End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May23_06-57-00_cba75054b839/events.out.tfevents.1747984700.cba75054b839.2385.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1135 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.5977
|
| 22 |
+
- Accuracy: 0.9029
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 56.0,
|
| 3 |
+
"eval_accuracy": 0.9028571428571428,
|
| 4 |
+
"eval_loss": 0.597748875617981,
|
| 5 |
+
"eval_runtime": 1.9786,
|
| 6 |
+
"eval_samples_per_second": 88.445,
|
| 7 |
+
"eval_steps_per_second": 5.559,
|
| 8 |
+
"total_flos": 1.7108328318259692e+18,
|
| 9 |
+
"train_loss": 0.19382851386354083,
|
| 10 |
+
"train_runtime": 1219.347,
|
| 11 |
+
"train_samples_per_second": 46.205,
|
| 12 |
+
"train_steps_per_second": 0.689
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 56.0,
|
| 3 |
+
"eval_accuracy": 0.9028571428571428,
|
| 4 |
+
"eval_loss": 0.597748875617981,
|
| 5 |
+
"eval_runtime": 1.9786,
|
| 6 |
+
"eval_samples_per_second": 88.445,
|
| 7 |
+
"eval_steps_per_second": 5.559
|
| 8 |
+
}
|
runs/May23_06-57-00_cba75054b839/events.out.tfevents.1747984700.cba75054b839.2385.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9f2c5aa30d6a473d28062d5e548e57e195db3d97f5aec9feeefb9177ac54c96
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 56.0,
|
| 3 |
+
"total_flos": 1.7108328318259692e+18,
|
| 4 |
+
"train_loss": 0.19382851386354083,
|
| 5 |
+
"train_runtime": 1219.347,
|
| 6 |
+
"train_samples_per_second": 46.205,
|
| 7 |
+
"train_steps_per_second": 0.689
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1135 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 750,
|
| 3 |
+
"best_metric": 0.9028571428571428,
|
| 4 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV71/checkpoint-750",
|
| 5 |
+
"epoch": 56.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 840,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.6779661016949152,
|
| 14 |
+
"grad_norm": 3.3897862434387207,
|
| 15 |
+
"learning_rate": 6.428571428571429e-06,
|
| 16 |
+
"loss": 1.1246,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.0,
|
| 21 |
+
"eval_accuracy": 0.41714285714285715,
|
| 22 |
+
"eval_loss": 1.05587899684906,
|
| 23 |
+
"eval_runtime": 2.0285,
|
| 24 |
+
"eval_samples_per_second": 86.269,
|
| 25 |
+
"eval_steps_per_second": 5.423,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.3389830508474576,
|
| 30 |
+
"grad_norm": 4.0562825202941895,
|
| 31 |
+
"learning_rate": 1.3571428571428572e-05,
|
| 32 |
+
"loss": 1.0075,
|
| 33 |
+
"step": 20
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"grad_norm": 4.147009372711182,
|
| 38 |
+
"learning_rate": 2.0714285714285715e-05,
|
| 39 |
+
"loss": 0.8728,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.0,
|
| 44 |
+
"eval_accuracy": 0.6971428571428572,
|
| 45 |
+
"eval_loss": 0.746044933795929,
|
| 46 |
+
"eval_runtime": 2.3392,
|
| 47 |
+
"eval_samples_per_second": 74.811,
|
| 48 |
+
"eval_steps_per_second": 4.702,
|
| 49 |
+
"step": 30
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 2.6779661016949152,
|
| 53 |
+
"grad_norm": 8.447574615478516,
|
| 54 |
+
"learning_rate": 2.7857142857142858e-05,
|
| 55 |
+
"loss": 0.6663,
|
| 56 |
+
"step": 40
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.0,
|
| 60 |
+
"eval_accuracy": 0.8057142857142857,
|
| 61 |
+
"eval_loss": 0.4562816619873047,
|
| 62 |
+
"eval_runtime": 1.9074,
|
| 63 |
+
"eval_samples_per_second": 91.749,
|
| 64 |
+
"eval_steps_per_second": 5.767,
|
| 65 |
+
"step": 45
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 3.3389830508474576,
|
| 69 |
+
"grad_norm": 11.307677268981934,
|
| 70 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 71 |
+
"loss": 0.5432,
|
| 72 |
+
"step": 50
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.0,
|
| 76 |
+
"grad_norm": 8.582294464111328,
|
| 77 |
+
"learning_rate": 4.214285714285714e-05,
|
| 78 |
+
"loss": 0.4632,
|
| 79 |
+
"step": 60
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 4.0,
|
| 83 |
+
"eval_accuracy": 0.8285714285714286,
|
| 84 |
+
"eval_loss": 0.40758973360061646,
|
| 85 |
+
"eval_runtime": 1.8904,
|
| 86 |
+
"eval_samples_per_second": 92.573,
|
| 87 |
+
"eval_steps_per_second": 5.819,
|
| 88 |
+
"step": 60
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 4.677966101694915,
|
| 92 |
+
"grad_norm": 9.387711524963379,
|
| 93 |
+
"learning_rate": 4.928571428571428e-05,
|
| 94 |
+
"loss": 0.4278,
|
| 95 |
+
"step": 70
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 5.0,
|
| 99 |
+
"eval_accuracy": 0.84,
|
| 100 |
+
"eval_loss": 0.3669876456260681,
|
| 101 |
+
"eval_runtime": 2.6198,
|
| 102 |
+
"eval_samples_per_second": 66.799,
|
| 103 |
+
"eval_steps_per_second": 4.199,
|
| 104 |
+
"step": 75
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 5.338983050847458,
|
| 108 |
+
"grad_norm": 8.6371431350708,
|
| 109 |
+
"learning_rate": 5.642857142857143e-05,
|
| 110 |
+
"loss": 0.3608,
|
| 111 |
+
"step": 80
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.0,
|
| 115 |
+
"grad_norm": 11.995199203491211,
|
| 116 |
+
"learning_rate": 5.96031746031746e-05,
|
| 117 |
+
"loss": 0.361,
|
| 118 |
+
"step": 90
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 6.0,
|
| 122 |
+
"eval_accuracy": 0.8457142857142858,
|
| 123 |
+
"eval_loss": 0.36241406202316284,
|
| 124 |
+
"eval_runtime": 1.9339,
|
| 125 |
+
"eval_samples_per_second": 90.49,
|
| 126 |
+
"eval_steps_per_second": 5.688,
|
| 127 |
+
"step": 90
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 6.677966101694915,
|
| 131 |
+
"grad_norm": 10.142115592956543,
|
| 132 |
+
"learning_rate": 5.880952380952381e-05,
|
| 133 |
+
"loss": 0.3742,
|
| 134 |
+
"step": 100
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 7.0,
|
| 138 |
+
"eval_accuracy": 0.8628571428571429,
|
| 139 |
+
"eval_loss": 0.3504450023174286,
|
| 140 |
+
"eval_runtime": 1.8843,
|
| 141 |
+
"eval_samples_per_second": 92.872,
|
| 142 |
+
"eval_steps_per_second": 5.838,
|
| 143 |
+
"step": 105
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 7.338983050847458,
|
| 147 |
+
"grad_norm": 11.923420906066895,
|
| 148 |
+
"learning_rate": 5.801587301587302e-05,
|
| 149 |
+
"loss": 0.2351,
|
| 150 |
+
"step": 110
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 8.0,
|
| 154 |
+
"grad_norm": 4.905794620513916,
|
| 155 |
+
"learning_rate": 5.722222222222223e-05,
|
| 156 |
+
"loss": 0.3313,
|
| 157 |
+
"step": 120
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 8.0,
|
| 161 |
+
"eval_accuracy": 0.8628571428571429,
|
| 162 |
+
"eval_loss": 0.2962282598018646,
|
| 163 |
+
"eval_runtime": 2.4213,
|
| 164 |
+
"eval_samples_per_second": 72.274,
|
| 165 |
+
"eval_steps_per_second": 4.543,
|
| 166 |
+
"step": 120
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 8.677966101694915,
|
| 170 |
+
"grad_norm": 8.893570899963379,
|
| 171 |
+
"learning_rate": 5.642857142857143e-05,
|
| 172 |
+
"loss": 0.2977,
|
| 173 |
+
"step": 130
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 9.0,
|
| 177 |
+
"eval_accuracy": 0.8685714285714285,
|
| 178 |
+
"eval_loss": 0.33206191658973694,
|
| 179 |
+
"eval_runtime": 1.9246,
|
| 180 |
+
"eval_samples_per_second": 90.929,
|
| 181 |
+
"eval_steps_per_second": 5.716,
|
| 182 |
+
"step": 135
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 9.338983050847457,
|
| 186 |
+
"grad_norm": 12.26343822479248,
|
| 187 |
+
"learning_rate": 5.563492063492064e-05,
|
| 188 |
+
"loss": 0.3156,
|
| 189 |
+
"step": 140
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 10.0,
|
| 193 |
+
"grad_norm": 6.5488362312316895,
|
| 194 |
+
"learning_rate": 5.4841269841269845e-05,
|
| 195 |
+
"loss": 0.2589,
|
| 196 |
+
"step": 150
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 10.0,
|
| 200 |
+
"eval_accuracy": 0.8628571428571429,
|
| 201 |
+
"eval_loss": 0.34251242876052856,
|
| 202 |
+
"eval_runtime": 1.9199,
|
| 203 |
+
"eval_samples_per_second": 91.152,
|
| 204 |
+
"eval_steps_per_second": 5.73,
|
| 205 |
+
"step": 150
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 10.677966101694915,
|
| 209 |
+
"grad_norm": 5.538010120391846,
|
| 210 |
+
"learning_rate": 5.404761904761905e-05,
|
| 211 |
+
"loss": 0.2477,
|
| 212 |
+
"step": 160
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 11.0,
|
| 216 |
+
"eval_accuracy": 0.8457142857142858,
|
| 217 |
+
"eval_loss": 0.39819851517677307,
|
| 218 |
+
"eval_runtime": 1.9172,
|
| 219 |
+
"eval_samples_per_second": 91.281,
|
| 220 |
+
"eval_steps_per_second": 5.738,
|
| 221 |
+
"step": 165
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 11.338983050847457,
|
| 225 |
+
"grad_norm": 6.982760906219482,
|
| 226 |
+
"learning_rate": 5.333333333333333e-05,
|
| 227 |
+
"loss": 0.2527,
|
| 228 |
+
"step": 170
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 12.0,
|
| 232 |
+
"grad_norm": 8.375303268432617,
|
| 233 |
+
"learning_rate": 5.253968253968254e-05,
|
| 234 |
+
"loss": 0.2187,
|
| 235 |
+
"step": 180
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 12.0,
|
| 239 |
+
"eval_accuracy": 0.8514285714285714,
|
| 240 |
+
"eval_loss": 0.5953956246376038,
|
| 241 |
+
"eval_runtime": 2.0286,
|
| 242 |
+
"eval_samples_per_second": 86.268,
|
| 243 |
+
"eval_steps_per_second": 5.423,
|
| 244 |
+
"step": 180
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 12.677966101694915,
|
| 248 |
+
"grad_norm": 7.4971723556518555,
|
| 249 |
+
"learning_rate": 5.174603174603175e-05,
|
| 250 |
+
"loss": 0.2342,
|
| 251 |
+
"step": 190
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 13.0,
|
| 255 |
+
"eval_accuracy": 0.8514285714285714,
|
| 256 |
+
"eval_loss": 0.3745245933532715,
|
| 257 |
+
"eval_runtime": 1.8701,
|
| 258 |
+
"eval_samples_per_second": 93.58,
|
| 259 |
+
"eval_steps_per_second": 5.882,
|
| 260 |
+
"step": 195
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 13.338983050847457,
|
| 264 |
+
"grad_norm": 6.758434772491455,
|
| 265 |
+
"learning_rate": 5.095238095238095e-05,
|
| 266 |
+
"loss": 0.2354,
|
| 267 |
+
"step": 200
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 14.0,
|
| 271 |
+
"grad_norm": 4.655900001525879,
|
| 272 |
+
"learning_rate": 5.015873015873016e-05,
|
| 273 |
+
"loss": 0.2444,
|
| 274 |
+
"step": 210
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 14.0,
|
| 278 |
+
"eval_accuracy": 0.8628571428571429,
|
| 279 |
+
"eval_loss": 0.5219993591308594,
|
| 280 |
+
"eval_runtime": 2.5324,
|
| 281 |
+
"eval_samples_per_second": 69.106,
|
| 282 |
+
"eval_steps_per_second": 4.344,
|
| 283 |
+
"step": 210
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 14.677966101694915,
|
| 287 |
+
"grad_norm": 8.788654327392578,
|
| 288 |
+
"learning_rate": 4.936507936507937e-05,
|
| 289 |
+
"loss": 0.2067,
|
| 290 |
+
"step": 220
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 15.0,
|
| 294 |
+
"eval_accuracy": 0.8457142857142858,
|
| 295 |
+
"eval_loss": 0.44333723187446594,
|
| 296 |
+
"eval_runtime": 1.9312,
|
| 297 |
+
"eval_samples_per_second": 90.617,
|
| 298 |
+
"eval_steps_per_second": 5.696,
|
| 299 |
+
"step": 225
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 15.338983050847457,
|
| 303 |
+
"grad_norm": 8.221491813659668,
|
| 304 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 305 |
+
"loss": 0.2097,
|
| 306 |
+
"step": 230
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 16.0,
|
| 310 |
+
"grad_norm": 5.007316589355469,
|
| 311 |
+
"learning_rate": 4.777777777777778e-05,
|
| 312 |
+
"loss": 0.1882,
|
| 313 |
+
"step": 240
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 16.0,
|
| 317 |
+
"eval_accuracy": 0.8628571428571429,
|
| 318 |
+
"eval_loss": 0.3937312960624695,
|
| 319 |
+
"eval_runtime": 1.922,
|
| 320 |
+
"eval_samples_per_second": 91.052,
|
| 321 |
+
"eval_steps_per_second": 5.723,
|
| 322 |
+
"step": 240
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 16.677966101694913,
|
| 326 |
+
"grad_norm": 9.630002975463867,
|
| 327 |
+
"learning_rate": 4.6984126984126986e-05,
|
| 328 |
+
"loss": 0.199,
|
| 329 |
+
"step": 250
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 17.0,
|
| 333 |
+
"eval_accuracy": 0.8628571428571429,
|
| 334 |
+
"eval_loss": 0.5102602243423462,
|
| 335 |
+
"eval_runtime": 2.5902,
|
| 336 |
+
"eval_samples_per_second": 67.563,
|
| 337 |
+
"eval_steps_per_second": 4.247,
|
| 338 |
+
"step": 255
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 17.338983050847457,
|
| 342 |
+
"grad_norm": 5.763312339782715,
|
| 343 |
+
"learning_rate": 4.6190476190476194e-05,
|
| 344 |
+
"loss": 0.174,
|
| 345 |
+
"step": 260
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 18.0,
|
| 349 |
+
"grad_norm": 2.7853012084960938,
|
| 350 |
+
"learning_rate": 4.53968253968254e-05,
|
| 351 |
+
"loss": 0.1565,
|
| 352 |
+
"step": 270
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 18.0,
|
| 356 |
+
"eval_accuracy": 0.8857142857142857,
|
| 357 |
+
"eval_loss": 0.36082425713539124,
|
| 358 |
+
"eval_runtime": 1.8955,
|
| 359 |
+
"eval_samples_per_second": 92.326,
|
| 360 |
+
"eval_steps_per_second": 5.803,
|
| 361 |
+
"step": 270
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 18.677966101694913,
|
| 365 |
+
"grad_norm": 7.840061187744141,
|
| 366 |
+
"learning_rate": 4.4603174603174604e-05,
|
| 367 |
+
"loss": 0.2068,
|
| 368 |
+
"step": 280
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 19.0,
|
| 372 |
+
"eval_accuracy": 0.88,
|
| 373 |
+
"eval_loss": 0.3678865134716034,
|
| 374 |
+
"eval_runtime": 1.914,
|
| 375 |
+
"eval_samples_per_second": 91.43,
|
| 376 |
+
"eval_steps_per_second": 5.747,
|
| 377 |
+
"step": 285
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 19.338983050847457,
|
| 381 |
+
"grad_norm": 10.269192695617676,
|
| 382 |
+
"learning_rate": 4.3809523809523805e-05,
|
| 383 |
+
"loss": 0.1742,
|
| 384 |
+
"step": 290
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 20.0,
|
| 388 |
+
"grad_norm": 11.602302551269531,
|
| 389 |
+
"learning_rate": 4.301587301587302e-05,
|
| 390 |
+
"loss": 0.194,
|
| 391 |
+
"step": 300
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 20.0,
|
| 395 |
+
"eval_accuracy": 0.8457142857142858,
|
| 396 |
+
"eval_loss": 0.5581231117248535,
|
| 397 |
+
"eval_runtime": 2.5904,
|
| 398 |
+
"eval_samples_per_second": 67.556,
|
| 399 |
+
"eval_steps_per_second": 4.246,
|
| 400 |
+
"step": 300
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 20.677966101694913,
|
| 404 |
+
"grad_norm": 4.199820518493652,
|
| 405 |
+
"learning_rate": 4.222222222222222e-05,
|
| 406 |
+
"loss": 0.1654,
|
| 407 |
+
"step": 310
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 21.0,
|
| 411 |
+
"eval_accuracy": 0.8685714285714285,
|
| 412 |
+
"eval_loss": 0.5074398517608643,
|
| 413 |
+
"eval_runtime": 1.9301,
|
| 414 |
+
"eval_samples_per_second": 90.667,
|
| 415 |
+
"eval_steps_per_second": 5.699,
|
| 416 |
+
"step": 315
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 21.338983050847457,
|
| 420 |
+
"grad_norm": 8.24092960357666,
|
| 421 |
+
"learning_rate": 4.142857142857143e-05,
|
| 422 |
+
"loss": 0.1393,
|
| 423 |
+
"step": 320
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 22.0,
|
| 427 |
+
"grad_norm": 6.02392053604126,
|
| 428 |
+
"learning_rate": 4.063492063492063e-05,
|
| 429 |
+
"loss": 0.1986,
|
| 430 |
+
"step": 330
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 22.0,
|
| 434 |
+
"eval_accuracy": 0.88,
|
| 435 |
+
"eval_loss": 0.4395482540130615,
|
| 436 |
+
"eval_runtime": 1.9058,
|
| 437 |
+
"eval_samples_per_second": 91.826,
|
| 438 |
+
"eval_steps_per_second": 5.772,
|
| 439 |
+
"step": 330
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 22.677966101694913,
|
| 443 |
+
"grad_norm": 5.646173000335693,
|
| 444 |
+
"learning_rate": 3.9841269841269846e-05,
|
| 445 |
+
"loss": 0.1257,
|
| 446 |
+
"step": 340
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 23.0,
|
| 450 |
+
"eval_accuracy": 0.8685714285714285,
|
| 451 |
+
"eval_loss": 0.42931947112083435,
|
| 452 |
+
"eval_runtime": 2.3278,
|
| 453 |
+
"eval_samples_per_second": 75.179,
|
| 454 |
+
"eval_steps_per_second": 4.726,
|
| 455 |
+
"step": 345
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 23.338983050847457,
|
| 459 |
+
"grad_norm": 7.199140548706055,
|
| 460 |
+
"learning_rate": 3.904761904761905e-05,
|
| 461 |
+
"loss": 0.1364,
|
| 462 |
+
"step": 350
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 24.0,
|
| 466 |
+
"grad_norm": 2.1807098388671875,
|
| 467 |
+
"learning_rate": 3.8253968253968256e-05,
|
| 468 |
+
"loss": 0.1976,
|
| 469 |
+
"step": 360
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 24.0,
|
| 473 |
+
"eval_accuracy": 0.8571428571428571,
|
| 474 |
+
"eval_loss": 0.4932045638561249,
|
| 475 |
+
"eval_runtime": 1.9503,
|
| 476 |
+
"eval_samples_per_second": 89.731,
|
| 477 |
+
"eval_steps_per_second": 5.64,
|
| 478 |
+
"step": 360
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 24.677966101694913,
|
| 482 |
+
"grad_norm": 3.774115562438965,
|
| 483 |
+
"learning_rate": 3.746031746031746e-05,
|
| 484 |
+
"loss": 0.1563,
|
| 485 |
+
"step": 370
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 25.0,
|
| 489 |
+
"eval_accuracy": 0.8857142857142857,
|
| 490 |
+
"eval_loss": 0.42544516921043396,
|
| 491 |
+
"eval_runtime": 1.9489,
|
| 492 |
+
"eval_samples_per_second": 89.793,
|
| 493 |
+
"eval_steps_per_second": 5.644,
|
| 494 |
+
"step": 375
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 25.338983050847457,
|
| 498 |
+
"grad_norm": 5.813924789428711,
|
| 499 |
+
"learning_rate": 3.666666666666667e-05,
|
| 500 |
+
"loss": 0.1537,
|
| 501 |
+
"step": 380
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 26.0,
|
| 505 |
+
"grad_norm": 5.670418739318848,
|
| 506 |
+
"learning_rate": 3.5873015873015874e-05,
|
| 507 |
+
"loss": 0.0985,
|
| 508 |
+
"step": 390
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 26.0,
|
| 512 |
+
"eval_accuracy": 0.8685714285714285,
|
| 513 |
+
"eval_loss": 0.5096610188484192,
|
| 514 |
+
"eval_runtime": 2.2632,
|
| 515 |
+
"eval_samples_per_second": 77.323,
|
| 516 |
+
"eval_steps_per_second": 4.86,
|
| 517 |
+
"step": 390
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 26.677966101694913,
|
| 521 |
+
"grad_norm": 7.973656177520752,
|
| 522 |
+
"learning_rate": 3.507936507936508e-05,
|
| 523 |
+
"loss": 0.1238,
|
| 524 |
+
"step": 400
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 27.0,
|
| 528 |
+
"eval_accuracy": 0.8514285714285714,
|
| 529 |
+
"eval_loss": 0.7264113426208496,
|
| 530 |
+
"eval_runtime": 2.2954,
|
| 531 |
+
"eval_samples_per_second": 76.239,
|
| 532 |
+
"eval_steps_per_second": 4.792,
|
| 533 |
+
"step": 405
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 27.338983050847457,
|
| 537 |
+
"grad_norm": 5.4732866287231445,
|
| 538 |
+
"learning_rate": 3.4285714285714284e-05,
|
| 539 |
+
"loss": 0.1385,
|
| 540 |
+
"step": 410
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 28.0,
|
| 544 |
+
"grad_norm": 4.48883056640625,
|
| 545 |
+
"learning_rate": 3.34920634920635e-05,
|
| 546 |
+
"loss": 0.1577,
|
| 547 |
+
"step": 420
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 28.0,
|
| 551 |
+
"eval_accuracy": 0.8571428571428571,
|
| 552 |
+
"eval_loss": 0.4826878011226654,
|
| 553 |
+
"eval_runtime": 1.9183,
|
| 554 |
+
"eval_samples_per_second": 91.227,
|
| 555 |
+
"eval_steps_per_second": 5.734,
|
| 556 |
+
"step": 420
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 28.677966101694913,
|
| 560 |
+
"grad_norm": 4.5706787109375,
|
| 561 |
+
"learning_rate": 3.26984126984127e-05,
|
| 562 |
+
"loss": 0.1271,
|
| 563 |
+
"step": 430
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 29.0,
|
| 567 |
+
"eval_accuracy": 0.8685714285714285,
|
| 568 |
+
"eval_loss": 0.530450165271759,
|
| 569 |
+
"eval_runtime": 2.147,
|
| 570 |
+
"eval_samples_per_second": 81.509,
|
| 571 |
+
"eval_steps_per_second": 5.123,
|
| 572 |
+
"step": 435
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 29.338983050847457,
|
| 576 |
+
"grad_norm": 8.417387962341309,
|
| 577 |
+
"learning_rate": 3.190476190476191e-05,
|
| 578 |
+
"loss": 0.1171,
|
| 579 |
+
"step": 440
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 30.0,
|
| 583 |
+
"grad_norm": 1.3924190998077393,
|
| 584 |
+
"learning_rate": 3.111111111111111e-05,
|
| 585 |
+
"loss": 0.1002,
|
| 586 |
+
"step": 450
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 30.0,
|
| 590 |
+
"eval_accuracy": 0.8628571428571429,
|
| 591 |
+
"eval_loss": 0.5888301134109497,
|
| 592 |
+
"eval_runtime": 1.8837,
|
| 593 |
+
"eval_samples_per_second": 92.902,
|
| 594 |
+
"eval_steps_per_second": 5.84,
|
| 595 |
+
"step": 450
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 30.677966101694913,
|
| 599 |
+
"grad_norm": 4.39676570892334,
|
| 600 |
+
"learning_rate": 3.031746031746032e-05,
|
| 601 |
+
"loss": 0.1268,
|
| 602 |
+
"step": 460
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 31.0,
|
| 606 |
+
"eval_accuracy": 0.8571428571428571,
|
| 607 |
+
"eval_loss": 0.6432701945304871,
|
| 608 |
+
"eval_runtime": 1.9204,
|
| 609 |
+
"eval_samples_per_second": 91.126,
|
| 610 |
+
"eval_steps_per_second": 5.728,
|
| 611 |
+
"step": 465
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 31.338983050847457,
|
| 615 |
+
"grad_norm": 6.017373561859131,
|
| 616 |
+
"learning_rate": 2.9523809523809523e-05,
|
| 617 |
+
"loss": 0.1077,
|
| 618 |
+
"step": 470
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 32.0,
|
| 622 |
+
"grad_norm": 5.33542013168335,
|
| 623 |
+
"learning_rate": 2.873015873015873e-05,
|
| 624 |
+
"loss": 0.1153,
|
| 625 |
+
"step": 480
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 32.0,
|
| 629 |
+
"eval_accuracy": 0.8342857142857143,
|
| 630 |
+
"eval_loss": 0.8394165635108948,
|
| 631 |
+
"eval_runtime": 2.2924,
|
| 632 |
+
"eval_samples_per_second": 76.34,
|
| 633 |
+
"eval_steps_per_second": 4.798,
|
| 634 |
+
"step": 480
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 32.67796610169491,
|
| 638 |
+
"grad_norm": 13.854134559631348,
|
| 639 |
+
"learning_rate": 2.7936507936507936e-05,
|
| 640 |
+
"loss": 0.1191,
|
| 641 |
+
"step": 490
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 33.0,
|
| 645 |
+
"eval_accuracy": 0.84,
|
| 646 |
+
"eval_loss": 0.747542142868042,
|
| 647 |
+
"eval_runtime": 1.9178,
|
| 648 |
+
"eval_samples_per_second": 91.249,
|
| 649 |
+
"eval_steps_per_second": 5.736,
|
| 650 |
+
"step": 495
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 33.33898305084746,
|
| 654 |
+
"grad_norm": 8.244441986083984,
|
| 655 |
+
"learning_rate": 2.7142857142857144e-05,
|
| 656 |
+
"loss": 0.1271,
|
| 657 |
+
"step": 500
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 34.0,
|
| 661 |
+
"grad_norm": 2.1750755310058594,
|
| 662 |
+
"learning_rate": 2.634920634920635e-05,
|
| 663 |
+
"loss": 0.1184,
|
| 664 |
+
"step": 510
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 34.0,
|
| 668 |
+
"eval_accuracy": 0.8742857142857143,
|
| 669 |
+
"eval_loss": 0.4883846938610077,
|
| 670 |
+
"eval_runtime": 1.8782,
|
| 671 |
+
"eval_samples_per_second": 93.173,
|
| 672 |
+
"eval_steps_per_second": 5.857,
|
| 673 |
+
"step": 510
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 34.67796610169491,
|
| 677 |
+
"grad_norm": 9.822646141052246,
|
| 678 |
+
"learning_rate": 2.5555555555555557e-05,
|
| 679 |
+
"loss": 0.1332,
|
| 680 |
+
"step": 520
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 35.0,
|
| 684 |
+
"eval_accuracy": 0.8857142857142857,
|
| 685 |
+
"eval_loss": 0.5834174752235413,
|
| 686 |
+
"eval_runtime": 2.1992,
|
| 687 |
+
"eval_samples_per_second": 79.573,
|
| 688 |
+
"eval_steps_per_second": 5.002,
|
| 689 |
+
"step": 525
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 35.33898305084746,
|
| 693 |
+
"grad_norm": 5.659489154815674,
|
| 694 |
+
"learning_rate": 2.4761904761904762e-05,
|
| 695 |
+
"loss": 0.0804,
|
| 696 |
+
"step": 530
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 36.0,
|
| 700 |
+
"grad_norm": 5.710267066955566,
|
| 701 |
+
"learning_rate": 2.396825396825397e-05,
|
| 702 |
+
"loss": 0.1071,
|
| 703 |
+
"step": 540
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 36.0,
|
| 707 |
+
"eval_accuracy": 0.8571428571428571,
|
| 708 |
+
"eval_loss": 0.627878725528717,
|
| 709 |
+
"eval_runtime": 1.9363,
|
| 710 |
+
"eval_samples_per_second": 90.377,
|
| 711 |
+
"eval_steps_per_second": 5.681,
|
| 712 |
+
"step": 540
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 36.67796610169491,
|
| 716 |
+
"grad_norm": 6.440234661102295,
|
| 717 |
+
"learning_rate": 2.3174603174603175e-05,
|
| 718 |
+
"loss": 0.0886,
|
| 719 |
+
"step": 550
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 37.0,
|
| 723 |
+
"eval_accuracy": 0.8628571428571429,
|
| 724 |
+
"eval_loss": 0.6998600363731384,
|
| 725 |
+
"eval_runtime": 1.9136,
|
| 726 |
+
"eval_samples_per_second": 91.451,
|
| 727 |
+
"eval_steps_per_second": 5.748,
|
| 728 |
+
"step": 555
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 37.33898305084746,
|
| 732 |
+
"grad_norm": 4.174771785736084,
|
| 733 |
+
"learning_rate": 2.238095238095238e-05,
|
| 734 |
+
"loss": 0.0845,
|
| 735 |
+
"step": 560
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 38.0,
|
| 739 |
+
"grad_norm": 0.6355146765708923,
|
| 740 |
+
"learning_rate": 2.1587301587301585e-05,
|
| 741 |
+
"loss": 0.0744,
|
| 742 |
+
"step": 570
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 38.0,
|
| 746 |
+
"eval_accuracy": 0.8685714285714285,
|
| 747 |
+
"eval_loss": 0.729452908039093,
|
| 748 |
+
"eval_runtime": 2.1953,
|
| 749 |
+
"eval_samples_per_second": 79.716,
|
| 750 |
+
"eval_steps_per_second": 5.011,
|
| 751 |
+
"step": 570
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 38.67796610169491,
|
| 755 |
+
"grad_norm": 12.643016815185547,
|
| 756 |
+
"learning_rate": 2.0793650793650793e-05,
|
| 757 |
+
"loss": 0.1274,
|
| 758 |
+
"step": 580
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 39.0,
|
| 762 |
+
"eval_accuracy": 0.8914285714285715,
|
| 763 |
+
"eval_loss": 0.6137147545814514,
|
| 764 |
+
"eval_runtime": 1.9072,
|
| 765 |
+
"eval_samples_per_second": 91.758,
|
| 766 |
+
"eval_steps_per_second": 5.768,
|
| 767 |
+
"step": 585
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 39.33898305084746,
|
| 771 |
+
"grad_norm": 8.217287063598633,
|
| 772 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 773 |
+
"loss": 0.0905,
|
| 774 |
+
"step": 590
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 40.0,
|
| 778 |
+
"grad_norm": 2.896934986114502,
|
| 779 |
+
"learning_rate": 1.9206349206349206e-05,
|
| 780 |
+
"loss": 0.0795,
|
| 781 |
+
"step": 600
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 40.0,
|
| 785 |
+
"eval_accuracy": 0.8742857142857143,
|
| 786 |
+
"eval_loss": 0.5706229209899902,
|
| 787 |
+
"eval_runtime": 2.3574,
|
| 788 |
+
"eval_samples_per_second": 74.235,
|
| 789 |
+
"eval_steps_per_second": 4.666,
|
| 790 |
+
"step": 600
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 40.67796610169491,
|
| 794 |
+
"grad_norm": 8.571166038513184,
|
| 795 |
+
"learning_rate": 1.841269841269841e-05,
|
| 796 |
+
"loss": 0.0962,
|
| 797 |
+
"step": 610
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 41.0,
|
| 801 |
+
"eval_accuracy": 0.8857142857142857,
|
| 802 |
+
"eval_loss": 0.6100188493728638,
|
| 803 |
+
"eval_runtime": 2.0136,
|
| 804 |
+
"eval_samples_per_second": 86.908,
|
| 805 |
+
"eval_steps_per_second": 5.463,
|
| 806 |
+
"step": 615
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 41.33898305084746,
|
| 810 |
+
"grad_norm": 4.700484752655029,
|
| 811 |
+
"learning_rate": 1.761904761904762e-05,
|
| 812 |
+
"loss": 0.0749,
|
| 813 |
+
"step": 620
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 42.0,
|
| 817 |
+
"grad_norm": 2.4115490913391113,
|
| 818 |
+
"learning_rate": 1.6825396825396824e-05,
|
| 819 |
+
"loss": 0.094,
|
| 820 |
+
"step": 630
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 42.0,
|
| 824 |
+
"eval_accuracy": 0.8742857142857143,
|
| 825 |
+
"eval_loss": 0.6148616075515747,
|
| 826 |
+
"eval_runtime": 1.9168,
|
| 827 |
+
"eval_samples_per_second": 91.298,
|
| 828 |
+
"eval_steps_per_second": 5.739,
|
| 829 |
+
"step": 630
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 42.67796610169491,
|
| 833 |
+
"grad_norm": 4.732550144195557,
|
| 834 |
+
"learning_rate": 1.6031746031746033e-05,
|
| 835 |
+
"loss": 0.0945,
|
| 836 |
+
"step": 640
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 43.0,
|
| 840 |
+
"eval_accuracy": 0.88,
|
| 841 |
+
"eval_loss": 0.5688998699188232,
|
| 842 |
+
"eval_runtime": 2.0595,
|
| 843 |
+
"eval_samples_per_second": 84.974,
|
| 844 |
+
"eval_steps_per_second": 5.341,
|
| 845 |
+
"step": 645
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 43.33898305084746,
|
| 849 |
+
"grad_norm": 4.677188396453857,
|
| 850 |
+
"learning_rate": 1.5238095238095238e-05,
|
| 851 |
+
"loss": 0.0851,
|
| 852 |
+
"step": 650
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 44.0,
|
| 856 |
+
"grad_norm": 6.458128452301025,
|
| 857 |
+
"learning_rate": 1.4444444444444444e-05,
|
| 858 |
+
"loss": 0.0584,
|
| 859 |
+
"step": 660
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 44.0,
|
| 863 |
+
"eval_accuracy": 0.8742857142857143,
|
| 864 |
+
"eval_loss": 0.7018650770187378,
|
| 865 |
+
"eval_runtime": 2.3309,
|
| 866 |
+
"eval_samples_per_second": 75.079,
|
| 867 |
+
"eval_steps_per_second": 4.719,
|
| 868 |
+
"step": 660
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 44.67796610169491,
|
| 872 |
+
"grad_norm": 4.308237552642822,
|
| 873 |
+
"learning_rate": 1.365079365079365e-05,
|
| 874 |
+
"loss": 0.0676,
|
| 875 |
+
"step": 670
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 45.0,
|
| 879 |
+
"eval_accuracy": 0.88,
|
| 880 |
+
"eval_loss": 0.6934124231338501,
|
| 881 |
+
"eval_runtime": 1.9251,
|
| 882 |
+
"eval_samples_per_second": 90.902,
|
| 883 |
+
"eval_steps_per_second": 5.714,
|
| 884 |
+
"step": 675
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 45.33898305084746,
|
| 888 |
+
"grad_norm": 2.8312790393829346,
|
| 889 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 890 |
+
"loss": 0.0893,
|
| 891 |
+
"step": 680
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 46.0,
|
| 895 |
+
"grad_norm": 7.0031328201293945,
|
| 896 |
+
"learning_rate": 1.2063492063492064e-05,
|
| 897 |
+
"loss": 0.0763,
|
| 898 |
+
"step": 690
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 46.0,
|
| 902 |
+
"eval_accuracy": 0.8914285714285715,
|
| 903 |
+
"eval_loss": 0.6047118902206421,
|
| 904 |
+
"eval_runtime": 2.0296,
|
| 905 |
+
"eval_samples_per_second": 86.224,
|
| 906 |
+
"eval_steps_per_second": 5.42,
|
| 907 |
+
"step": 690
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 46.67796610169491,
|
| 911 |
+
"grad_norm": 8.401297569274902,
|
| 912 |
+
"learning_rate": 1.126984126984127e-05,
|
| 913 |
+
"loss": 0.0762,
|
| 914 |
+
"step": 700
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 47.0,
|
| 918 |
+
"eval_accuracy": 0.88,
|
| 919 |
+
"eval_loss": 0.6063617467880249,
|
| 920 |
+
"eval_runtime": 1.8566,
|
| 921 |
+
"eval_samples_per_second": 94.259,
|
| 922 |
+
"eval_steps_per_second": 5.925,
|
| 923 |
+
"step": 705
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 47.33898305084746,
|
| 927 |
+
"grad_norm": 1.462274432182312,
|
| 928 |
+
"learning_rate": 1.0476190476190475e-05,
|
| 929 |
+
"loss": 0.0563,
|
| 930 |
+
"step": 710
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 48.0,
|
| 934 |
+
"grad_norm": 1.8739376068115234,
|
| 935 |
+
"learning_rate": 9.682539682539682e-06,
|
| 936 |
+
"loss": 0.0696,
|
| 937 |
+
"step": 720
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 48.0,
|
| 941 |
+
"eval_accuracy": 0.8685714285714285,
|
| 942 |
+
"eval_loss": 0.7335702776908875,
|
| 943 |
+
"eval_runtime": 1.911,
|
| 944 |
+
"eval_samples_per_second": 91.576,
|
| 945 |
+
"eval_steps_per_second": 5.756,
|
| 946 |
+
"step": 720
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 48.67796610169491,
|
| 950 |
+
"grad_norm": 2.9189000129699707,
|
| 951 |
+
"learning_rate": 8.888888888888888e-06,
|
| 952 |
+
"loss": 0.0555,
|
| 953 |
+
"step": 730
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 49.0,
|
| 957 |
+
"eval_accuracy": 0.8742857142857143,
|
| 958 |
+
"eval_loss": 0.6598544120788574,
|
| 959 |
+
"eval_runtime": 1.9333,
|
| 960 |
+
"eval_samples_per_second": 90.519,
|
| 961 |
+
"eval_steps_per_second": 5.69,
|
| 962 |
+
"step": 735
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 49.33898305084746,
|
| 966 |
+
"grad_norm": 3.1225035190582275,
|
| 967 |
+
"learning_rate": 8.095238095238095e-06,
|
| 968 |
+
"loss": 0.1129,
|
| 969 |
+
"step": 740
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 50.0,
|
| 973 |
+
"grad_norm": 2.0588467121124268,
|
| 974 |
+
"learning_rate": 7.301587301587301e-06,
|
| 975 |
+
"loss": 0.0572,
|
| 976 |
+
"step": 750
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 50.0,
|
| 980 |
+
"eval_accuracy": 0.9028571428571428,
|
| 981 |
+
"eval_loss": 0.597748875617981,
|
| 982 |
+
"eval_runtime": 2.541,
|
| 983 |
+
"eval_samples_per_second": 68.87,
|
| 984 |
+
"eval_steps_per_second": 4.329,
|
| 985 |
+
"step": 750
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 50.67796610169491,
|
| 989 |
+
"grad_norm": 2.578906536102295,
|
| 990 |
+
"learning_rate": 6.507936507936508e-06,
|
| 991 |
+
"loss": 0.0648,
|
| 992 |
+
"step": 760
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 51.0,
|
| 996 |
+
"eval_accuracy": 0.88,
|
| 997 |
+
"eval_loss": 0.6257001757621765,
|
| 998 |
+
"eval_runtime": 1.8911,
|
| 999 |
+
"eval_samples_per_second": 92.541,
|
| 1000 |
+
"eval_steps_per_second": 5.817,
|
| 1001 |
+
"step": 765
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 51.33898305084746,
|
| 1005 |
+
"grad_norm": 3.871882677078247,
|
| 1006 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 1007 |
+
"loss": 0.0521,
|
| 1008 |
+
"step": 770
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 52.0,
|
| 1012 |
+
"grad_norm": 3.6807923316955566,
|
| 1013 |
+
"learning_rate": 4.92063492063492e-06,
|
| 1014 |
+
"loss": 0.0705,
|
| 1015 |
+
"step": 780
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 52.0,
|
| 1019 |
+
"eval_accuracy": 0.8857142857142857,
|
| 1020 |
+
"eval_loss": 0.6653619408607483,
|
| 1021 |
+
"eval_runtime": 1.8778,
|
| 1022 |
+
"eval_samples_per_second": 93.193,
|
| 1023 |
+
"eval_steps_per_second": 5.858,
|
| 1024 |
+
"step": 780
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 52.67796610169491,
|
| 1028 |
+
"grad_norm": 6.3525519371032715,
|
| 1029 |
+
"learning_rate": 4.126984126984127e-06,
|
| 1030 |
+
"loss": 0.0646,
|
| 1031 |
+
"step": 790
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 53.0,
|
| 1035 |
+
"eval_accuracy": 0.8685714285714285,
|
| 1036 |
+
"eval_loss": 0.6813338994979858,
|
| 1037 |
+
"eval_runtime": 1.8783,
|
| 1038 |
+
"eval_samples_per_second": 93.171,
|
| 1039 |
+
"eval_steps_per_second": 5.856,
|
| 1040 |
+
"step": 795
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 53.33898305084746,
|
| 1044 |
+
"grad_norm": 5.389460563659668,
|
| 1045 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 1046 |
+
"loss": 0.0463,
|
| 1047 |
+
"step": 800
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 54.0,
|
| 1051 |
+
"grad_norm": 0.5001619458198547,
|
| 1052 |
+
"learning_rate": 2.5396825396825395e-06,
|
| 1053 |
+
"loss": 0.0795,
|
| 1054 |
+
"step": 810
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 54.0,
|
| 1058 |
+
"eval_accuracy": 0.8742857142857143,
|
| 1059 |
+
"eval_loss": 0.6209337711334229,
|
| 1060 |
+
"eval_runtime": 2.6137,
|
| 1061 |
+
"eval_samples_per_second": 66.955,
|
| 1062 |
+
"eval_steps_per_second": 4.209,
|
| 1063 |
+
"step": 810
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 54.67796610169491,
|
| 1067 |
+
"grad_norm": 9.643752098083496,
|
| 1068 |
+
"learning_rate": 1.746031746031746e-06,
|
| 1069 |
+
"loss": 0.0828,
|
| 1070 |
+
"step": 820
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 55.0,
|
| 1074 |
+
"eval_accuracy": 0.8742857142857143,
|
| 1075 |
+
"eval_loss": 0.6456648707389832,
|
| 1076 |
+
"eval_runtime": 1.8674,
|
| 1077 |
+
"eval_samples_per_second": 93.711,
|
| 1078 |
+
"eval_steps_per_second": 5.89,
|
| 1079 |
+
"step": 825
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 55.33898305084746,
|
| 1083 |
+
"grad_norm": 7.717844486236572,
|
| 1084 |
+
"learning_rate": 9.523809523809523e-07,
|
| 1085 |
+
"loss": 0.0916,
|
| 1086 |
+
"step": 830
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 56.0,
|
| 1090 |
+
"grad_norm": 3.563279390335083,
|
| 1091 |
+
"learning_rate": 1.5873015873015872e-07,
|
| 1092 |
+
"loss": 0.0674,
|
| 1093 |
+
"step": 840
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 56.0,
|
| 1097 |
+
"eval_accuracy": 0.88,
|
| 1098 |
+
"eval_loss": 0.6521316766738892,
|
| 1099 |
+
"eval_runtime": 1.8761,
|
| 1100 |
+
"eval_samples_per_second": 93.278,
|
| 1101 |
+
"eval_steps_per_second": 5.863,
|
| 1102 |
+
"step": 840
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 56.0,
|
| 1106 |
+
"step": 840,
|
| 1107 |
+
"total_flos": 1.7108328318259692e+18,
|
| 1108 |
+
"train_loss": 0.19382851386354083,
|
| 1109 |
+
"train_runtime": 1219.347,
|
| 1110 |
+
"train_samples_per_second": 46.205,
|
| 1111 |
+
"train_steps_per_second": 0.689
|
| 1112 |
+
}
|
| 1113 |
+
],
|
| 1114 |
+
"logging_steps": 10,
|
| 1115 |
+
"max_steps": 840,
|
| 1116 |
+
"num_input_tokens_seen": 0,
|
| 1117 |
+
"num_train_epochs": 60,
|
| 1118 |
+
"save_steps": 500,
|
| 1119 |
+
"stateful_callbacks": {
|
| 1120 |
+
"TrainerControl": {
|
| 1121 |
+
"args": {
|
| 1122 |
+
"should_epoch_stop": false,
|
| 1123 |
+
"should_evaluate": false,
|
| 1124 |
+
"should_log": false,
|
| 1125 |
+
"should_save": true,
|
| 1126 |
+
"should_training_stop": true
|
| 1127 |
+
},
|
| 1128 |
+
"attributes": {}
|
| 1129 |
+
}
|
| 1130 |
+
},
|
| 1131 |
+
"total_flos": 1.7108328318259692e+18,
|
| 1132 |
+
"train_batch_size": 16,
|
| 1133 |
+
"trial_name": null,
|
| 1134 |
+
"trial_params": null
|
| 1135 |
+
}
|