🍻 cheers
Browse files- README.md +7 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- runs/Mar19_05-57-04_6492c5bf3fae/events.out.tfevents.1710829161.6492c5bf3fae.6515.7 +3 -0
- train_results.json +8 -0
- trainer_state.json +1424 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
# vit-lr-inverse-sqrt
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.
|
| 24 |
-
- Accuracy: 0.
|
| 25 |
-
- Precision: 0.
|
| 26 |
-
- Recall: 0.
|
| 27 |
-
- F1: 0.
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 19 |
|
| 20 |
# vit-lr-inverse-sqrt
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.4469
|
| 25 |
+
- Accuracy: 0.8499
|
| 26 |
+
- Precision: 0.8565
|
| 27 |
+
- Recall: 0.8499
|
| 28 |
+
- F1: 0.8516
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.3,
|
| 3 |
+
"eval_accuracy": 0.8498613037447988,
|
| 4 |
+
"eval_f1": 0.8515512179522667,
|
| 5 |
+
"eval_loss": 0.44692692160606384,
|
| 6 |
+
"eval_precision": 0.856522763832034,
|
| 7 |
+
"eval_recall": 0.8498613037447988,
|
| 8 |
+
"eval_runtime": 37.3402,
|
| 9 |
+
"eval_samples_per_second": 77.236,
|
| 10 |
+
"eval_steps_per_second": 9.668,
|
| 11 |
+
"total_flos": 2.1047767559471923e+18,
|
| 12 |
+
"train_loss": 0.21456260421779005,
|
| 13 |
+
"train_runtime": 1227.9557,
|
| 14 |
+
"train_samples_per_second": 417.605,
|
| 15 |
+
"train_steps_per_second": 26.141
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.3,
|
| 3 |
+
"eval_accuracy": 0.8498613037447988,
|
| 4 |
+
"eval_f1": 0.8515512179522667,
|
| 5 |
+
"eval_loss": 0.44692692160606384,
|
| 6 |
+
"eval_precision": 0.856522763832034,
|
| 7 |
+
"eval_recall": 0.8498613037447988,
|
| 8 |
+
"eval_runtime": 37.3402,
|
| 9 |
+
"eval_samples_per_second": 77.236,
|
| 10 |
+
"eval_steps_per_second": 9.668
|
| 11 |
+
}
|
runs/Mar19_05-57-04_6492c5bf3fae/events.out.tfevents.1710829161.6492c5bf3fae.6515.7
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:722b3ad70bd73a57d10c963d9971a604b3a146f600a192ab87fd18c1ec60a615
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.3,
|
| 3 |
+
"total_flos": 2.1047767559471923e+18,
|
| 4 |
+
"train_loss": 0.21456260421779005,
|
| 5 |
+
"train_runtime": 1227.9557,
|
| 6 |
+
"train_samples_per_second": 417.605,
|
| 7 |
+
"train_steps_per_second": 26.141
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1424 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.44692692160606384,
|
| 3 |
+
"best_model_checkpoint": "./vit-lr-inverse-sqrt/checkpoint-700",
|
| 4 |
+
"epoch": 5.29595015576324,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1700,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 17.940969467163086,
|
| 14 |
+
"learning_rate": 1.125e-05,
|
| 15 |
+
"loss": 2.0172,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06,
|
| 20 |
+
"grad_norm": 4.308961391448975,
|
| 21 |
+
"learning_rate": 2.375e-05,
|
| 22 |
+
"loss": 1.1159,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09,
|
| 27 |
+
"grad_norm": 5.38205099105835,
|
| 28 |
+
"learning_rate": 3.625e-05,
|
| 29 |
+
"loss": 1.1398,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.12,
|
| 34 |
+
"grad_norm": 5.569328308105469,
|
| 35 |
+
"learning_rate": 4.875e-05,
|
| 36 |
+
"loss": 1.0508,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16,
|
| 41 |
+
"grad_norm": 5.870121002197266,
|
| 42 |
+
"learning_rate": 6.125000000000001e-05,
|
| 43 |
+
"loss": 0.8095,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19,
|
| 48 |
+
"grad_norm": 6.100069046020508,
|
| 49 |
+
"learning_rate": 7.375e-05,
|
| 50 |
+
"loss": 0.8756,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22,
|
| 55 |
+
"grad_norm": 4.655179023742676,
|
| 56 |
+
"learning_rate": 8.625000000000001e-05,
|
| 57 |
+
"loss": 0.9221,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.25,
|
| 62 |
+
"grad_norm": 4.762995719909668,
|
| 63 |
+
"learning_rate": 9.875000000000002e-05,
|
| 64 |
+
"loss": 0.6852,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.28,
|
| 69 |
+
"grad_norm": 5.286139011383057,
|
| 70 |
+
"learning_rate": 9.480909262799545e-05,
|
| 71 |
+
"loss": 0.6655,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.31,
|
| 76 |
+
"grad_norm": 4.113163471221924,
|
| 77 |
+
"learning_rate": 8.989331499509895e-05,
|
| 78 |
+
"loss": 0.6694,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.31,
|
| 83 |
+
"eval_accuracy": 0.7631761442441054,
|
| 84 |
+
"eval_f1": 0.7499001805196772,
|
| 85 |
+
"eval_loss": 0.6511249542236328,
|
| 86 |
+
"eval_precision": 0.7557663051217941,
|
| 87 |
+
"eval_recall": 0.7631761442441054,
|
| 88 |
+
"eval_runtime": 36.5742,
|
| 89 |
+
"eval_samples_per_second": 78.853,
|
| 90 |
+
"eval_steps_per_second": 9.87,
|
| 91 |
+
"step": 100
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.34,
|
| 95 |
+
"grad_norm": 3.8087217807769775,
|
| 96 |
+
"learning_rate": 8.567058737562387e-05,
|
| 97 |
+
"loss": 0.6742,
|
| 98 |
+
"step": 110
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.37,
|
| 102 |
+
"grad_norm": 8.133487701416016,
|
| 103 |
+
"learning_rate": 8.199200616907878e-05,
|
| 104 |
+
"loss": 0.6546,
|
| 105 |
+
"step": 120
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.4,
|
| 109 |
+
"grad_norm": 7.749859809875488,
|
| 110 |
+
"learning_rate": 7.874992309581578e-05,
|
| 111 |
+
"loss": 0.6175,
|
| 112 |
+
"step": 130
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.44,
|
| 116 |
+
"grad_norm": 3.3360700607299805,
|
| 117 |
+
"learning_rate": 7.586432418108816e-05,
|
| 118 |
+
"loss": 0.4904,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.47,
|
| 123 |
+
"grad_norm": 4.42198371887207,
|
| 124 |
+
"learning_rate": 7.327433054473117e-05,
|
| 125 |
+
"loss": 0.5958,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.5,
|
| 130 |
+
"grad_norm": 6.188783168792725,
|
| 131 |
+
"learning_rate": 7.093269021319087e-05,
|
| 132 |
+
"loss": 0.5494,
|
| 133 |
+
"step": 160
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.53,
|
| 137 |
+
"grad_norm": 4.339505672454834,
|
| 138 |
+
"learning_rate": 6.880209161537815e-05,
|
| 139 |
+
"loss": 0.5787,
|
| 140 |
+
"step": 170
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.56,
|
| 144 |
+
"grad_norm": 7.221861362457275,
|
| 145 |
+
"learning_rate": 6.685262704648756e-05,
|
| 146 |
+
"loss": 0.6344,
|
| 147 |
+
"step": 180
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.59,
|
| 151 |
+
"grad_norm": 6.194063663482666,
|
| 152 |
+
"learning_rate": 6.506000486323555e-05,
|
| 153 |
+
"loss": 0.5987,
|
| 154 |
+
"step": 190
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.62,
|
| 158 |
+
"grad_norm": 5.874014854431152,
|
| 159 |
+
"learning_rate": 6.340426249482415e-05,
|
| 160 |
+
"loss": 0.5468,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.62,
|
| 165 |
+
"eval_accuracy": 0.7617891816920943,
|
| 166 |
+
"eval_f1": 0.7109169676846827,
|
| 167 |
+
"eval_loss": 0.6536840796470642,
|
| 168 |
+
"eval_precision": 0.7713261180604432,
|
| 169 |
+
"eval_recall": 0.7617891816920943,
|
| 170 |
+
"eval_runtime": 36.698,
|
| 171 |
+
"eval_samples_per_second": 78.587,
|
| 172 |
+
"eval_steps_per_second": 9.837,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.65,
|
| 177 |
+
"grad_norm": 4.4315361976623535,
|
| 178 |
+
"learning_rate": 6.18688224889746e-05,
|
| 179 |
+
"loss": 0.6049,
|
| 180 |
+
"step": 210
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.69,
|
| 184 |
+
"grad_norm": 6.3696699142456055,
|
| 185 |
+
"learning_rate": 6.043978852154994e-05,
|
| 186 |
+
"loss": 0.6266,
|
| 187 |
+
"step": 220
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.72,
|
| 191 |
+
"grad_norm": 4.233091354370117,
|
| 192 |
+
"learning_rate": 5.910541245655418e-05,
|
| 193 |
+
"loss": 0.3634,
|
| 194 |
+
"step": 230
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.75,
|
| 198 |
+
"grad_norm": 2.4452743530273438,
|
| 199 |
+
"learning_rate": 5.7855685414037173e-05,
|
| 200 |
+
"loss": 0.5039,
|
| 201 |
+
"step": 240
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.78,
|
| 205 |
+
"grad_norm": 6.689948558807373,
|
| 206 |
+
"learning_rate": 5.6796183424706484e-05,
|
| 207 |
+
"loss": 0.6249,
|
| 208 |
+
"step": 250
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.81,
|
| 212 |
+
"grad_norm": 5.241004467010498,
|
| 213 |
+
"learning_rate": 5.568460463897046e-05,
|
| 214 |
+
"loss": 0.4695,
|
| 215 |
+
"step": 260
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.84,
|
| 219 |
+
"grad_norm": 3.0335886478424072,
|
| 220 |
+
"learning_rate": 5.46358364708153e-05,
|
| 221 |
+
"loss": 0.4405,
|
| 222 |
+
"step": 270
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.87,
|
| 226 |
+
"grad_norm": 4.185052394866943,
|
| 227 |
+
"learning_rate": 5.364417807858201e-05,
|
| 228 |
+
"loss": 0.4555,
|
| 229 |
+
"step": 280
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.9,
|
| 233 |
+
"grad_norm": 7.301243305206299,
|
| 234 |
+
"learning_rate": 5.270462766947299e-05,
|
| 235 |
+
"loss": 0.5954,
|
| 236 |
+
"step": 290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.93,
|
| 240 |
+
"grad_norm": 5.7551798820495605,
|
| 241 |
+
"learning_rate": 5.181277601508398e-05,
|
| 242 |
+
"loss": 0.6132,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.93,
|
| 247 |
+
"eval_accuracy": 0.8144937586685159,
|
| 248 |
+
"eval_f1": 0.813842282595786,
|
| 249 |
+
"eval_loss": 0.5131940841674805,
|
| 250 |
+
"eval_precision": 0.8261935513216945,
|
| 251 |
+
"eval_recall": 0.8144937586685159,
|
| 252 |
+
"eval_runtime": 35.4672,
|
| 253 |
+
"eval_samples_per_second": 81.314,
|
| 254 |
+
"eval_steps_per_second": 10.178,
|
| 255 |
+
"step": 300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.97,
|
| 259 |
+
"grad_norm": 3.989379405975342,
|
| 260 |
+
"learning_rate": 5.0964719143762554e-05,
|
| 261 |
+
"loss": 0.4919,
|
| 262 |
+
"step": 310
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.0,
|
| 266 |
+
"grad_norm": 6.611064910888672,
|
| 267 |
+
"learning_rate": 5.015698625755192e-05,
|
| 268 |
+
"loss": 0.4443,
|
| 269 |
+
"step": 320
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.03,
|
| 273 |
+
"grad_norm": 3.822683334350586,
|
| 274 |
+
"learning_rate": 4.9386479832479486e-05,
|
| 275 |
+
"loss": 0.4072,
|
| 276 |
+
"step": 330
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.06,
|
| 280 |
+
"grad_norm": 3.768205165863037,
|
| 281 |
+
"learning_rate": 4.865042554105199e-05,
|
| 282 |
+
"loss": 0.2584,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.09,
|
| 287 |
+
"grad_norm": 1.0489230155944824,
|
| 288 |
+
"learning_rate": 4.794633014853842e-05,
|
| 289 |
+
"loss": 0.3173,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.12,
|
| 294 |
+
"grad_norm": 3.1184098720550537,
|
| 295 |
+
"learning_rate": 4.727194592470655e-05,
|
| 296 |
+
"loss": 0.2359,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.15,
|
| 301 |
+
"grad_norm": 2.9331960678100586,
|
| 302 |
+
"learning_rate": 4.662524041201569e-05,
|
| 303 |
+
"loss": 0.2935,
|
| 304 |
+
"step": 370
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.18,
|
| 308 |
+
"grad_norm": 10.099534034729004,
|
| 309 |
+
"learning_rate": 4.600437062282362e-05,
|
| 310 |
+
"loss": 0.2998,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.21,
|
| 315 |
+
"grad_norm": 4.351399898529053,
|
| 316 |
+
"learning_rate": 4.540766091864998e-05,
|
| 317 |
+
"loss": 0.4389,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.25,
|
| 322 |
+
"grad_norm": 2.582808256149292,
|
| 323 |
+
"learning_rate": 4.4833583966222034e-05,
|
| 324 |
+
"loss": 0.3319,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.25,
|
| 329 |
+
"eval_accuracy": 0.8307905686546463,
|
| 330 |
+
"eval_f1": 0.829309725170082,
|
| 331 |
+
"eval_loss": 0.47056975960731506,
|
| 332 |
+
"eval_precision": 0.8327103233714703,
|
| 333 |
+
"eval_recall": 0.8307905686546463,
|
| 334 |
+
"eval_runtime": 36.3423,
|
| 335 |
+
"eval_samples_per_second": 79.357,
|
| 336 |
+
"eval_steps_per_second": 9.933,
|
| 337 |
+
"step": 400
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.28,
|
| 341 |
+
"grad_norm": 5.291513919830322,
|
| 342 |
+
"learning_rate": 4.428074427700477e-05,
|
| 343 |
+
"loss": 0.2839,
|
| 344 |
+
"step": 410
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.31,
|
| 348 |
+
"grad_norm": 5.357462406158447,
|
| 349 |
+
"learning_rate": 4.3747863925980715e-05,
|
| 350 |
+
"loss": 0.3073,
|
| 351 |
+
"step": 420
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.34,
|
| 355 |
+
"grad_norm": 4.056901454925537,
|
| 356 |
+
"learning_rate": 4.32337701167117e-05,
|
| 357 |
+
"loss": 0.3814,
|
| 358 |
+
"step": 430
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.37,
|
| 362 |
+
"grad_norm": 4.224523544311523,
|
| 363 |
+
"learning_rate": 4.273738431706883e-05,
|
| 364 |
+
"loss": 0.2582,
|
| 365 |
+
"step": 440
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.4,
|
| 369 |
+
"grad_norm": 5.398918628692627,
|
| 370 |
+
"learning_rate": 4.225771273642583e-05,
|
| 371 |
+
"loss": 0.3279,
|
| 372 |
+
"step": 450
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.43,
|
| 376 |
+
"grad_norm": 4.845595359802246,
|
| 377 |
+
"learning_rate": 4.179383795285729e-05,
|
| 378 |
+
"loss": 0.2912,
|
| 379 |
+
"step": 460
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.46,
|
| 383 |
+
"grad_norm": 3.754013776779175,
|
| 384 |
+
"learning_rate": 4.1344911529736155e-05,
|
| 385 |
+
"loss": 0.224,
|
| 386 |
+
"step": 470
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.5,
|
| 390 |
+
"grad_norm": 9.241665840148926,
|
| 391 |
+
"learning_rate": 4.0910147486461317e-05,
|
| 392 |
+
"loss": 0.3102,
|
| 393 |
+
"step": 480
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.53,
|
| 397 |
+
"grad_norm": 6.650195598602295,
|
| 398 |
+
"learning_rate": 4.0488816508945806e-05,
|
| 399 |
+
"loss": 0.3613,
|
| 400 |
+
"step": 490
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 1.56,
|
| 404 |
+
"grad_norm": 2.6371705532073975,
|
| 405 |
+
"learning_rate": 4.008024080281012e-05,
|
| 406 |
+
"loss": 0.2286,
|
| 407 |
+
"step": 500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.56,
|
| 411 |
+
"eval_accuracy": 0.8352981969486823,
|
| 412 |
+
"eval_f1": 0.8225709280993477,
|
| 413 |
+
"eval_loss": 0.4952014982700348,
|
| 414 |
+
"eval_precision": 0.8446619811727021,
|
| 415 |
+
"eval_recall": 0.8352981969486823,
|
| 416 |
+
"eval_runtime": 37.026,
|
| 417 |
+
"eval_samples_per_second": 77.891,
|
| 418 |
+
"eval_steps_per_second": 9.75,
|
| 419 |
+
"step": 500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.59,
|
| 423 |
+
"grad_norm": 2.9743077754974365,
|
| 424 |
+
"learning_rate": 3.9683789506627256e-05,
|
| 425 |
+
"loss": 0.3705,
|
| 426 |
+
"step": 510
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.62,
|
| 430 |
+
"grad_norm": 3.7261903285980225,
|
| 431 |
+
"learning_rate": 3.929887459459297e-05,
|
| 432 |
+
"loss": 0.305,
|
| 433 |
+
"step": 520
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.65,
|
| 437 |
+
"grad_norm": 6.446900367736816,
|
| 438 |
+
"learning_rate": 3.892494720807615e-05,
|
| 439 |
+
"loss": 0.278,
|
| 440 |
+
"step": 530
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.68,
|
| 444 |
+
"grad_norm": 2.1284918785095215,
|
| 445 |
+
"learning_rate": 3.856149436398495e-05,
|
| 446 |
+
"loss": 0.2053,
|
| 447 |
+
"step": 540
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.71,
|
| 451 |
+
"grad_norm": 4.4590277671813965,
|
| 452 |
+
"learning_rate": 3.8208035995043505e-05,
|
| 453 |
+
"loss": 0.3642,
|
| 454 |
+
"step": 550
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 1.74,
|
| 458 |
+
"grad_norm": 4.997034549713135,
|
| 459 |
+
"learning_rate": 3.786412228313765e-05,
|
| 460 |
+
"loss": 0.3187,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.78,
|
| 465 |
+
"grad_norm": 2.7240359783172607,
|
| 466 |
+
"learning_rate": 3.752933125204008e-05,
|
| 467 |
+
"loss": 0.2594,
|
| 468 |
+
"step": 570
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 1.81,
|
| 472 |
+
"grad_norm": 4.143340110778809,
|
| 473 |
+
"learning_rate": 3.720326659021623e-05,
|
| 474 |
+
"loss": 0.2151,
|
| 475 |
+
"step": 580
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.84,
|
| 479 |
+
"grad_norm": 5.146928787231445,
|
| 480 |
+
"learning_rate": 3.688555567816587e-05,
|
| 481 |
+
"loss": 0.3299,
|
| 482 |
+
"step": 590
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.87,
|
| 486 |
+
"grad_norm": 2.214599132537842,
|
| 487 |
+
"learning_rate": 3.6575847797972757e-05,
|
| 488 |
+
"loss": 0.2299,
|
| 489 |
+
"step": 600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.87,
|
| 493 |
+
"eval_accuracy": 0.8366851595006934,
|
| 494 |
+
"eval_f1": 0.8357987474414161,
|
| 495 |
+
"eval_loss": 0.46962958574295044,
|
| 496 |
+
"eval_precision": 0.8516509479787296,
|
| 497 |
+
"eval_recall": 0.8366851595006934,
|
| 498 |
+
"eval_runtime": 35.6912,
|
| 499 |
+
"eval_samples_per_second": 80.804,
|
| 500 |
+
"eval_steps_per_second": 10.115,
|
| 501 |
+
"step": 600
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.9,
|
| 505 |
+
"grad_norm": 3.345350980758667,
|
| 506 |
+
"learning_rate": 3.627381250550059e-05,
|
| 507 |
+
"loss": 0.273,
|
| 508 |
+
"step": 610
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.93,
|
| 512 |
+
"grad_norm": 5.383599758148193,
|
| 513 |
+
"learning_rate": 3.597913814805773e-05,
|
| 514 |
+
"loss": 0.2903,
|
| 515 |
+
"step": 620
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.96,
|
| 519 |
+
"grad_norm": 3.014591932296753,
|
| 520 |
+
"learning_rate": 3.5691530512412484e-05,
|
| 521 |
+
"loss": 0.2079,
|
| 522 |
+
"step": 630
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.99,
|
| 526 |
+
"grad_norm": 3.041154146194458,
|
| 527 |
+
"learning_rate": 3.541071158982556e-05,
|
| 528 |
+
"loss": 0.2614,
|
| 529 |
+
"step": 640
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.02,
|
| 533 |
+
"grad_norm": 3.956765651702881,
|
| 534 |
+
"learning_rate": 3.513641844631533e-05,
|
| 535 |
+
"loss": 0.1976,
|
| 536 |
+
"step": 650
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.06,
|
| 540 |
+
"grad_norm": 4.5174736976623535,
|
| 541 |
+
"learning_rate": 3.4868402187720335e-05,
|
| 542 |
+
"loss": 0.1183,
|
| 543 |
+
"step": 660
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.09,
|
| 547 |
+
"grad_norm": 1.5246050357818604,
|
| 548 |
+
"learning_rate": 3.460642701029914e-05,
|
| 549 |
+
"loss": 0.1483,
|
| 550 |
+
"step": 670
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.12,
|
| 554 |
+
"grad_norm": 4.209073066711426,
|
| 555 |
+
"learning_rate": 3.435026932863631e-05,
|
| 556 |
+
"loss": 0.1136,
|
| 557 |
+
"step": 680
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.15,
|
| 561 |
+
"grad_norm": 3.5777525901794434,
|
| 562 |
+
"learning_rate": 3.4099716973523676e-05,
|
| 563 |
+
"loss": 0.0993,
|
| 564 |
+
"step": 690
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.18,
|
| 568 |
+
"grad_norm": 4.207378387451172,
|
| 569 |
+
"learning_rate": 3.385456845327663e-05,
|
| 570 |
+
"loss": 0.0542,
|
| 571 |
+
"step": 700
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.18,
|
| 575 |
+
"eval_accuracy": 0.8498613037447988,
|
| 576 |
+
"eval_f1": 0.8515512179522667,
|
| 577 |
+
"eval_loss": 0.44692692160606384,
|
| 578 |
+
"eval_precision": 0.856522763832034,
|
| 579 |
+
"eval_recall": 0.8498613037447988,
|
| 580 |
+
"eval_runtime": 35.9166,
|
| 581 |
+
"eval_samples_per_second": 80.297,
|
| 582 |
+
"eval_steps_per_second": 10.051,
|
| 583 |
+
"step": 700
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.21,
|
| 587 |
+
"grad_norm": 2.78676438331604,
|
| 588 |
+
"learning_rate": 3.361463227264072e-05,
|
| 589 |
+
"loss": 0.1083,
|
| 590 |
+
"step": 710
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.24,
|
| 594 |
+
"grad_norm": 3.2999024391174316,
|
| 595 |
+
"learning_rate": 3.337972630405625e-05,
|
| 596 |
+
"loss": 0.1755,
|
| 597 |
+
"step": 720
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.27,
|
| 601 |
+
"grad_norm": 4.462038040161133,
|
| 602 |
+
"learning_rate": 3.3149677206589793e-05,
|
| 603 |
+
"loss": 0.1548,
|
| 604 |
+
"step": 730
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.31,
|
| 608 |
+
"grad_norm": 2.0917258262634277,
|
| 609 |
+
"learning_rate": 3.2924319888319655e-05,
|
| 610 |
+
"loss": 0.0877,
|
| 611 |
+
"step": 740
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.34,
|
| 615 |
+
"grad_norm": 4.596163272857666,
|
| 616 |
+
"learning_rate": 3.2703497008386434e-05,
|
| 617 |
+
"loss": 0.1115,
|
| 618 |
+
"step": 750
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.37,
|
| 622 |
+
"grad_norm": 0.25166478753089905,
|
| 623 |
+
"learning_rate": 3.24870585152958e-05,
|
| 624 |
+
"loss": 0.1126,
|
| 625 |
+
"step": 760
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.4,
|
| 629 |
+
"grad_norm": 1.1833806037902832,
|
| 630 |
+
"learning_rate": 3.2274861218395145e-05,
|
| 631 |
+
"loss": 0.2144,
|
| 632 |
+
"step": 770
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.43,
|
| 636 |
+
"grad_norm": 3.9837536811828613,
|
| 637 |
+
"learning_rate": 3.206676838974329e-05,
|
| 638 |
+
"loss": 0.1465,
|
| 639 |
+
"step": 780
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.46,
|
| 643 |
+
"grad_norm": 0.4924279451370239,
|
| 644 |
+
"learning_rate": 3.1862649393858316e-05,
|
| 645 |
+
"loss": 0.1795,
|
| 646 |
+
"step": 790
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.49,
|
| 650 |
+
"grad_norm": 7.527681350708008,
|
| 651 |
+
"learning_rate": 3.166237934306518e-05,
|
| 652 |
+
"loss": 0.198,
|
| 653 |
+
"step": 800
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.49,
|
| 657 |
+
"eval_accuracy": 0.8224687933425797,
|
| 658 |
+
"eval_f1": 0.8322706898840134,
|
| 659 |
+
"eval_loss": 0.5284662246704102,
|
| 660 |
+
"eval_precision": 0.8615898706530778,
|
| 661 |
+
"eval_recall": 0.8224687933425797,
|
| 662 |
+
"eval_runtime": 36.8194,
|
| 663 |
+
"eval_samples_per_second": 78.328,
|
| 664 |
+
"eval_steps_per_second": 9.805,
|
| 665 |
+
"step": 800
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 2.52,
|
| 669 |
+
"grad_norm": 2.07871150970459,
|
| 670 |
+
"learning_rate": 3.146583877637763e-05,
|
| 671 |
+
"loss": 0.1241,
|
| 672 |
+
"step": 810
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.55,
|
| 676 |
+
"grad_norm": 8.270013809204102,
|
| 677 |
+
"learning_rate": 3.127291336003811e-05,
|
| 678 |
+
"loss": 0.1626,
|
| 679 |
+
"step": 820
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 2.59,
|
| 683 |
+
"grad_norm": 4.030840873718262,
|
| 684 |
+
"learning_rate": 3.1083493608010464e-05,
|
| 685 |
+
"loss": 0.0771,
|
| 686 |
+
"step": 830
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.62,
|
| 690 |
+
"grad_norm": 3.374802589416504,
|
| 691 |
+
"learning_rate": 3.0897474620873045e-05,
|
| 692 |
+
"loss": 0.0761,
|
| 693 |
+
"step": 840
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 2.65,
|
| 697 |
+
"grad_norm": 0.3307746946811676,
|
| 698 |
+
"learning_rate": 3.0714755841697564e-05,
|
| 699 |
+
"loss": 0.0477,
|
| 700 |
+
"step": 850
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 2.68,
|
| 704 |
+
"grad_norm": 5.573653221130371,
|
| 705 |
+
"learning_rate": 3.0535240827622965e-05,
|
| 706 |
+
"loss": 0.0799,
|
| 707 |
+
"step": 860
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 2.71,
|
| 711 |
+
"grad_norm": 1.2958701848983765,
|
| 712 |
+
"learning_rate": 3.035883703594582e-05,
|
| 713 |
+
"loss": 0.0984,
|
| 714 |
+
"step": 870
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.74,
|
| 718 |
+
"grad_norm": 3.165799379348755,
|
| 719 |
+
"learning_rate": 3.0185455623649106e-05,
|
| 720 |
+
"loss": 0.1073,
|
| 721 |
+
"step": 880
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.77,
|
| 725 |
+
"grad_norm": 0.7759466767311096,
|
| 726 |
+
"learning_rate": 3.0015011259383213e-05,
|
| 727 |
+
"loss": 0.0734,
|
| 728 |
+
"step": 890
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.8,
|
| 732 |
+
"grad_norm": 1.0856852531433105,
|
| 733 |
+
"learning_rate": 2.9847421946995018e-05,
|
| 734 |
+
"loss": 0.0311,
|
| 735 |
+
"step": 900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 2.8,
|
| 739 |
+
"eval_accuracy": 0.8651178918169209,
|
| 740 |
+
"eval_f1": 0.8661898070442219,
|
| 741 |
+
"eval_loss": 0.4723583161830902,
|
| 742 |
+
"eval_precision": 0.8686654723138755,
|
| 743 |
+
"eval_recall": 0.8651178918169209,
|
| 744 |
+
"eval_runtime": 36.0484,
|
| 745 |
+
"eval_samples_per_second": 80.004,
|
| 746 |
+
"eval_steps_per_second": 10.014,
|
| 747 |
+
"step": 900
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 2.83,
|
| 751 |
+
"grad_norm": 0.283659428358078,
|
| 752 |
+
"learning_rate": 2.968260885977624e-05,
|
| 753 |
+
"loss": 0.0772,
|
| 754 |
+
"step": 910
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 2.87,
|
| 758 |
+
"grad_norm": 6.567378520965576,
|
| 759 |
+
"learning_rate": 2.9520496184669844e-05,
|
| 760 |
+
"loss": 0.1693,
|
| 761 |
+
"step": 920
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.9,
|
| 765 |
+
"grad_norm": 0.2684701681137085,
|
| 766 |
+
"learning_rate": 2.9361010975735175e-05,
|
| 767 |
+
"loss": 0.2045,
|
| 768 |
+
"step": 930
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.93,
|
| 772 |
+
"grad_norm": 3.3332574367523193,
|
| 773 |
+
"learning_rate": 2.9204083016228457e-05,
|
| 774 |
+
"loss": 0.0739,
|
| 775 |
+
"step": 940
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 2.96,
|
| 779 |
+
"grad_norm": 1.2437993288040161,
|
| 780 |
+
"learning_rate": 2.904964468870634e-05,
|
| 781 |
+
"loss": 0.0674,
|
| 782 |
+
"step": 950
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 2.99,
|
| 786 |
+
"grad_norm": 12.299838066101074,
|
| 787 |
+
"learning_rate": 2.8897630852606727e-05,
|
| 788 |
+
"loss": 0.1677,
|
| 789 |
+
"step": 960
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.02,
|
| 793 |
+
"grad_norm": 5.6871442794799805,
|
| 794 |
+
"learning_rate": 2.8747978728803455e-05,
|
| 795 |
+
"loss": 0.0329,
|
| 796 |
+
"step": 970
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 3.05,
|
| 800 |
+
"grad_norm": 0.1499292105436325,
|
| 801 |
+
"learning_rate": 2.8600627790670087e-05,
|
| 802 |
+
"loss": 0.019,
|
| 803 |
+
"step": 980
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 3.08,
|
| 807 |
+
"grad_norm": 0.24846801161766052,
|
| 808 |
+
"learning_rate": 2.8455519661223613e-05,
|
| 809 |
+
"loss": 0.0075,
|
| 810 |
+
"step": 990
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.12,
|
| 814 |
+
"grad_norm": 0.13459719717502594,
|
| 815 |
+
"learning_rate": 2.8312598015950882e-05,
|
| 816 |
+
"loss": 0.0543,
|
| 817 |
+
"step": 1000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.12,
|
| 821 |
+
"eval_accuracy": 0.866504854368932,
|
| 822 |
+
"eval_f1": 0.8610792630184371,
|
| 823 |
+
"eval_loss": 0.4949225187301636,
|
| 824 |
+
"eval_precision": 0.8612246115926523,
|
| 825 |
+
"eval_recall": 0.866504854368932,
|
| 826 |
+
"eval_runtime": 36.1317,
|
| 827 |
+
"eval_samples_per_second": 79.819,
|
| 828 |
+
"eval_steps_per_second": 9.991,
|
| 829 |
+
"step": 1000
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 3.15,
|
| 833 |
+
"grad_norm": 5.778144836425781,
|
| 834 |
+
"learning_rate": 2.817180849095055e-05,
|
| 835 |
+
"loss": 0.0448,
|
| 836 |
+
"step": 1010
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 3.18,
|
| 840 |
+
"grad_norm": 0.5647971630096436,
|
| 841 |
+
"learning_rate": 2.803309859605025e-05,
|
| 842 |
+
"loss": 0.0279,
|
| 843 |
+
"step": 1020
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 3.21,
|
| 847 |
+
"grad_norm": 0.3868822157382965,
|
| 848 |
+
"learning_rate": 2.7896417632583534e-05,
|
| 849 |
+
"loss": 0.0175,
|
| 850 |
+
"step": 1030
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 3.24,
|
| 854 |
+
"grad_norm": 0.2270067036151886,
|
| 855 |
+
"learning_rate": 2.77617166155343e-05,
|
| 856 |
+
"loss": 0.0073,
|
| 857 |
+
"step": 1040
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.27,
|
| 861 |
+
"grad_norm": 0.43895870447158813,
|
| 862 |
+
"learning_rate": 2.762894819977688e-05,
|
| 863 |
+
"loss": 0.0243,
|
| 864 |
+
"step": 1050
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 3.3,
|
| 868 |
+
"grad_norm": 3.83394455909729,
|
| 869 |
+
"learning_rate": 2.749806661015982e-05,
|
| 870 |
+
"loss": 0.0217,
|
| 871 |
+
"step": 1060
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 3.33,
|
| 875 |
+
"grad_norm": 0.19903437793254852,
|
| 876 |
+
"learning_rate": 2.736902757519867e-05,
|
| 877 |
+
"loss": 0.0235,
|
| 878 |
+
"step": 1070
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 3.36,
|
| 882 |
+
"grad_norm": 0.2069879025220871,
|
| 883 |
+
"learning_rate": 2.724178826415978e-05,
|
| 884 |
+
"loss": 0.0398,
|
| 885 |
+
"step": 1080
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.4,
|
| 889 |
+
"grad_norm": 9.80591106414795,
|
| 890 |
+
"learning_rate": 2.711630722733202e-05,
|
| 891 |
+
"loss": 0.0649,
|
| 892 |
+
"step": 1090
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 3.43,
|
| 896 |
+
"grad_norm": 0.3856530487537384,
|
| 897 |
+
"learning_rate": 2.69925443392972e-05,
|
| 898 |
+
"loss": 0.0242,
|
| 899 |
+
"step": 1100
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 3.43,
|
| 903 |
+
"eval_accuracy": 0.8623439667128987,
|
| 904 |
+
"eval_f1": 0.8510243678619469,
|
| 905 |
+
"eval_loss": 0.6283301115036011,
|
| 906 |
+
"eval_precision": 0.8661193391176473,
|
| 907 |
+
"eval_recall": 0.8623439667128987,
|
| 908 |
+
"eval_runtime": 36.1111,
|
| 909 |
+
"eval_samples_per_second": 79.865,
|
| 910 |
+
"eval_steps_per_second": 9.997,
|
| 911 |
+
"step": 1100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 3.46,
|
| 915 |
+
"grad_norm": 0.18870727717876434,
|
| 916 |
+
"learning_rate": 2.687046074502295e-05,
|
| 917 |
+
"loss": 0.0271,
|
| 918 |
+
"step": 1110
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 3.49,
|
| 922 |
+
"grad_norm": 1.5470973253250122,
|
| 923 |
+
"learning_rate": 2.675001880861359e-05,
|
| 924 |
+
"loss": 0.0398,
|
| 925 |
+
"step": 1120
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 3.52,
|
| 929 |
+
"grad_norm": 0.26312679052352905,
|
| 930 |
+
"learning_rate": 2.6631182064565375e-05,
|
| 931 |
+
"loss": 0.0141,
|
| 932 |
+
"step": 1130
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 3.55,
|
| 936 |
+
"grad_norm": 0.15462630987167358,
|
| 937 |
+
"learning_rate": 2.6513915171382936e-05,
|
| 938 |
+
"loss": 0.0761,
|
| 939 |
+
"step": 1140
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 3.58,
|
| 943 |
+
"grad_norm": 0.10996732115745544,
|
| 944 |
+
"learning_rate": 2.6398183867422732e-05,
|
| 945 |
+
"loss": 0.0048,
|
| 946 |
+
"step": 1150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 3.61,
|
| 950 |
+
"grad_norm": 0.38109445571899414,
|
| 951 |
+
"learning_rate": 2.6283954928838412e-05,
|
| 952 |
+
"loss": 0.0243,
|
| 953 |
+
"step": 1160
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 3.64,
|
| 957 |
+
"grad_norm": 0.5097649097442627,
|
| 958 |
+
"learning_rate": 2.6171196129510684e-05,
|
| 959 |
+
"loss": 0.0362,
|
| 960 |
+
"step": 1170
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 3.68,
|
| 964 |
+
"grad_norm": 0.05090225115418434,
|
| 965 |
+
"learning_rate": 2.605987620285215e-05,
|
| 966 |
+
"loss": 0.0261,
|
| 967 |
+
"step": 1180
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 3.71,
|
| 971 |
+
"grad_norm": 0.6788052320480347,
|
| 972 |
+
"learning_rate": 2.5949964805384102e-05,
|
| 973 |
+
"loss": 0.0045,
|
| 974 |
+
"step": 1190
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 3.74,
|
| 978 |
+
"grad_norm": 0.026266297325491905,
|
| 979 |
+
"learning_rate": 2.5841432481989113e-05,
|
| 980 |
+
"loss": 0.0179,
|
| 981 |
+
"step": 1200
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 3.74,
|
| 985 |
+
"eval_accuracy": 0.8723994452149791,
|
| 986 |
+
"eval_f1": 0.8674518149970984,
|
| 987 |
+
"eval_loss": 0.5766238570213318,
|
| 988 |
+
"eval_precision": 0.8681300166641136,
|
| 989 |
+
"eval_recall": 0.8723994452149791,
|
| 990 |
+
"eval_runtime": 36.8086,
|
| 991 |
+
"eval_samples_per_second": 78.351,
|
| 992 |
+
"eval_steps_per_second": 9.807,
|
| 993 |
+
"step": 1200
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 3.77,
|
| 997 |
+
"grad_norm": 4.131288528442383,
|
| 998 |
+
"learning_rate": 2.573425063274894e-05,
|
| 999 |
+
"loss": 0.0311,
|
| 1000 |
+
"step": 1210
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 3.8,
|
| 1004 |
+
"grad_norm": 0.09610464423894882,
|
| 1005 |
+
"learning_rate": 2.5628391481282988e-05,
|
| 1006 |
+
"loss": 0.0247,
|
| 1007 |
+
"step": 1220
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 3.83,
|
| 1011 |
+
"grad_norm": 0.39180612564086914,
|
| 1012 |
+
"learning_rate": 2.5523828044507798e-05,
|
| 1013 |
+
"loss": 0.0185,
|
| 1014 |
+
"step": 1230
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 3.86,
|
| 1018 |
+
"grad_norm": 0.017242716625332832,
|
| 1019 |
+
"learning_rate": 2.5420534103742737e-05,
|
| 1020 |
+
"loss": 0.0258,
|
| 1021 |
+
"step": 1240
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 3.89,
|
| 1025 |
+
"grad_norm": 4.467873573303223,
|
| 1026 |
+
"learning_rate": 2.5318484177091666e-05,
|
| 1027 |
+
"loss": 0.0137,
|
| 1028 |
+
"step": 1250
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 3.93,
|
| 1032 |
+
"grad_norm": 0.21439874172210693,
|
| 1033 |
+
"learning_rate": 2.5217653493034472e-05,
|
| 1034 |
+
"loss": 0.032,
|
| 1035 |
+
"step": 1260
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 3.96,
|
| 1039 |
+
"grad_norm": 0.06672481447458267,
|
| 1040 |
+
"learning_rate": 2.511801796516642e-05,
|
| 1041 |
+
"loss": 0.0061,
|
| 1042 |
+
"step": 1270
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 3.99,
|
| 1046 |
+
"grad_norm": 0.050495993345975876,
|
| 1047 |
+
"learning_rate": 2.501955416802672e-05,
|
| 1048 |
+
"loss": 0.0194,
|
| 1049 |
+
"step": 1280
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 4.02,
|
| 1053 |
+
"grad_norm": 0.430034875869751,
|
| 1054 |
+
"learning_rate": 2.492223931396134e-05,
|
| 1055 |
+
"loss": 0.0138,
|
| 1056 |
+
"step": 1290
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 4.05,
|
| 1060 |
+
"grad_norm": 0.020715204998850822,
|
| 1061 |
+
"learning_rate": 2.482605123096805e-05,
|
| 1062 |
+
"loss": 0.01,
|
| 1063 |
+
"step": 1300
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 4.05,
|
| 1067 |
+
"eval_accuracy": 0.8595700416088765,
|
| 1068 |
+
"eval_f1": 0.8534690367784884,
|
| 1069 |
+
"eval_loss": 0.6232466697692871,
|
| 1070 |
+
"eval_precision": 0.8523064665048566,
|
| 1071 |
+
"eval_recall": 0.8595700416088765,
|
| 1072 |
+
"eval_runtime": 36.0575,
|
| 1073 |
+
"eval_samples_per_second": 79.983,
|
| 1074 |
+
"eval_steps_per_second": 10.012,
|
| 1075 |
+
"step": 1300
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 4.08,
|
| 1079 |
+
"grad_norm": 0.04595523327589035,
|
| 1080 |
+
"learning_rate": 2.47309683414749e-05,
|
| 1081 |
+
"loss": 0.0065,
|
| 1082 |
+
"step": 1310
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 4.11,
|
| 1086 |
+
"grad_norm": 0.03214849531650543,
|
| 1087 |
+
"learning_rate": 2.4636969642005952e-05,
|
| 1088 |
+
"loss": 0.0019,
|
| 1089 |
+
"step": 1320
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 4.14,
|
| 1093 |
+
"grad_norm": 0.10033855587244034,
|
| 1094 |
+
"learning_rate": 2.45440346836908e-05,
|
| 1095 |
+
"loss": 0.0039,
|
| 1096 |
+
"step": 1330
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 4.17,
|
| 1100 |
+
"grad_norm": 2.8943097591400146,
|
| 1101 |
+
"learning_rate": 2.4452143553576716e-05,
|
| 1102 |
+
"loss": 0.0023,
|
| 1103 |
+
"step": 1340
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 4.21,
|
| 1107 |
+
"grad_norm": 0.9686470627784729,
|
| 1108 |
+
"learning_rate": 2.4361276856704794e-05,
|
| 1109 |
+
"loss": 0.0037,
|
| 1110 |
+
"step": 1350
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 4.24,
|
| 1114 |
+
"grad_norm": 0.03016245923936367,
|
| 1115 |
+
"learning_rate": 2.4271415698913302e-05,
|
| 1116 |
+
"loss": 0.0018,
|
| 1117 |
+
"step": 1360
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 4.27,
|
| 1121 |
+
"grad_norm": 0.10285181552171707,
|
| 1122 |
+
"learning_rate": 2.4182541670333722e-05,
|
| 1123 |
+
"loss": 0.0017,
|
| 1124 |
+
"step": 1370
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 4.3,
|
| 1128 |
+
"grad_norm": 0.1811099499464035,
|
| 1129 |
+
"learning_rate": 2.4094636829546745e-05,
|
| 1130 |
+
"loss": 0.0011,
|
| 1131 |
+
"step": 1380
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 4.33,
|
| 1135 |
+
"grad_norm": 0.03486869856715202,
|
| 1136 |
+
"learning_rate": 2.4007683688367184e-05,
|
| 1137 |
+
"loss": 0.0013,
|
| 1138 |
+
"step": 1390
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 4.36,
|
| 1142 |
+
"grad_norm": 0.019455306231975555,
|
| 1143 |
+
"learning_rate": 2.3921665197228592e-05,
|
| 1144 |
+
"loss": 0.0018,
|
| 1145 |
+
"step": 1400
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 4.36,
|
| 1149 |
+
"eval_accuracy": 0.874133148404993,
|
| 1150 |
+
"eval_f1": 0.8710107373815618,
|
| 1151 |
+
"eval_loss": 0.6012547612190247,
|
| 1152 |
+
"eval_precision": 0.8707403535526532,
|
| 1153 |
+
"eval_recall": 0.874133148404993,
|
| 1154 |
+
"eval_runtime": 35.9997,
|
| 1155 |
+
"eval_samples_per_second": 80.112,
|
| 1156 |
+
"eval_steps_per_second": 10.028,
|
| 1157 |
+
"step": 1400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 4.39,
|
| 1161 |
+
"grad_norm": 3.699671745300293,
|
| 1162 |
+
"learning_rate": 2.383656473113981e-05,
|
| 1163 |
+
"loss": 0.0054,
|
| 1164 |
+
"step": 1410
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 4.42,
|
| 1168 |
+
"grad_norm": 0.07088273018598557,
|
| 1169 |
+
"learning_rate": 2.3752366076187175e-05,
|
| 1170 |
+
"loss": 0.0012,
|
| 1171 |
+
"step": 1420
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 4.45,
|
| 1175 |
+
"grad_norm": 0.006407163105905056,
|
| 1176 |
+
"learning_rate": 2.3669053416557544e-05,
|
| 1177 |
+
"loss": 0.0058,
|
| 1178 |
+
"step": 1430
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 4.49,
|
| 1182 |
+
"grad_norm": 0.24147652089595795,
|
| 1183 |
+
"learning_rate": 2.35866113220585e-05,
|
| 1184 |
+
"loss": 0.0288,
|
| 1185 |
+
"step": 1440
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 4.52,
|
| 1189 |
+
"grad_norm": 0.0284121036529541,
|
| 1190 |
+
"learning_rate": 2.3505024736113422e-05,
|
| 1191 |
+
"loss": 0.0012,
|
| 1192 |
+
"step": 1450
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 4.55,
|
| 1196 |
+
"grad_norm": 0.10920742154121399,
|
| 1197 |
+
"learning_rate": 2.3424278964210216e-05,
|
| 1198 |
+
"loss": 0.0171,
|
| 1199 |
+
"step": 1460
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 4.58,
|
| 1203 |
+
"grad_norm": 0.02022623084485531,
|
| 1204 |
+
"learning_rate": 2.334435966278354e-05,
|
| 1205 |
+
"loss": 0.0013,
|
| 1206 |
+
"step": 1470
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 4.61,
|
| 1210 |
+
"grad_norm": 1.6955662965774536,
|
| 1211 |
+
"learning_rate": 2.3265252828511455e-05,
|
| 1212 |
+
"loss": 0.0043,
|
| 1213 |
+
"step": 1480
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 4.64,
|
| 1217 |
+
"grad_norm": 0.2646436095237732,
|
| 1218 |
+
"learning_rate": 2.3186944788008412e-05,
|
| 1219 |
+
"loss": 0.007,
|
| 1220 |
+
"step": 1490
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 4.67,
|
| 1224 |
+
"grad_norm": 0.02134719491004944,
|
| 1225 |
+
"learning_rate": 2.3109422187897257e-05,
|
| 1226 |
+
"loss": 0.0019,
|
| 1227 |
+
"step": 1500
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 4.67,
|
| 1231 |
+
"eval_accuracy": 0.8682385575589459,
|
| 1232 |
+
"eval_f1": 0.8643329503142037,
|
| 1233 |
+
"eval_loss": 0.6553735136985779,
|
| 1234 |
+
"eval_precision": 0.8688611568946607,
|
| 1235 |
+
"eval_recall": 0.8682385575589459,
|
| 1236 |
+
"eval_runtime": 36.5058,
|
| 1237 |
+
"eval_samples_per_second": 79.001,
|
| 1238 |
+
"eval_steps_per_second": 9.889,
|
| 1239 |
+
"step": 1500
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 4.7,
|
| 1243 |
+
"grad_norm": 0.10854582488536835,
|
| 1244 |
+
"learning_rate": 2.3032671985243938e-05,
|
| 1245 |
+
"loss": 0.0035,
|
| 1246 |
+
"step": 1510
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 4.74,
|
| 1250 |
+
"grad_norm": 0.023413635790348053,
|
| 1251 |
+
"learning_rate": 2.2956681438339396e-05,
|
| 1252 |
+
"loss": 0.0014,
|
| 1253 |
+
"step": 1520
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 4.77,
|
| 1257 |
+
"grad_norm": 0.03499768301844597,
|
| 1258 |
+
"learning_rate": 2.2881438097813777e-05,
|
| 1259 |
+
"loss": 0.001,
|
| 1260 |
+
"step": 1530
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 4.8,
|
| 1264 |
+
"grad_norm": 0.28913193941116333,
|
| 1265 |
+
"learning_rate": 2.2806929798068923e-05,
|
| 1266 |
+
"loss": 0.0016,
|
| 1267 |
+
"step": 1540
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 4.83,
|
| 1271 |
+
"grad_norm": 0.02359519526362419,
|
| 1272 |
+
"learning_rate": 2.273314464901578e-05,
|
| 1273 |
+
"loss": 0.017,
|
| 1274 |
+
"step": 1550
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 4.86,
|
| 1278 |
+
"grad_norm": 0.20512829720973969,
|
| 1279 |
+
"learning_rate": 2.2660071028103958e-05,
|
| 1280 |
+
"loss": 0.0011,
|
| 1281 |
+
"step": 1560
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 4.89,
|
| 1285 |
+
"grad_norm": 0.017939003184437752,
|
| 1286 |
+
"learning_rate": 2.2587697572631283e-05,
|
| 1287 |
+
"loss": 0.0114,
|
| 1288 |
+
"step": 1570
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 4.92,
|
| 1292 |
+
"grad_norm": 0.05149148404598236,
|
| 1293 |
+
"learning_rate": 2.2516013172321875e-05,
|
| 1294 |
+
"loss": 0.0008,
|
| 1295 |
+
"step": 1580
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 4.95,
|
| 1299 |
+
"grad_norm": 0.03117392770946026,
|
| 1300 |
+
"learning_rate": 2.2445006962161678e-05,
|
| 1301 |
+
"loss": 0.001,
|
| 1302 |
+
"step": 1590
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 4.98,
|
| 1306 |
+
"grad_norm": 0.03808211162686348,
|
| 1307 |
+
"learning_rate": 2.2374668315480894e-05,
|
| 1308 |
+
"loss": 0.0024,
|
| 1309 |
+
"step": 1600
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 4.98,
|
| 1313 |
+
"eval_accuracy": 0.8713592233009708,
|
| 1314 |
+
"eval_f1": 0.8718548085138199,
|
| 1315 |
+
"eval_loss": 0.6107261776924133,
|
| 1316 |
+
"eval_precision": 0.8729661517370155,
|
| 1317 |
+
"eval_recall": 0.8713592233009708,
|
| 1318 |
+
"eval_runtime": 35.9356,
|
| 1319 |
+
"eval_samples_per_second": 80.255,
|
| 1320 |
+
"eval_steps_per_second": 10.046,
|
| 1321 |
+
"step": 1600
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 5.02,
|
| 1325 |
+
"grad_norm": 0.016484640538692474,
|
| 1326 |
+
"learning_rate": 2.2304986837273524e-05,
|
| 1327 |
+
"loss": 0.004,
|
| 1328 |
+
"step": 1610
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 5.05,
|
| 1332 |
+
"grad_norm": 0.026364829391241074,
|
| 1333 |
+
"learning_rate": 2.2235952357744237e-05,
|
| 1334 |
+
"loss": 0.001,
|
| 1335 |
+
"step": 1620
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 5.08,
|
| 1339 |
+
"grad_norm": 0.007881557568907738,
|
| 1340 |
+
"learning_rate": 2.2167554926073632e-05,
|
| 1341 |
+
"loss": 0.0005,
|
| 1342 |
+
"step": 1630
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 5.11,
|
| 1346 |
+
"grad_norm": 0.02089390717446804,
|
| 1347 |
+
"learning_rate": 2.2099784804393198e-05,
|
| 1348 |
+
"loss": 0.0005,
|
| 1349 |
+
"step": 1640
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 5.14,
|
| 1353 |
+
"grad_norm": 0.01271623745560646,
|
| 1354 |
+
"learning_rate": 2.2032632461961585e-05,
|
| 1355 |
+
"loss": 0.0008,
|
| 1356 |
+
"step": 1650
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 5.17,
|
| 1360 |
+
"grad_norm": 0.006357505917549133,
|
| 1361 |
+
"learning_rate": 2.196608856953445e-05,
|
| 1362 |
+
"loss": 0.0005,
|
| 1363 |
+
"step": 1660
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 5.2,
|
| 1367 |
+
"grad_norm": 0.0172483678907156,
|
| 1368 |
+
"learning_rate": 2.1900143993920144e-05,
|
| 1369 |
+
"loss": 0.0005,
|
| 1370 |
+
"step": 1670
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 5.23,
|
| 1374 |
+
"grad_norm": 0.009964230470359325,
|
| 1375 |
+
"learning_rate": 2.1834789792714154e-05,
|
| 1376 |
+
"loss": 0.0005,
|
| 1377 |
+
"step": 1680
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 5.26,
|
| 1381 |
+
"grad_norm": 0.017695054411888123,
|
| 1382 |
+
"learning_rate": 2.1770017209205408e-05,
|
| 1383 |
+
"loss": 0.0015,
|
| 1384 |
+
"step": 1690
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 5.3,
|
| 1388 |
+
"grad_norm": 0.058045148849487305,
|
| 1389 |
+
"learning_rate": 2.170581766744771e-05,
|
| 1390 |
+
"loss": 0.0006,
|
| 1391 |
+
"step": 1700
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 5.3,
|
| 1395 |
+
"eval_accuracy": 0.8755201109570042,
|
| 1396 |
+
"eval_f1": 0.8725140598764357,
|
| 1397 |
+
"eval_loss": 0.6352503299713135,
|
| 1398 |
+
"eval_precision": 0.8751361058141811,
|
| 1399 |
+
"eval_recall": 0.8755201109570042,
|
| 1400 |
+
"eval_runtime": 35.8249,
|
| 1401 |
+
"eval_samples_per_second": 80.503,
|
| 1402 |
+
"eval_steps_per_second": 10.077,
|
| 1403 |
+
"step": 1700
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 5.3,
|
| 1407 |
+
"step": 1700,
|
| 1408 |
+
"total_flos": 2.1047767559471923e+18,
|
| 1409 |
+
"train_loss": 0.21456260421779005,
|
| 1410 |
+
"train_runtime": 1227.9557,
|
| 1411 |
+
"train_samples_per_second": 417.605,
|
| 1412 |
+
"train_steps_per_second": 26.141
|
| 1413 |
+
}
|
| 1414 |
+
],
|
| 1415 |
+
"logging_steps": 10,
|
| 1416 |
+
"max_steps": 32100,
|
| 1417 |
+
"num_input_tokens_seen": 0,
|
| 1418 |
+
"num_train_epochs": 100,
|
| 1419 |
+
"save_steps": 100,
|
| 1420 |
+
"total_flos": 2.1047767559471923e+18,
|
| 1421 |
+
"train_batch_size": 16,
|
| 1422 |
+
"trial_name": null,
|
| 1423 |
+
"trial_params": null
|
| 1424 |
+
}
|