End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Oct16_13-41-41_5015a13a2e4f/events.out.tfevents.1729139133.5015a13a2e4f.341.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1962 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1249
|
| 22 |
+
- Accuracy: 0.9682
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.9681742043551089,
|
| 4 |
+
"eval_loss": 0.12486154586076736,
|
| 5 |
+
"eval_runtime": 96.2308,
|
| 6 |
+
"eval_samples_per_second": 6.204,
|
| 7 |
+
"eval_steps_per_second": 0.197,
|
| 8 |
+
"total_flos": 2.0803097508518707e+19,
|
| 9 |
+
"train_loss": 0.1363242975941726,
|
| 10 |
+
"train_runtime": 52918.3087,
|
| 11 |
+
"train_samples_per_second": 5.073,
|
| 12 |
+
"train_steps_per_second": 0.04
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.9681742043551089,
|
| 4 |
+
"eval_loss": 0.12486154586076736,
|
| 5 |
+
"eval_runtime": 96.2308,
|
| 6 |
+
"eval_samples_per_second": 6.204,
|
| 7 |
+
"eval_steps_per_second": 0.197
|
| 8 |
+
}
|
runs/Oct16_13-41-41_5015a13a2e4f/events.out.tfevents.1729139133.5015a13a2e4f.341.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d98fb10756aada0ba28e528bbd0a301b35957a7e615aa81c2b0697afdde1201
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"total_flos": 2.0803097508518707e+19,
|
| 4 |
+
"train_loss": 0.1363242975941726,
|
| 5 |
+
"train_runtime": 52918.3087,
|
| 6 |
+
"train_samples_per_second": 5.073,
|
| 7 |
+
"train_steps_per_second": 0.04
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1962 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9681742043551089,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-biopsy/checkpoint-1596",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.23809523809523808,
|
| 13 |
+
"grad_norm": 1.7908698320388794,
|
| 14 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 15 |
+
"loss": 1.3741,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.47619047619047616,
|
| 20 |
+
"grad_norm": 1.7948147058486938,
|
| 21 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 22 |
+
"loss": 1.3269,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7142857142857143,
|
| 27 |
+
"grad_norm": 1.6937719583511353,
|
| 28 |
+
"learning_rate": 7.142857142857143e-06,
|
| 29 |
+
"loss": 1.2328,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9523809523809523,
|
| 34 |
+
"grad_norm": 0.9957559108734131,
|
| 35 |
+
"learning_rate": 9.523809523809523e-06,
|
| 36 |
+
"loss": 1.1332,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_accuracy": 0.5494137353433836,
|
| 42 |
+
"eval_loss": 1.071189522743225,
|
| 43 |
+
"eval_runtime": 102.6887,
|
| 44 |
+
"eval_samples_per_second": 5.814,
|
| 45 |
+
"eval_steps_per_second": 0.185,
|
| 46 |
+
"step": 42
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.1904761904761905,
|
| 50 |
+
"grad_norm": 0.9576915502548218,
|
| 51 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 52 |
+
"loss": 1.0191,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.4285714285714286,
|
| 57 |
+
"grad_norm": 1.0786559581756592,
|
| 58 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 59 |
+
"loss": 0.9104,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.6666666666666665,
|
| 64 |
+
"grad_norm": 1.0609694719314575,
|
| 65 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 66 |
+
"loss": 0.7667,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.9047619047619047,
|
| 71 |
+
"grad_norm": 1.6637555360794067,
|
| 72 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 73 |
+
"loss": 0.6556,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.0,
|
| 78 |
+
"eval_accuracy": 0.8492462311557789,
|
| 79 |
+
"eval_loss": 0.5741624236106873,
|
| 80 |
+
"eval_runtime": 102.8217,
|
| 81 |
+
"eval_samples_per_second": 5.806,
|
| 82 |
+
"eval_steps_per_second": 0.185,
|
| 83 |
+
"step": 84
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.142857142857143,
|
| 87 |
+
"grad_norm": 2.9285330772399902,
|
| 88 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 89 |
+
"loss": 0.5688,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.380952380952381,
|
| 94 |
+
"grad_norm": 1.4198558330535889,
|
| 95 |
+
"learning_rate": 2.380952380952381e-05,
|
| 96 |
+
"loss": 0.4944,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.619047619047619,
|
| 101 |
+
"grad_norm": 1.1169885396957397,
|
| 102 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 103 |
+
"loss": 0.4229,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.857142857142857,
|
| 108 |
+
"grad_norm": 1.9193533658981323,
|
| 109 |
+
"learning_rate": 2.857142857142857e-05,
|
| 110 |
+
"loss": 0.3987,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.0,
|
| 115 |
+
"eval_accuracy": 0.8894472361809045,
|
| 116 |
+
"eval_loss": 0.3950396776199341,
|
| 117 |
+
"eval_runtime": 102.779,
|
| 118 |
+
"eval_samples_per_second": 5.809,
|
| 119 |
+
"eval_steps_per_second": 0.185,
|
| 120 |
+
"step": 126
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.0952380952380953,
|
| 124 |
+
"grad_norm": 4.524629592895508,
|
| 125 |
+
"learning_rate": 3.095238095238095e-05,
|
| 126 |
+
"loss": 0.3845,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.3333333333333335,
|
| 131 |
+
"grad_norm": 2.838951587677002,
|
| 132 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 133 |
+
"loss": 0.3616,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.571428571428571,
|
| 138 |
+
"grad_norm": 2.4979946613311768,
|
| 139 |
+
"learning_rate": 3.571428571428572e-05,
|
| 140 |
+
"loss": 0.3479,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.8095238095238093,
|
| 145 |
+
"grad_norm": 4.321163177490234,
|
| 146 |
+
"learning_rate": 3.809523809523809e-05,
|
| 147 |
+
"loss": 0.2825,
|
| 148 |
+
"step": 160
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.0,
|
| 152 |
+
"eval_accuracy": 0.8777219430485762,
|
| 153 |
+
"eval_loss": 0.39236369729042053,
|
| 154 |
+
"eval_runtime": 102.4215,
|
| 155 |
+
"eval_samples_per_second": 5.829,
|
| 156 |
+
"eval_steps_per_second": 0.186,
|
| 157 |
+
"step": 168
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.0476190476190474,
|
| 161 |
+
"grad_norm": 2.3643758296966553,
|
| 162 |
+
"learning_rate": 4.047619047619048e-05,
|
| 163 |
+
"loss": 0.3009,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.285714285714286,
|
| 168 |
+
"grad_norm": 6.605370998382568,
|
| 169 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 170 |
+
"loss": 0.3091,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.523809523809524,
|
| 175 |
+
"grad_norm": 4.201854228973389,
|
| 176 |
+
"learning_rate": 4.523809523809524e-05,
|
| 177 |
+
"loss": 0.277,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.761904761904762,
|
| 182 |
+
"grad_norm": 1.754778265953064,
|
| 183 |
+
"learning_rate": 4.761904761904762e-05,
|
| 184 |
+
"loss": 0.2548,
|
| 185 |
+
"step": 200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.0,
|
| 189 |
+
"grad_norm": 3.1804463863372803,
|
| 190 |
+
"learning_rate": 5e-05,
|
| 191 |
+
"loss": 0.3662,
|
| 192 |
+
"step": 210
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.0,
|
| 196 |
+
"eval_accuracy": 0.8860971524288107,
|
| 197 |
+
"eval_loss": 0.36218151450157166,
|
| 198 |
+
"eval_runtime": 101.769,
|
| 199 |
+
"eval_samples_per_second": 5.866,
|
| 200 |
+
"eval_steps_per_second": 0.187,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.238095238095238,
|
| 205 |
+
"grad_norm": 1.8210253715515137,
|
| 206 |
+
"learning_rate": 4.973544973544973e-05,
|
| 207 |
+
"loss": 0.3025,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.476190476190476,
|
| 212 |
+
"grad_norm": 2.0568735599517822,
|
| 213 |
+
"learning_rate": 4.9470899470899475e-05,
|
| 214 |
+
"loss": 0.2431,
|
| 215 |
+
"step": 230
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.714285714285714,
|
| 219 |
+
"grad_norm": 5.020086765289307,
|
| 220 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 221 |
+
"loss": 0.2462,
|
| 222 |
+
"step": 240
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.9523809523809526,
|
| 226 |
+
"grad_norm": 2.455803632736206,
|
| 227 |
+
"learning_rate": 4.894179894179895e-05,
|
| 228 |
+
"loss": 0.2218,
|
| 229 |
+
"step": 250
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 6.0,
|
| 233 |
+
"eval_accuracy": 0.9246231155778895,
|
| 234 |
+
"eval_loss": 0.2705938518047333,
|
| 235 |
+
"eval_runtime": 102.0179,
|
| 236 |
+
"eval_samples_per_second": 5.852,
|
| 237 |
+
"eval_steps_per_second": 0.186,
|
| 238 |
+
"step": 252
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.190476190476191,
|
| 242 |
+
"grad_norm": 1.1127841472625732,
|
| 243 |
+
"learning_rate": 4.8677248677248676e-05,
|
| 244 |
+
"loss": 0.2424,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.428571428571429,
|
| 249 |
+
"grad_norm": 5.003824234008789,
|
| 250 |
+
"learning_rate": 4.841269841269841e-05,
|
| 251 |
+
"loss": 0.2726,
|
| 252 |
+
"step": 270
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.666666666666667,
|
| 256 |
+
"grad_norm": 1.7662922143936157,
|
| 257 |
+
"learning_rate": 4.814814814814815e-05,
|
| 258 |
+
"loss": 0.2482,
|
| 259 |
+
"step": 280
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.904761904761905,
|
| 263 |
+
"grad_norm": 4.009079933166504,
|
| 264 |
+
"learning_rate": 4.7883597883597884e-05,
|
| 265 |
+
"loss": 0.2236,
|
| 266 |
+
"step": 290
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 7.0,
|
| 270 |
+
"eval_accuracy": 0.9346733668341709,
|
| 271 |
+
"eval_loss": 0.2283419966697693,
|
| 272 |
+
"eval_runtime": 101.6184,
|
| 273 |
+
"eval_samples_per_second": 5.875,
|
| 274 |
+
"eval_steps_per_second": 0.187,
|
| 275 |
+
"step": 294
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.142857142857143,
|
| 279 |
+
"grad_norm": 1.4519894123077393,
|
| 280 |
+
"learning_rate": 4.761904761904762e-05,
|
| 281 |
+
"loss": 0.2111,
|
| 282 |
+
"step": 300
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.380952380952381,
|
| 286 |
+
"grad_norm": 2.688704490661621,
|
| 287 |
+
"learning_rate": 4.7354497354497356e-05,
|
| 288 |
+
"loss": 0.2187,
|
| 289 |
+
"step": 310
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.619047619047619,
|
| 293 |
+
"grad_norm": 2.2603533267974854,
|
| 294 |
+
"learning_rate": 4.708994708994709e-05,
|
| 295 |
+
"loss": 0.1934,
|
| 296 |
+
"step": 320
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.857142857142857,
|
| 300 |
+
"grad_norm": 2.341301679611206,
|
| 301 |
+
"learning_rate": 4.682539682539683e-05,
|
| 302 |
+
"loss": 0.2224,
|
| 303 |
+
"step": 330
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.0,
|
| 307 |
+
"eval_accuracy": 0.931323283082077,
|
| 308 |
+
"eval_loss": 0.23672878742218018,
|
| 309 |
+
"eval_runtime": 100.0543,
|
| 310 |
+
"eval_samples_per_second": 5.967,
|
| 311 |
+
"eval_steps_per_second": 0.19,
|
| 312 |
+
"step": 336
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.095238095238095,
|
| 316 |
+
"grad_norm": 1.5976060628890991,
|
| 317 |
+
"learning_rate": 4.656084656084656e-05,
|
| 318 |
+
"loss": 0.1954,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.333333333333334,
|
| 323 |
+
"grad_norm": 1.360371708869934,
|
| 324 |
+
"learning_rate": 4.62962962962963e-05,
|
| 325 |
+
"loss": 0.1607,
|
| 326 |
+
"step": 350
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.571428571428571,
|
| 330 |
+
"grad_norm": 1.7058426141738892,
|
| 331 |
+
"learning_rate": 4.603174603174603e-05,
|
| 332 |
+
"loss": 0.1804,
|
| 333 |
+
"step": 360
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.80952380952381,
|
| 337 |
+
"grad_norm": 2.388390302658081,
|
| 338 |
+
"learning_rate": 4.576719576719577e-05,
|
| 339 |
+
"loss": 0.1754,
|
| 340 |
+
"step": 370
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 9.0,
|
| 344 |
+
"eval_accuracy": 0.9296482412060302,
|
| 345 |
+
"eval_loss": 0.21387061476707458,
|
| 346 |
+
"eval_runtime": 101.5101,
|
| 347 |
+
"eval_samples_per_second": 5.881,
|
| 348 |
+
"eval_steps_per_second": 0.187,
|
| 349 |
+
"step": 378
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.047619047619047,
|
| 353 |
+
"grad_norm": 2.2065589427948,
|
| 354 |
+
"learning_rate": 4.55026455026455e-05,
|
| 355 |
+
"loss": 0.1695,
|
| 356 |
+
"step": 380
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.285714285714286,
|
| 360 |
+
"grad_norm": 1.8705061674118042,
|
| 361 |
+
"learning_rate": 4.523809523809524e-05,
|
| 362 |
+
"loss": 0.1655,
|
| 363 |
+
"step": 390
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.523809523809524,
|
| 367 |
+
"grad_norm": 1.9261928796768188,
|
| 368 |
+
"learning_rate": 4.4973544973544974e-05,
|
| 369 |
+
"loss": 0.1584,
|
| 370 |
+
"step": 400
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 9.761904761904763,
|
| 374 |
+
"grad_norm": 2.372283697128296,
|
| 375 |
+
"learning_rate": 4.470899470899471e-05,
|
| 376 |
+
"loss": 0.1711,
|
| 377 |
+
"step": 410
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 10.0,
|
| 381 |
+
"grad_norm": 2.0945327281951904,
|
| 382 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 383 |
+
"loss": 0.1707,
|
| 384 |
+
"step": 420
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 10.0,
|
| 388 |
+
"eval_accuracy": 0.949748743718593,
|
| 389 |
+
"eval_loss": 0.18289239704608917,
|
| 390 |
+
"eval_runtime": 100.379,
|
| 391 |
+
"eval_samples_per_second": 5.947,
|
| 392 |
+
"eval_steps_per_second": 0.189,
|
| 393 |
+
"step": 420
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 10.238095238095237,
|
| 397 |
+
"grad_norm": 2.2974770069122314,
|
| 398 |
+
"learning_rate": 4.417989417989418e-05,
|
| 399 |
+
"loss": 0.1446,
|
| 400 |
+
"step": 430
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 10.476190476190476,
|
| 404 |
+
"grad_norm": 3.294903039932251,
|
| 405 |
+
"learning_rate": 4.391534391534391e-05,
|
| 406 |
+
"loss": 0.1848,
|
| 407 |
+
"step": 440
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 10.714285714285714,
|
| 411 |
+
"grad_norm": 2.059147834777832,
|
| 412 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 413 |
+
"loss": 0.1923,
|
| 414 |
+
"step": 450
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 10.952380952380953,
|
| 418 |
+
"grad_norm": 1.4008002281188965,
|
| 419 |
+
"learning_rate": 4.3386243386243384e-05,
|
| 420 |
+
"loss": 0.1619,
|
| 421 |
+
"step": 460
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 11.0,
|
| 425 |
+
"eval_accuracy": 0.9463986599664992,
|
| 426 |
+
"eval_loss": 0.21719853579998016,
|
| 427 |
+
"eval_runtime": 100.686,
|
| 428 |
+
"eval_samples_per_second": 5.929,
|
| 429 |
+
"eval_steps_per_second": 0.189,
|
| 430 |
+
"step": 462
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 11.19047619047619,
|
| 434 |
+
"grad_norm": 3.9857630729675293,
|
| 435 |
+
"learning_rate": 4.312169312169313e-05,
|
| 436 |
+
"loss": 0.1479,
|
| 437 |
+
"step": 470
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 11.428571428571429,
|
| 441 |
+
"grad_norm": 1.8493316173553467,
|
| 442 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 443 |
+
"loss": 0.1399,
|
| 444 |
+
"step": 480
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 11.666666666666666,
|
| 448 |
+
"grad_norm": 3.7601022720336914,
|
| 449 |
+
"learning_rate": 4.259259259259259e-05,
|
| 450 |
+
"loss": 0.1693,
|
| 451 |
+
"step": 490
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 11.904761904761905,
|
| 455 |
+
"grad_norm": 2.0924296379089355,
|
| 456 |
+
"learning_rate": 4.232804232804233e-05,
|
| 457 |
+
"loss": 0.1547,
|
| 458 |
+
"step": 500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 12.0,
|
| 462 |
+
"eval_accuracy": 0.9380234505862647,
|
| 463 |
+
"eval_loss": 0.19601453840732574,
|
| 464 |
+
"eval_runtime": 100.959,
|
| 465 |
+
"eval_samples_per_second": 5.913,
|
| 466 |
+
"eval_steps_per_second": 0.188,
|
| 467 |
+
"step": 504
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 12.142857142857142,
|
| 471 |
+
"grad_norm": 0.876983106136322,
|
| 472 |
+
"learning_rate": 4.2063492063492065e-05,
|
| 473 |
+
"loss": 0.1467,
|
| 474 |
+
"step": 510
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 12.380952380952381,
|
| 478 |
+
"grad_norm": 1.4364930391311646,
|
| 479 |
+
"learning_rate": 4.17989417989418e-05,
|
| 480 |
+
"loss": 0.1293,
|
| 481 |
+
"step": 520
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 12.619047619047619,
|
| 485 |
+
"grad_norm": 2.0503087043762207,
|
| 486 |
+
"learning_rate": 4.153439153439154e-05,
|
| 487 |
+
"loss": 0.123,
|
| 488 |
+
"step": 530
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 12.857142857142858,
|
| 492 |
+
"grad_norm": 4.358797550201416,
|
| 493 |
+
"learning_rate": 4.126984126984127e-05,
|
| 494 |
+
"loss": 0.1213,
|
| 495 |
+
"step": 540
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 13.0,
|
| 499 |
+
"eval_accuracy": 0.9581239530988275,
|
| 500 |
+
"eval_loss": 0.14836081862449646,
|
| 501 |
+
"eval_runtime": 100.2262,
|
| 502 |
+
"eval_samples_per_second": 5.957,
|
| 503 |
+
"eval_steps_per_second": 0.19,
|
| 504 |
+
"step": 546
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 13.095238095238095,
|
| 508 |
+
"grad_norm": 5.3521199226379395,
|
| 509 |
+
"learning_rate": 4.100529100529101e-05,
|
| 510 |
+
"loss": 0.1213,
|
| 511 |
+
"step": 550
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 13.333333333333334,
|
| 515 |
+
"grad_norm": 1.1555789709091187,
|
| 516 |
+
"learning_rate": 4.074074074074074e-05,
|
| 517 |
+
"loss": 0.1353,
|
| 518 |
+
"step": 560
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 13.571428571428571,
|
| 522 |
+
"grad_norm": 2.3314497470855713,
|
| 523 |
+
"learning_rate": 4.047619047619048e-05,
|
| 524 |
+
"loss": 0.1392,
|
| 525 |
+
"step": 570
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 13.80952380952381,
|
| 529 |
+
"grad_norm": 2.4166135787963867,
|
| 530 |
+
"learning_rate": 4.021164021164021e-05,
|
| 531 |
+
"loss": 0.1388,
|
| 532 |
+
"step": 580
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 14.0,
|
| 536 |
+
"eval_accuracy": 0.9581239530988275,
|
| 537 |
+
"eval_loss": 0.16891160607337952,
|
| 538 |
+
"eval_runtime": 101.124,
|
| 539 |
+
"eval_samples_per_second": 5.904,
|
| 540 |
+
"eval_steps_per_second": 0.188,
|
| 541 |
+
"step": 588
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 14.047619047619047,
|
| 545 |
+
"grad_norm": 1.2457020282745361,
|
| 546 |
+
"learning_rate": 3.9947089947089946e-05,
|
| 547 |
+
"loss": 0.1075,
|
| 548 |
+
"step": 590
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 14.285714285714286,
|
| 552 |
+
"grad_norm": 1.428454041481018,
|
| 553 |
+
"learning_rate": 3.968253968253968e-05,
|
| 554 |
+
"loss": 0.1149,
|
| 555 |
+
"step": 600
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 14.523809523809524,
|
| 559 |
+
"grad_norm": 1.9853533506393433,
|
| 560 |
+
"learning_rate": 3.941798941798942e-05,
|
| 561 |
+
"loss": 0.1232,
|
| 562 |
+
"step": 610
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 14.761904761904763,
|
| 566 |
+
"grad_norm": 2.842386484146118,
|
| 567 |
+
"learning_rate": 3.9153439153439155e-05,
|
| 568 |
+
"loss": 0.1123,
|
| 569 |
+
"step": 620
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 15.0,
|
| 573 |
+
"grad_norm": 1.4970998764038086,
|
| 574 |
+
"learning_rate": 3.888888888888889e-05,
|
| 575 |
+
"loss": 0.1009,
|
| 576 |
+
"step": 630
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 15.0,
|
| 580 |
+
"eval_accuracy": 0.9581239530988275,
|
| 581 |
+
"eval_loss": 0.1493997424840927,
|
| 582 |
+
"eval_runtime": 103.3785,
|
| 583 |
+
"eval_samples_per_second": 5.775,
|
| 584 |
+
"eval_steps_per_second": 0.184,
|
| 585 |
+
"step": 630
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 15.238095238095237,
|
| 589 |
+
"grad_norm": 2.815284490585327,
|
| 590 |
+
"learning_rate": 3.862433862433863e-05,
|
| 591 |
+
"loss": 0.1301,
|
| 592 |
+
"step": 640
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 15.476190476190476,
|
| 596 |
+
"grad_norm": 2.5713961124420166,
|
| 597 |
+
"learning_rate": 3.835978835978836e-05,
|
| 598 |
+
"loss": 0.0979,
|
| 599 |
+
"step": 650
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 15.714285714285714,
|
| 603 |
+
"grad_norm": 1.771735668182373,
|
| 604 |
+
"learning_rate": 3.809523809523809e-05,
|
| 605 |
+
"loss": 0.12,
|
| 606 |
+
"step": 660
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 15.952380952380953,
|
| 610 |
+
"grad_norm": 2.1721739768981934,
|
| 611 |
+
"learning_rate": 3.7830687830687835e-05,
|
| 612 |
+
"loss": 0.124,
|
| 613 |
+
"step": 670
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 16.0,
|
| 617 |
+
"eval_accuracy": 0.9581239530988275,
|
| 618 |
+
"eval_loss": 0.15640056133270264,
|
| 619 |
+
"eval_runtime": 101.1931,
|
| 620 |
+
"eval_samples_per_second": 5.9,
|
| 621 |
+
"eval_steps_per_second": 0.188,
|
| 622 |
+
"step": 672
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 16.19047619047619,
|
| 626 |
+
"grad_norm": 3.04827880859375,
|
| 627 |
+
"learning_rate": 3.7566137566137564e-05,
|
| 628 |
+
"loss": 0.1047,
|
| 629 |
+
"step": 680
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 16.428571428571427,
|
| 633 |
+
"grad_norm": 1.7996746301651,
|
| 634 |
+
"learning_rate": 3.730158730158731e-05,
|
| 635 |
+
"loss": 0.0987,
|
| 636 |
+
"step": 690
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 16.666666666666668,
|
| 640 |
+
"grad_norm": 2.0100674629211426,
|
| 641 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 642 |
+
"loss": 0.1132,
|
| 643 |
+
"step": 700
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 16.904761904761905,
|
| 647 |
+
"grad_norm": 3.321587562561035,
|
| 648 |
+
"learning_rate": 3.677248677248677e-05,
|
| 649 |
+
"loss": 0.1078,
|
| 650 |
+
"step": 710
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 17.0,
|
| 654 |
+
"eval_accuracy": 0.9514237855946399,
|
| 655 |
+
"eval_loss": 0.17282074689865112,
|
| 656 |
+
"eval_runtime": 103.2232,
|
| 657 |
+
"eval_samples_per_second": 5.784,
|
| 658 |
+
"eval_steps_per_second": 0.184,
|
| 659 |
+
"step": 714
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 17.142857142857142,
|
| 663 |
+
"grad_norm": 1.6693847179412842,
|
| 664 |
+
"learning_rate": 3.650793650793651e-05,
|
| 665 |
+
"loss": 0.0762,
|
| 666 |
+
"step": 720
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 17.38095238095238,
|
| 670 |
+
"grad_norm": 1.9719501733779907,
|
| 671 |
+
"learning_rate": 3.6243386243386245e-05,
|
| 672 |
+
"loss": 0.0961,
|
| 673 |
+
"step": 730
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 17.61904761904762,
|
| 677 |
+
"grad_norm": 1.4737845659255981,
|
| 678 |
+
"learning_rate": 3.597883597883598e-05,
|
| 679 |
+
"loss": 0.0777,
|
| 680 |
+
"step": 740
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 17.857142857142858,
|
| 684 |
+
"grad_norm": 5.473972797393799,
|
| 685 |
+
"learning_rate": 3.571428571428572e-05,
|
| 686 |
+
"loss": 0.102,
|
| 687 |
+
"step": 750
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 18.0,
|
| 691 |
+
"eval_accuracy": 0.9447236180904522,
|
| 692 |
+
"eval_loss": 0.16686658561229706,
|
| 693 |
+
"eval_runtime": 100.8047,
|
| 694 |
+
"eval_samples_per_second": 5.922,
|
| 695 |
+
"eval_steps_per_second": 0.188,
|
| 696 |
+
"step": 756
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 18.095238095238095,
|
| 700 |
+
"grad_norm": 2.001469612121582,
|
| 701 |
+
"learning_rate": 3.5449735449735446e-05,
|
| 702 |
+
"loss": 0.0987,
|
| 703 |
+
"step": 760
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 18.333333333333332,
|
| 707 |
+
"grad_norm": 1.7356058359146118,
|
| 708 |
+
"learning_rate": 3.518518518518519e-05,
|
| 709 |
+
"loss": 0.0933,
|
| 710 |
+
"step": 770
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 18.571428571428573,
|
| 714 |
+
"grad_norm": 2.967357635498047,
|
| 715 |
+
"learning_rate": 3.492063492063492e-05,
|
| 716 |
+
"loss": 0.122,
|
| 717 |
+
"step": 780
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 18.80952380952381,
|
| 721 |
+
"grad_norm": 1.3743668794631958,
|
| 722 |
+
"learning_rate": 3.465608465608466e-05,
|
| 723 |
+
"loss": 0.1006,
|
| 724 |
+
"step": 790
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 19.0,
|
| 728 |
+
"eval_accuracy": 0.9581239530988275,
|
| 729 |
+
"eval_loss": 0.1404878944158554,
|
| 730 |
+
"eval_runtime": 101.9768,
|
| 731 |
+
"eval_samples_per_second": 5.854,
|
| 732 |
+
"eval_steps_per_second": 0.186,
|
| 733 |
+
"step": 798
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 19.047619047619047,
|
| 737 |
+
"grad_norm": 1.777043342590332,
|
| 738 |
+
"learning_rate": 3.439153439153439e-05,
|
| 739 |
+
"loss": 0.0782,
|
| 740 |
+
"step": 800
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 19.285714285714285,
|
| 744 |
+
"grad_norm": 0.9696865677833557,
|
| 745 |
+
"learning_rate": 3.412698412698413e-05,
|
| 746 |
+
"loss": 0.08,
|
| 747 |
+
"step": 810
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 19.523809523809526,
|
| 751 |
+
"grad_norm": 2.0073580741882324,
|
| 752 |
+
"learning_rate": 3.386243386243386e-05,
|
| 753 |
+
"loss": 0.095,
|
| 754 |
+
"step": 820
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 19.761904761904763,
|
| 758 |
+
"grad_norm": 2.404144048690796,
|
| 759 |
+
"learning_rate": 3.35978835978836e-05,
|
| 760 |
+
"loss": 0.116,
|
| 761 |
+
"step": 830
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 20.0,
|
| 765 |
+
"grad_norm": 0.8298864960670471,
|
| 766 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 767 |
+
"loss": 0.0791,
|
| 768 |
+
"step": 840
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 20.0,
|
| 772 |
+
"eval_accuracy": 0.966499162479062,
|
| 773 |
+
"eval_loss": 0.11790105700492859,
|
| 774 |
+
"eval_runtime": 102.1963,
|
| 775 |
+
"eval_samples_per_second": 5.842,
|
| 776 |
+
"eval_steps_per_second": 0.186,
|
| 777 |
+
"step": 840
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 20.238095238095237,
|
| 781 |
+
"grad_norm": 1.1204876899719238,
|
| 782 |
+
"learning_rate": 3.306878306878307e-05,
|
| 783 |
+
"loss": 0.0827,
|
| 784 |
+
"step": 850
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 20.476190476190474,
|
| 788 |
+
"grad_norm": 1.7101030349731445,
|
| 789 |
+
"learning_rate": 3.280423280423281e-05,
|
| 790 |
+
"loss": 0.0809,
|
| 791 |
+
"step": 860
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 20.714285714285715,
|
| 795 |
+
"grad_norm": 1.100553035736084,
|
| 796 |
+
"learning_rate": 3.253968253968254e-05,
|
| 797 |
+
"loss": 0.0768,
|
| 798 |
+
"step": 870
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 20.952380952380953,
|
| 802 |
+
"grad_norm": 1.7138772010803223,
|
| 803 |
+
"learning_rate": 3.227513227513227e-05,
|
| 804 |
+
"loss": 0.0694,
|
| 805 |
+
"step": 880
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 21.0,
|
| 809 |
+
"eval_accuracy": 0.9631490787269682,
|
| 810 |
+
"eval_loss": 0.14235423505306244,
|
| 811 |
+
"eval_runtime": 100.0469,
|
| 812 |
+
"eval_samples_per_second": 5.967,
|
| 813 |
+
"eval_steps_per_second": 0.19,
|
| 814 |
+
"step": 882
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 21.19047619047619,
|
| 818 |
+
"grad_norm": 2.3028171062469482,
|
| 819 |
+
"learning_rate": 3.2010582010582015e-05,
|
| 820 |
+
"loss": 0.0691,
|
| 821 |
+
"step": 890
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 21.428571428571427,
|
| 825 |
+
"grad_norm": 1.7097909450531006,
|
| 826 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 827 |
+
"loss": 0.0748,
|
| 828 |
+
"step": 900
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 21.666666666666668,
|
| 832 |
+
"grad_norm": 1.1676493883132935,
|
| 833 |
+
"learning_rate": 3.148148148148148e-05,
|
| 834 |
+
"loss": 0.0991,
|
| 835 |
+
"step": 910
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 21.904761904761905,
|
| 839 |
+
"grad_norm": 1.6388928890228271,
|
| 840 |
+
"learning_rate": 3.121693121693122e-05,
|
| 841 |
+
"loss": 0.0627,
|
| 842 |
+
"step": 920
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 22.0,
|
| 846 |
+
"eval_accuracy": 0.966499162479062,
|
| 847 |
+
"eval_loss": 0.12243743985891342,
|
| 848 |
+
"eval_runtime": 101.3465,
|
| 849 |
+
"eval_samples_per_second": 5.891,
|
| 850 |
+
"eval_steps_per_second": 0.187,
|
| 851 |
+
"step": 924
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 22.142857142857142,
|
| 855 |
+
"grad_norm": 6.947789192199707,
|
| 856 |
+
"learning_rate": 3.095238095238095e-05,
|
| 857 |
+
"loss": 0.0952,
|
| 858 |
+
"step": 930
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 22.38095238095238,
|
| 862 |
+
"grad_norm": 2.1520142555236816,
|
| 863 |
+
"learning_rate": 3.068783068783069e-05,
|
| 864 |
+
"loss": 0.0697,
|
| 865 |
+
"step": 940
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 22.61904761904762,
|
| 869 |
+
"grad_norm": 2.1891136169433594,
|
| 870 |
+
"learning_rate": 3.0423280423280425e-05,
|
| 871 |
+
"loss": 0.0727,
|
| 872 |
+
"step": 950
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 22.857142857142858,
|
| 876 |
+
"grad_norm": 1.557528018951416,
|
| 877 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 878 |
+
"loss": 0.0883,
|
| 879 |
+
"step": 960
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 23.0,
|
| 883 |
+
"eval_accuracy": 0.9447236180904522,
|
| 884 |
+
"eval_loss": 0.16018958389759064,
|
| 885 |
+
"eval_runtime": 101.009,
|
| 886 |
+
"eval_samples_per_second": 5.91,
|
| 887 |
+
"eval_steps_per_second": 0.188,
|
| 888 |
+
"step": 966
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 23.095238095238095,
|
| 892 |
+
"grad_norm": 0.4970981478691101,
|
| 893 |
+
"learning_rate": 2.9894179894179897e-05,
|
| 894 |
+
"loss": 0.0727,
|
| 895 |
+
"step": 970
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 23.333333333333332,
|
| 899 |
+
"grad_norm": 0.9330291748046875,
|
| 900 |
+
"learning_rate": 2.962962962962963e-05,
|
| 901 |
+
"loss": 0.0776,
|
| 902 |
+
"step": 980
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 23.571428571428573,
|
| 906 |
+
"grad_norm": 0.5385302305221558,
|
| 907 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 908 |
+
"loss": 0.068,
|
| 909 |
+
"step": 990
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 23.80952380952381,
|
| 913 |
+
"grad_norm": 1.3941255807876587,
|
| 914 |
+
"learning_rate": 2.91005291005291e-05,
|
| 915 |
+
"loss": 0.074,
|
| 916 |
+
"step": 1000
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 24.0,
|
| 920 |
+
"eval_accuracy": 0.9614740368509213,
|
| 921 |
+
"eval_loss": 0.13146798312664032,
|
| 922 |
+
"eval_runtime": 101.4893,
|
| 923 |
+
"eval_samples_per_second": 5.882,
|
| 924 |
+
"eval_steps_per_second": 0.187,
|
| 925 |
+
"step": 1008
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 24.047619047619047,
|
| 929 |
+
"grad_norm": 1.8402540683746338,
|
| 930 |
+
"learning_rate": 2.8835978835978838e-05,
|
| 931 |
+
"loss": 0.0529,
|
| 932 |
+
"step": 1010
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 24.285714285714285,
|
| 936 |
+
"grad_norm": 1.1238616704940796,
|
| 937 |
+
"learning_rate": 2.857142857142857e-05,
|
| 938 |
+
"loss": 0.0645,
|
| 939 |
+
"step": 1020
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 24.523809523809526,
|
| 943 |
+
"grad_norm": 2.4385814666748047,
|
| 944 |
+
"learning_rate": 2.830687830687831e-05,
|
| 945 |
+
"loss": 0.0891,
|
| 946 |
+
"step": 1030
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 24.761904761904763,
|
| 950 |
+
"grad_norm": 0.7520690560340881,
|
| 951 |
+
"learning_rate": 2.8042328042328043e-05,
|
| 952 |
+
"loss": 0.0856,
|
| 953 |
+
"step": 1040
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 25.0,
|
| 957 |
+
"grad_norm": 1.3286396265029907,
|
| 958 |
+
"learning_rate": 2.777777777777778e-05,
|
| 959 |
+
"loss": 0.0708,
|
| 960 |
+
"step": 1050
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 25.0,
|
| 964 |
+
"eval_accuracy": 0.9631490787269682,
|
| 965 |
+
"eval_loss": 0.1330818086862564,
|
| 966 |
+
"eval_runtime": 101.2634,
|
| 967 |
+
"eval_samples_per_second": 5.896,
|
| 968 |
+
"eval_steps_per_second": 0.188,
|
| 969 |
+
"step": 1050
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 25.238095238095237,
|
| 973 |
+
"grad_norm": 1.7304377555847168,
|
| 974 |
+
"learning_rate": 2.7513227513227512e-05,
|
| 975 |
+
"loss": 0.0685,
|
| 976 |
+
"step": 1060
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 25.476190476190474,
|
| 980 |
+
"grad_norm": 3.330090284347534,
|
| 981 |
+
"learning_rate": 2.724867724867725e-05,
|
| 982 |
+
"loss": 0.0664,
|
| 983 |
+
"step": 1070
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 25.714285714285715,
|
| 987 |
+
"grad_norm": 0.801770806312561,
|
| 988 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 989 |
+
"loss": 0.0641,
|
| 990 |
+
"step": 1080
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 25.952380952380953,
|
| 994 |
+
"grad_norm": 1.6211001873016357,
|
| 995 |
+
"learning_rate": 2.6719576719576723e-05,
|
| 996 |
+
"loss": 0.06,
|
| 997 |
+
"step": 1090
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 26.0,
|
| 1001 |
+
"eval_accuracy": 0.966499162479062,
|
| 1002 |
+
"eval_loss": 0.11908143013715744,
|
| 1003 |
+
"eval_runtime": 100.9095,
|
| 1004 |
+
"eval_samples_per_second": 5.916,
|
| 1005 |
+
"eval_steps_per_second": 0.188,
|
| 1006 |
+
"step": 1092
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 26.19047619047619,
|
| 1010 |
+
"grad_norm": 1.944045066833496,
|
| 1011 |
+
"learning_rate": 2.6455026455026456e-05,
|
| 1012 |
+
"loss": 0.0636,
|
| 1013 |
+
"step": 1100
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 26.428571428571427,
|
| 1017 |
+
"grad_norm": 1.0481789112091064,
|
| 1018 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 1019 |
+
"loss": 0.0612,
|
| 1020 |
+
"step": 1110
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 26.666666666666668,
|
| 1024 |
+
"grad_norm": 2.2019481658935547,
|
| 1025 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 1026 |
+
"loss": 0.0637,
|
| 1027 |
+
"step": 1120
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 26.904761904761905,
|
| 1031 |
+
"grad_norm": 3.6049156188964844,
|
| 1032 |
+
"learning_rate": 2.5661375661375664e-05,
|
| 1033 |
+
"loss": 0.083,
|
| 1034 |
+
"step": 1130
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 27.0,
|
| 1038 |
+
"eval_accuracy": 0.9530988274706867,
|
| 1039 |
+
"eval_loss": 0.1582762449979782,
|
| 1040 |
+
"eval_runtime": 100.0407,
|
| 1041 |
+
"eval_samples_per_second": 5.968,
|
| 1042 |
+
"eval_steps_per_second": 0.19,
|
| 1043 |
+
"step": 1134
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 27.142857142857142,
|
| 1047 |
+
"grad_norm": 2.5039260387420654,
|
| 1048 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 1049 |
+
"loss": 0.0606,
|
| 1050 |
+
"step": 1140
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 27.38095238095238,
|
| 1054 |
+
"grad_norm": 1.956338882446289,
|
| 1055 |
+
"learning_rate": 2.5132275132275137e-05,
|
| 1056 |
+
"loss": 0.077,
|
| 1057 |
+
"step": 1150
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 27.61904761904762,
|
| 1061 |
+
"grad_norm": 1.0798221826553345,
|
| 1062 |
+
"learning_rate": 2.4867724867724866e-05,
|
| 1063 |
+
"loss": 0.0427,
|
| 1064 |
+
"step": 1160
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 27.857142857142858,
|
| 1068 |
+
"grad_norm": 3.189793348312378,
|
| 1069 |
+
"learning_rate": 2.4603174603174602e-05,
|
| 1070 |
+
"loss": 0.0584,
|
| 1071 |
+
"step": 1170
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 28.0,
|
| 1075 |
+
"eval_accuracy": 0.9564489112227805,
|
| 1076 |
+
"eval_loss": 0.13481402397155762,
|
| 1077 |
+
"eval_runtime": 100.6312,
|
| 1078 |
+
"eval_samples_per_second": 5.933,
|
| 1079 |
+
"eval_steps_per_second": 0.189,
|
| 1080 |
+
"step": 1176
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 28.095238095238095,
|
| 1084 |
+
"grad_norm": 1.0723867416381836,
|
| 1085 |
+
"learning_rate": 2.4338624338624338e-05,
|
| 1086 |
+
"loss": 0.0432,
|
| 1087 |
+
"step": 1180
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 28.333333333333332,
|
| 1091 |
+
"grad_norm": 0.41823190450668335,
|
| 1092 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 1093 |
+
"loss": 0.0747,
|
| 1094 |
+
"step": 1190
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 28.571428571428573,
|
| 1098 |
+
"grad_norm": 1.465440034866333,
|
| 1099 |
+
"learning_rate": 2.380952380952381e-05,
|
| 1100 |
+
"loss": 0.0451,
|
| 1101 |
+
"step": 1200
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 28.80952380952381,
|
| 1105 |
+
"grad_norm": 1.9235646724700928,
|
| 1106 |
+
"learning_rate": 2.3544973544973546e-05,
|
| 1107 |
+
"loss": 0.0627,
|
| 1108 |
+
"step": 1210
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 29.0,
|
| 1112 |
+
"eval_accuracy": 0.9564489112227805,
|
| 1113 |
+
"eval_loss": 0.12703652679920197,
|
| 1114 |
+
"eval_runtime": 101.0473,
|
| 1115 |
+
"eval_samples_per_second": 5.908,
|
| 1116 |
+
"eval_steps_per_second": 0.188,
|
| 1117 |
+
"step": 1218
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 29.047619047619047,
|
| 1121 |
+
"grad_norm": 1.0633718967437744,
|
| 1122 |
+
"learning_rate": 2.328042328042328e-05,
|
| 1123 |
+
"loss": 0.0453,
|
| 1124 |
+
"step": 1220
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 29.285714285714285,
|
| 1128 |
+
"grad_norm": 1.1838030815124512,
|
| 1129 |
+
"learning_rate": 2.3015873015873015e-05,
|
| 1130 |
+
"loss": 0.0549,
|
| 1131 |
+
"step": 1230
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 29.523809523809526,
|
| 1135 |
+
"grad_norm": 1.2892875671386719,
|
| 1136 |
+
"learning_rate": 2.275132275132275e-05,
|
| 1137 |
+
"loss": 0.0506,
|
| 1138 |
+
"step": 1240
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 29.761904761904763,
|
| 1142 |
+
"grad_norm": 0.46934711933135986,
|
| 1143 |
+
"learning_rate": 2.2486772486772487e-05,
|
| 1144 |
+
"loss": 0.0586,
|
| 1145 |
+
"step": 1250
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 30.0,
|
| 1149 |
+
"grad_norm": 0.5484461784362793,
|
| 1150 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 1151 |
+
"loss": 0.0627,
|
| 1152 |
+
"step": 1260
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 30.0,
|
| 1156 |
+
"eval_accuracy": 0.9564489112227805,
|
| 1157 |
+
"eval_loss": 0.14105232059955597,
|
| 1158 |
+
"eval_runtime": 100.5476,
|
| 1159 |
+
"eval_samples_per_second": 5.937,
|
| 1160 |
+
"eval_steps_per_second": 0.189,
|
| 1161 |
+
"step": 1260
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 30.238095238095237,
|
| 1165 |
+
"grad_norm": 1.0722055435180664,
|
| 1166 |
+
"learning_rate": 2.1957671957671956e-05,
|
| 1167 |
+
"loss": 0.0436,
|
| 1168 |
+
"step": 1270
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 30.476190476190474,
|
| 1172 |
+
"grad_norm": 3.5860719680786133,
|
| 1173 |
+
"learning_rate": 2.1693121693121692e-05,
|
| 1174 |
+
"loss": 0.0477,
|
| 1175 |
+
"step": 1280
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 30.714285714285715,
|
| 1179 |
+
"grad_norm": 0.9268338680267334,
|
| 1180 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 1181 |
+
"loss": 0.0464,
|
| 1182 |
+
"step": 1290
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 30.952380952380953,
|
| 1186 |
+
"grad_norm": 0.41101688146591187,
|
| 1187 |
+
"learning_rate": 2.1164021164021164e-05,
|
| 1188 |
+
"loss": 0.038,
|
| 1189 |
+
"step": 1300
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 31.0,
|
| 1193 |
+
"eval_accuracy": 0.966499162479062,
|
| 1194 |
+
"eval_loss": 0.12079473584890366,
|
| 1195 |
+
"eval_runtime": 101.8569,
|
| 1196 |
+
"eval_samples_per_second": 5.861,
|
| 1197 |
+
"eval_steps_per_second": 0.187,
|
| 1198 |
+
"step": 1302
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 31.19047619047619,
|
| 1202 |
+
"grad_norm": 1.7808339595794678,
|
| 1203 |
+
"learning_rate": 2.08994708994709e-05,
|
| 1204 |
+
"loss": 0.0485,
|
| 1205 |
+
"step": 1310
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 31.428571428571427,
|
| 1209 |
+
"grad_norm": 2.029599666595459,
|
| 1210 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 1211 |
+
"loss": 0.0391,
|
| 1212 |
+
"step": 1320
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 31.666666666666668,
|
| 1216 |
+
"grad_norm": 0.1751292645931244,
|
| 1217 |
+
"learning_rate": 2.037037037037037e-05,
|
| 1218 |
+
"loss": 0.0393,
|
| 1219 |
+
"step": 1330
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 31.904761904761905,
|
| 1223 |
+
"grad_norm": 1.3681291341781616,
|
| 1224 |
+
"learning_rate": 2.0105820105820105e-05,
|
| 1225 |
+
"loss": 0.0569,
|
| 1226 |
+
"step": 1340
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 32.0,
|
| 1230 |
+
"eval_accuracy": 0.9514237855946399,
|
| 1231 |
+
"eval_loss": 0.15873192250728607,
|
| 1232 |
+
"eval_runtime": 101.2347,
|
| 1233 |
+
"eval_samples_per_second": 5.897,
|
| 1234 |
+
"eval_steps_per_second": 0.188,
|
| 1235 |
+
"step": 1344
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 32.142857142857146,
|
| 1239 |
+
"grad_norm": 1.053713321685791,
|
| 1240 |
+
"learning_rate": 1.984126984126984e-05,
|
| 1241 |
+
"loss": 0.0385,
|
| 1242 |
+
"step": 1350
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 32.38095238095238,
|
| 1246 |
+
"grad_norm": 1.0178366899490356,
|
| 1247 |
+
"learning_rate": 1.9576719576719577e-05,
|
| 1248 |
+
"loss": 0.0561,
|
| 1249 |
+
"step": 1360
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 32.61904761904762,
|
| 1253 |
+
"grad_norm": 1.472675085067749,
|
| 1254 |
+
"learning_rate": 1.9312169312169313e-05,
|
| 1255 |
+
"loss": 0.0434,
|
| 1256 |
+
"step": 1370
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 32.857142857142854,
|
| 1260 |
+
"grad_norm": 1.232201099395752,
|
| 1261 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 1262 |
+
"loss": 0.0502,
|
| 1263 |
+
"step": 1380
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 33.0,
|
| 1267 |
+
"eval_accuracy": 0.949748743718593,
|
| 1268 |
+
"eval_loss": 0.1500607430934906,
|
| 1269 |
+
"eval_runtime": 102.5738,
|
| 1270 |
+
"eval_samples_per_second": 5.82,
|
| 1271 |
+
"eval_steps_per_second": 0.185,
|
| 1272 |
+
"step": 1386
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 33.095238095238095,
|
| 1276 |
+
"grad_norm": 4.194485664367676,
|
| 1277 |
+
"learning_rate": 1.8783068783068782e-05,
|
| 1278 |
+
"loss": 0.0589,
|
| 1279 |
+
"step": 1390
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 33.333333333333336,
|
| 1283 |
+
"grad_norm": 0.24997162818908691,
|
| 1284 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 1285 |
+
"loss": 0.0389,
|
| 1286 |
+
"step": 1400
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 33.57142857142857,
|
| 1290 |
+
"grad_norm": 1.9876259565353394,
|
| 1291 |
+
"learning_rate": 1.8253968253968254e-05,
|
| 1292 |
+
"loss": 0.039,
|
| 1293 |
+
"step": 1410
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 33.80952380952381,
|
| 1297 |
+
"grad_norm": 1.257222056388855,
|
| 1298 |
+
"learning_rate": 1.798941798941799e-05,
|
| 1299 |
+
"loss": 0.0464,
|
| 1300 |
+
"step": 1420
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 34.0,
|
| 1304 |
+
"eval_accuracy": 0.9614740368509213,
|
| 1305 |
+
"eval_loss": 0.15079863369464874,
|
| 1306 |
+
"eval_runtime": 103.3413,
|
| 1307 |
+
"eval_samples_per_second": 5.777,
|
| 1308 |
+
"eval_steps_per_second": 0.184,
|
| 1309 |
+
"step": 1428
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 34.04761904761905,
|
| 1313 |
+
"grad_norm": 1.1036677360534668,
|
| 1314 |
+
"learning_rate": 1.7724867724867723e-05,
|
| 1315 |
+
"loss": 0.0542,
|
| 1316 |
+
"step": 1430
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 34.285714285714285,
|
| 1320 |
+
"grad_norm": 1.4602792263031006,
|
| 1321 |
+
"learning_rate": 1.746031746031746e-05,
|
| 1322 |
+
"loss": 0.0599,
|
| 1323 |
+
"step": 1440
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 34.523809523809526,
|
| 1327 |
+
"grad_norm": 1.127734899520874,
|
| 1328 |
+
"learning_rate": 1.7195767195767195e-05,
|
| 1329 |
+
"loss": 0.0456,
|
| 1330 |
+
"step": 1450
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 34.76190476190476,
|
| 1334 |
+
"grad_norm": 2.489420175552368,
|
| 1335 |
+
"learning_rate": 1.693121693121693e-05,
|
| 1336 |
+
"loss": 0.034,
|
| 1337 |
+
"step": 1460
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 35.0,
|
| 1341 |
+
"grad_norm": 1.582891821861267,
|
| 1342 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 1343 |
+
"loss": 0.0317,
|
| 1344 |
+
"step": 1470
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 35.0,
|
| 1348 |
+
"eval_accuracy": 0.9631490787269682,
|
| 1349 |
+
"eval_loss": 0.13086940348148346,
|
| 1350 |
+
"eval_runtime": 102.9849,
|
| 1351 |
+
"eval_samples_per_second": 5.797,
|
| 1352 |
+
"eval_steps_per_second": 0.184,
|
| 1353 |
+
"step": 1470
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 35.23809523809524,
|
| 1357 |
+
"grad_norm": 1.932721495628357,
|
| 1358 |
+
"learning_rate": 1.6402116402116404e-05,
|
| 1359 |
+
"loss": 0.0387,
|
| 1360 |
+
"step": 1480
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 35.476190476190474,
|
| 1364 |
+
"grad_norm": 3.2499399185180664,
|
| 1365 |
+
"learning_rate": 1.6137566137566136e-05,
|
| 1366 |
+
"loss": 0.0556,
|
| 1367 |
+
"step": 1490
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 35.714285714285715,
|
| 1371 |
+
"grad_norm": 1.335181713104248,
|
| 1372 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 1373 |
+
"loss": 0.0514,
|
| 1374 |
+
"step": 1500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 35.95238095238095,
|
| 1378 |
+
"grad_norm": 0.5096970200538635,
|
| 1379 |
+
"learning_rate": 1.560846560846561e-05,
|
| 1380 |
+
"loss": 0.0552,
|
| 1381 |
+
"step": 1510
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 36.0,
|
| 1385 |
+
"eval_accuracy": 0.9597989949748744,
|
| 1386 |
+
"eval_loss": 0.13715249300003052,
|
| 1387 |
+
"eval_runtime": 101.916,
|
| 1388 |
+
"eval_samples_per_second": 5.858,
|
| 1389 |
+
"eval_steps_per_second": 0.186,
|
| 1390 |
+
"step": 1512
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 36.19047619047619,
|
| 1394 |
+
"grad_norm": 0.6248478293418884,
|
| 1395 |
+
"learning_rate": 1.5343915343915344e-05,
|
| 1396 |
+
"loss": 0.0634,
|
| 1397 |
+
"step": 1520
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 36.42857142857143,
|
| 1401 |
+
"grad_norm": 0.9641430377960205,
|
| 1402 |
+
"learning_rate": 1.5079365079365079e-05,
|
| 1403 |
+
"loss": 0.0369,
|
| 1404 |
+
"step": 1530
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 36.666666666666664,
|
| 1408 |
+
"grad_norm": 0.7052914500236511,
|
| 1409 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 1410 |
+
"loss": 0.0429,
|
| 1411 |
+
"step": 1540
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 36.904761904761905,
|
| 1415 |
+
"grad_norm": 0.45987239480018616,
|
| 1416 |
+
"learning_rate": 1.455026455026455e-05,
|
| 1417 |
+
"loss": 0.031,
|
| 1418 |
+
"step": 1550
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 37.0,
|
| 1422 |
+
"eval_accuracy": 0.9597989949748744,
|
| 1423 |
+
"eval_loss": 0.12577290832996368,
|
| 1424 |
+
"eval_runtime": 102.0267,
|
| 1425 |
+
"eval_samples_per_second": 5.851,
|
| 1426 |
+
"eval_steps_per_second": 0.186,
|
| 1427 |
+
"step": 1554
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 37.142857142857146,
|
| 1431 |
+
"grad_norm": 1.7245473861694336,
|
| 1432 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 1433 |
+
"loss": 0.0519,
|
| 1434 |
+
"step": 1560
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 37.38095238095238,
|
| 1438 |
+
"grad_norm": 1.6640981435775757,
|
| 1439 |
+
"learning_rate": 1.4021164021164022e-05,
|
| 1440 |
+
"loss": 0.0448,
|
| 1441 |
+
"step": 1570
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 37.61904761904762,
|
| 1445 |
+
"grad_norm": 0.358934611082077,
|
| 1446 |
+
"learning_rate": 1.3756613756613756e-05,
|
| 1447 |
+
"loss": 0.0315,
|
| 1448 |
+
"step": 1580
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 37.857142857142854,
|
| 1452 |
+
"grad_norm": 3.198369026184082,
|
| 1453 |
+
"learning_rate": 1.3492063492063492e-05,
|
| 1454 |
+
"loss": 0.0383,
|
| 1455 |
+
"step": 1590
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 38.0,
|
| 1459 |
+
"eval_accuracy": 0.9681742043551089,
|
| 1460 |
+
"eval_loss": 0.12486154586076736,
|
| 1461 |
+
"eval_runtime": 101.5464,
|
| 1462 |
+
"eval_samples_per_second": 5.879,
|
| 1463 |
+
"eval_steps_per_second": 0.187,
|
| 1464 |
+
"step": 1596
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 38.095238095238095,
|
| 1468 |
+
"grad_norm": 0.6694173216819763,
|
| 1469 |
+
"learning_rate": 1.3227513227513228e-05,
|
| 1470 |
+
"loss": 0.0289,
|
| 1471 |
+
"step": 1600
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 38.333333333333336,
|
| 1475 |
+
"grad_norm": 2.153737783432007,
|
| 1476 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 1477 |
+
"loss": 0.038,
|
| 1478 |
+
"step": 1610
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 38.57142857142857,
|
| 1482 |
+
"grad_norm": 0.9305407404899597,
|
| 1483 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 1484 |
+
"loss": 0.0467,
|
| 1485 |
+
"step": 1620
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 38.80952380952381,
|
| 1489 |
+
"grad_norm": 1.1094582080841064,
|
| 1490 |
+
"learning_rate": 1.2433862433862433e-05,
|
| 1491 |
+
"loss": 0.036,
|
| 1492 |
+
"step": 1630
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 39.0,
|
| 1496 |
+
"eval_accuracy": 0.966499162479062,
|
| 1497 |
+
"eval_loss": 0.13118697702884674,
|
| 1498 |
+
"eval_runtime": 101.001,
|
| 1499 |
+
"eval_samples_per_second": 5.911,
|
| 1500 |
+
"eval_steps_per_second": 0.188,
|
| 1501 |
+
"step": 1638
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 39.04761904761905,
|
| 1505 |
+
"grad_norm": 0.5792038440704346,
|
| 1506 |
+
"learning_rate": 1.2169312169312169e-05,
|
| 1507 |
+
"loss": 0.0499,
|
| 1508 |
+
"step": 1640
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 39.285714285714285,
|
| 1512 |
+
"grad_norm": 0.9286515116691589,
|
| 1513 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 1514 |
+
"loss": 0.0366,
|
| 1515 |
+
"step": 1650
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 39.523809523809526,
|
| 1519 |
+
"grad_norm": 0.1906491369009018,
|
| 1520 |
+
"learning_rate": 1.164021164021164e-05,
|
| 1521 |
+
"loss": 0.0479,
|
| 1522 |
+
"step": 1660
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 39.76190476190476,
|
| 1526 |
+
"grad_norm": 0.3992154896259308,
|
| 1527 |
+
"learning_rate": 1.1375661375661376e-05,
|
| 1528 |
+
"loss": 0.037,
|
| 1529 |
+
"step": 1670
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 40.0,
|
| 1533 |
+
"grad_norm": 0.3215673863887787,
|
| 1534 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1535 |
+
"loss": 0.0405,
|
| 1536 |
+
"step": 1680
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 40.0,
|
| 1540 |
+
"eval_accuracy": 0.966499162479062,
|
| 1541 |
+
"eval_loss": 0.1207132637500763,
|
| 1542 |
+
"eval_runtime": 101.9503,
|
| 1543 |
+
"eval_samples_per_second": 5.856,
|
| 1544 |
+
"eval_steps_per_second": 0.186,
|
| 1545 |
+
"step": 1680
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 40.23809523809524,
|
| 1549 |
+
"grad_norm": 0.8763807415962219,
|
| 1550 |
+
"learning_rate": 1.0846560846560846e-05,
|
| 1551 |
+
"loss": 0.0258,
|
| 1552 |
+
"step": 1690
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 40.476190476190474,
|
| 1556 |
+
"grad_norm": 1.2383830547332764,
|
| 1557 |
+
"learning_rate": 1.0582010582010582e-05,
|
| 1558 |
+
"loss": 0.0373,
|
| 1559 |
+
"step": 1700
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 40.714285714285715,
|
| 1563 |
+
"grad_norm": 1.2714569568634033,
|
| 1564 |
+
"learning_rate": 1.0317460317460318e-05,
|
| 1565 |
+
"loss": 0.0345,
|
| 1566 |
+
"step": 1710
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 40.95238095238095,
|
| 1570 |
+
"grad_norm": 1.3528138399124146,
|
| 1571 |
+
"learning_rate": 1.0052910052910053e-05,
|
| 1572 |
+
"loss": 0.0343,
|
| 1573 |
+
"step": 1720
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 41.0,
|
| 1577 |
+
"eval_accuracy": 0.964824120603015,
|
| 1578 |
+
"eval_loss": 0.12328309565782547,
|
| 1579 |
+
"eval_runtime": 102.5699,
|
| 1580 |
+
"eval_samples_per_second": 5.82,
|
| 1581 |
+
"eval_steps_per_second": 0.185,
|
| 1582 |
+
"step": 1722
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 41.19047619047619,
|
| 1586 |
+
"grad_norm": 0.28864073753356934,
|
| 1587 |
+
"learning_rate": 9.788359788359789e-06,
|
| 1588 |
+
"loss": 0.0306,
|
| 1589 |
+
"step": 1730
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 41.42857142857143,
|
| 1593 |
+
"grad_norm": 0.915897011756897,
|
| 1594 |
+
"learning_rate": 9.523809523809523e-06,
|
| 1595 |
+
"loss": 0.0335,
|
| 1596 |
+
"step": 1740
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 41.666666666666664,
|
| 1600 |
+
"grad_norm": 1.0583347082138062,
|
| 1601 |
+
"learning_rate": 9.259259259259259e-06,
|
| 1602 |
+
"loss": 0.036,
|
| 1603 |
+
"step": 1750
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 41.904761904761905,
|
| 1607 |
+
"grad_norm": 1.7799439430236816,
|
| 1608 |
+
"learning_rate": 8.994708994708995e-06,
|
| 1609 |
+
"loss": 0.0325,
|
| 1610 |
+
"step": 1760
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 42.0,
|
| 1614 |
+
"eval_accuracy": 0.9631490787269682,
|
| 1615 |
+
"eval_loss": 0.12857390940189362,
|
| 1616 |
+
"eval_runtime": 104.7251,
|
| 1617 |
+
"eval_samples_per_second": 5.701,
|
| 1618 |
+
"eval_steps_per_second": 0.181,
|
| 1619 |
+
"step": 1764
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 42.142857142857146,
|
| 1623 |
+
"grad_norm": 1.5153943300247192,
|
| 1624 |
+
"learning_rate": 8.73015873015873e-06,
|
| 1625 |
+
"loss": 0.0305,
|
| 1626 |
+
"step": 1770
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 42.38095238095238,
|
| 1630 |
+
"grad_norm": 1.5341339111328125,
|
| 1631 |
+
"learning_rate": 8.465608465608466e-06,
|
| 1632 |
+
"loss": 0.0422,
|
| 1633 |
+
"step": 1780
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 42.61904761904762,
|
| 1637 |
+
"grad_norm": 0.7007244825363159,
|
| 1638 |
+
"learning_rate": 8.201058201058202e-06,
|
| 1639 |
+
"loss": 0.0462,
|
| 1640 |
+
"step": 1790
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 42.857142857142854,
|
| 1644 |
+
"grad_norm": 1.075047254562378,
|
| 1645 |
+
"learning_rate": 7.936507936507936e-06,
|
| 1646 |
+
"loss": 0.0293,
|
| 1647 |
+
"step": 1800
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 43.0,
|
| 1651 |
+
"eval_accuracy": 0.9681742043551089,
|
| 1652 |
+
"eval_loss": 0.11350709199905396,
|
| 1653 |
+
"eval_runtime": 105.2069,
|
| 1654 |
+
"eval_samples_per_second": 5.675,
|
| 1655 |
+
"eval_steps_per_second": 0.181,
|
| 1656 |
+
"step": 1806
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 43.095238095238095,
|
| 1660 |
+
"grad_norm": 0.27737095952033997,
|
| 1661 |
+
"learning_rate": 7.671957671957672e-06,
|
| 1662 |
+
"loss": 0.0308,
|
| 1663 |
+
"step": 1810
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 43.333333333333336,
|
| 1667 |
+
"grad_norm": 0.4645295739173889,
|
| 1668 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 1669 |
+
"loss": 0.0296,
|
| 1670 |
+
"step": 1820
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 43.57142857142857,
|
| 1674 |
+
"grad_norm": 0.7459697723388672,
|
| 1675 |
+
"learning_rate": 7.142857142857143e-06,
|
| 1676 |
+
"loss": 0.039,
|
| 1677 |
+
"step": 1830
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 43.80952380952381,
|
| 1681 |
+
"grad_norm": 1.326521396636963,
|
| 1682 |
+
"learning_rate": 6.878306878306878e-06,
|
| 1683 |
+
"loss": 0.0306,
|
| 1684 |
+
"step": 1840
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 44.0,
|
| 1688 |
+
"eval_accuracy": 0.9614740368509213,
|
| 1689 |
+
"eval_loss": 0.12575660645961761,
|
| 1690 |
+
"eval_runtime": 104.9936,
|
| 1691 |
+
"eval_samples_per_second": 5.686,
|
| 1692 |
+
"eval_steps_per_second": 0.181,
|
| 1693 |
+
"step": 1848
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 44.04761904761905,
|
| 1697 |
+
"grad_norm": 0.11777963489294052,
|
| 1698 |
+
"learning_rate": 6.613756613756614e-06,
|
| 1699 |
+
"loss": 0.0355,
|
| 1700 |
+
"step": 1850
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 44.285714285714285,
|
| 1704 |
+
"grad_norm": 0.9748639464378357,
|
| 1705 |
+
"learning_rate": 6.349206349206349e-06,
|
| 1706 |
+
"loss": 0.0281,
|
| 1707 |
+
"step": 1860
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 44.523809523809526,
|
| 1711 |
+
"grad_norm": 1.6148064136505127,
|
| 1712 |
+
"learning_rate": 6.0846560846560845e-06,
|
| 1713 |
+
"loss": 0.0251,
|
| 1714 |
+
"step": 1870
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 44.76190476190476,
|
| 1718 |
+
"grad_norm": 0.1060701385140419,
|
| 1719 |
+
"learning_rate": 5.82010582010582e-06,
|
| 1720 |
+
"loss": 0.0242,
|
| 1721 |
+
"step": 1880
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 45.0,
|
| 1725 |
+
"grad_norm": 1.808620572090149,
|
| 1726 |
+
"learning_rate": 5.555555555555556e-06,
|
| 1727 |
+
"loss": 0.0267,
|
| 1728 |
+
"step": 1890
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 45.0,
|
| 1732 |
+
"eval_accuracy": 0.964824120603015,
|
| 1733 |
+
"eval_loss": 0.12606783211231232,
|
| 1734 |
+
"eval_runtime": 106.1357,
|
| 1735 |
+
"eval_samples_per_second": 5.625,
|
| 1736 |
+
"eval_steps_per_second": 0.179,
|
| 1737 |
+
"step": 1890
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 45.23809523809524,
|
| 1741 |
+
"grad_norm": 1.064568042755127,
|
| 1742 |
+
"learning_rate": 5.291005291005291e-06,
|
| 1743 |
+
"loss": 0.0295,
|
| 1744 |
+
"step": 1900
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 45.476190476190474,
|
| 1748 |
+
"grad_norm": 0.4456542730331421,
|
| 1749 |
+
"learning_rate": 5.026455026455026e-06,
|
| 1750 |
+
"loss": 0.0341,
|
| 1751 |
+
"step": 1910
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 45.714285714285715,
|
| 1755 |
+
"grad_norm": 1.5678577423095703,
|
| 1756 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 1757 |
+
"loss": 0.0323,
|
| 1758 |
+
"step": 1920
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 45.95238095238095,
|
| 1762 |
+
"grad_norm": 1.6082887649536133,
|
| 1763 |
+
"learning_rate": 4.497354497354498e-06,
|
| 1764 |
+
"loss": 0.0338,
|
| 1765 |
+
"step": 1930
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 46.0,
|
| 1769 |
+
"eval_accuracy": 0.966499162479062,
|
| 1770 |
+
"eval_loss": 0.12089279294013977,
|
| 1771 |
+
"eval_runtime": 105.3726,
|
| 1772 |
+
"eval_samples_per_second": 5.666,
|
| 1773 |
+
"eval_steps_per_second": 0.18,
|
| 1774 |
+
"step": 1932
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 46.19047619047619,
|
| 1778 |
+
"grad_norm": 0.628757894039154,
|
| 1779 |
+
"learning_rate": 4.232804232804233e-06,
|
| 1780 |
+
"loss": 0.039,
|
| 1781 |
+
"step": 1940
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 46.42857142857143,
|
| 1785 |
+
"grad_norm": 0.13868111371994019,
|
| 1786 |
+
"learning_rate": 3.968253968253968e-06,
|
| 1787 |
+
"loss": 0.0264,
|
| 1788 |
+
"step": 1950
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 46.666666666666664,
|
| 1792 |
+
"grad_norm": 1.5409092903137207,
|
| 1793 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 1794 |
+
"loss": 0.0312,
|
| 1795 |
+
"step": 1960
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 46.904761904761905,
|
| 1799 |
+
"grad_norm": 1.158074975013733,
|
| 1800 |
+
"learning_rate": 3.439153439153439e-06,
|
| 1801 |
+
"loss": 0.0213,
|
| 1802 |
+
"step": 1970
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 47.0,
|
| 1806 |
+
"eval_accuracy": 0.966499162479062,
|
| 1807 |
+
"eval_loss": 0.11565249413251877,
|
| 1808 |
+
"eval_runtime": 104.0226,
|
| 1809 |
+
"eval_samples_per_second": 5.739,
|
| 1810 |
+
"eval_steps_per_second": 0.183,
|
| 1811 |
+
"step": 1974
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 47.142857142857146,
|
| 1815 |
+
"grad_norm": 1.6575576066970825,
|
| 1816 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 1817 |
+
"loss": 0.0262,
|
| 1818 |
+
"step": 1980
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 47.38095238095238,
|
| 1822 |
+
"grad_norm": 1.0997730493545532,
|
| 1823 |
+
"learning_rate": 2.91005291005291e-06,
|
| 1824 |
+
"loss": 0.0263,
|
| 1825 |
+
"step": 1990
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 47.61904761904762,
|
| 1829 |
+
"grad_norm": 0.41330787539482117,
|
| 1830 |
+
"learning_rate": 2.6455026455026455e-06,
|
| 1831 |
+
"loss": 0.031,
|
| 1832 |
+
"step": 2000
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 47.857142857142854,
|
| 1836 |
+
"grad_norm": 0.409661203622818,
|
| 1837 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 1838 |
+
"loss": 0.0285,
|
| 1839 |
+
"step": 2010
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 48.0,
|
| 1843 |
+
"eval_accuracy": 0.9631490787269682,
|
| 1844 |
+
"eval_loss": 0.12032893300056458,
|
| 1845 |
+
"eval_runtime": 103.9176,
|
| 1846 |
+
"eval_samples_per_second": 5.745,
|
| 1847 |
+
"eval_steps_per_second": 0.183,
|
| 1848 |
+
"step": 2016
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 48.095238095238095,
|
| 1852 |
+
"grad_norm": 1.2258131504058838,
|
| 1853 |
+
"learning_rate": 2.1164021164021164e-06,
|
| 1854 |
+
"loss": 0.0379,
|
| 1855 |
+
"step": 2020
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 48.333333333333336,
|
| 1859 |
+
"grad_norm": 1.0911734104156494,
|
| 1860 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1861 |
+
"loss": 0.0267,
|
| 1862 |
+
"step": 2030
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 48.57142857142857,
|
| 1866 |
+
"grad_norm": 1.2295212745666504,
|
| 1867 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 1868 |
+
"loss": 0.0194,
|
| 1869 |
+
"step": 2040
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 48.80952380952381,
|
| 1873 |
+
"grad_norm": 0.645757257938385,
|
| 1874 |
+
"learning_rate": 1.3227513227513228e-06,
|
| 1875 |
+
"loss": 0.0287,
|
| 1876 |
+
"step": 2050
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 49.0,
|
| 1880 |
+
"eval_accuracy": 0.964824120603015,
|
| 1881 |
+
"eval_loss": 0.12397059053182602,
|
| 1882 |
+
"eval_runtime": 104.3655,
|
| 1883 |
+
"eval_samples_per_second": 5.72,
|
| 1884 |
+
"eval_steps_per_second": 0.182,
|
| 1885 |
+
"step": 2058
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 49.04761904761905,
|
| 1889 |
+
"grad_norm": 1.8496387004852295,
|
| 1890 |
+
"learning_rate": 1.0582010582010582e-06,
|
| 1891 |
+
"loss": 0.0338,
|
| 1892 |
+
"step": 2060
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 49.285714285714285,
|
| 1896 |
+
"grad_norm": 0.7916781306266785,
|
| 1897 |
+
"learning_rate": 7.936507936507937e-07,
|
| 1898 |
+
"loss": 0.0271,
|
| 1899 |
+
"step": 2070
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 49.523809523809526,
|
| 1903 |
+
"grad_norm": 1.60358464717865,
|
| 1904 |
+
"learning_rate": 5.291005291005291e-07,
|
| 1905 |
+
"loss": 0.0329,
|
| 1906 |
+
"step": 2080
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 49.76190476190476,
|
| 1910 |
+
"grad_norm": 1.0516901016235352,
|
| 1911 |
+
"learning_rate": 2.6455026455026455e-07,
|
| 1912 |
+
"loss": 0.0373,
|
| 1913 |
+
"step": 2090
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 50.0,
|
| 1917 |
+
"grad_norm": 1.148287296295166,
|
| 1918 |
+
"learning_rate": 0.0,
|
| 1919 |
+
"loss": 0.0183,
|
| 1920 |
+
"step": 2100
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 50.0,
|
| 1924 |
+
"eval_accuracy": 0.966499162479062,
|
| 1925 |
+
"eval_loss": 0.12244007736444473,
|
| 1926 |
+
"eval_runtime": 105.8827,
|
| 1927 |
+
"eval_samples_per_second": 5.638,
|
| 1928 |
+
"eval_steps_per_second": 0.179,
|
| 1929 |
+
"step": 2100
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 50.0,
|
| 1933 |
+
"step": 2100,
|
| 1934 |
+
"total_flos": 2.0803097508518707e+19,
|
| 1935 |
+
"train_loss": 0.1363242975941726,
|
| 1936 |
+
"train_runtime": 52918.3087,
|
| 1937 |
+
"train_samples_per_second": 5.073,
|
| 1938 |
+
"train_steps_per_second": 0.04
|
| 1939 |
+
}
|
| 1940 |
+
],
|
| 1941 |
+
"logging_steps": 10,
|
| 1942 |
+
"max_steps": 2100,
|
| 1943 |
+
"num_input_tokens_seen": 0,
|
| 1944 |
+
"num_train_epochs": 50,
|
| 1945 |
+
"save_steps": 500,
|
| 1946 |
+
"stateful_callbacks": {
|
| 1947 |
+
"TrainerControl": {
|
| 1948 |
+
"args": {
|
| 1949 |
+
"should_epoch_stop": false,
|
| 1950 |
+
"should_evaluate": false,
|
| 1951 |
+
"should_log": false,
|
| 1952 |
+
"should_save": true,
|
| 1953 |
+
"should_training_stop": true
|
| 1954 |
+
},
|
| 1955 |
+
"attributes": {}
|
| 1956 |
+
}
|
| 1957 |
+
},
|
| 1958 |
+
"total_flos": 2.0803097508518707e+19,
|
| 1959 |
+
"train_batch_size": 32,
|
| 1960 |
+
"trial_name": null,
|
| 1961 |
+
"trial_params": null
|
| 1962 |
+
}
|