🍻 cheers
Browse files- README.md +4 -3
- all_results.json +14 -0
- eval_results.json +9 -0
- runs/Dec14_08-45-43_592abeefaa6c/events.out.tfevents.1734166523.592abeefaa6c.3097.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1037 -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
|
|
@@ -16,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# hurricane_model
|
| 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 |
- Model Preparation Time: 0.0051
|
| 23 |
-
- Accuracy: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 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 |
# hurricane_model
|
| 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 jonathan-roberts1/Satellite-Images-of-Hurricane-Damage dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.0224
|
| 23 |
- Model Preparation Time: 0.0051
|
| 24 |
+
- Accuracy: 0.9948
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9948,
|
| 4 |
+
"eval_loss": 0.02237132005393505,
|
| 5 |
+
"eval_model_preparation_time": 0.0051,
|
| 6 |
+
"eval_runtime": 12.589,
|
| 7 |
+
"eval_samples_per_second": 198.586,
|
| 8 |
+
"eval_steps_per_second": 24.863,
|
| 9 |
+
"total_flos": 1.54983979229184e+18,
|
| 10 |
+
"train_loss": 0.055932876091605174,
|
| 11 |
+
"train_runtime": 533.4054,
|
| 12 |
+
"train_samples_per_second": 37.495,
|
| 13 |
+
"train_steps_per_second": 2.347
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9948,
|
| 4 |
+
"eval_loss": 0.02237132005393505,
|
| 5 |
+
"eval_model_preparation_time": 0.0051,
|
| 6 |
+
"eval_runtime": 12.589,
|
| 7 |
+
"eval_samples_per_second": 198.586,
|
| 8 |
+
"eval_steps_per_second": 24.863
|
| 9 |
+
}
|
runs/Dec14_08-45-43_592abeefaa6c/events.out.tfevents.1734166523.592abeefaa6c.3097.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1ebb2174c437722e40eb8b0ebc8d636a0024f4366289293bae21edc477590fe
|
| 3 |
+
size 477
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.54983979229184e+18,
|
| 4 |
+
"train_loss": 0.055932876091605174,
|
| 5 |
+
"train_runtime": 533.4054,
|
| 6 |
+
"train_samples_per_second": 37.495,
|
| 7 |
+
"train_steps_per_second": 2.347
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1037 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.02237132005393505,
|
| 3 |
+
"best_model_checkpoint": "hurricane_model/checkpoint-1000",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1252,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03194888178913738,
|
| 13 |
+
"grad_norm": 1.788046956062317,
|
| 14 |
+
"learning_rate": 0.00019840255591054313,
|
| 15 |
+
"loss": 0.5871,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06389776357827476,
|
| 20 |
+
"grad_norm": 1.5475027561187744,
|
| 21 |
+
"learning_rate": 0.00019680511182108628,
|
| 22 |
+
"loss": 0.283,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09584664536741214,
|
| 27 |
+
"grad_norm": 6.500663757324219,
|
| 28 |
+
"learning_rate": 0.0001952076677316294,
|
| 29 |
+
"loss": 0.1913,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.12779552715654952,
|
| 34 |
+
"grad_norm": 11.902435302734375,
|
| 35 |
+
"learning_rate": 0.00019361022364217253,
|
| 36 |
+
"loss": 0.2797,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1597444089456869,
|
| 41 |
+
"grad_norm": 0.10494810342788696,
|
| 42 |
+
"learning_rate": 0.00019201277955271565,
|
| 43 |
+
"loss": 0.1443,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19169329073482427,
|
| 48 |
+
"grad_norm": 3.824563980102539,
|
| 49 |
+
"learning_rate": 0.0001904153354632588,
|
| 50 |
+
"loss": 0.1807,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22364217252396165,
|
| 55 |
+
"grad_norm": 5.14697790145874,
|
| 56 |
+
"learning_rate": 0.00018881789137380192,
|
| 57 |
+
"loss": 0.2322,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.25559105431309903,
|
| 62 |
+
"grad_norm": 1.0805331468582153,
|
| 63 |
+
"learning_rate": 0.00018722044728434505,
|
| 64 |
+
"loss": 0.1309,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.28753993610223644,
|
| 69 |
+
"grad_norm": 0.8772349953651428,
|
| 70 |
+
"learning_rate": 0.0001856230031948882,
|
| 71 |
+
"loss": 0.2684,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3194888178913738,
|
| 76 |
+
"grad_norm": 0.09347983449697495,
|
| 77 |
+
"learning_rate": 0.00018402555910543132,
|
| 78 |
+
"loss": 0.1118,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3194888178913738,
|
| 83 |
+
"eval_accuracy": 0.9476,
|
| 84 |
+
"eval_loss": 0.14858174324035645,
|
| 85 |
+
"eval_model_preparation_time": 0.0051,
|
| 86 |
+
"eval_runtime": 14.0146,
|
| 87 |
+
"eval_samples_per_second": 178.385,
|
| 88 |
+
"eval_steps_per_second": 22.334,
|
| 89 |
+
"step": 100
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.3514376996805112,
|
| 93 |
+
"grad_norm": 0.8523539304733276,
|
| 94 |
+
"learning_rate": 0.00018242811501597444,
|
| 95 |
+
"loss": 0.1224,
|
| 96 |
+
"step": 110
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.38338658146964855,
|
| 100 |
+
"grad_norm": 26.86109733581543,
|
| 101 |
+
"learning_rate": 0.00018083067092651756,
|
| 102 |
+
"loss": 0.1597,
|
| 103 |
+
"step": 120
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.41533546325878595,
|
| 107 |
+
"grad_norm": 0.07714489102363586,
|
| 108 |
+
"learning_rate": 0.00017923322683706071,
|
| 109 |
+
"loss": 0.3385,
|
| 110 |
+
"step": 130
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.4472843450479233,
|
| 114 |
+
"grad_norm": 0.46879705786705017,
|
| 115 |
+
"learning_rate": 0.00017763578274760384,
|
| 116 |
+
"loss": 0.1338,
|
| 117 |
+
"step": 140
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.4792332268370607,
|
| 121 |
+
"grad_norm": 0.19195349514484406,
|
| 122 |
+
"learning_rate": 0.000176038338658147,
|
| 123 |
+
"loss": 0.1792,
|
| 124 |
+
"step": 150
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.5111821086261981,
|
| 128 |
+
"grad_norm": 1.775911569595337,
|
| 129 |
+
"learning_rate": 0.0001744408945686901,
|
| 130 |
+
"loss": 0.142,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.5431309904153354,
|
| 135 |
+
"grad_norm": 3.7460222244262695,
|
| 136 |
+
"learning_rate": 0.00017284345047923323,
|
| 137 |
+
"loss": 0.2113,
|
| 138 |
+
"step": 170
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.5750798722044729,
|
| 142 |
+
"grad_norm": 1.6586716175079346,
|
| 143 |
+
"learning_rate": 0.00017124600638977638,
|
| 144 |
+
"loss": 0.2379,
|
| 145 |
+
"step": 180
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.6070287539936102,
|
| 149 |
+
"grad_norm": 0.37085050344467163,
|
| 150 |
+
"learning_rate": 0.00016964856230031948,
|
| 151 |
+
"loss": 0.0626,
|
| 152 |
+
"step": 190
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.6389776357827476,
|
| 156 |
+
"grad_norm": 1.5378907918930054,
|
| 157 |
+
"learning_rate": 0.00016805111821086263,
|
| 158 |
+
"loss": 0.1112,
|
| 159 |
+
"step": 200
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.6389776357827476,
|
| 163 |
+
"eval_accuracy": 0.9752,
|
| 164 |
+
"eval_loss": 0.0701194703578949,
|
| 165 |
+
"eval_model_preparation_time": 0.0051,
|
| 166 |
+
"eval_runtime": 12.3233,
|
| 167 |
+
"eval_samples_per_second": 202.868,
|
| 168 |
+
"eval_steps_per_second": 25.399,
|
| 169 |
+
"step": 200
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.670926517571885,
|
| 173 |
+
"grad_norm": 0.3154517114162445,
|
| 174 |
+
"learning_rate": 0.00016645367412140575,
|
| 175 |
+
"loss": 0.051,
|
| 176 |
+
"step": 210
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.7028753993610224,
|
| 180 |
+
"grad_norm": 0.05547064542770386,
|
| 181 |
+
"learning_rate": 0.0001648562300319489,
|
| 182 |
+
"loss": 0.0512,
|
| 183 |
+
"step": 220
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.7348242811501597,
|
| 187 |
+
"grad_norm": 0.0428142175078392,
|
| 188 |
+
"learning_rate": 0.00016325878594249202,
|
| 189 |
+
"loss": 0.1203,
|
| 190 |
+
"step": 230
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.7667731629392971,
|
| 194 |
+
"grad_norm": 0.07227767258882523,
|
| 195 |
+
"learning_rate": 0.00016166134185303515,
|
| 196 |
+
"loss": 0.1993,
|
| 197 |
+
"step": 240
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.7987220447284346,
|
| 201 |
+
"grad_norm": 4.76859188079834,
|
| 202 |
+
"learning_rate": 0.0001600638977635783,
|
| 203 |
+
"loss": 0.0644,
|
| 204 |
+
"step": 250
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.8306709265175719,
|
| 208 |
+
"grad_norm": 0.06700027734041214,
|
| 209 |
+
"learning_rate": 0.00015846645367412142,
|
| 210 |
+
"loss": 0.0898,
|
| 211 |
+
"step": 260
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.8626198083067093,
|
| 215 |
+
"grad_norm": 0.14176583290100098,
|
| 216 |
+
"learning_rate": 0.00015686900958466454,
|
| 217 |
+
"loss": 0.1146,
|
| 218 |
+
"step": 270
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.8945686900958466,
|
| 222 |
+
"grad_norm": 0.23704954981803894,
|
| 223 |
+
"learning_rate": 0.00015527156549520767,
|
| 224 |
+
"loss": 0.1933,
|
| 225 |
+
"step": 280
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.9265175718849841,
|
| 229 |
+
"grad_norm": 0.32876431941986084,
|
| 230 |
+
"learning_rate": 0.00015367412140575082,
|
| 231 |
+
"loss": 0.1404,
|
| 232 |
+
"step": 290
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.9584664536741214,
|
| 236 |
+
"grad_norm": 0.569324791431427,
|
| 237 |
+
"learning_rate": 0.00015207667731629394,
|
| 238 |
+
"loss": 0.0694,
|
| 239 |
+
"step": 300
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.9584664536741214,
|
| 243 |
+
"eval_accuracy": 0.9808,
|
| 244 |
+
"eval_loss": 0.06084931641817093,
|
| 245 |
+
"eval_model_preparation_time": 0.0051,
|
| 246 |
+
"eval_runtime": 12.4406,
|
| 247 |
+
"eval_samples_per_second": 200.956,
|
| 248 |
+
"eval_steps_per_second": 25.16,
|
| 249 |
+
"step": 300
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.9904153354632588,
|
| 253 |
+
"grad_norm": 2.446889877319336,
|
| 254 |
+
"learning_rate": 0.00015047923322683706,
|
| 255 |
+
"loss": 0.0315,
|
| 256 |
+
"step": 310
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.0223642172523961,
|
| 260 |
+
"grad_norm": 0.13555611670017242,
|
| 261 |
+
"learning_rate": 0.0001488817891373802,
|
| 262 |
+
"loss": 0.0476,
|
| 263 |
+
"step": 320
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.0543130990415335,
|
| 267 |
+
"grad_norm": 0.076418437063694,
|
| 268 |
+
"learning_rate": 0.00014728434504792333,
|
| 269 |
+
"loss": 0.0388,
|
| 270 |
+
"step": 330
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.0862619808306708,
|
| 274 |
+
"grad_norm": 0.034216273576021194,
|
| 275 |
+
"learning_rate": 0.00014568690095846646,
|
| 276 |
+
"loss": 0.0514,
|
| 277 |
+
"step": 340
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.1182108626198084,
|
| 281 |
+
"grad_norm": 0.8253958821296692,
|
| 282 |
+
"learning_rate": 0.00014408945686900958,
|
| 283 |
+
"loss": 0.0734,
|
| 284 |
+
"step": 350
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.1501597444089458,
|
| 288 |
+
"grad_norm": 0.5345448851585388,
|
| 289 |
+
"learning_rate": 0.00014249201277955273,
|
| 290 |
+
"loss": 0.0159,
|
| 291 |
+
"step": 360
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.182108626198083,
|
| 295 |
+
"grad_norm": 0.031145131215453148,
|
| 296 |
+
"learning_rate": 0.00014089456869009585,
|
| 297 |
+
"loss": 0.1411,
|
| 298 |
+
"step": 370
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.2140575079872205,
|
| 302 |
+
"grad_norm": 2.8567800521850586,
|
| 303 |
+
"learning_rate": 0.000139297124600639,
|
| 304 |
+
"loss": 0.0221,
|
| 305 |
+
"step": 380
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.2460063897763578,
|
| 309 |
+
"grad_norm": 0.06364869326353073,
|
| 310 |
+
"learning_rate": 0.00013769968051118212,
|
| 311 |
+
"loss": 0.0326,
|
| 312 |
+
"step": 390
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.2779552715654952,
|
| 316 |
+
"grad_norm": 0.024139605462551117,
|
| 317 |
+
"learning_rate": 0.00013610223642172525,
|
| 318 |
+
"loss": 0.0048,
|
| 319 |
+
"step": 400
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.2779552715654952,
|
| 323 |
+
"eval_accuracy": 0.9744,
|
| 324 |
+
"eval_loss": 0.09166774898767471,
|
| 325 |
+
"eval_model_preparation_time": 0.0051,
|
| 326 |
+
"eval_runtime": 12.5348,
|
| 327 |
+
"eval_samples_per_second": 199.445,
|
| 328 |
+
"eval_steps_per_second": 24.971,
|
| 329 |
+
"step": 400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.3099041533546325,
|
| 333 |
+
"grad_norm": 0.061418768018484116,
|
| 334 |
+
"learning_rate": 0.00013450479233226837,
|
| 335 |
+
"loss": 0.0336,
|
| 336 |
+
"step": 410
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.34185303514377,
|
| 340 |
+
"grad_norm": 0.04032498970627785,
|
| 341 |
+
"learning_rate": 0.0001329073482428115,
|
| 342 |
+
"loss": 0.0272,
|
| 343 |
+
"step": 420
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.3738019169329074,
|
| 347 |
+
"grad_norm": 0.05345893278717995,
|
| 348 |
+
"learning_rate": 0.00013130990415335464,
|
| 349 |
+
"loss": 0.0232,
|
| 350 |
+
"step": 430
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.4057507987220448,
|
| 354 |
+
"grad_norm": 0.1362222582101822,
|
| 355 |
+
"learning_rate": 0.00012971246006389777,
|
| 356 |
+
"loss": 0.0234,
|
| 357 |
+
"step": 440
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.4376996805111821,
|
| 361 |
+
"grad_norm": 2.3406131267547607,
|
| 362 |
+
"learning_rate": 0.00012811501597444092,
|
| 363 |
+
"loss": 0.0339,
|
| 364 |
+
"step": 450
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.4696485623003195,
|
| 368 |
+
"grad_norm": 0.01726055145263672,
|
| 369 |
+
"learning_rate": 0.00012651757188498404,
|
| 370 |
+
"loss": 0.0329,
|
| 371 |
+
"step": 460
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.5015974440894568,
|
| 375 |
+
"grad_norm": 0.021831806749105453,
|
| 376 |
+
"learning_rate": 0.00012492012779552716,
|
| 377 |
+
"loss": 0.0539,
|
| 378 |
+
"step": 470
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.5335463258785942,
|
| 382 |
+
"grad_norm": 0.01614479534327984,
|
| 383 |
+
"learning_rate": 0.00012332268370607028,
|
| 384 |
+
"loss": 0.071,
|
| 385 |
+
"step": 480
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.5654952076677318,
|
| 389 |
+
"grad_norm": 0.2092825472354889,
|
| 390 |
+
"learning_rate": 0.00012172523961661342,
|
| 391 |
+
"loss": 0.0503,
|
| 392 |
+
"step": 490
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.5974440894568689,
|
| 396 |
+
"grad_norm": 0.02242388017475605,
|
| 397 |
+
"learning_rate": 0.00012012779552715656,
|
| 398 |
+
"loss": 0.036,
|
| 399 |
+
"step": 500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.5974440894568689,
|
| 403 |
+
"eval_accuracy": 0.9836,
|
| 404 |
+
"eval_loss": 0.05519802123308182,
|
| 405 |
+
"eval_model_preparation_time": 0.0051,
|
| 406 |
+
"eval_runtime": 12.5164,
|
| 407 |
+
"eval_samples_per_second": 199.738,
|
| 408 |
+
"eval_steps_per_second": 25.007,
|
| 409 |
+
"step": 500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.6293929712460065,
|
| 413 |
+
"grad_norm": 1.8705270290374756,
|
| 414 |
+
"learning_rate": 0.00011853035143769968,
|
| 415 |
+
"loss": 0.0691,
|
| 416 |
+
"step": 510
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.6613418530351438,
|
| 420 |
+
"grad_norm": 0.12965114414691925,
|
| 421 |
+
"learning_rate": 0.00011693290734824283,
|
| 422 |
+
"loss": 0.0254,
|
| 423 |
+
"step": 520
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.6932907348242812,
|
| 427 |
+
"grad_norm": 0.2740132212638855,
|
| 428 |
+
"learning_rate": 0.00011533546325878595,
|
| 429 |
+
"loss": 0.0508,
|
| 430 |
+
"step": 530
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.7252396166134185,
|
| 434 |
+
"grad_norm": 0.01778402552008629,
|
| 435 |
+
"learning_rate": 0.00011373801916932908,
|
| 436 |
+
"loss": 0.0145,
|
| 437 |
+
"step": 540
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.7571884984025559,
|
| 441 |
+
"grad_norm": 0.014663617126643658,
|
| 442 |
+
"learning_rate": 0.00011214057507987221,
|
| 443 |
+
"loss": 0.0315,
|
| 444 |
+
"step": 550
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.7891373801916934,
|
| 448 |
+
"grad_norm": 0.015493140555918217,
|
| 449 |
+
"learning_rate": 0.00011054313099041533,
|
| 450 |
+
"loss": 0.1852,
|
| 451 |
+
"step": 560
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.8210862619808306,
|
| 455 |
+
"grad_norm": 0.05500736087560654,
|
| 456 |
+
"learning_rate": 0.00010894568690095847,
|
| 457 |
+
"loss": 0.0292,
|
| 458 |
+
"step": 570
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.8530351437699681,
|
| 462 |
+
"grad_norm": 0.014051590114831924,
|
| 463 |
+
"learning_rate": 0.0001073482428115016,
|
| 464 |
+
"loss": 0.0193,
|
| 465 |
+
"step": 580
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.8849840255591053,
|
| 469 |
+
"grad_norm": 0.050858091562986374,
|
| 470 |
+
"learning_rate": 0.00010575079872204474,
|
| 471 |
+
"loss": 0.0825,
|
| 472 |
+
"step": 590
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.9169329073482428,
|
| 476 |
+
"grad_norm": 0.08435127884149551,
|
| 477 |
+
"learning_rate": 0.00010415335463258787,
|
| 478 |
+
"loss": 0.0594,
|
| 479 |
+
"step": 600
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.9169329073482428,
|
| 483 |
+
"eval_accuracy": 0.9808,
|
| 484 |
+
"eval_loss": 0.05469144880771637,
|
| 485 |
+
"eval_model_preparation_time": 0.0051,
|
| 486 |
+
"eval_runtime": 12.5149,
|
| 487 |
+
"eval_samples_per_second": 199.761,
|
| 488 |
+
"eval_steps_per_second": 25.01,
|
| 489 |
+
"step": 600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.9488817891373802,
|
| 493 |
+
"grad_norm": 0.016737263649702072,
|
| 494 |
+
"learning_rate": 0.000102555910543131,
|
| 495 |
+
"loss": 0.0463,
|
| 496 |
+
"step": 610
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.9808306709265175,
|
| 500 |
+
"grad_norm": 0.018503542989492416,
|
| 501 |
+
"learning_rate": 0.00010095846645367413,
|
| 502 |
+
"loss": 0.0213,
|
| 503 |
+
"step": 620
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.012779552715655,
|
| 507 |
+
"grad_norm": 0.11741246283054352,
|
| 508 |
+
"learning_rate": 9.936102236421726e-05,
|
| 509 |
+
"loss": 0.006,
|
| 510 |
+
"step": 630
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.0447284345047922,
|
| 514 |
+
"grad_norm": 0.015154512599110603,
|
| 515 |
+
"learning_rate": 9.77635782747604e-05,
|
| 516 |
+
"loss": 0.003,
|
| 517 |
+
"step": 640
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.07667731629393,
|
| 521 |
+
"grad_norm": 0.017609527334570885,
|
| 522 |
+
"learning_rate": 9.616613418530351e-05,
|
| 523 |
+
"loss": 0.0017,
|
| 524 |
+
"step": 650
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.108626198083067,
|
| 528 |
+
"grad_norm": 0.012800313532352448,
|
| 529 |
+
"learning_rate": 9.456869009584664e-05,
|
| 530 |
+
"loss": 0.0042,
|
| 531 |
+
"step": 660
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 2.1405750798722045,
|
| 535 |
+
"grad_norm": 1.593684434890747,
|
| 536 |
+
"learning_rate": 9.297124600638978e-05,
|
| 537 |
+
"loss": 0.0132,
|
| 538 |
+
"step": 670
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 2.1725239616613417,
|
| 542 |
+
"grad_norm": 0.01244130078703165,
|
| 543 |
+
"learning_rate": 9.137380191693292e-05,
|
| 544 |
+
"loss": 0.0015,
|
| 545 |
+
"step": 680
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 2.2044728434504792,
|
| 549 |
+
"grad_norm": 0.011427606455981731,
|
| 550 |
+
"learning_rate": 8.977635782747604e-05,
|
| 551 |
+
"loss": 0.0355,
|
| 552 |
+
"step": 690
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.236421725239617,
|
| 556 |
+
"grad_norm": 0.017172975465655327,
|
| 557 |
+
"learning_rate": 8.817891373801918e-05,
|
| 558 |
+
"loss": 0.0115,
|
| 559 |
+
"step": 700
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.236421725239617,
|
| 563 |
+
"eval_accuracy": 0.9844,
|
| 564 |
+
"eval_loss": 0.0627300888299942,
|
| 565 |
+
"eval_model_preparation_time": 0.0051,
|
| 566 |
+
"eval_runtime": 12.3578,
|
| 567 |
+
"eval_samples_per_second": 202.301,
|
| 568 |
+
"eval_steps_per_second": 25.328,
|
| 569 |
+
"step": 700
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.268370607028754,
|
| 573 |
+
"grad_norm": 0.010955709032714367,
|
| 574 |
+
"learning_rate": 8.658146964856231e-05,
|
| 575 |
+
"loss": 0.0135,
|
| 576 |
+
"step": 710
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.3003194888178915,
|
| 580 |
+
"grad_norm": 0.013627716340124607,
|
| 581 |
+
"learning_rate": 8.498402555910544e-05,
|
| 582 |
+
"loss": 0.0328,
|
| 583 |
+
"step": 720
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.3322683706070286,
|
| 587 |
+
"grad_norm": 0.024327559396624565,
|
| 588 |
+
"learning_rate": 8.338658146964856e-05,
|
| 589 |
+
"loss": 0.0049,
|
| 590 |
+
"step": 730
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.364217252396166,
|
| 594 |
+
"grad_norm": 0.012321406975388527,
|
| 595 |
+
"learning_rate": 8.17891373801917e-05,
|
| 596 |
+
"loss": 0.0096,
|
| 597 |
+
"step": 740
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.3961661341853033,
|
| 601 |
+
"grad_norm": 0.010169831104576588,
|
| 602 |
+
"learning_rate": 8.019169329073483e-05,
|
| 603 |
+
"loss": 0.0014,
|
| 604 |
+
"step": 750
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.428115015974441,
|
| 608 |
+
"grad_norm": 0.010039562359452248,
|
| 609 |
+
"learning_rate": 7.859424920127795e-05,
|
| 610 |
+
"loss": 0.0482,
|
| 611 |
+
"step": 760
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.460063897763578,
|
| 615 |
+
"grad_norm": 0.010927636176347733,
|
| 616 |
+
"learning_rate": 7.699680511182109e-05,
|
| 617 |
+
"loss": 0.0149,
|
| 618 |
+
"step": 770
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.4920127795527156,
|
| 622 |
+
"grad_norm": 0.010604580864310265,
|
| 623 |
+
"learning_rate": 7.539936102236423e-05,
|
| 624 |
+
"loss": 0.0028,
|
| 625 |
+
"step": 780
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.523961661341853,
|
| 629 |
+
"grad_norm": 0.009957361966371536,
|
| 630 |
+
"learning_rate": 7.380191693290735e-05,
|
| 631 |
+
"loss": 0.0042,
|
| 632 |
+
"step": 790
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.5559105431309903,
|
| 636 |
+
"grad_norm": 0.01096680574119091,
|
| 637 |
+
"learning_rate": 7.220447284345049e-05,
|
| 638 |
+
"loss": 0.0016,
|
| 639 |
+
"step": 800
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.5559105431309903,
|
| 643 |
+
"eval_accuracy": 0.9936,
|
| 644 |
+
"eval_loss": 0.02957286313176155,
|
| 645 |
+
"eval_model_preparation_time": 0.0051,
|
| 646 |
+
"eval_runtime": 12.3992,
|
| 647 |
+
"eval_samples_per_second": 201.627,
|
| 648 |
+
"eval_steps_per_second": 25.244,
|
| 649 |
+
"step": 800
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.587859424920128,
|
| 653 |
+
"grad_norm": 0.010381845757365227,
|
| 654 |
+
"learning_rate": 7.060702875399361e-05,
|
| 655 |
+
"loss": 0.0013,
|
| 656 |
+
"step": 810
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.619808306709265,
|
| 660 |
+
"grad_norm": 0.03203699365258217,
|
| 661 |
+
"learning_rate": 6.900958466453674e-05,
|
| 662 |
+
"loss": 0.0209,
|
| 663 |
+
"step": 820
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.6517571884984026,
|
| 667 |
+
"grad_norm": 0.009094555862247944,
|
| 668 |
+
"learning_rate": 6.741214057507987e-05,
|
| 669 |
+
"loss": 0.0011,
|
| 670 |
+
"step": 830
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.68370607028754,
|
| 674 |
+
"grad_norm": 0.008862023241817951,
|
| 675 |
+
"learning_rate": 6.5814696485623e-05,
|
| 676 |
+
"loss": 0.0287,
|
| 677 |
+
"step": 840
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 2.7156549520766773,
|
| 681 |
+
"grad_norm": 0.00914779119193554,
|
| 682 |
+
"learning_rate": 6.421725239616614e-05,
|
| 683 |
+
"loss": 0.0012,
|
| 684 |
+
"step": 850
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.747603833865815,
|
| 688 |
+
"grad_norm": 0.009615966118872166,
|
| 689 |
+
"learning_rate": 6.261980830670928e-05,
|
| 690 |
+
"loss": 0.0011,
|
| 691 |
+
"step": 860
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.779552715654952,
|
| 695 |
+
"grad_norm": 0.035701438784599304,
|
| 696 |
+
"learning_rate": 6.1022364217252406e-05,
|
| 697 |
+
"loss": 0.0089,
|
| 698 |
+
"step": 870
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.8115015974440896,
|
| 702 |
+
"grad_norm": 0.012455513700842857,
|
| 703 |
+
"learning_rate": 5.942492012779552e-05,
|
| 704 |
+
"loss": 0.0011,
|
| 705 |
+
"step": 880
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.8434504792332267,
|
| 709 |
+
"grad_norm": 0.009292890317738056,
|
| 710 |
+
"learning_rate": 5.782747603833866e-05,
|
| 711 |
+
"loss": 0.0091,
|
| 712 |
+
"step": 890
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.8753993610223643,
|
| 716 |
+
"grad_norm": 0.008451340720057487,
|
| 717 |
+
"learning_rate": 5.623003194888179e-05,
|
| 718 |
+
"loss": 0.004,
|
| 719 |
+
"step": 900
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 2.8753993610223643,
|
| 723 |
+
"eval_accuracy": 0.9916,
|
| 724 |
+
"eval_loss": 0.032514333724975586,
|
| 725 |
+
"eval_model_preparation_time": 0.0051,
|
| 726 |
+
"eval_runtime": 12.4929,
|
| 727 |
+
"eval_samples_per_second": 200.114,
|
| 728 |
+
"eval_steps_per_second": 25.054,
|
| 729 |
+
"step": 900
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.9073482428115014,
|
| 733 |
+
"grad_norm": 0.7223944067955017,
|
| 734 |
+
"learning_rate": 5.4632587859424925e-05,
|
| 735 |
+
"loss": 0.0019,
|
| 736 |
+
"step": 910
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.939297124600639,
|
| 740 |
+
"grad_norm": 0.007820590399205685,
|
| 741 |
+
"learning_rate": 5.3035143769968054e-05,
|
| 742 |
+
"loss": 0.001,
|
| 743 |
+
"step": 920
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 2.9712460063897765,
|
| 747 |
+
"grad_norm": 0.011096654459834099,
|
| 748 |
+
"learning_rate": 5.1437699680511184e-05,
|
| 749 |
+
"loss": 0.0011,
|
| 750 |
+
"step": 930
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 3.0031948881789137,
|
| 754 |
+
"grad_norm": 0.007654030807316303,
|
| 755 |
+
"learning_rate": 4.984025559105431e-05,
|
| 756 |
+
"loss": 0.0011,
|
| 757 |
+
"step": 940
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 3.0351437699680512,
|
| 761 |
+
"grad_norm": 0.010331162251532078,
|
| 762 |
+
"learning_rate": 4.824281150159744e-05,
|
| 763 |
+
"loss": 0.0009,
|
| 764 |
+
"step": 950
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 3.0670926517571884,
|
| 768 |
+
"grad_norm": 0.007492523640394211,
|
| 769 |
+
"learning_rate": 4.664536741214058e-05,
|
| 770 |
+
"loss": 0.0009,
|
| 771 |
+
"step": 960
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 3.099041533546326,
|
| 775 |
+
"grad_norm": 0.007593994960188866,
|
| 776 |
+
"learning_rate": 4.504792332268371e-05,
|
| 777 |
+
"loss": 0.001,
|
| 778 |
+
"step": 970
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 3.130990415335463,
|
| 782 |
+
"grad_norm": 0.007749509997665882,
|
| 783 |
+
"learning_rate": 4.345047923322684e-05,
|
| 784 |
+
"loss": 0.0009,
|
| 785 |
+
"step": 980
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 3.1629392971246006,
|
| 789 |
+
"grad_norm": 0.007726697251200676,
|
| 790 |
+
"learning_rate": 4.185303514376997e-05,
|
| 791 |
+
"loss": 0.0009,
|
| 792 |
+
"step": 990
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 3.194888178913738,
|
| 796 |
+
"grad_norm": 0.007175728678703308,
|
| 797 |
+
"learning_rate": 4.0255591054313104e-05,
|
| 798 |
+
"loss": 0.0009,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 3.194888178913738,
|
| 803 |
+
"eval_accuracy": 0.9948,
|
| 804 |
+
"eval_loss": 0.02237132005393505,
|
| 805 |
+
"eval_model_preparation_time": 0.0051,
|
| 806 |
+
"eval_runtime": 12.3213,
|
| 807 |
+
"eval_samples_per_second": 202.9,
|
| 808 |
+
"eval_steps_per_second": 25.403,
|
| 809 |
+
"step": 1000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 3.2268370607028753,
|
| 813 |
+
"grad_norm": 0.007404172793030739,
|
| 814 |
+
"learning_rate": 3.8658146964856234e-05,
|
| 815 |
+
"loss": 0.0019,
|
| 816 |
+
"step": 1010
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.258785942492013,
|
| 820 |
+
"grad_norm": 0.007146772928535938,
|
| 821 |
+
"learning_rate": 3.7060702875399364e-05,
|
| 822 |
+
"loss": 0.0009,
|
| 823 |
+
"step": 1020
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 3.29073482428115,
|
| 827 |
+
"grad_norm": 0.02829296886920929,
|
| 828 |
+
"learning_rate": 3.546325878594249e-05,
|
| 829 |
+
"loss": 0.0009,
|
| 830 |
+
"step": 1030
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 3.3226837060702876,
|
| 834 |
+
"grad_norm": 0.007179423235356808,
|
| 835 |
+
"learning_rate": 3.386581469648562e-05,
|
| 836 |
+
"loss": 0.0009,
|
| 837 |
+
"step": 1040
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.3546325878594248,
|
| 841 |
+
"grad_norm": 0.0070446510799229145,
|
| 842 |
+
"learning_rate": 3.226837060702875e-05,
|
| 843 |
+
"loss": 0.0009,
|
| 844 |
+
"step": 1050
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 3.3865814696485623,
|
| 848 |
+
"grad_norm": 0.007006002124398947,
|
| 849 |
+
"learning_rate": 3.067092651757188e-05,
|
| 850 |
+
"loss": 0.0009,
|
| 851 |
+
"step": 1060
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.4185303514377,
|
| 855 |
+
"grad_norm": 0.006977152545005083,
|
| 856 |
+
"learning_rate": 2.907348242811502e-05,
|
| 857 |
+
"loss": 0.0008,
|
| 858 |
+
"step": 1070
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 3.450479233226837,
|
| 862 |
+
"grad_norm": 0.0067046028561890125,
|
| 863 |
+
"learning_rate": 2.747603833865815e-05,
|
| 864 |
+
"loss": 0.0009,
|
| 865 |
+
"step": 1080
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.4824281150159746,
|
| 869 |
+
"grad_norm": 0.0067406343296170235,
|
| 870 |
+
"learning_rate": 2.5878594249201278e-05,
|
| 871 |
+
"loss": 0.0012,
|
| 872 |
+
"step": 1090
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.5143769968051117,
|
| 876 |
+
"grad_norm": 0.006645725108683109,
|
| 877 |
+
"learning_rate": 2.428115015974441e-05,
|
| 878 |
+
"loss": 0.0008,
|
| 879 |
+
"step": 1100
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.5143769968051117,
|
| 883 |
+
"eval_accuracy": 0.9936,
|
| 884 |
+
"eval_loss": 0.027039185166358948,
|
| 885 |
+
"eval_model_preparation_time": 0.0051,
|
| 886 |
+
"eval_runtime": 13.3116,
|
| 887 |
+
"eval_samples_per_second": 187.807,
|
| 888 |
+
"eval_steps_per_second": 23.513,
|
| 889 |
+
"step": 1100
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 3.5463258785942493,
|
| 893 |
+
"grad_norm": 0.0066969566978514194,
|
| 894 |
+
"learning_rate": 2.268370607028754e-05,
|
| 895 |
+
"loss": 0.0008,
|
| 896 |
+
"step": 1110
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 3.5782747603833864,
|
| 900 |
+
"grad_norm": 0.0066523440182209015,
|
| 901 |
+
"learning_rate": 2.108626198083067e-05,
|
| 902 |
+
"loss": 0.0008,
|
| 903 |
+
"step": 1120
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 3.610223642172524,
|
| 907 |
+
"grad_norm": 0.0067514642141759396,
|
| 908 |
+
"learning_rate": 1.9488817891373803e-05,
|
| 909 |
+
"loss": 0.0008,
|
| 910 |
+
"step": 1130
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 3.642172523961661,
|
| 914 |
+
"grad_norm": 0.0066239056177437305,
|
| 915 |
+
"learning_rate": 1.7891373801916932e-05,
|
| 916 |
+
"loss": 0.001,
|
| 917 |
+
"step": 1140
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.6741214057507987,
|
| 921 |
+
"grad_norm": 0.006579084321856499,
|
| 922 |
+
"learning_rate": 1.6293929712460065e-05,
|
| 923 |
+
"loss": 0.0008,
|
| 924 |
+
"step": 1150
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.7060702875399363,
|
| 928 |
+
"grad_norm": 0.006953445728868246,
|
| 929 |
+
"learning_rate": 1.4696485623003195e-05,
|
| 930 |
+
"loss": 0.0009,
|
| 931 |
+
"step": 1160
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 3.7380191693290734,
|
| 935 |
+
"grad_norm": 0.006420870777219534,
|
| 936 |
+
"learning_rate": 1.3099041533546328e-05,
|
| 937 |
+
"loss": 0.0008,
|
| 938 |
+
"step": 1170
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 3.769968051118211,
|
| 942 |
+
"grad_norm": 0.006650346331298351,
|
| 943 |
+
"learning_rate": 1.1501597444089457e-05,
|
| 944 |
+
"loss": 0.0008,
|
| 945 |
+
"step": 1180
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 3.801916932907348,
|
| 949 |
+
"grad_norm": 0.006530832499265671,
|
| 950 |
+
"learning_rate": 9.904153354632589e-06,
|
| 951 |
+
"loss": 0.0018,
|
| 952 |
+
"step": 1190
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 3.8338658146964857,
|
| 956 |
+
"grad_norm": 0.007318571675568819,
|
| 957 |
+
"learning_rate": 8.306709265175718e-06,
|
| 958 |
+
"loss": 0.0008,
|
| 959 |
+
"step": 1200
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.8338658146964857,
|
| 963 |
+
"eval_accuracy": 0.994,
|
| 964 |
+
"eval_loss": 0.02559490129351616,
|
| 965 |
+
"eval_model_preparation_time": 0.0051,
|
| 966 |
+
"eval_runtime": 12.5362,
|
| 967 |
+
"eval_samples_per_second": 199.423,
|
| 968 |
+
"eval_steps_per_second": 24.968,
|
| 969 |
+
"step": 1200
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 3.8658146964856233,
|
| 973 |
+
"grad_norm": 0.007055082358419895,
|
| 974 |
+
"learning_rate": 6.70926517571885e-06,
|
| 975 |
+
"loss": 0.001,
|
| 976 |
+
"step": 1210
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 3.8977635782747604,
|
| 980 |
+
"grad_norm": 0.006507499609142542,
|
| 981 |
+
"learning_rate": 5.111821086261981e-06,
|
| 982 |
+
"loss": 0.0008,
|
| 983 |
+
"step": 1220
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 3.9297124600638975,
|
| 987 |
+
"grad_norm": 0.006634117104113102,
|
| 988 |
+
"learning_rate": 3.5143769968051118e-06,
|
| 989 |
+
"loss": 0.0007,
|
| 990 |
+
"step": 1230
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 3.961661341853035,
|
| 994 |
+
"grad_norm": 0.006404239218682051,
|
| 995 |
+
"learning_rate": 1.9169329073482426e-06,
|
| 996 |
+
"loss": 0.0008,
|
| 997 |
+
"step": 1240
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 3.9936102236421727,
|
| 1001 |
+
"grad_norm": 0.006795979104936123,
|
| 1002 |
+
"learning_rate": 3.194888178913738e-07,
|
| 1003 |
+
"loss": 0.0008,
|
| 1004 |
+
"step": 1250
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 4.0,
|
| 1008 |
+
"step": 1252,
|
| 1009 |
+
"total_flos": 1.54983979229184e+18,
|
| 1010 |
+
"train_loss": 0.055932876091605174,
|
| 1011 |
+
"train_runtime": 533.4054,
|
| 1012 |
+
"train_samples_per_second": 37.495,
|
| 1013 |
+
"train_steps_per_second": 2.347
|
| 1014 |
+
}
|
| 1015 |
+
],
|
| 1016 |
+
"logging_steps": 10,
|
| 1017 |
+
"max_steps": 1252,
|
| 1018 |
+
"num_input_tokens_seen": 0,
|
| 1019 |
+
"num_train_epochs": 4,
|
| 1020 |
+
"save_steps": 100,
|
| 1021 |
+
"stateful_callbacks": {
|
| 1022 |
+
"TrainerControl": {
|
| 1023 |
+
"args": {
|
| 1024 |
+
"should_epoch_stop": false,
|
| 1025 |
+
"should_evaluate": false,
|
| 1026 |
+
"should_log": false,
|
| 1027 |
+
"should_save": true,
|
| 1028 |
+
"should_training_stop": true
|
| 1029 |
+
},
|
| 1030 |
+
"attributes": {}
|
| 1031 |
+
}
|
| 1032 |
+
},
|
| 1033 |
+
"total_flos": 1.54983979229184e+18,
|
| 1034 |
+
"train_batch_size": 16,
|
| 1035 |
+
"trial_name": null,
|
| 1036 |
+
"trial_params": null
|
| 1037 |
+
}
|