919,
|
| 669 |
+
"step": 128
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 65.0,
|
| 673 |
+
"grad_norm": 3.3198602199554443,
|
| 674 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 675 |
+
"loss": 0.1481,
|
| 676 |
+
"step": 130
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 65.0,
|
| 680 |
+
"eval_accuracy": 0.9090909090909091,
|
| 681 |
+
"eval_loss": 0.23272232711315155,
|
| 682 |
+
"eval_runtime": 0.6764,
|
| 683 |
+
"eval_samples_per_second": 65.054,
|
| 684 |
+
"eval_steps_per_second": 2.957,
|
| 685 |
+
"step": 130
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 66.0,
|
| 689 |
+
"eval_accuracy": 0.9090909090909091,
|
| 690 |
+
"eval_loss": 0.28850874304771423,
|
| 691 |
+
"eval_runtime": 0.6864,
|
| 692 |
+
"eval_samples_per_second": 64.098,
|
| 693 |
+
"eval_steps_per_second": 2.914,
|
| 694 |
+
"step": 132
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 67.0,
|
| 698 |
+
"eval_accuracy": 0.9090909090909091,
|
| 699 |
+
"eval_loss": 0.3575742840766907,
|
| 700 |
+
"eval_runtime": 0.692,
|
| 701 |
+
"eval_samples_per_second": 63.587,
|
| 702 |
+
"eval_steps_per_second": 2.89,
|
| 703 |
+
"step": 134
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 68.0,
|
| 707 |
+
"eval_accuracy": 0.9318181818181818,
|
| 708 |
+
"eval_loss": 0.26858440041542053,
|
| 709 |
+
"eval_runtime": 0.6743,
|
| 710 |
+
"eval_samples_per_second": 65.257,
|
| 711 |
+
"eval_steps_per_second": 2.966,
|
| 712 |
+
"step": 136
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 69.0,
|
| 716 |
+
"eval_accuracy": 0.9545454545454546,
|
| 717 |
+
"eval_loss": 0.1716589480638504,
|
| 718 |
+
"eval_runtime": 0.6829,
|
| 719 |
+
"eval_samples_per_second": 64.43,
|
| 720 |
+
"eval_steps_per_second": 2.929,
|
| 721 |
+
"step": 138
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 70.0,
|
| 725 |
+
"grad_norm": 4.273709774017334,
|
| 726 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 727 |
+
"loss": 0.1237,
|
| 728 |
+
"step": 140
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 70.0,
|
| 732 |
+
"eval_accuracy": 0.9545454545454546,
|
| 733 |
+
"eval_loss": 0.14934584498405457,
|
| 734 |
+
"eval_runtime": 0.6712,
|
| 735 |
+
"eval_samples_per_second": 65.551,
|
| 736 |
+
"eval_steps_per_second": 2.98,
|
| 737 |
+
"step": 140
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 71.0,
|
| 741 |
+
"eval_accuracy": 0.9318181818181818,
|
| 742 |
+
"eval_loss": 0.1428576558828354,
|
| 743 |
+
"eval_runtime": 0.6789,
|
| 744 |
+
"eval_samples_per_second": 64.814,
|
| 745 |
+
"eval_steps_per_second": 2.946,
|
| 746 |
+
"step": 142
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 72.0,
|
| 750 |
+
"eval_accuracy": 0.9318181818181818,
|
| 751 |
+
"eval_loss": 0.1789616346359253,
|
| 752 |
+
"eval_runtime": 0.6831,
|
| 753 |
+
"eval_samples_per_second": 64.412,
|
| 754 |
+
"eval_steps_per_second": 2.928,
|
| 755 |
+
"step": 144
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 73.0,
|
| 759 |
+
"eval_accuracy": 0.9318181818181818,
|
| 760 |
+
"eval_loss": 0.1590358167886734,
|
| 761 |
+
"eval_runtime": 0.6785,
|
| 762 |
+
"eval_samples_per_second": 64.853,
|
| 763 |
+
"eval_steps_per_second": 2.948,
|
| 764 |
+
"step": 146
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 74.0,
|
| 768 |
+
"eval_accuracy": 0.8863636363636364,
|
| 769 |
+
"eval_loss": 0.19707775115966797,
|
| 770 |
+
"eval_runtime": 0.6779,
|
| 771 |
+
"eval_samples_per_second": 64.91,
|
| 772 |
+
"eval_steps_per_second": 2.95,
|
| 773 |
+
"step": 148
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 75.0,
|
| 777 |
+
"grad_norm": 3.867281436920166,
|
| 778 |
+
"learning_rate": 1.388888888888889e-05,
|
| 779 |
+
"loss": 0.105,
|
| 780 |
+
"step": 150
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 75.0,
|
| 784 |
+
"eval_accuracy": 0.9318181818181818,
|
| 785 |
+
"eval_loss": 0.22291280329227448,
|
| 786 |
+
"eval_runtime": 0.6754,
|
| 787 |
+
"eval_samples_per_second": 65.149,
|
| 788 |
+
"eval_steps_per_second": 2.961,
|
| 789 |
+
"step": 150
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 76.0,
|
| 793 |
+
"eval_accuracy": 0.8863636363636364,
|
| 794 |
+
"eval_loss": 0.17888131737709045,
|
| 795 |
+
"eval_runtime": 0.6771,
|
| 796 |
+
"eval_samples_per_second": 64.982,
|
| 797 |
+
"eval_steps_per_second": 2.954,
|
| 798 |
+
"step": 152
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 77.0,
|
| 802 |
+
"eval_accuracy": 0.9545454545454546,
|
| 803 |
+
"eval_loss": 0.16707843542099,
|
| 804 |
+
"eval_runtime": 0.6843,
|
| 805 |
+
"eval_samples_per_second": 64.299,
|
| 806 |
+
"eval_steps_per_second": 2.923,
|
| 807 |
+
"step": 154
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 78.0,
|
| 811 |
+
"eval_accuracy": 0.9318181818181818,
|
| 812 |
+
"eval_loss": 0.24351829290390015,
|
| 813 |
+
"eval_runtime": 0.6914,
|
| 814 |
+
"eval_samples_per_second": 63.642,
|
| 815 |
+
"eval_steps_per_second": 2.893,
|
| 816 |
+
"step": 156
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 79.0,
|
| 820 |
+
"eval_accuracy": 0.9318181818181818,
|
| 821 |
+
"eval_loss": 0.2658483386039734,
|
| 822 |
+
"eval_runtime": 0.6856,
|
| 823 |
+
"eval_samples_per_second": 64.176,
|
| 824 |
+
"eval_steps_per_second": 2.917,
|
| 825 |
+
"step": 158
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 80.0,
|
| 829 |
+
"grad_norm": 2.501020908355713,
|
| 830 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 831 |
+
"loss": 0.0923,
|
| 832 |
+
"step": 160
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 80.0,
|
| 836 |
+
"eval_accuracy": 0.9318181818181818,
|
| 837 |
+
"eval_loss": 0.20919661223888397,
|
| 838 |
+
"eval_runtime": 0.6861,
|
| 839 |
+
"eval_samples_per_second": 64.132,
|
| 840 |
+
"eval_steps_per_second": 2.915,
|
| 841 |
+
"step": 160
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 81.0,
|
| 845 |
+
"eval_accuracy": 0.9318181818181818,
|
| 846 |
+
"eval_loss": 0.17477592825889587,
|
| 847 |
+
"eval_runtime": 0.6775,
|
| 848 |
+
"eval_samples_per_second": 64.946,
|
| 849 |
+
"eval_steps_per_second": 2.952,
|
| 850 |
+
"step": 162
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 82.0,
|
| 854 |
+
"eval_accuracy": 0.9318181818181818,
|
| 855 |
+
"eval_loss": 0.1727462261915207,
|
| 856 |
+
"eval_runtime": 0.6757,
|
| 857 |
+
"eval_samples_per_second": 65.116,
|
| 858 |
+
"eval_steps_per_second": 2.96,
|
| 859 |
+
"step": 164
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 83.0,
|
| 863 |
+
"eval_accuracy": 0.9090909090909091,
|
| 864 |
+
"eval_loss": 0.1945178508758545,
|
| 865 |
+
"eval_runtime": 0.6723,
|
| 866 |
+
"eval_samples_per_second": 65.443,
|
| 867 |
+
"eval_steps_per_second": 2.975,
|
| 868 |
+
"step": 166
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 84.0,
|
| 872 |
+
"eval_accuracy": 0.9318181818181818,
|
| 873 |
+
"eval_loss": 0.24294686317443848,
|
| 874 |
+
"eval_runtime": 0.6772,
|
| 875 |
+
"eval_samples_per_second": 64.978,
|
| 876 |
+
"eval_steps_per_second": 2.954,
|
| 877 |
+
"step": 168
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 85.0,
|
| 881 |
+
"grad_norm": 3.9196958541870117,
|
| 882 |
+
"learning_rate": 8.333333333333334e-06,
|
| 883 |
+
"loss": 0.1033,
|
| 884 |
+
"step": 170
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 85.0,
|
| 888 |
+
"eval_accuracy": 0.9318181818181818,
|
| 889 |
+
"eval_loss": 0.2795504331588745,
|
| 890 |
+
"eval_runtime": 0.6951,
|
| 891 |
+
"eval_samples_per_second": 63.299,
|
| 892 |
+
"eval_steps_per_second": 2.877,
|
| 893 |
+
"step": 170
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 86.0,
|
| 897 |
+
"eval_accuracy": 0.9318181818181818,
|
| 898 |
+
"eval_loss": 0.2548198401927948,
|
| 899 |
+
"eval_runtime": 0.6876,
|
| 900 |
+
"eval_samples_per_second": 63.99,
|
| 901 |
+
"eval_steps_per_second": 2.909,
|
| 902 |
+
"step": 172
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 87.0,
|
| 906 |
+
"eval_accuracy": 0.9090909090909091,
|
| 907 |
+
"eval_loss": 0.2379317283630371,
|
| 908 |
+
"eval_runtime": 0.6842,
|
| 909 |
+
"eval_samples_per_second": 64.312,
|
| 910 |
+
"eval_steps_per_second": 2.923,
|
| 911 |
+
"step": 174
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 88.0,
|
| 915 |
+
"eval_accuracy": 0.9090909090909091,
|
| 916 |
+
"eval_loss": 0.2408948391675949,
|
| 917 |
+
"eval_runtime": 0.6803,
|
| 918 |
+
"eval_samples_per_second": 64.68,
|
| 919 |
+
"eval_steps_per_second": 2.94,
|
| 920 |
+
"step": 176
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 89.0,
|
| 924 |
+
"eval_accuracy": 0.9090909090909091,
|
| 925 |
+
"eval_loss": 0.24209485948085785,
|
| 926 |
+
"eval_runtime": 0.6754,
|
| 927 |
+
"eval_samples_per_second": 65.142,
|
| 928 |
+
"eval_steps_per_second": 2.961,
|
| 929 |
+
"step": 178
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 90.0,
|
| 933 |
+
"grad_norm": 2.0967917442321777,
|
| 934 |
+
"learning_rate": 5.555555555555556e-06,
|
| 935 |
+
"loss": 0.1073,
|
| 936 |
+
"step": 180
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 90.0,
|
| 940 |
+
"eval_accuracy": 0.9090909090909091,
|
| 941 |
+
"eval_loss": 0.23317375779151917,
|
| 942 |
+
"eval_runtime": 0.6747,
|
| 943 |
+
"eval_samples_per_second": 65.214,
|
| 944 |
+
"eval_steps_per_second": 2.964,
|
| 945 |
+
"step": 180
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 91.0,
|
| 949 |
+
"eval_accuracy": 0.9090909090909091,
|
| 950 |
+
"eval_loss": 0.22311018407344818,
|
| 951 |
+
"eval_runtime": 0.6758,
|
| 952 |
+
"eval_samples_per_second": 65.111,
|
| 953 |
+
"eval_steps_per_second": 2.96,
|
| 954 |
+
"step": 182
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 92.0,
|
| 958 |
+
"eval_accuracy": 0.9318181818181818,
|
| 959 |
+
"eval_loss": 0.21525071561336517,
|
| 960 |
+
"eval_runtime": 0.6873,
|
| 961 |
+
"eval_samples_per_second": 64.018,
|
| 962 |
+
"eval_steps_per_second": 2.91,
|
| 963 |
+
"step": 184
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 93.0,
|
| 967 |
+
"eval_accuracy": 0.9318181818181818,
|
| 968 |
+
"eval_loss": 0.20876212418079376,
|
| 969 |
+
"eval_runtime": 0.6844,
|
| 970 |
+
"eval_samples_per_second": 64.295,
|
| 971 |
+
"eval_steps_per_second": 2.922,
|
| 972 |
+
"step": 186
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 94.0,
|
| 976 |
+
"eval_accuracy": 0.9318181818181818,
|
| 977 |
+
"eval_loss": 0.20578283071517944,
|
| 978 |
+
"eval_runtime": 0.6983,
|
| 979 |
+
"eval_samples_per_second": 63.011,
|
| 980 |
+
"eval_steps_per_second": 2.864,
|
| 981 |
+
"step": 188
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 95.0,
|
| 985 |
+
"grad_norm": 2.9749207496643066,
|
| 986 |
+
"learning_rate": 2.777777777777778e-06,
|
| 987 |
+
"loss": 0.104,
|
| 988 |
+
"step": 190
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 95.0,
|
| 992 |
+
"eval_accuracy": 0.9318181818181818,
|
| 993 |
+
"eval_loss": 0.20398248732089996,
|
| 994 |
+
"eval_runtime": 0.6745,
|
| 995 |
+
"eval_samples_per_second": 65.237,
|
| 996 |
+
"eval_steps_per_second": 2.965,
|
| 997 |
+
"step": 190
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 96.0,
|
| 1001 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1002 |
+
"eval_loss": 0.20461073517799377,
|
| 1003 |
+
"eval_runtime": 0.673,
|
| 1004 |
+
"eval_samples_per_second": 65.381,
|
| 1005 |
+
"eval_steps_per_second": 2.972,
|
| 1006 |
+
"step": 192
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 97.0,
|
| 1010 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1011 |
+
"eval_loss": 0.20431600511074066,
|
| 1012 |
+
"eval_runtime": 0.6746,
|
| 1013 |
+
"eval_samples_per_second": 65.228,
|
| 1014 |
+
"eval_steps_per_second": 2.965,
|
| 1015 |
+
"step": 194
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 98.0,
|
| 1019 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1020 |
+
"eval_loss": 0.20564264059066772,
|
| 1021 |
+
"eval_runtime": 0.6795,
|
| 1022 |
+
"eval_samples_per_second": 64.751,
|
| 1023 |
+
"eval_steps_per_second": 2.943,
|
| 1024 |
+
"step": 196
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 99.0,
|
| 1028 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1029 |
+
"eval_loss": 0.20811475813388824,
|
| 1030 |
+
"eval_runtime": 0.6891,
|
| 1031 |
+
"eval_samples_per_second": 63.849,
|
| 1032 |
+
"eval_steps_per_second": 2.902,
|
| 1033 |
+
"step": 198
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 100.0,
|
| 1037 |
+
"grad_norm": 2.3249521255493164,
|
| 1038 |
+
"learning_rate": 0.0,
|
| 1039 |
+
"loss": 0.0896,
|
| 1040 |
+
"step": 200
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 100.0,
|
| 1044 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1045 |
+
"eval_loss": 0.2097294181585312,
|
| 1046 |
+
"eval_runtime": 0.686,
|
| 1047 |
+
"eval_samples_per_second": 64.143,
|
| 1048 |
+
"eval_steps_per_second": 2.916,
|
| 1049 |
+
"step": 200
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 100.0,
|
| 1053 |
+
"step": 200,
|
| 1054 |
+
"total_flos": 1.8822505078960128e+18,
|
| 1055 |
+
"train_loss": 0.21107831209897995,
|
| 1056 |
+
"train_runtime": 1239.3905,
|
| 1057 |
+
"train_samples_per_second": 19.606,
|
| 1058 |
+
"train_steps_per_second": 0.161
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 100.0,
|
| 1062 |
+
"eval_accuracy": 0.9772727272727273,
|
| 1063 |
+
"eval_loss": 0.14295423030853271,
|
| 1064 |
+
"eval_runtime": 0.7189,
|
| 1065 |
+
"eval_samples_per_second": 61.203,
|
| 1066 |
+
"eval_steps_per_second": 2.782,
|
| 1067 |
+
"step": 200
|
| 1068 |
+
}
|
| 1069 |
+
],
|
| 1070 |
+
"logging_steps": 10,
|
| 1071 |
+
"max_steps": 200,
|
| 1072 |
+
"num_input_tokens_seen": 0,
|
| 1073 |
+
"num_train_epochs": 100,
|
| 1074 |
+
"save_steps": 500,
|
| 1075 |
+
"total_flos": 1.8822505078960128e+18,
|
| 1076 |
+
"train_batch_size": 32,
|
| 1077 |
+
"trial_name": null,
|
| 1078 |
+
"trial_params": null
|
| 1079 |
+
}
|