Cheers 🎉
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec08_13-06-01_18af7fa0d1da/events.out.tfevents.1733666434.18af7fa0d1da.837.2 +3 -0
- test_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1743 -0
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# ViT-NIH-Chest-X-ray-dataset-small
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 17 |
|
| 18 |
# ViT-NIH-Chest-X-ray-dataset-small
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.2988
|
| 23 |
+
- Accuracy: 0.2299
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_accuracy": 0.22986708365910868,
|
| 4 |
+
"eval_loss": 0.2988394498825073,
|
| 5 |
+
"eval_runtime": 39.1477,
|
| 6 |
+
"eval_samples_per_second": 32.671,
|
| 7 |
+
"eval_steps_per_second": 4.087,
|
| 8 |
+
"total_flos": 2.6821552511927255e+18,
|
| 9 |
+
"train_loss": 0.16216810325304962,
|
| 10 |
+
"train_runtime": 3143.5173,
|
| 11 |
+
"train_samples_per_second": 11.009,
|
| 12 |
+
"train_steps_per_second": 0.69
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_accuracy": 0.6526011560693642,
|
| 4 |
+
"eval_loss": 0.0923289805650711,
|
| 5 |
+
"eval_runtime": 56.6908,
|
| 6 |
+
"eval_samples_per_second": 30.516,
|
| 7 |
+
"eval_steps_per_second": 3.828
|
| 8 |
+
}
|
runs/Dec08_13-06-01_18af7fa0d1da/events.out.tfevents.1733666434.18af7fa0d1da.837.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fedcd6fef977b6723b6a556e2363f7a526dff034073a085c72a0301e5d95c4a
|
| 3 |
+
size 734
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_accuracy": 0.22986708365910868,
|
| 4 |
+
"eval_loss": 0.2988394498825073,
|
| 5 |
+
"eval_runtime": 39.1477,
|
| 6 |
+
"eval_samples_per_second": 32.671,
|
| 7 |
+
"eval_steps_per_second": 4.087
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"total_flos": 2.6821552511927255e+18,
|
| 4 |
+
"train_loss": 0.16216810325304962,
|
| 5 |
+
"train_runtime": 3143.5173,
|
| 6 |
+
"train_samples_per_second": 11.009,
|
| 7 |
+
"train_steps_per_second": 0.69
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1743 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.0923289805650711,
|
| 3 |
+
"best_model_checkpoint": "./ViT-NIH-Chest-X-ray-dataset-small/checkpoint-2100",
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2168,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03690036900369004,
|
| 13 |
+
"grad_norm": 0.38048115372657776,
|
| 14 |
+
"learning_rate": 0.00019907749077490775,
|
| 15 |
+
"loss": 0.5131,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07380073800738007,
|
| 20 |
+
"grad_norm": 0.3416444957256317,
|
| 21 |
+
"learning_rate": 0.00019815498154981552,
|
| 22 |
+
"loss": 0.3661,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.11070110701107011,
|
| 27 |
+
"grad_norm": 0.268430233001709,
|
| 28 |
+
"learning_rate": 0.00019723247232472326,
|
| 29 |
+
"loss": 0.3122,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14760147601476015,
|
| 34 |
+
"grad_norm": 0.20281535387039185,
|
| 35 |
+
"learning_rate": 0.000196309963099631,
|
| 36 |
+
"loss": 0.2862,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.18450184501845018,
|
| 41 |
+
"grad_norm": 0.17638804018497467,
|
| 42 |
+
"learning_rate": 0.00019538745387453877,
|
| 43 |
+
"loss": 0.2582,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.22140221402214022,
|
| 48 |
+
"grad_norm": 0.13983863592147827,
|
| 49 |
+
"learning_rate": 0.00019446494464944652,
|
| 50 |
+
"loss": 0.2621,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.25830258302583026,
|
| 55 |
+
"grad_norm": 0.1306193470954895,
|
| 56 |
+
"learning_rate": 0.00019354243542435426,
|
| 57 |
+
"loss": 0.2361,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2952029520295203,
|
| 62 |
+
"grad_norm": 0.12178371101617813,
|
| 63 |
+
"learning_rate": 0.000192619926199262,
|
| 64 |
+
"loss": 0.2271,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.33210332103321033,
|
| 69 |
+
"grad_norm": 0.09625957161188126,
|
| 70 |
+
"learning_rate": 0.00019169741697416974,
|
| 71 |
+
"loss": 0.2275,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.36900369003690037,
|
| 76 |
+
"grad_norm": 0.12183211743831635,
|
| 77 |
+
"learning_rate": 0.00019077490774907748,
|
| 78 |
+
"loss": 0.2128,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.36900369003690037,
|
| 83 |
+
"eval_accuracy": 0.0,
|
| 84 |
+
"eval_loss": 0.20924170315265656,
|
| 85 |
+
"eval_runtime": 51.5235,
|
| 86 |
+
"eval_samples_per_second": 33.577,
|
| 87 |
+
"eval_steps_per_second": 4.212,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.4059040590405904,
|
| 92 |
+
"grad_norm": 0.08246201276779175,
|
| 93 |
+
"learning_rate": 0.00018985239852398525,
|
| 94 |
+
"loss": 0.2089,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.44280442804428044,
|
| 99 |
+
"grad_norm": 0.11126121133565903,
|
| 100 |
+
"learning_rate": 0.000188929889298893,
|
| 101 |
+
"loss": 0.2147,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.4797047970479705,
|
| 106 |
+
"grad_norm": 0.11177172511816025,
|
| 107 |
+
"learning_rate": 0.00018800738007380074,
|
| 108 |
+
"loss": 0.2044,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5166051660516605,
|
| 113 |
+
"grad_norm": 0.10021921247243881,
|
| 114 |
+
"learning_rate": 0.0001870848708487085,
|
| 115 |
+
"loss": 0.1839,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.5535055350553506,
|
| 120 |
+
"grad_norm": 0.185609832406044,
|
| 121 |
+
"learning_rate": 0.00018616236162361625,
|
| 122 |
+
"loss": 0.2209,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.5904059040590406,
|
| 127 |
+
"grad_norm": 0.17880532145500183,
|
| 128 |
+
"learning_rate": 0.000185239852398524,
|
| 129 |
+
"loss": 0.2023,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.6273062730627307,
|
| 134 |
+
"grad_norm": 0.09112340211868286,
|
| 135 |
+
"learning_rate": 0.00018431734317343173,
|
| 136 |
+
"loss": 0.1952,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.6642066420664207,
|
| 141 |
+
"grad_norm": 0.14496631920337677,
|
| 142 |
+
"learning_rate": 0.0001833948339483395,
|
| 143 |
+
"loss": 0.2154,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.7011070110701108,
|
| 148 |
+
"grad_norm": 0.1129971593618393,
|
| 149 |
+
"learning_rate": 0.00018247232472324724,
|
| 150 |
+
"loss": 0.1865,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.7380073800738007,
|
| 155 |
+
"grad_norm": 0.1439884454011917,
|
| 156 |
+
"learning_rate": 0.00018154981549815499,
|
| 157 |
+
"loss": 0.1848,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.7380073800738007,
|
| 162 |
+
"eval_accuracy": 0.38208092485549133,
|
| 163 |
+
"eval_loss": 0.1909271478652954,
|
| 164 |
+
"eval_runtime": 51.554,
|
| 165 |
+
"eval_samples_per_second": 33.557,
|
| 166 |
+
"eval_steps_per_second": 4.209,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.7749077490774908,
|
| 171 |
+
"grad_norm": 0.1933569312095642,
|
| 172 |
+
"learning_rate": 0.00018062730627306276,
|
| 173 |
+
"loss": 0.1921,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.8118081180811808,
|
| 178 |
+
"grad_norm": 0.19441623985767365,
|
| 179 |
+
"learning_rate": 0.0001797047970479705,
|
| 180 |
+
"loss": 0.2087,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.8487084870848709,
|
| 185 |
+
"grad_norm": 0.10429559648036957,
|
| 186 |
+
"learning_rate": 0.00017878228782287824,
|
| 187 |
+
"loss": 0.2016,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.8856088560885609,
|
| 192 |
+
"grad_norm": 0.11976602673530579,
|
| 193 |
+
"learning_rate": 0.00017785977859778598,
|
| 194 |
+
"loss": 0.1871,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.922509225092251,
|
| 199 |
+
"grad_norm": 0.13647252321243286,
|
| 200 |
+
"learning_rate": 0.00017693726937269372,
|
| 201 |
+
"loss": 0.1951,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.959409594095941,
|
| 206 |
+
"grad_norm": 0.13491246104240417,
|
| 207 |
+
"learning_rate": 0.00017601476014760147,
|
| 208 |
+
"loss": 0.1987,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.996309963099631,
|
| 213 |
+
"grad_norm": 0.10877779126167297,
|
| 214 |
+
"learning_rate": 0.00017509225092250923,
|
| 215 |
+
"loss": 0.2067,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.033210332103321,
|
| 220 |
+
"grad_norm": 0.10195717215538025,
|
| 221 |
+
"learning_rate": 0.00017416974169741698,
|
| 222 |
+
"loss": 0.1829,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.070110701107011,
|
| 227 |
+
"grad_norm": 0.11287475377321243,
|
| 228 |
+
"learning_rate": 0.00017324723247232472,
|
| 229 |
+
"loss": 0.1771,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.1070110701107012,
|
| 234 |
+
"grad_norm": 0.10177090764045715,
|
| 235 |
+
"learning_rate": 0.0001723247232472325,
|
| 236 |
+
"loss": 0.171,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.1070110701107012,
|
| 241 |
+
"eval_accuracy": 0.5387283236994219,
|
| 242 |
+
"eval_loss": 0.1967419981956482,
|
| 243 |
+
"eval_runtime": 50.6773,
|
| 244 |
+
"eval_samples_per_second": 34.138,
|
| 245 |
+
"eval_steps_per_second": 4.282,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.1439114391143912,
|
| 250 |
+
"grad_norm": 0.14216077327728271,
|
| 251 |
+
"learning_rate": 0.00017140221402214023,
|
| 252 |
+
"loss": 0.2104,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.1808118081180812,
|
| 257 |
+
"grad_norm": 0.0753447487950325,
|
| 258 |
+
"learning_rate": 0.00017047970479704797,
|
| 259 |
+
"loss": 0.201,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.2177121771217712,
|
| 264 |
+
"grad_norm": 0.13927125930786133,
|
| 265 |
+
"learning_rate": 0.00016955719557195574,
|
| 266 |
+
"loss": 0.1923,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.2546125461254611,
|
| 271 |
+
"grad_norm": 0.14396004378795624,
|
| 272 |
+
"learning_rate": 0.00016863468634686348,
|
| 273 |
+
"loss": 0.1996,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.2915129151291513,
|
| 278 |
+
"grad_norm": 0.09631673246622086,
|
| 279 |
+
"learning_rate": 0.00016771217712177123,
|
| 280 |
+
"loss": 0.1672,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.3284132841328413,
|
| 285 |
+
"grad_norm": 0.1342993974685669,
|
| 286 |
+
"learning_rate": 0.00016678966789667897,
|
| 287 |
+
"loss": 0.1792,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.3653136531365313,
|
| 292 |
+
"grad_norm": 0.20273268222808838,
|
| 293 |
+
"learning_rate": 0.00016586715867158674,
|
| 294 |
+
"loss": 0.1753,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.4022140221402215,
|
| 299 |
+
"grad_norm": 0.1452128291130066,
|
| 300 |
+
"learning_rate": 0.00016494464944649448,
|
| 301 |
+
"loss": 0.1801,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.4391143911439115,
|
| 306 |
+
"grad_norm": 0.0973893478512764,
|
| 307 |
+
"learning_rate": 0.00016402214022140222,
|
| 308 |
+
"loss": 0.2136,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.4760147601476015,
|
| 313 |
+
"grad_norm": 0.2589876651763916,
|
| 314 |
+
"learning_rate": 0.00016309963099630996,
|
| 315 |
+
"loss": 0.1772,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.4760147601476015,
|
| 320 |
+
"eval_accuracy": 0.5450867052023122,
|
| 321 |
+
"eval_loss": 0.19317613542079926,
|
| 322 |
+
"eval_runtime": 50.5026,
|
| 323 |
+
"eval_samples_per_second": 34.256,
|
| 324 |
+
"eval_steps_per_second": 4.297,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.5129151291512914,
|
| 329 |
+
"grad_norm": 0.16841156780719757,
|
| 330 |
+
"learning_rate": 0.0001621771217712177,
|
| 331 |
+
"loss": 0.1894,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.5498154981549814,
|
| 336 |
+
"grad_norm": 0.17766626179218292,
|
| 337 |
+
"learning_rate": 0.00016125461254612547,
|
| 338 |
+
"loss": 0.204,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.5867158671586716,
|
| 343 |
+
"grad_norm": 0.13253839313983917,
|
| 344 |
+
"learning_rate": 0.00016033210332103322,
|
| 345 |
+
"loss": 0.2141,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.6236162361623616,
|
| 350 |
+
"grad_norm": 0.10587523877620697,
|
| 351 |
+
"learning_rate": 0.00015940959409594096,
|
| 352 |
+
"loss": 0.2192,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6605166051660518,
|
| 357 |
+
"grad_norm": 0.10049675405025482,
|
| 358 |
+
"learning_rate": 0.0001584870848708487,
|
| 359 |
+
"loss": 0.1939,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6974169741697418,
|
| 364 |
+
"grad_norm": 0.1287400722503662,
|
| 365 |
+
"learning_rate": 0.00015756457564575647,
|
| 366 |
+
"loss": 0.1955,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.7343173431734318,
|
| 371 |
+
"grad_norm": 0.15120339393615723,
|
| 372 |
+
"learning_rate": 0.0001566420664206642,
|
| 373 |
+
"loss": 0.1863,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.7712177121771218,
|
| 378 |
+
"grad_norm": 0.1202373206615448,
|
| 379 |
+
"learning_rate": 0.00015571955719557195,
|
| 380 |
+
"loss": 0.1832,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.8081180811808117,
|
| 385 |
+
"grad_norm": 0.1368759572505951,
|
| 386 |
+
"learning_rate": 0.00015479704797047972,
|
| 387 |
+
"loss": 0.2007,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.8450184501845017,
|
| 392 |
+
"grad_norm": 0.08591968566179276,
|
| 393 |
+
"learning_rate": 0.00015387453874538746,
|
| 394 |
+
"loss": 0.1629,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.8450184501845017,
|
| 399 |
+
"eval_accuracy": 0.4485549132947977,
|
| 400 |
+
"eval_loss": 0.1842162311077118,
|
| 401 |
+
"eval_runtime": 50.5861,
|
| 402 |
+
"eval_samples_per_second": 34.199,
|
| 403 |
+
"eval_steps_per_second": 4.29,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.881918819188192,
|
| 408 |
+
"grad_norm": 0.12317466735839844,
|
| 409 |
+
"learning_rate": 0.0001529520295202952,
|
| 410 |
+
"loss": 0.213,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.918819188191882,
|
| 415 |
+
"grad_norm": 0.15092293918132782,
|
| 416 |
+
"learning_rate": 0.00015202952029520298,
|
| 417 |
+
"loss": 0.2004,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.9557195571955721,
|
| 422 |
+
"grad_norm": 0.20343895256519318,
|
| 423 |
+
"learning_rate": 0.00015110701107011072,
|
| 424 |
+
"loss": 0.1835,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.992619926199262,
|
| 429 |
+
"grad_norm": 0.32362422347068787,
|
| 430 |
+
"learning_rate": 0.00015018450184501846,
|
| 431 |
+
"loss": 0.1915,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.029520295202952,
|
| 436 |
+
"grad_norm": 0.14631719887256622,
|
| 437 |
+
"learning_rate": 0.00014926199261992623,
|
| 438 |
+
"loss": 0.2113,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.066420664206642,
|
| 443 |
+
"grad_norm": 0.12011805921792984,
|
| 444 |
+
"learning_rate": 0.00014833948339483394,
|
| 445 |
+
"loss": 0.1781,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.103321033210332,
|
| 450 |
+
"grad_norm": 0.14479252696037292,
|
| 451 |
+
"learning_rate": 0.00014741697416974169,
|
| 452 |
+
"loss": 0.182,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.140221402214022,
|
| 457 |
+
"grad_norm": 0.18806347250938416,
|
| 458 |
+
"learning_rate": 0.00014649446494464946,
|
| 459 |
+
"loss": 0.1682,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.177121771217712,
|
| 464 |
+
"grad_norm": 0.17025235295295715,
|
| 465 |
+
"learning_rate": 0.0001455719557195572,
|
| 466 |
+
"loss": 0.1824,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.2140221402214024,
|
| 471 |
+
"grad_norm": 0.16879422962665558,
|
| 472 |
+
"learning_rate": 0.00014464944649446494,
|
| 473 |
+
"loss": 0.1942,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.2140221402214024,
|
| 478 |
+
"eval_accuracy": 0.41965317919075146,
|
| 479 |
+
"eval_loss": 0.17699980735778809,
|
| 480 |
+
"eval_runtime": 50.0976,
|
| 481 |
+
"eval_samples_per_second": 34.533,
|
| 482 |
+
"eval_steps_per_second": 4.332,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.2509225092250924,
|
| 487 |
+
"grad_norm": 0.168411523103714,
|
| 488 |
+
"learning_rate": 0.0001437269372693727,
|
| 489 |
+
"loss": 0.1732,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.2878228782287824,
|
| 494 |
+
"grad_norm": 0.21096496284008026,
|
| 495 |
+
"learning_rate": 0.00014280442804428045,
|
| 496 |
+
"loss": 0.1842,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.3247232472324724,
|
| 501 |
+
"grad_norm": 0.18110381066799164,
|
| 502 |
+
"learning_rate": 0.0001418819188191882,
|
| 503 |
+
"loss": 0.1772,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.3616236162361623,
|
| 508 |
+
"grad_norm": 0.15854766964912415,
|
| 509 |
+
"learning_rate": 0.00014095940959409593,
|
| 510 |
+
"loss": 0.1709,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.3985239852398523,
|
| 515 |
+
"grad_norm": 0.19320182502269745,
|
| 516 |
+
"learning_rate": 0.0001400369003690037,
|
| 517 |
+
"loss": 0.2,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.4354243542435423,
|
| 522 |
+
"grad_norm": 0.16658619046211243,
|
| 523 |
+
"learning_rate": 0.00013911439114391145,
|
| 524 |
+
"loss": 0.2061,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.4723247232472323,
|
| 529 |
+
"grad_norm": 0.14636483788490295,
|
| 530 |
+
"learning_rate": 0.0001381918819188192,
|
| 531 |
+
"loss": 0.1946,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.5092250922509223,
|
| 536 |
+
"grad_norm": 0.1587982028722763,
|
| 537 |
+
"learning_rate": 0.00013726937269372696,
|
| 538 |
+
"loss": 0.1963,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.5461254612546127,
|
| 543 |
+
"grad_norm": 0.2764102518558502,
|
| 544 |
+
"learning_rate": 0.0001363468634686347,
|
| 545 |
+
"loss": 0.1702,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.5830258302583027,
|
| 550 |
+
"grad_norm": 0.14449751377105713,
|
| 551 |
+
"learning_rate": 0.00013542435424354244,
|
| 552 |
+
"loss": 0.1714,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.5830258302583027,
|
| 557 |
+
"eval_accuracy": 0.5023121387283237,
|
| 558 |
+
"eval_loss": 0.17974236607551575,
|
| 559 |
+
"eval_runtime": 50.796,
|
| 560 |
+
"eval_samples_per_second": 34.058,
|
| 561 |
+
"eval_steps_per_second": 4.272,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.6199261992619927,
|
| 566 |
+
"grad_norm": 0.15392902493476868,
|
| 567 |
+
"learning_rate": 0.0001345018450184502,
|
| 568 |
+
"loss": 0.191,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.6568265682656826,
|
| 573 |
+
"grad_norm": 0.15529021620750427,
|
| 574 |
+
"learning_rate": 0.00013357933579335793,
|
| 575 |
+
"loss": 0.193,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.6937269372693726,
|
| 580 |
+
"grad_norm": 0.18234789371490479,
|
| 581 |
+
"learning_rate": 0.00013265682656826567,
|
| 582 |
+
"loss": 0.1836,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.7306273062730626,
|
| 587 |
+
"grad_norm": 0.19954174757003784,
|
| 588 |
+
"learning_rate": 0.00013173431734317344,
|
| 589 |
+
"loss": 0.2176,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.767527675276753,
|
| 594 |
+
"grad_norm": 0.13893257081508636,
|
| 595 |
+
"learning_rate": 0.00013081180811808118,
|
| 596 |
+
"loss": 0.1699,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.804428044280443,
|
| 601 |
+
"grad_norm": 0.16896647214889526,
|
| 602 |
+
"learning_rate": 0.00012988929889298892,
|
| 603 |
+
"loss": 0.168,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.841328413284133,
|
| 608 |
+
"grad_norm": 0.20796014368534088,
|
| 609 |
+
"learning_rate": 0.0001289667896678967,
|
| 610 |
+
"loss": 0.2141,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.878228782287823,
|
| 615 |
+
"grad_norm": 0.2690466046333313,
|
| 616 |
+
"learning_rate": 0.00012804428044280443,
|
| 617 |
+
"loss": 0.1778,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.915129151291513,
|
| 622 |
+
"grad_norm": 0.14259500801563263,
|
| 623 |
+
"learning_rate": 0.00012712177121771217,
|
| 624 |
+
"loss": 0.1748,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.952029520295203,
|
| 629 |
+
"grad_norm": 0.14488738775253296,
|
| 630 |
+
"learning_rate": 0.00012619926199261994,
|
| 631 |
+
"loss": 0.1832,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.952029520295203,
|
| 636 |
+
"eval_accuracy": 0.36878612716763004,
|
| 637 |
+
"eval_loss": 0.17303667962551117,
|
| 638 |
+
"eval_runtime": 50.3505,
|
| 639 |
+
"eval_samples_per_second": 34.359,
|
| 640 |
+
"eval_steps_per_second": 4.31,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.988929889298893,
|
| 645 |
+
"grad_norm": 0.1963815540075302,
|
| 646 |
+
"learning_rate": 0.00012527675276752769,
|
| 647 |
+
"loss": 0.1603,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 3.025830258302583,
|
| 652 |
+
"grad_norm": 0.18811728060245514,
|
| 653 |
+
"learning_rate": 0.00012435424354243543,
|
| 654 |
+
"loss": 0.1668,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 3.062730627306273,
|
| 659 |
+
"grad_norm": 0.3115330636501312,
|
| 660 |
+
"learning_rate": 0.0001234317343173432,
|
| 661 |
+
"loss": 0.1764,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 3.0996309963099633,
|
| 666 |
+
"grad_norm": 0.15212470293045044,
|
| 667 |
+
"learning_rate": 0.00012250922509225094,
|
| 668 |
+
"loss": 0.1668,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 3.1365313653136533,
|
| 673 |
+
"grad_norm": 0.17935976386070251,
|
| 674 |
+
"learning_rate": 0.00012158671586715868,
|
| 675 |
+
"loss": 0.1807,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.1734317343173433,
|
| 680 |
+
"grad_norm": 0.23978868126869202,
|
| 681 |
+
"learning_rate": 0.00012066420664206644,
|
| 682 |
+
"loss": 0.168,
|
| 683 |
+
"step": 860
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 3.2103321033210332,
|
| 687 |
+
"grad_norm": 0.29603224992752075,
|
| 688 |
+
"learning_rate": 0.00011974169741697419,
|
| 689 |
+
"loss": 0.1827,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.2472324723247232,
|
| 694 |
+
"grad_norm": 0.1385461539030075,
|
| 695 |
+
"learning_rate": 0.00011881918819188192,
|
| 696 |
+
"loss": 0.1794,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 3.284132841328413,
|
| 701 |
+
"grad_norm": 0.20920993387699127,
|
| 702 |
+
"learning_rate": 0.00011789667896678966,
|
| 703 |
+
"loss": 0.1738,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.321033210332103,
|
| 708 |
+
"grad_norm": 0.31590428948402405,
|
| 709 |
+
"learning_rate": 0.00011697416974169742,
|
| 710 |
+
"loss": 0.1766,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.321033210332103,
|
| 715 |
+
"eval_accuracy": 0.34277456647398846,
|
| 716 |
+
"eval_loss": 0.17552779614925385,
|
| 717 |
+
"eval_runtime": 50.5381,
|
| 718 |
+
"eval_samples_per_second": 34.232,
|
| 719 |
+
"eval_steps_per_second": 4.294,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 3.357933579335793,
|
| 724 |
+
"grad_norm": 0.22194945812225342,
|
| 725 |
+
"learning_rate": 0.00011605166051660516,
|
| 726 |
+
"loss": 0.1814,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 3.3948339483394836,
|
| 731 |
+
"grad_norm": 0.22071777284145355,
|
| 732 |
+
"learning_rate": 0.00011512915129151292,
|
| 733 |
+
"loss": 0.1629,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.4317343173431736,
|
| 738 |
+
"grad_norm": 0.44112759828567505,
|
| 739 |
+
"learning_rate": 0.00011420664206642067,
|
| 740 |
+
"loss": 0.1914,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.4686346863468636,
|
| 745 |
+
"grad_norm": 0.20971660315990448,
|
| 746 |
+
"learning_rate": 0.00011328413284132841,
|
| 747 |
+
"loss": 0.1691,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 3.5055350553505535,
|
| 752 |
+
"grad_norm": 0.23813588917255402,
|
| 753 |
+
"learning_rate": 0.00011236162361623617,
|
| 754 |
+
"loss": 0.1919,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 3.5424354243542435,
|
| 759 |
+
"grad_norm": 0.19610780477523804,
|
| 760 |
+
"learning_rate": 0.00011143911439114391,
|
| 761 |
+
"loss": 0.1631,
|
| 762 |
+
"step": 960
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 3.5793357933579335,
|
| 766 |
+
"grad_norm": 0.29578620195388794,
|
| 767 |
+
"learning_rate": 0.00011051660516605167,
|
| 768 |
+
"loss": 0.1721,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 3.6162361623616235,
|
| 773 |
+
"grad_norm": 0.15876761078834534,
|
| 774 |
+
"learning_rate": 0.00010959409594095942,
|
| 775 |
+
"loss": 0.1869,
|
| 776 |
+
"step": 980
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 3.6531365313653135,
|
| 780 |
+
"grad_norm": 0.19575054943561554,
|
| 781 |
+
"learning_rate": 0.00010867158671586716,
|
| 782 |
+
"loss": 0.1676,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 3.6900369003690034,
|
| 787 |
+
"grad_norm": 0.12657958269119263,
|
| 788 |
+
"learning_rate": 0.00010774907749077492,
|
| 789 |
+
"loss": 0.1697,
|
| 790 |
+
"step": 1000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 3.6900369003690034,
|
| 794 |
+
"eval_accuracy": 0.5167630057803468,
|
| 795 |
+
"eval_loss": 0.1601094752550125,
|
| 796 |
+
"eval_runtime": 50.1373,
|
| 797 |
+
"eval_samples_per_second": 34.505,
|
| 798 |
+
"eval_steps_per_second": 4.328,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 3.726937269372694,
|
| 803 |
+
"grad_norm": 0.2477671205997467,
|
| 804 |
+
"learning_rate": 0.00010682656826568268,
|
| 805 |
+
"loss": 0.1745,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 3.763837638376384,
|
| 810 |
+
"grad_norm": 0.21879136562347412,
|
| 811 |
+
"learning_rate": 0.00010590405904059042,
|
| 812 |
+
"loss": 0.1617,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 3.800738007380074,
|
| 817 |
+
"grad_norm": 0.195592001080513,
|
| 818 |
+
"learning_rate": 0.00010498154981549817,
|
| 819 |
+
"loss": 0.1534,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 3.837638376383764,
|
| 824 |
+
"grad_norm": 0.35998597741127014,
|
| 825 |
+
"learning_rate": 0.0001040590405904059,
|
| 826 |
+
"loss": 0.1606,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 3.874538745387454,
|
| 831 |
+
"grad_norm": 0.30765026807785034,
|
| 832 |
+
"learning_rate": 0.00010313653136531364,
|
| 833 |
+
"loss": 0.2019,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 3.911439114391144,
|
| 838 |
+
"grad_norm": 0.16130860149860382,
|
| 839 |
+
"learning_rate": 0.0001022140221402214,
|
| 840 |
+
"loss": 0.1738,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 3.948339483394834,
|
| 845 |
+
"grad_norm": 0.1843736171722412,
|
| 846 |
+
"learning_rate": 0.00010129151291512916,
|
| 847 |
+
"loss": 0.1941,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 3.985239852398524,
|
| 852 |
+
"grad_norm": 0.21090315282344818,
|
| 853 |
+
"learning_rate": 0.0001003690036900369,
|
| 854 |
+
"loss": 0.1695,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 4.022140221402214,
|
| 859 |
+
"grad_norm": 0.19030509889125824,
|
| 860 |
+
"learning_rate": 9.944649446494465e-05,
|
| 861 |
+
"loss": 0.1711,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 4.059040590405904,
|
| 866 |
+
"grad_norm": 0.12992843985557556,
|
| 867 |
+
"learning_rate": 9.85239852398524e-05,
|
| 868 |
+
"loss": 0.1568,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.059040590405904,
|
| 873 |
+
"eval_accuracy": 0.5352601156069364,
|
| 874 |
+
"eval_loss": 0.15768703818321228,
|
| 875 |
+
"eval_runtime": 50.6123,
|
| 876 |
+
"eval_samples_per_second": 34.181,
|
| 877 |
+
"eval_steps_per_second": 4.287,
|
| 878 |
+
"step": 1100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 4.095940959409594,
|
| 882 |
+
"grad_norm": 0.23201997578144073,
|
| 883 |
+
"learning_rate": 9.760147601476015e-05,
|
| 884 |
+
"loss": 0.1484,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 4.132841328413284,
|
| 889 |
+
"grad_norm": 0.3783067762851715,
|
| 890 |
+
"learning_rate": 9.66789667896679e-05,
|
| 891 |
+
"loss": 0.1597,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 4.169741697416974,
|
| 896 |
+
"grad_norm": 0.27165931463241577,
|
| 897 |
+
"learning_rate": 9.575645756457565e-05,
|
| 898 |
+
"loss": 0.156,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 4.206642066420664,
|
| 903 |
+
"grad_norm": 0.2932455241680145,
|
| 904 |
+
"learning_rate": 9.48339483394834e-05,
|
| 905 |
+
"loss": 0.1353,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 4.243542435424354,
|
| 910 |
+
"grad_norm": 0.27856454253196716,
|
| 911 |
+
"learning_rate": 9.391143911439116e-05,
|
| 912 |
+
"loss": 0.1555,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 4.280442804428044,
|
| 917 |
+
"grad_norm": 0.2609305679798126,
|
| 918 |
+
"learning_rate": 9.298892988929889e-05,
|
| 919 |
+
"loss": 0.1549,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 4.317343173431734,
|
| 924 |
+
"grad_norm": 0.4013775587081909,
|
| 925 |
+
"learning_rate": 9.206642066420664e-05,
|
| 926 |
+
"loss": 0.1555,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 4.354243542435424,
|
| 931 |
+
"grad_norm": 0.24482858180999756,
|
| 932 |
+
"learning_rate": 9.11439114391144e-05,
|
| 933 |
+
"loss": 0.1583,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 4.391143911439114,
|
| 938 |
+
"grad_norm": 0.2422870397567749,
|
| 939 |
+
"learning_rate": 9.022140221402214e-05,
|
| 940 |
+
"loss": 0.1663,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 4.428044280442805,
|
| 945 |
+
"grad_norm": 0.2710004448890686,
|
| 946 |
+
"learning_rate": 8.92988929889299e-05,
|
| 947 |
+
"loss": 0.1484,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 4.428044280442805,
|
| 952 |
+
"eval_accuracy": 0.49190751445086706,
|
| 953 |
+
"eval_loss": 0.1513577699661255,
|
| 954 |
+
"eval_runtime": 50.7754,
|
| 955 |
+
"eval_samples_per_second": 34.072,
|
| 956 |
+
"eval_steps_per_second": 4.274,
|
| 957 |
+
"step": 1200
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 4.464944649446495,
|
| 961 |
+
"grad_norm": 0.3608151972293854,
|
| 962 |
+
"learning_rate": 8.837638376383764e-05,
|
| 963 |
+
"loss": 0.1595,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 4.501845018450185,
|
| 968 |
+
"grad_norm": 0.14578911662101746,
|
| 969 |
+
"learning_rate": 8.74538745387454e-05,
|
| 970 |
+
"loss": 0.1841,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 4.538745387453875,
|
| 975 |
+
"grad_norm": 0.2544012665748596,
|
| 976 |
+
"learning_rate": 8.653136531365315e-05,
|
| 977 |
+
"loss": 0.1576,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 4.575645756457565,
|
| 982 |
+
"grad_norm": 0.3130911886692047,
|
| 983 |
+
"learning_rate": 8.560885608856088e-05,
|
| 984 |
+
"loss": 0.1626,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 4.612546125461255,
|
| 989 |
+
"grad_norm": 0.31136009097099304,
|
| 990 |
+
"learning_rate": 8.468634686346863e-05,
|
| 991 |
+
"loss": 0.1715,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 4.649446494464945,
|
| 996 |
+
"grad_norm": 0.20172053575515747,
|
| 997 |
+
"learning_rate": 8.376383763837639e-05,
|
| 998 |
+
"loss": 0.1476,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 4.686346863468635,
|
| 1003 |
+
"grad_norm": 0.2550618648529053,
|
| 1004 |
+
"learning_rate": 8.284132841328413e-05,
|
| 1005 |
+
"loss": 0.1376,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 4.723247232472325,
|
| 1010 |
+
"grad_norm": 0.16149303317070007,
|
| 1011 |
+
"learning_rate": 8.191881918819189e-05,
|
| 1012 |
+
"loss": 0.1461,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 4.760147601476015,
|
| 1017 |
+
"grad_norm": 0.27109894156455994,
|
| 1018 |
+
"learning_rate": 8.099630996309964e-05,
|
| 1019 |
+
"loss": 0.1556,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 4.797047970479705,
|
| 1024 |
+
"grad_norm": 0.26436206698417664,
|
| 1025 |
+
"learning_rate": 8.007380073800739e-05,
|
| 1026 |
+
"loss": 0.1483,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 4.797047970479705,
|
| 1031 |
+
"eval_accuracy": 0.5699421965317919,
|
| 1032 |
+
"eval_loss": 0.14818404614925385,
|
| 1033 |
+
"eval_runtime": 49.9368,
|
| 1034 |
+
"eval_samples_per_second": 34.644,
|
| 1035 |
+
"eval_steps_per_second": 4.345,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 4.833948339483395,
|
| 1040 |
+
"grad_norm": 0.39457815885543823,
|
| 1041 |
+
"learning_rate": 7.915129151291514e-05,
|
| 1042 |
+
"loss": 0.1731,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 4.870848708487085,
|
| 1047 |
+
"grad_norm": 0.1614658087491989,
|
| 1048 |
+
"learning_rate": 7.822878228782288e-05,
|
| 1049 |
+
"loss": 0.1525,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 4.907749077490775,
|
| 1054 |
+
"grad_norm": 0.26091647148132324,
|
| 1055 |
+
"learning_rate": 7.730627306273062e-05,
|
| 1056 |
+
"loss": 0.1854,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 4.944649446494465,
|
| 1061 |
+
"grad_norm": 0.33017560839653015,
|
| 1062 |
+
"learning_rate": 7.638376383763838e-05,
|
| 1063 |
+
"loss": 0.1695,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 4.9815498154981555,
|
| 1068 |
+
"grad_norm": 0.3637866973876953,
|
| 1069 |
+
"learning_rate": 7.546125461254612e-05,
|
| 1070 |
+
"loss": 0.1666,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 5.018450184501845,
|
| 1075 |
+
"grad_norm": 0.3373745083808899,
|
| 1076 |
+
"learning_rate": 7.453874538745388e-05,
|
| 1077 |
+
"loss": 0.1525,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 5.055350553505535,
|
| 1082 |
+
"grad_norm": 0.25333917140960693,
|
| 1083 |
+
"learning_rate": 7.361623616236163e-05,
|
| 1084 |
+
"loss": 0.1356,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 5.092250922509225,
|
| 1089 |
+
"grad_norm": 0.1722867488861084,
|
| 1090 |
+
"learning_rate": 7.269372693726938e-05,
|
| 1091 |
+
"loss": 0.1357,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 5.129151291512915,
|
| 1096 |
+
"grad_norm": 0.13959679007530212,
|
| 1097 |
+
"learning_rate": 7.177121771217713e-05,
|
| 1098 |
+
"loss": 0.1285,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 5.166051660516605,
|
| 1103 |
+
"grad_norm": 0.17668481171131134,
|
| 1104 |
+
"learning_rate": 7.084870848708487e-05,
|
| 1105 |
+
"loss": 0.1301,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 5.166051660516605,
|
| 1110 |
+
"eval_accuracy": 0.5433526011560693,
|
| 1111 |
+
"eval_loss": 0.13149897754192352,
|
| 1112 |
+
"eval_runtime": 50.8795,
|
| 1113 |
+
"eval_samples_per_second": 34.002,
|
| 1114 |
+
"eval_steps_per_second": 4.265,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 5.202952029520295,
|
| 1119 |
+
"grad_norm": 0.1745270937681198,
|
| 1120 |
+
"learning_rate": 6.992619926199262e-05,
|
| 1121 |
+
"loss": 0.1516,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 5.239852398523985,
|
| 1126 |
+
"grad_norm": 0.5758349299430847,
|
| 1127 |
+
"learning_rate": 6.900369003690037e-05,
|
| 1128 |
+
"loss": 0.1294,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 5.276752767527675,
|
| 1133 |
+
"grad_norm": 0.2458232194185257,
|
| 1134 |
+
"learning_rate": 6.808118081180813e-05,
|
| 1135 |
+
"loss": 0.1385,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 5.313653136531365,
|
| 1140 |
+
"grad_norm": 0.3469581604003906,
|
| 1141 |
+
"learning_rate": 6.715867158671587e-05,
|
| 1142 |
+
"loss": 0.1394,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 5.350553505535055,
|
| 1147 |
+
"grad_norm": 0.267447292804718,
|
| 1148 |
+
"learning_rate": 6.623616236162362e-05,
|
| 1149 |
+
"loss": 0.1432,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 5.387453874538745,
|
| 1154 |
+
"grad_norm": 0.24406275153160095,
|
| 1155 |
+
"learning_rate": 6.531365313653137e-05,
|
| 1156 |
+
"loss": 0.1396,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 5.424354243542435,
|
| 1161 |
+
"grad_norm": 0.7067885994911194,
|
| 1162 |
+
"learning_rate": 6.439114391143912e-05,
|
| 1163 |
+
"loss": 0.1456,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 5.461254612546125,
|
| 1168 |
+
"grad_norm": 0.2915806174278259,
|
| 1169 |
+
"learning_rate": 6.346863468634686e-05,
|
| 1170 |
+
"loss": 0.1366,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 5.498154981549815,
|
| 1175 |
+
"grad_norm": 0.22377534210681915,
|
| 1176 |
+
"learning_rate": 6.25461254612546e-05,
|
| 1177 |
+
"loss": 0.1273,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 5.535055350553505,
|
| 1182 |
+
"grad_norm": 0.3705073893070221,
|
| 1183 |
+
"learning_rate": 6.162361623616236e-05,
|
| 1184 |
+
"loss": 0.1149,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 5.535055350553505,
|
| 1189 |
+
"eval_accuracy": 0.5583815028901734,
|
| 1190 |
+
"eval_loss": 0.12937474250793457,
|
| 1191 |
+
"eval_runtime": 50.8394,
|
| 1192 |
+
"eval_samples_per_second": 34.029,
|
| 1193 |
+
"eval_steps_per_second": 4.268,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 5.571955719557195,
|
| 1198 |
+
"grad_norm": 0.13345371186733246,
|
| 1199 |
+
"learning_rate": 6.070110701107011e-05,
|
| 1200 |
+
"loss": 0.1229,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 5.608856088560886,
|
| 1205 |
+
"grad_norm": 0.4122871160507202,
|
| 1206 |
+
"learning_rate": 5.9778597785977866e-05,
|
| 1207 |
+
"loss": 0.1689,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 5.645756457564576,
|
| 1212 |
+
"grad_norm": 0.14905782043933868,
|
| 1213 |
+
"learning_rate": 5.8856088560885615e-05,
|
| 1214 |
+
"loss": 0.1365,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 5.682656826568266,
|
| 1219 |
+
"grad_norm": 0.21198387444019318,
|
| 1220 |
+
"learning_rate": 5.7933579335793364e-05,
|
| 1221 |
+
"loss": 0.1453,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 5.719557195571956,
|
| 1226 |
+
"grad_norm": 0.3941808044910431,
|
| 1227 |
+
"learning_rate": 5.701107011070111e-05,
|
| 1228 |
+
"loss": 0.1584,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 5.756457564575646,
|
| 1233 |
+
"grad_norm": 0.1366042047739029,
|
| 1234 |
+
"learning_rate": 5.6088560885608855e-05,
|
| 1235 |
+
"loss": 0.1219,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 5.793357933579336,
|
| 1240 |
+
"grad_norm": 0.1590586006641388,
|
| 1241 |
+
"learning_rate": 5.5166051660516604e-05,
|
| 1242 |
+
"loss": 0.1482,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 5.830258302583026,
|
| 1247 |
+
"grad_norm": 0.3574014902114868,
|
| 1248 |
+
"learning_rate": 5.424354243542435e-05,
|
| 1249 |
+
"loss": 0.1241,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 5.867158671586716,
|
| 1254 |
+
"grad_norm": 0.2934325039386749,
|
| 1255 |
+
"learning_rate": 5.332103321033211e-05,
|
| 1256 |
+
"loss": 0.1397,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 5.904059040590406,
|
| 1261 |
+
"grad_norm": 0.2349650263786316,
|
| 1262 |
+
"learning_rate": 5.239852398523986e-05,
|
| 1263 |
+
"loss": 0.1448,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 5.904059040590406,
|
| 1268 |
+
"eval_accuracy": 0.5416184971098266,
|
| 1269 |
+
"eval_loss": 0.12662799656391144,
|
| 1270 |
+
"eval_runtime": 50.1064,
|
| 1271 |
+
"eval_samples_per_second": 34.527,
|
| 1272 |
+
"eval_steps_per_second": 4.331,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 5.940959409594096,
|
| 1277 |
+
"grad_norm": 0.39207130670547485,
|
| 1278 |
+
"learning_rate": 5.1476014760147606e-05,
|
| 1279 |
+
"loss": 0.1491,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 5.977859778597786,
|
| 1284 |
+
"grad_norm": 0.21359127759933472,
|
| 1285 |
+
"learning_rate": 5.0553505535055354e-05,
|
| 1286 |
+
"loss": 0.1367,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 6.014760147601476,
|
| 1291 |
+
"grad_norm": 0.17874382436275482,
|
| 1292 |
+
"learning_rate": 4.96309963099631e-05,
|
| 1293 |
+
"loss": 0.1276,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 6.051660516605166,
|
| 1298 |
+
"grad_norm": 0.15224817395210266,
|
| 1299 |
+
"learning_rate": 4.870848708487085e-05,
|
| 1300 |
+
"loss": 0.1223,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 6.088560885608856,
|
| 1305 |
+
"grad_norm": 0.28657016158103943,
|
| 1306 |
+
"learning_rate": 4.77859778597786e-05,
|
| 1307 |
+
"loss": 0.1327,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 6.125461254612546,
|
| 1312 |
+
"grad_norm": 0.16251201927661896,
|
| 1313 |
+
"learning_rate": 4.686346863468635e-05,
|
| 1314 |
+
"loss": 0.1318,
|
| 1315 |
+
"step": 1660
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 6.162361623616236,
|
| 1319 |
+
"grad_norm": 0.3002704381942749,
|
| 1320 |
+
"learning_rate": 4.59409594095941e-05,
|
| 1321 |
+
"loss": 0.1188,
|
| 1322 |
+
"step": 1670
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 6.199261992619927,
|
| 1326 |
+
"grad_norm": 0.4188823103904724,
|
| 1327 |
+
"learning_rate": 4.501845018450185e-05,
|
| 1328 |
+
"loss": 0.1004,
|
| 1329 |
+
"step": 1680
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 6.236162361623617,
|
| 1333 |
+
"grad_norm": 0.18772590160369873,
|
| 1334 |
+
"learning_rate": 4.4095940959409596e-05,
|
| 1335 |
+
"loss": 0.1002,
|
| 1336 |
+
"step": 1690
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 6.273062730627307,
|
| 1340 |
+
"grad_norm": 0.30921700596809387,
|
| 1341 |
+
"learning_rate": 4.3173431734317345e-05,
|
| 1342 |
+
"loss": 0.1035,
|
| 1343 |
+
"step": 1700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 6.273062730627307,
|
| 1347 |
+
"eval_accuracy": 0.6017341040462427,
|
| 1348 |
+
"eval_loss": 0.11507368832826614,
|
| 1349 |
+
"eval_runtime": 50.6734,
|
| 1350 |
+
"eval_samples_per_second": 34.14,
|
| 1351 |
+
"eval_steps_per_second": 4.282,
|
| 1352 |
+
"step": 1700
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 6.3099630996309966,
|
| 1356 |
+
"grad_norm": 0.42562779784202576,
|
| 1357 |
+
"learning_rate": 4.2250922509225094e-05,
|
| 1358 |
+
"loss": 0.1071,
|
| 1359 |
+
"step": 1710
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 6.3468634686346865,
|
| 1363 |
+
"grad_norm": 0.36547404527664185,
|
| 1364 |
+
"learning_rate": 4.132841328413284e-05,
|
| 1365 |
+
"loss": 0.12,
|
| 1366 |
+
"step": 1720
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 6.3837638376383765,
|
| 1370 |
+
"grad_norm": 0.12006784975528717,
|
| 1371 |
+
"learning_rate": 4.040590405904059e-05,
|
| 1372 |
+
"loss": 0.1107,
|
| 1373 |
+
"step": 1730
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 6.4206642066420665,
|
| 1377 |
+
"grad_norm": 0.1983233392238617,
|
| 1378 |
+
"learning_rate": 3.948339483394834e-05,
|
| 1379 |
+
"loss": 0.1206,
|
| 1380 |
+
"step": 1740
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 6.4575645756457565,
|
| 1384 |
+
"grad_norm": 0.17691943049430847,
|
| 1385 |
+
"learning_rate": 3.856088560885609e-05,
|
| 1386 |
+
"loss": 0.1252,
|
| 1387 |
+
"step": 1750
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 6.4944649446494465,
|
| 1391 |
+
"grad_norm": 0.39386728405952454,
|
| 1392 |
+
"learning_rate": 3.763837638376384e-05,
|
| 1393 |
+
"loss": 0.1314,
|
| 1394 |
+
"step": 1760
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 6.531365313653136,
|
| 1398 |
+
"grad_norm": 0.607455313205719,
|
| 1399 |
+
"learning_rate": 3.6715867158671594e-05,
|
| 1400 |
+
"loss": 0.1095,
|
| 1401 |
+
"step": 1770
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 6.568265682656826,
|
| 1405 |
+
"grad_norm": 0.21057389676570892,
|
| 1406 |
+
"learning_rate": 3.5793357933579336e-05,
|
| 1407 |
+
"loss": 0.1223,
|
| 1408 |
+
"step": 1780
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 6.605166051660516,
|
| 1412 |
+
"grad_norm": 0.27539491653442383,
|
| 1413 |
+
"learning_rate": 3.4870848708487085e-05,
|
| 1414 |
+
"loss": 0.1163,
|
| 1415 |
+
"step": 1790
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 6.642066420664206,
|
| 1419 |
+
"grad_norm": 0.24495290219783783,
|
| 1420 |
+
"learning_rate": 3.3948339483394833e-05,
|
| 1421 |
+
"loss": 0.1048,
|
| 1422 |
+
"step": 1800
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 6.642066420664206,
|
| 1426 |
+
"eval_accuracy": 0.6046242774566474,
|
| 1427 |
+
"eval_loss": 0.10599144548177719,
|
| 1428 |
+
"eval_runtime": 50.9957,
|
| 1429 |
+
"eval_samples_per_second": 33.924,
|
| 1430 |
+
"eval_steps_per_second": 4.255,
|
| 1431 |
+
"step": 1800
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 6.678966789667896,
|
| 1435 |
+
"grad_norm": 0.38892611861228943,
|
| 1436 |
+
"learning_rate": 3.302583025830259e-05,
|
| 1437 |
+
"loss": 0.1352,
|
| 1438 |
+
"step": 1810
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 6.715867158671586,
|
| 1442 |
+
"grad_norm": 0.2850606143474579,
|
| 1443 |
+
"learning_rate": 3.210332103321033e-05,
|
| 1444 |
+
"loss": 0.1153,
|
| 1445 |
+
"step": 1820
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 6.752767527675276,
|
| 1449 |
+
"grad_norm": 0.16241934895515442,
|
| 1450 |
+
"learning_rate": 3.118081180811808e-05,
|
| 1451 |
+
"loss": 0.1074,
|
| 1452 |
+
"step": 1830
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 6.789667896678967,
|
| 1456 |
+
"grad_norm": 0.36088794469833374,
|
| 1457 |
+
"learning_rate": 3.0258302583025832e-05,
|
| 1458 |
+
"loss": 0.1219,
|
| 1459 |
+
"step": 1840
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 6.826568265682657,
|
| 1463 |
+
"grad_norm": 0.21467632055282593,
|
| 1464 |
+
"learning_rate": 2.9335793357933584e-05,
|
| 1465 |
+
"loss": 0.1083,
|
| 1466 |
+
"step": 1850
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 6.863468634686347,
|
| 1470 |
+
"grad_norm": 0.4730125069618225,
|
| 1471 |
+
"learning_rate": 2.8413284132841326e-05,
|
| 1472 |
+
"loss": 0.1227,
|
| 1473 |
+
"step": 1860
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 6.900369003690037,
|
| 1477 |
+
"grad_norm": 0.20842638611793518,
|
| 1478 |
+
"learning_rate": 2.749077490774908e-05,
|
| 1479 |
+
"loss": 0.128,
|
| 1480 |
+
"step": 1870
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 6.937269372693727,
|
| 1484 |
+
"grad_norm": 0.1885102540254593,
|
| 1485 |
+
"learning_rate": 2.6568265682656828e-05,
|
| 1486 |
+
"loss": 0.0923,
|
| 1487 |
+
"step": 1880
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 6.974169741697417,
|
| 1491 |
+
"grad_norm": 0.48948994278907776,
|
| 1492 |
+
"learning_rate": 2.564575645756458e-05,
|
| 1493 |
+
"loss": 0.1165,
|
| 1494 |
+
"step": 1890
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 7.011070110701107,
|
| 1498 |
+
"grad_norm": 0.4080180525779724,
|
| 1499 |
+
"learning_rate": 2.472324723247233e-05,
|
| 1500 |
+
"loss": 0.1168,
|
| 1501 |
+
"step": 1900
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 7.011070110701107,
|
| 1505 |
+
"eval_accuracy": 0.6173410404624278,
|
| 1506 |
+
"eval_loss": 0.10073487460613251,
|
| 1507 |
+
"eval_runtime": 50.8605,
|
| 1508 |
+
"eval_samples_per_second": 34.015,
|
| 1509 |
+
"eval_steps_per_second": 4.267,
|
| 1510 |
+
"step": 1900
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 7.047970479704797,
|
| 1514 |
+
"grad_norm": 0.19095434248447418,
|
| 1515 |
+
"learning_rate": 2.3800738007380074e-05,
|
| 1516 |
+
"loss": 0.1131,
|
| 1517 |
+
"step": 1910
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 7.084870848708487,
|
| 1521 |
+
"grad_norm": 0.23603685200214386,
|
| 1522 |
+
"learning_rate": 2.2878228782287826e-05,
|
| 1523 |
+
"loss": 0.089,
|
| 1524 |
+
"step": 1920
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 7.121771217712177,
|
| 1528 |
+
"grad_norm": 0.09547635912895203,
|
| 1529 |
+
"learning_rate": 2.195571955719557e-05,
|
| 1530 |
+
"loss": 0.1032,
|
| 1531 |
+
"step": 1930
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 7.158671586715867,
|
| 1535 |
+
"grad_norm": 0.18442951142787933,
|
| 1536 |
+
"learning_rate": 2.1033210332103324e-05,
|
| 1537 |
+
"loss": 0.0909,
|
| 1538 |
+
"step": 1940
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 7.195571955719557,
|
| 1542 |
+
"grad_norm": 0.2125350534915924,
|
| 1543 |
+
"learning_rate": 2.011070110701107e-05,
|
| 1544 |
+
"loss": 0.0922,
|
| 1545 |
+
"step": 1950
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 7.232472324723247,
|
| 1549 |
+
"grad_norm": 0.15140217542648315,
|
| 1550 |
+
"learning_rate": 1.918819188191882e-05,
|
| 1551 |
+
"loss": 0.1155,
|
| 1552 |
+
"step": 1960
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 7.269372693726937,
|
| 1556 |
+
"grad_norm": 0.25905662775039673,
|
| 1557 |
+
"learning_rate": 1.826568265682657e-05,
|
| 1558 |
+
"loss": 0.1194,
|
| 1559 |
+
"step": 1970
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 7.306273062730627,
|
| 1563 |
+
"grad_norm": 0.18217885494232178,
|
| 1564 |
+
"learning_rate": 1.734317343173432e-05,
|
| 1565 |
+
"loss": 0.1,
|
| 1566 |
+
"step": 1980
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 7.343173431734318,
|
| 1570 |
+
"grad_norm": 0.21871539950370789,
|
| 1571 |
+
"learning_rate": 1.6420664206642068e-05,
|
| 1572 |
+
"loss": 0.1022,
|
| 1573 |
+
"step": 1990
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 7.380073800738008,
|
| 1577 |
+
"grad_norm": 0.4127865731716156,
|
| 1578 |
+
"learning_rate": 1.5498154981549817e-05,
|
| 1579 |
+
"loss": 0.1104,
|
| 1580 |
+
"step": 2000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 7.380073800738008,
|
| 1584 |
+
"eval_accuracy": 0.6445086705202312,
|
| 1585 |
+
"eval_loss": 0.09489033371210098,
|
| 1586 |
+
"eval_runtime": 50.8042,
|
| 1587 |
+
"eval_samples_per_second": 34.052,
|
| 1588 |
+
"eval_steps_per_second": 4.271,
|
| 1589 |
+
"step": 2000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 7.416974169741698,
|
| 1593 |
+
"grad_norm": 0.20000500977039337,
|
| 1594 |
+
"learning_rate": 1.4575645756457566e-05,
|
| 1595 |
+
"loss": 0.1031,
|
| 1596 |
+
"step": 2010
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 7.453874538745388,
|
| 1600 |
+
"grad_norm": 0.5234202742576599,
|
| 1601 |
+
"learning_rate": 1.3653136531365315e-05,
|
| 1602 |
+
"loss": 0.1197,
|
| 1603 |
+
"step": 2020
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 7.490774907749078,
|
| 1607 |
+
"grad_norm": 0.16442282497882843,
|
| 1608 |
+
"learning_rate": 1.2730627306273063e-05,
|
| 1609 |
+
"loss": 0.1036,
|
| 1610 |
+
"step": 2030
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 7.527675276752768,
|
| 1614 |
+
"grad_norm": 0.19210496544837952,
|
| 1615 |
+
"learning_rate": 1.1808118081180812e-05,
|
| 1616 |
+
"loss": 0.0993,
|
| 1617 |
+
"step": 2040
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 7.564575645756458,
|
| 1621 |
+
"grad_norm": 0.1562729775905609,
|
| 1622 |
+
"learning_rate": 1.0885608856088561e-05,
|
| 1623 |
+
"loss": 0.0941,
|
| 1624 |
+
"step": 2050
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 7.601476014760148,
|
| 1628 |
+
"grad_norm": 0.29051193594932556,
|
| 1629 |
+
"learning_rate": 9.96309963099631e-06,
|
| 1630 |
+
"loss": 0.092,
|
| 1631 |
+
"step": 2060
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 7.638376383763838,
|
| 1635 |
+
"grad_norm": 0.21477282047271729,
|
| 1636 |
+
"learning_rate": 9.040590405904059e-06,
|
| 1637 |
+
"loss": 0.1123,
|
| 1638 |
+
"step": 2070
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 7.675276752767528,
|
| 1642 |
+
"grad_norm": 0.03506307676434517,
|
| 1643 |
+
"learning_rate": 8.118081180811808e-06,
|
| 1644 |
+
"loss": 0.0927,
|
| 1645 |
+
"step": 2080
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 7.712177121771218,
|
| 1649 |
+
"grad_norm": 0.21280255913734436,
|
| 1650 |
+
"learning_rate": 7.195571955719557e-06,
|
| 1651 |
+
"loss": 0.084,
|
| 1652 |
+
"step": 2090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 7.749077490774908,
|
| 1656 |
+
"grad_norm": 0.19547449052333832,
|
| 1657 |
+
"learning_rate": 6.273062730627306e-06,
|
| 1658 |
+
"loss": 0.0873,
|
| 1659 |
+
"step": 2100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 7.749077490774908,
|
| 1663 |
+
"eval_accuracy": 0.6526011560693642,
|
| 1664 |
+
"eval_loss": 0.0923289805650711,
|
| 1665 |
+
"eval_runtime": 50.6448,
|
| 1666 |
+
"eval_samples_per_second": 34.159,
|
| 1667 |
+
"eval_steps_per_second": 4.285,
|
| 1668 |
+
"step": 2100
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 7.785977859778598,
|
| 1672 |
+
"grad_norm": 0.22129392623901367,
|
| 1673 |
+
"learning_rate": 5.350553505535055e-06,
|
| 1674 |
+
"loss": 0.1,
|
| 1675 |
+
"step": 2110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 7.822878228782288,
|
| 1679 |
+
"grad_norm": 0.2631789445877075,
|
| 1680 |
+
"learning_rate": 4.428044280442805e-06,
|
| 1681 |
+
"loss": 0.0811,
|
| 1682 |
+
"step": 2120
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 7.8597785977859775,
|
| 1686 |
+
"grad_norm": 0.16971804201602936,
|
| 1687 |
+
"learning_rate": 3.5055350553505534e-06,
|
| 1688 |
+
"loss": 0.0977,
|
| 1689 |
+
"step": 2130
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 7.8966789667896675,
|
| 1693 |
+
"grad_norm": 0.10247929394245148,
|
| 1694 |
+
"learning_rate": 2.5830258302583027e-06,
|
| 1695 |
+
"loss": 0.0966,
|
| 1696 |
+
"step": 2140
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 7.9335793357933575,
|
| 1700 |
+
"grad_norm": 0.15865936875343323,
|
| 1701 |
+
"learning_rate": 1.6605166051660517e-06,
|
| 1702 |
+
"loss": 0.1039,
|
| 1703 |
+
"step": 2150
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 7.970479704797048,
|
| 1707 |
+
"grad_norm": 0.506331205368042,
|
| 1708 |
+
"learning_rate": 7.380073800738008e-07,
|
| 1709 |
+
"loss": 0.0977,
|
| 1710 |
+
"step": 2160
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 8.0,
|
| 1714 |
+
"step": 2168,
|
| 1715 |
+
"total_flos": 2.6821552511927255e+18,
|
| 1716 |
+
"train_loss": 0.16216810325304962,
|
| 1717 |
+
"train_runtime": 3143.5173,
|
| 1718 |
+
"train_samples_per_second": 11.009,
|
| 1719 |
+
"train_steps_per_second": 0.69
|
| 1720 |
+
}
|
| 1721 |
+
],
|
| 1722 |
+
"logging_steps": 10,
|
| 1723 |
+
"max_steps": 2168,
|
| 1724 |
+
"num_input_tokens_seen": 0,
|
| 1725 |
+
"num_train_epochs": 8,
|
| 1726 |
+
"save_steps": 100,
|
| 1727 |
+
"stateful_callbacks": {
|
| 1728 |
+
"TrainerControl": {
|
| 1729 |
+
"args": {
|
| 1730 |
+
"should_epoch_stop": false,
|
| 1731 |
+
"should_evaluate": false,
|
| 1732 |
+
"should_log": false,
|
| 1733 |
+
"should_save": true,
|
| 1734 |
+
"should_training_stop": true
|
| 1735 |
+
},
|
| 1736 |
+
"attributes": {}
|
| 1737 |
+
}
|
| 1738 |
+
},
|
| 1739 |
+
"total_flos": 2.6821552511927255e+18,
|
| 1740 |
+
"train_batch_size": 16,
|
| 1741 |
+
"trial_name": null,
|
| 1742 |
+
"trial_params": null
|
| 1743 |
+
}
|