🍻 cheers
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar29_00-56-56_5b2ae90c8915/events.out.tfevents.1711677406.5b2ae90c8915.192.3 +3 -0
- train_results.json +8 -0
- trainer_state.json +1530 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: nsugianto/vit-base-lcdoctypev1_session2
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: validation
|
|
@@ -30,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# vit-base-lcdoctypev1_session3
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [nsugianto/vit-base-lcdoctypev1_session2](https://huggingface.co/nsugianto/vit-base-lcdoctypev1_session2) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
- Accuracy: 0.9669
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: nsugianto/vit-base-lcdoctypev1_session2
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: doctype_v1
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: validation
|
|
|
|
| 31 |
|
| 32 |
# vit-base-lcdoctypev1_session3
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [nsugianto/vit-base-lcdoctypev1_session2](https://huggingface.co/nsugianto/vit-base-lcdoctypev1_session2) on the doctype_v1 dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.1050
|
| 37 |
- Accuracy: 0.9669
|
| 38 |
|
| 39 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9669421487603306,
|
| 4 |
+
"eval_loss": 0.10501641035079956,
|
| 5 |
+
"eval_runtime": 8.768,
|
| 6 |
+
"eval_samples_per_second": 13.8,
|
| 7 |
+
"eval_steps_per_second": 1.825,
|
| 8 |
+
"total_flos": 7.39286832673751e+17,
|
| 9 |
+
"train_loss": 0.06743512197087208,
|
| 10 |
+
"train_runtime": 3546.8991,
|
| 11 |
+
"train_samples_per_second": 2.69,
|
| 12 |
+
"train_steps_per_second": 0.169
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9669421487603306,
|
| 4 |
+
"eval_loss": 0.10501641035079956,
|
| 5 |
+
"eval_runtime": 8.768,
|
| 6 |
+
"eval_samples_per_second": 13.8,
|
| 7 |
+
"eval_steps_per_second": 1.825
|
| 8 |
+
}
|
runs/Mar29_00-56-56_5b2ae90c8915/events.out.tfevents.1711677406.5b2ae90c8915.192.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da8b2c7ff7f30718159307c053a596ae10df2fb1033e13652514c3c4b848b77e
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.39286832673751e+17,
|
| 4 |
+
"train_loss": 0.06743512197087208,
|
| 5 |
+
"train_runtime": 3546.8991,
|
| 6 |
+
"train_samples_per_second": 2.69,
|
| 7 |
+
"train_steps_per_second": 0.169
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1530 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.10501641035079956,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-lcdoctypev1_session3/checkpoint-335",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08,
|
| 13 |
+
"eval_accuracy": 0.9090909090909091,
|
| 14 |
+
"eval_loss": 0.3158922493457794,
|
| 15 |
+
"eval_runtime": 9.7781,
|
| 16 |
+
"eval_samples_per_second": 12.375,
|
| 17 |
+
"eval_steps_per_second": 1.636,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.17,
|
| 22 |
+
"grad_norm": 6.890113353729248,
|
| 23 |
+
"learning_rate": 0.00019666666666666666,
|
| 24 |
+
"loss": 0.1798,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.17,
|
| 29 |
+
"eval_accuracy": 0.9338842975206612,
|
| 30 |
+
"eval_loss": 0.22616656124591827,
|
| 31 |
+
"eval_runtime": 8.5759,
|
| 32 |
+
"eval_samples_per_second": 14.109,
|
| 33 |
+
"eval_steps_per_second": 1.866,
|
| 34 |
+
"step": 10
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.25,
|
| 38 |
+
"eval_accuracy": 0.7768595041322314,
|
| 39 |
+
"eval_loss": 0.9910252690315247,
|
| 40 |
+
"eval_runtime": 9.2477,
|
| 41 |
+
"eval_samples_per_second": 13.084,
|
| 42 |
+
"eval_steps_per_second": 1.73,
|
| 43 |
+
"step": 15
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.33,
|
| 47 |
+
"grad_norm": 5.934413909912109,
|
| 48 |
+
"learning_rate": 0.0001936666666666667,
|
| 49 |
+
"loss": 0.3815,
|
| 50 |
+
"step": 20
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.33,
|
| 54 |
+
"eval_accuracy": 0.9008264462809917,
|
| 55 |
+
"eval_loss": 0.3035498857498169,
|
| 56 |
+
"eval_runtime": 8.7721,
|
| 57 |
+
"eval_samples_per_second": 13.794,
|
| 58 |
+
"eval_steps_per_second": 1.824,
|
| 59 |
+
"step": 20
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.42,
|
| 63 |
+
"eval_accuracy": 0.9338842975206612,
|
| 64 |
+
"eval_loss": 0.21773552894592285,
|
| 65 |
+
"eval_runtime": 9.0833,
|
| 66 |
+
"eval_samples_per_second": 13.321,
|
| 67 |
+
"eval_steps_per_second": 1.761,
|
| 68 |
+
"step": 25
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.5,
|
| 72 |
+
"grad_norm": 0.3238607347011566,
|
| 73 |
+
"learning_rate": 0.00019033333333333334,
|
| 74 |
+
"loss": 0.1429,
|
| 75 |
+
"step": 30
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.5,
|
| 79 |
+
"eval_accuracy": 0.8842975206611571,
|
| 80 |
+
"eval_loss": 0.4908874034881592,
|
| 81 |
+
"eval_runtime": 8.456,
|
| 82 |
+
"eval_samples_per_second": 14.309,
|
| 83 |
+
"eval_steps_per_second": 1.892,
|
| 84 |
+
"step": 30
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.58,
|
| 88 |
+
"eval_accuracy": 0.9256198347107438,
|
| 89 |
+
"eval_loss": 0.3096350133419037,
|
| 90 |
+
"eval_runtime": 10.0423,
|
| 91 |
+
"eval_samples_per_second": 12.049,
|
| 92 |
+
"eval_steps_per_second": 1.593,
|
| 93 |
+
"step": 35
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.67,
|
| 97 |
+
"grad_norm": 5.55199670791626,
|
| 98 |
+
"learning_rate": 0.00018700000000000002,
|
| 99 |
+
"loss": 0.2424,
|
| 100 |
+
"step": 40
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.67,
|
| 104 |
+
"eval_accuracy": 0.9173553719008265,
|
| 105 |
+
"eval_loss": 0.32702526450157166,
|
| 106 |
+
"eval_runtime": 8.357,
|
| 107 |
+
"eval_samples_per_second": 14.479,
|
| 108 |
+
"eval_steps_per_second": 1.915,
|
| 109 |
+
"step": 40
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.75,
|
| 113 |
+
"eval_accuracy": 0.9173553719008265,
|
| 114 |
+
"eval_loss": 0.2554916441440582,
|
| 115 |
+
"eval_runtime": 9.2174,
|
| 116 |
+
"eval_samples_per_second": 13.127,
|
| 117 |
+
"eval_steps_per_second": 1.736,
|
| 118 |
+
"step": 45
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.83,
|
| 122 |
+
"grad_norm": 14.463412284851074,
|
| 123 |
+
"learning_rate": 0.00018366666666666667,
|
| 124 |
+
"loss": 0.1172,
|
| 125 |
+
"step": 50
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.83,
|
| 129 |
+
"eval_accuracy": 0.9173553719008265,
|
| 130 |
+
"eval_loss": 0.2309066504240036,
|
| 131 |
+
"eval_runtime": 9.1271,
|
| 132 |
+
"eval_samples_per_second": 13.257,
|
| 133 |
+
"eval_steps_per_second": 1.753,
|
| 134 |
+
"step": 50
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.92,
|
| 138 |
+
"eval_accuracy": 0.9173553719008265,
|
| 139 |
+
"eval_loss": 0.2952496409416199,
|
| 140 |
+
"eval_runtime": 9.6618,
|
| 141 |
+
"eval_samples_per_second": 12.523,
|
| 142 |
+
"eval_steps_per_second": 1.656,
|
| 143 |
+
"step": 55
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.0,
|
| 147 |
+
"grad_norm": 0.9041020274162292,
|
| 148 |
+
"learning_rate": 0.00018033333333333334,
|
| 149 |
+
"loss": 0.1185,
|
| 150 |
+
"step": 60
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.0,
|
| 154 |
+
"eval_accuracy": 0.9173553719008265,
|
| 155 |
+
"eval_loss": 0.295705646276474,
|
| 156 |
+
"eval_runtime": 8.6323,
|
| 157 |
+
"eval_samples_per_second": 14.017,
|
| 158 |
+
"eval_steps_per_second": 1.854,
|
| 159 |
+
"step": 60
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.08,
|
| 163 |
+
"eval_accuracy": 0.8925619834710744,
|
| 164 |
+
"eval_loss": 0.3724129796028137,
|
| 165 |
+
"eval_runtime": 8.5924,
|
| 166 |
+
"eval_samples_per_second": 14.082,
|
| 167 |
+
"eval_steps_per_second": 1.862,
|
| 168 |
+
"step": 65
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.17,
|
| 172 |
+
"grad_norm": 24.987598419189453,
|
| 173 |
+
"learning_rate": 0.00017700000000000002,
|
| 174 |
+
"loss": 0.1594,
|
| 175 |
+
"step": 70
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.17,
|
| 179 |
+
"eval_accuracy": 0.8842975206611571,
|
| 180 |
+
"eval_loss": 0.4216250777244568,
|
| 181 |
+
"eval_runtime": 9.0834,
|
| 182 |
+
"eval_samples_per_second": 13.321,
|
| 183 |
+
"eval_steps_per_second": 1.761,
|
| 184 |
+
"step": 70
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.25,
|
| 188 |
+
"eval_accuracy": 0.9173553719008265,
|
| 189 |
+
"eval_loss": 0.347516268491745,
|
| 190 |
+
"eval_runtime": 8.9447,
|
| 191 |
+
"eval_samples_per_second": 13.528,
|
| 192 |
+
"eval_steps_per_second": 1.789,
|
| 193 |
+
"step": 75
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.33,
|
| 197 |
+
"grad_norm": 0.3783435523509979,
|
| 198 |
+
"learning_rate": 0.00017366666666666667,
|
| 199 |
+
"loss": 0.1231,
|
| 200 |
+
"step": 80
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.33,
|
| 204 |
+
"eval_accuracy": 0.8925619834710744,
|
| 205 |
+
"eval_loss": 0.323406845331192,
|
| 206 |
+
"eval_runtime": 9.122,
|
| 207 |
+
"eval_samples_per_second": 13.265,
|
| 208 |
+
"eval_steps_per_second": 1.754,
|
| 209 |
+
"step": 80
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.42,
|
| 213 |
+
"eval_accuracy": 0.8842975206611571,
|
| 214 |
+
"eval_loss": 0.4309641718864441,
|
| 215 |
+
"eval_runtime": 8.8847,
|
| 216 |
+
"eval_samples_per_second": 13.619,
|
| 217 |
+
"eval_steps_per_second": 1.801,
|
| 218 |
+
"step": 85
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.5,
|
| 222 |
+
"grad_norm": 5.4656147956848145,
|
| 223 |
+
"learning_rate": 0.00017033333333333334,
|
| 224 |
+
"loss": 0.0875,
|
| 225 |
+
"step": 90
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.5,
|
| 229 |
+
"eval_accuracy": 0.9256198347107438,
|
| 230 |
+
"eval_loss": 0.3598105013370514,
|
| 231 |
+
"eval_runtime": 9.086,
|
| 232 |
+
"eval_samples_per_second": 13.317,
|
| 233 |
+
"eval_steps_per_second": 1.761,
|
| 234 |
+
"step": 90
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.58,
|
| 238 |
+
"eval_accuracy": 0.9256198347107438,
|
| 239 |
+
"eval_loss": 0.3038037419319153,
|
| 240 |
+
"eval_runtime": 8.8538,
|
| 241 |
+
"eval_samples_per_second": 13.666,
|
| 242 |
+
"eval_steps_per_second": 1.807,
|
| 243 |
+
"step": 95
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.67,
|
| 247 |
+
"grad_norm": 0.0762423649430275,
|
| 248 |
+
"learning_rate": 0.000167,
|
| 249 |
+
"loss": 0.0897,
|
| 250 |
+
"step": 100
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.67,
|
| 254 |
+
"eval_accuracy": 0.9338842975206612,
|
| 255 |
+
"eval_loss": 0.25987809896469116,
|
| 256 |
+
"eval_runtime": 8.7997,
|
| 257 |
+
"eval_samples_per_second": 13.751,
|
| 258 |
+
"eval_steps_per_second": 1.818,
|
| 259 |
+
"step": 100
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 1.75,
|
| 263 |
+
"eval_accuracy": 0.9586776859504132,
|
| 264 |
+
"eval_loss": 0.1683536171913147,
|
| 265 |
+
"eval_runtime": 8.0037,
|
| 266 |
+
"eval_samples_per_second": 15.118,
|
| 267 |
+
"eval_steps_per_second": 1.999,
|
| 268 |
+
"step": 105
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.83,
|
| 272 |
+
"grad_norm": 1.4948017597198486,
|
| 273 |
+
"learning_rate": 0.00016366666666666667,
|
| 274 |
+
"loss": 0.1797,
|
| 275 |
+
"step": 110
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.83,
|
| 279 |
+
"eval_accuracy": 0.9504132231404959,
|
| 280 |
+
"eval_loss": 0.1412244588136673,
|
| 281 |
+
"eval_runtime": 8.3997,
|
| 282 |
+
"eval_samples_per_second": 14.405,
|
| 283 |
+
"eval_steps_per_second": 1.905,
|
| 284 |
+
"step": 110
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.92,
|
| 288 |
+
"eval_accuracy": 0.9586776859504132,
|
| 289 |
+
"eval_loss": 0.14531350135803223,
|
| 290 |
+
"eval_runtime": 8.6407,
|
| 291 |
+
"eval_samples_per_second": 14.003,
|
| 292 |
+
"eval_steps_per_second": 1.852,
|
| 293 |
+
"step": 115
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.0,
|
| 297 |
+
"grad_norm": 6.408501148223877,
|
| 298 |
+
"learning_rate": 0.00016033333333333335,
|
| 299 |
+
"loss": 0.1178,
|
| 300 |
+
"step": 120
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.0,
|
| 304 |
+
"eval_accuracy": 0.8925619834710744,
|
| 305 |
+
"eval_loss": 0.3830728530883789,
|
| 306 |
+
"eval_runtime": 9.0064,
|
| 307 |
+
"eval_samples_per_second": 13.435,
|
| 308 |
+
"eval_steps_per_second": 1.777,
|
| 309 |
+
"step": 120
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.08,
|
| 313 |
+
"eval_accuracy": 0.9090909090909091,
|
| 314 |
+
"eval_loss": 0.3321413993835449,
|
| 315 |
+
"eval_runtime": 8.8462,
|
| 316 |
+
"eval_samples_per_second": 13.678,
|
| 317 |
+
"eval_steps_per_second": 1.809,
|
| 318 |
+
"step": 125
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 2.17,
|
| 322 |
+
"grad_norm": 0.7101590633392334,
|
| 323 |
+
"learning_rate": 0.00015700000000000002,
|
| 324 |
+
"loss": 0.1969,
|
| 325 |
+
"step": 130
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 2.17,
|
| 329 |
+
"eval_accuracy": 0.9090909090909091,
|
| 330 |
+
"eval_loss": 0.25461918115615845,
|
| 331 |
+
"eval_runtime": 8.969,
|
| 332 |
+
"eval_samples_per_second": 13.491,
|
| 333 |
+
"eval_steps_per_second": 1.784,
|
| 334 |
+
"step": 130
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 2.25,
|
| 338 |
+
"eval_accuracy": 0.9504132231404959,
|
| 339 |
+
"eval_loss": 0.18391890823841095,
|
| 340 |
+
"eval_runtime": 8.644,
|
| 341 |
+
"eval_samples_per_second": 13.998,
|
| 342 |
+
"eval_steps_per_second": 1.851,
|
| 343 |
+
"step": 135
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 2.33,
|
| 347 |
+
"grad_norm": 0.9350224733352661,
|
| 348 |
+
"learning_rate": 0.00015366666666666667,
|
| 349 |
+
"loss": 0.0362,
|
| 350 |
+
"step": 140
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.33,
|
| 354 |
+
"eval_accuracy": 0.9586776859504132,
|
| 355 |
+
"eval_loss": 0.20266053080558777,
|
| 356 |
+
"eval_runtime": 8.5546,
|
| 357 |
+
"eval_samples_per_second": 14.144,
|
| 358 |
+
"eval_steps_per_second": 1.87,
|
| 359 |
+
"step": 140
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.42,
|
| 363 |
+
"eval_accuracy": 0.9090909090909091,
|
| 364 |
+
"eval_loss": 0.28766530752182007,
|
| 365 |
+
"eval_runtime": 8.0124,
|
| 366 |
+
"eval_samples_per_second": 15.102,
|
| 367 |
+
"eval_steps_per_second": 1.997,
|
| 368 |
+
"step": 145
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.5,
|
| 372 |
+
"grad_norm": 8.35146427154541,
|
| 373 |
+
"learning_rate": 0.00015033333333333335,
|
| 374 |
+
"loss": 0.1047,
|
| 375 |
+
"step": 150
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.5,
|
| 379 |
+
"eval_accuracy": 0.8925619834710744,
|
| 380 |
+
"eval_loss": 0.4503512978553772,
|
| 381 |
+
"eval_runtime": 8.454,
|
| 382 |
+
"eval_samples_per_second": 14.313,
|
| 383 |
+
"eval_steps_per_second": 1.893,
|
| 384 |
+
"step": 150
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.58,
|
| 388 |
+
"eval_accuracy": 0.9504132231404959,
|
| 389 |
+
"eval_loss": 0.1810603141784668,
|
| 390 |
+
"eval_runtime": 9.2694,
|
| 391 |
+
"eval_samples_per_second": 13.054,
|
| 392 |
+
"eval_steps_per_second": 1.726,
|
| 393 |
+
"step": 155
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.67,
|
| 397 |
+
"grad_norm": 0.6059785485267639,
|
| 398 |
+
"learning_rate": 0.000147,
|
| 399 |
+
"loss": 0.1232,
|
| 400 |
+
"step": 160
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.67,
|
| 404 |
+
"eval_accuracy": 0.9421487603305785,
|
| 405 |
+
"eval_loss": 0.21074515581130981,
|
| 406 |
+
"eval_runtime": 8.8489,
|
| 407 |
+
"eval_samples_per_second": 13.674,
|
| 408 |
+
"eval_steps_per_second": 1.808,
|
| 409 |
+
"step": 160
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.75,
|
| 413 |
+
"eval_accuracy": 0.9504132231404959,
|
| 414 |
+
"eval_loss": 0.20863419771194458,
|
| 415 |
+
"eval_runtime": 8.9684,
|
| 416 |
+
"eval_samples_per_second": 13.492,
|
| 417 |
+
"eval_steps_per_second": 1.784,
|
| 418 |
+
"step": 165
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 2.83,
|
| 422 |
+
"grad_norm": 0.04269757494330406,
|
| 423 |
+
"learning_rate": 0.00014366666666666667,
|
| 424 |
+
"loss": 0.0611,
|
| 425 |
+
"step": 170
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 2.83,
|
| 429 |
+
"eval_accuracy": 0.9338842975206612,
|
| 430 |
+
"eval_loss": 0.2971450686454773,
|
| 431 |
+
"eval_runtime": 9.1231,
|
| 432 |
+
"eval_samples_per_second": 13.263,
|
| 433 |
+
"eval_steps_per_second": 1.754,
|
| 434 |
+
"step": 170
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.92,
|
| 438 |
+
"eval_accuracy": 0.9338842975206612,
|
| 439 |
+
"eval_loss": 0.2731765806674957,
|
| 440 |
+
"eval_runtime": 8.9974,
|
| 441 |
+
"eval_samples_per_second": 13.448,
|
| 442 |
+
"eval_steps_per_second": 1.778,
|
| 443 |
+
"step": 175
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.0,
|
| 447 |
+
"grad_norm": 0.9442762732505798,
|
| 448 |
+
"learning_rate": 0.00014033333333333335,
|
| 449 |
+
"loss": 0.0815,
|
| 450 |
+
"step": 180
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 3.0,
|
| 454 |
+
"eval_accuracy": 0.9586776859504132,
|
| 455 |
+
"eval_loss": 0.16794723272323608,
|
| 456 |
+
"eval_runtime": 8.9455,
|
| 457 |
+
"eval_samples_per_second": 13.526,
|
| 458 |
+
"eval_steps_per_second": 1.789,
|
| 459 |
+
"step": 180
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.08,
|
| 463 |
+
"eval_accuracy": 0.9338842975206612,
|
| 464 |
+
"eval_loss": 0.24155646562576294,
|
| 465 |
+
"eval_runtime": 8.3613,
|
| 466 |
+
"eval_samples_per_second": 14.471,
|
| 467 |
+
"eval_steps_per_second": 1.914,
|
| 468 |
+
"step": 185
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 3.17,
|
| 472 |
+
"grad_norm": 2.08608078956604,
|
| 473 |
+
"learning_rate": 0.00013700000000000002,
|
| 474 |
+
"loss": 0.0469,
|
| 475 |
+
"step": 190
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 3.17,
|
| 479 |
+
"eval_accuracy": 0.9256198347107438,
|
| 480 |
+
"eval_loss": 0.29269692301750183,
|
| 481 |
+
"eval_runtime": 8.9407,
|
| 482 |
+
"eval_samples_per_second": 13.534,
|
| 483 |
+
"eval_steps_per_second": 1.79,
|
| 484 |
+
"step": 190
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 3.25,
|
| 488 |
+
"eval_accuracy": 0.9338842975206612,
|
| 489 |
+
"eval_loss": 0.28314581513404846,
|
| 490 |
+
"eval_runtime": 8.7918,
|
| 491 |
+
"eval_samples_per_second": 13.763,
|
| 492 |
+
"eval_steps_per_second": 1.82,
|
| 493 |
+
"step": 195
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 3.33,
|
| 497 |
+
"grad_norm": 0.8798501491546631,
|
| 498 |
+
"learning_rate": 0.00013366666666666667,
|
| 499 |
+
"loss": 0.0443,
|
| 500 |
+
"step": 200
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 3.33,
|
| 504 |
+
"eval_accuracy": 0.9421487603305785,
|
| 505 |
+
"eval_loss": 0.2744951546192169,
|
| 506 |
+
"eval_runtime": 8.612,
|
| 507 |
+
"eval_samples_per_second": 14.05,
|
| 508 |
+
"eval_steps_per_second": 1.858,
|
| 509 |
+
"step": 200
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.42,
|
| 513 |
+
"eval_accuracy": 0.8925619834710744,
|
| 514 |
+
"eval_loss": 0.4193201959133148,
|
| 515 |
+
"eval_runtime": 8.9147,
|
| 516 |
+
"eval_samples_per_second": 13.573,
|
| 517 |
+
"eval_steps_per_second": 1.795,
|
| 518 |
+
"step": 205
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 3.5,
|
| 522 |
+
"grad_norm": 0.03738969564437866,
|
| 523 |
+
"learning_rate": 0.00013033333333333332,
|
| 524 |
+
"loss": 0.0823,
|
| 525 |
+
"step": 210
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.5,
|
| 529 |
+
"eval_accuracy": 0.9173553719008265,
|
| 530 |
+
"eval_loss": 0.3746081292629242,
|
| 531 |
+
"eval_runtime": 8.5854,
|
| 532 |
+
"eval_samples_per_second": 14.094,
|
| 533 |
+
"eval_steps_per_second": 1.864,
|
| 534 |
+
"step": 210
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 3.58,
|
| 538 |
+
"eval_accuracy": 0.9421487603305785,
|
| 539 |
+
"eval_loss": 0.30296453833580017,
|
| 540 |
+
"eval_runtime": 8.8651,
|
| 541 |
+
"eval_samples_per_second": 13.649,
|
| 542 |
+
"eval_steps_per_second": 1.805,
|
| 543 |
+
"step": 215
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.67,
|
| 547 |
+
"grad_norm": 0.03203318268060684,
|
| 548 |
+
"learning_rate": 0.000127,
|
| 549 |
+
"loss": 0.0101,
|
| 550 |
+
"step": 220
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 3.67,
|
| 554 |
+
"eval_accuracy": 0.9504132231404959,
|
| 555 |
+
"eval_loss": 0.21464580297470093,
|
| 556 |
+
"eval_runtime": 8.8029,
|
| 557 |
+
"eval_samples_per_second": 13.745,
|
| 558 |
+
"eval_steps_per_second": 1.818,
|
| 559 |
+
"step": 220
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 3.75,
|
| 563 |
+
"eval_accuracy": 0.9421487603305785,
|
| 564 |
+
"eval_loss": 0.2514008581638336,
|
| 565 |
+
"eval_runtime": 9.2073,
|
| 566 |
+
"eval_samples_per_second": 13.142,
|
| 567 |
+
"eval_steps_per_second": 1.738,
|
| 568 |
+
"step": 225
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 3.83,
|
| 572 |
+
"grad_norm": 0.04610053077340126,
|
| 573 |
+
"learning_rate": 0.00012366666666666667,
|
| 574 |
+
"loss": 0.16,
|
| 575 |
+
"step": 230
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 3.83,
|
| 579 |
+
"eval_accuracy": 0.9421487603305785,
|
| 580 |
+
"eval_loss": 0.25517505407333374,
|
| 581 |
+
"eval_runtime": 8.8885,
|
| 582 |
+
"eval_samples_per_second": 13.613,
|
| 583 |
+
"eval_steps_per_second": 1.8,
|
| 584 |
+
"step": 230
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.92,
|
| 588 |
+
"eval_accuracy": 0.9421487603305785,
|
| 589 |
+
"eval_loss": 0.22389596700668335,
|
| 590 |
+
"eval_runtime": 8.5203,
|
| 591 |
+
"eval_samples_per_second": 14.201,
|
| 592 |
+
"eval_steps_per_second": 1.878,
|
| 593 |
+
"step": 235
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 4.0,
|
| 597 |
+
"grad_norm": 3.5699214935302734,
|
| 598 |
+
"learning_rate": 0.00012033333333333335,
|
| 599 |
+
"loss": 0.1687,
|
| 600 |
+
"step": 240
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 4.0,
|
| 604 |
+
"eval_accuracy": 0.9256198347107438,
|
| 605 |
+
"eval_loss": 0.25712552666664124,
|
| 606 |
+
"eval_runtime": 8.7329,
|
| 607 |
+
"eval_samples_per_second": 13.856,
|
| 608 |
+
"eval_steps_per_second": 1.832,
|
| 609 |
+
"step": 240
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.08,
|
| 613 |
+
"eval_accuracy": 0.9752066115702479,
|
| 614 |
+
"eval_loss": 0.13568438589572906,
|
| 615 |
+
"eval_runtime": 8.642,
|
| 616 |
+
"eval_samples_per_second": 14.001,
|
| 617 |
+
"eval_steps_per_second": 1.851,
|
| 618 |
+
"step": 245
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 4.17,
|
| 622 |
+
"grad_norm": 2.446256637573242,
|
| 623 |
+
"learning_rate": 0.000117,
|
| 624 |
+
"loss": 0.0758,
|
| 625 |
+
"step": 250
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 4.17,
|
| 629 |
+
"eval_accuracy": 0.9504132231404959,
|
| 630 |
+
"eval_loss": 0.17341962456703186,
|
| 631 |
+
"eval_runtime": 8.0872,
|
| 632 |
+
"eval_samples_per_second": 14.962,
|
| 633 |
+
"eval_steps_per_second": 1.978,
|
| 634 |
+
"step": 250
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 4.25,
|
| 638 |
+
"eval_accuracy": 0.9752066115702479,
|
| 639 |
+
"eval_loss": 0.11965644359588623,
|
| 640 |
+
"eval_runtime": 9.0168,
|
| 641 |
+
"eval_samples_per_second": 13.419,
|
| 642 |
+
"eval_steps_per_second": 1.774,
|
| 643 |
+
"step": 255
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 4.33,
|
| 647 |
+
"grad_norm": 0.048665851354599,
|
| 648 |
+
"learning_rate": 0.00011366666666666667,
|
| 649 |
+
"loss": 0.042,
|
| 650 |
+
"step": 260
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 4.33,
|
| 654 |
+
"eval_accuracy": 0.9421487603305785,
|
| 655 |
+
"eval_loss": 0.23387375473976135,
|
| 656 |
+
"eval_runtime": 8.7974,
|
| 657 |
+
"eval_samples_per_second": 13.754,
|
| 658 |
+
"eval_steps_per_second": 1.819,
|
| 659 |
+
"step": 260
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 4.42,
|
| 663 |
+
"eval_accuracy": 0.9173553719008265,
|
| 664 |
+
"eval_loss": 0.2923980951309204,
|
| 665 |
+
"eval_runtime": 8.7236,
|
| 666 |
+
"eval_samples_per_second": 13.87,
|
| 667 |
+
"eval_steps_per_second": 1.834,
|
| 668 |
+
"step": 265
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 4.5,
|
| 672 |
+
"grad_norm": 0.07859846204519272,
|
| 673 |
+
"learning_rate": 0.00011033333333333334,
|
| 674 |
+
"loss": 0.0114,
|
| 675 |
+
"step": 270
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.5,
|
| 679 |
+
"eval_accuracy": 0.9504132231404959,
|
| 680 |
+
"eval_loss": 0.23183898627758026,
|
| 681 |
+
"eval_runtime": 8.7284,
|
| 682 |
+
"eval_samples_per_second": 13.863,
|
| 683 |
+
"eval_steps_per_second": 1.833,
|
| 684 |
+
"step": 270
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 4.58,
|
| 688 |
+
"eval_accuracy": 0.9586776859504132,
|
| 689 |
+
"eval_loss": 0.17654092609882355,
|
| 690 |
+
"eval_runtime": 8.8357,
|
| 691 |
+
"eval_samples_per_second": 13.694,
|
| 692 |
+
"eval_steps_per_second": 1.811,
|
| 693 |
+
"step": 275
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 4.67,
|
| 697 |
+
"grad_norm": 0.266812801361084,
|
| 698 |
+
"learning_rate": 0.00010700000000000001,
|
| 699 |
+
"loss": 0.0197,
|
| 700 |
+
"step": 280
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 4.67,
|
| 704 |
+
"eval_accuracy": 0.9669421487603306,
|
| 705 |
+
"eval_loss": 0.12631382048130035,
|
| 706 |
+
"eval_runtime": 7.9448,
|
| 707 |
+
"eval_samples_per_second": 15.23,
|
| 708 |
+
"eval_steps_per_second": 2.014,
|
| 709 |
+
"step": 280
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 4.75,
|
| 713 |
+
"eval_accuracy": 0.9669421487603306,
|
| 714 |
+
"eval_loss": 0.12528358399868011,
|
| 715 |
+
"eval_runtime": 8.1841,
|
| 716 |
+
"eval_samples_per_second": 14.785,
|
| 717 |
+
"eval_steps_per_second": 1.955,
|
| 718 |
+
"step": 285
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 4.83,
|
| 722 |
+
"grad_norm": 0.8625770807266235,
|
| 723 |
+
"learning_rate": 0.00010366666666666666,
|
| 724 |
+
"loss": 0.0283,
|
| 725 |
+
"step": 290
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 4.83,
|
| 729 |
+
"eval_accuracy": 0.9669421487603306,
|
| 730 |
+
"eval_loss": 0.1239379420876503,
|
| 731 |
+
"eval_runtime": 9.1766,
|
| 732 |
+
"eval_samples_per_second": 13.186,
|
| 733 |
+
"eval_steps_per_second": 1.744,
|
| 734 |
+
"step": 290
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.92,
|
| 738 |
+
"eval_accuracy": 0.9669421487603306,
|
| 739 |
+
"eval_loss": 0.12782499194145203,
|
| 740 |
+
"eval_runtime": 8.9807,
|
| 741 |
+
"eval_samples_per_second": 13.473,
|
| 742 |
+
"eval_steps_per_second": 1.782,
|
| 743 |
+
"step": 295
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 5.0,
|
| 747 |
+
"grad_norm": 0.08663175255060196,
|
| 748 |
+
"learning_rate": 0.00010033333333333335,
|
| 749 |
+
"loss": 0.1115,
|
| 750 |
+
"step": 300
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 5.0,
|
| 754 |
+
"eval_accuracy": 0.9338842975206612,
|
| 755 |
+
"eval_loss": 0.2527827322483063,
|
| 756 |
+
"eval_runtime": 8.5733,
|
| 757 |
+
"eval_samples_per_second": 14.114,
|
| 758 |
+
"eval_steps_per_second": 1.866,
|
| 759 |
+
"step": 300
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 5.08,
|
| 763 |
+
"eval_accuracy": 0.9338842975206612,
|
| 764 |
+
"eval_loss": 0.3164093792438507,
|
| 765 |
+
"eval_runtime": 8.5647,
|
| 766 |
+
"eval_samples_per_second": 14.128,
|
| 767 |
+
"eval_steps_per_second": 1.868,
|
| 768 |
+
"step": 305
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 5.17,
|
| 772 |
+
"grad_norm": 0.11408742517232895,
|
| 773 |
+
"learning_rate": 9.7e-05,
|
| 774 |
+
"loss": 0.0404,
|
| 775 |
+
"step": 310
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 5.17,
|
| 779 |
+
"eval_accuracy": 0.9338842975206612,
|
| 780 |
+
"eval_loss": 0.2841833829879761,
|
| 781 |
+
"eval_runtime": 8.7395,
|
| 782 |
+
"eval_samples_per_second": 13.845,
|
| 783 |
+
"eval_steps_per_second": 1.831,
|
| 784 |
+
"step": 310
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 5.25,
|
| 788 |
+
"eval_accuracy": 0.9504132231404959,
|
| 789 |
+
"eval_loss": 0.17133790254592896,
|
| 790 |
+
"eval_runtime": 8.8895,
|
| 791 |
+
"eval_samples_per_second": 13.612,
|
| 792 |
+
"eval_steps_per_second": 1.8,
|
| 793 |
+
"step": 315
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 5.33,
|
| 797 |
+
"grad_norm": 0.08482904732227325,
|
| 798 |
+
"learning_rate": 9.366666666666668e-05,
|
| 799 |
+
"loss": 0.0719,
|
| 800 |
+
"step": 320
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 5.33,
|
| 804 |
+
"eval_accuracy": 0.9338842975206612,
|
| 805 |
+
"eval_loss": 0.18959270417690277,
|
| 806 |
+
"eval_runtime": 8.641,
|
| 807 |
+
"eval_samples_per_second": 14.003,
|
| 808 |
+
"eval_steps_per_second": 1.852,
|
| 809 |
+
"step": 320
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 5.42,
|
| 813 |
+
"eval_accuracy": 0.9256198347107438,
|
| 814 |
+
"eval_loss": 0.18550463020801544,
|
| 815 |
+
"eval_runtime": 8.219,
|
| 816 |
+
"eval_samples_per_second": 14.722,
|
| 817 |
+
"eval_steps_per_second": 1.947,
|
| 818 |
+
"step": 325
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 5.5,
|
| 822 |
+
"grad_norm": 4.354619026184082,
|
| 823 |
+
"learning_rate": 9.033333333333334e-05,
|
| 824 |
+
"loss": 0.0435,
|
| 825 |
+
"step": 330
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.5,
|
| 829 |
+
"eval_accuracy": 0.9669421487603306,
|
| 830 |
+
"eval_loss": 0.15409986674785614,
|
| 831 |
+
"eval_runtime": 8.5474,
|
| 832 |
+
"eval_samples_per_second": 14.156,
|
| 833 |
+
"eval_steps_per_second": 1.872,
|
| 834 |
+
"step": 330
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 5.58,
|
| 838 |
+
"eval_accuracy": 0.9669421487603306,
|
| 839 |
+
"eval_loss": 0.10501641035079956,
|
| 840 |
+
"eval_runtime": 8.6962,
|
| 841 |
+
"eval_samples_per_second": 13.914,
|
| 842 |
+
"eval_steps_per_second": 1.84,
|
| 843 |
+
"step": 335
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 5.67,
|
| 847 |
+
"grad_norm": 0.05306649208068848,
|
| 848 |
+
"learning_rate": 8.7e-05,
|
| 849 |
+
"loss": 0.0129,
|
| 850 |
+
"step": 340
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 5.67,
|
| 854 |
+
"eval_accuracy": 0.9586776859504132,
|
| 855 |
+
"eval_loss": 0.10632016509771347,
|
| 856 |
+
"eval_runtime": 8.7849,
|
| 857 |
+
"eval_samples_per_second": 13.774,
|
| 858 |
+
"eval_steps_per_second": 1.821,
|
| 859 |
+
"step": 340
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 5.75,
|
| 863 |
+
"eval_accuracy": 0.9586776859504132,
|
| 864 |
+
"eval_loss": 0.11378511786460876,
|
| 865 |
+
"eval_runtime": 8.0973,
|
| 866 |
+
"eval_samples_per_second": 14.943,
|
| 867 |
+
"eval_steps_per_second": 1.976,
|
| 868 |
+
"step": 345
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.83,
|
| 872 |
+
"grad_norm": 0.02987060882151127,
|
| 873 |
+
"learning_rate": 8.366666666666668e-05,
|
| 874 |
+
"loss": 0.0222,
|
| 875 |
+
"step": 350
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 5.83,
|
| 879 |
+
"eval_accuracy": 0.9586776859504132,
|
| 880 |
+
"eval_loss": 0.11444854736328125,
|
| 881 |
+
"eval_runtime": 8.7513,
|
| 882 |
+
"eval_samples_per_second": 13.827,
|
| 883 |
+
"eval_steps_per_second": 1.828,
|
| 884 |
+
"step": 350
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 5.92,
|
| 888 |
+
"eval_accuracy": 0.9669421487603306,
|
| 889 |
+
"eval_loss": 0.12378235161304474,
|
| 890 |
+
"eval_runtime": 9.2818,
|
| 891 |
+
"eval_samples_per_second": 13.036,
|
| 892 |
+
"eval_steps_per_second": 1.724,
|
| 893 |
+
"step": 355
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 6.0,
|
| 897 |
+
"grad_norm": 0.02307078428566456,
|
| 898 |
+
"learning_rate": 8.033333333333334e-05,
|
| 899 |
+
"loss": 0.0431,
|
| 900 |
+
"step": 360
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 6.0,
|
| 904 |
+
"eval_accuracy": 0.9752066115702479,
|
| 905 |
+
"eval_loss": 0.1342514306306839,
|
| 906 |
+
"eval_runtime": 7.9783,
|
| 907 |
+
"eval_samples_per_second": 15.166,
|
| 908 |
+
"eval_steps_per_second": 2.005,
|
| 909 |
+
"step": 360
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.08,
|
| 913 |
+
"eval_accuracy": 0.9669421487603306,
|
| 914 |
+
"eval_loss": 0.144140362739563,
|
| 915 |
+
"eval_runtime": 8.763,
|
| 916 |
+
"eval_samples_per_second": 13.808,
|
| 917 |
+
"eval_steps_per_second": 1.826,
|
| 918 |
+
"step": 365
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 6.17,
|
| 922 |
+
"grad_norm": 0.016979066655039787,
|
| 923 |
+
"learning_rate": 7.7e-05,
|
| 924 |
+
"loss": 0.0064,
|
| 925 |
+
"step": 370
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 6.17,
|
| 929 |
+
"eval_accuracy": 0.9669421487603306,
|
| 930 |
+
"eval_loss": 0.1470753401517868,
|
| 931 |
+
"eval_runtime": 8.9895,
|
| 932 |
+
"eval_samples_per_second": 13.46,
|
| 933 |
+
"eval_steps_per_second": 1.78,
|
| 934 |
+
"step": 370
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 6.25,
|
| 938 |
+
"eval_accuracy": 0.9752066115702479,
|
| 939 |
+
"eval_loss": 0.1360587477684021,
|
| 940 |
+
"eval_runtime": 8.9513,
|
| 941 |
+
"eval_samples_per_second": 13.518,
|
| 942 |
+
"eval_steps_per_second": 1.787,
|
| 943 |
+
"step": 375
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 6.33,
|
| 947 |
+
"grad_norm": 0.022434255108237267,
|
| 948 |
+
"learning_rate": 7.366666666666668e-05,
|
| 949 |
+
"loss": 0.0576,
|
| 950 |
+
"step": 380
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 6.33,
|
| 954 |
+
"eval_accuracy": 0.9752066115702479,
|
| 955 |
+
"eval_loss": 0.13161548972129822,
|
| 956 |
+
"eval_runtime": 8.7711,
|
| 957 |
+
"eval_samples_per_second": 13.795,
|
| 958 |
+
"eval_steps_per_second": 1.824,
|
| 959 |
+
"step": 380
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 6.42,
|
| 963 |
+
"eval_accuracy": 0.9669421487603306,
|
| 964 |
+
"eval_loss": 0.12319940328598022,
|
| 965 |
+
"eval_runtime": 8.3275,
|
| 966 |
+
"eval_samples_per_second": 14.53,
|
| 967 |
+
"eval_steps_per_second": 1.921,
|
| 968 |
+
"step": 385
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 6.5,
|
| 972 |
+
"grad_norm": 0.2556796371936798,
|
| 973 |
+
"learning_rate": 7.033333333333334e-05,
|
| 974 |
+
"loss": 0.0298,
|
| 975 |
+
"step": 390
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 6.5,
|
| 979 |
+
"eval_accuracy": 0.9669421487603306,
|
| 980 |
+
"eval_loss": 0.1254538893699646,
|
| 981 |
+
"eval_runtime": 8.6034,
|
| 982 |
+
"eval_samples_per_second": 14.064,
|
| 983 |
+
"eval_steps_per_second": 1.86,
|
| 984 |
+
"step": 390
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 6.58,
|
| 988 |
+
"eval_accuracy": 0.9669421487603306,
|
| 989 |
+
"eval_loss": 0.13591305911540985,
|
| 990 |
+
"eval_runtime": 8.8244,
|
| 991 |
+
"eval_samples_per_second": 13.712,
|
| 992 |
+
"eval_steps_per_second": 1.813,
|
| 993 |
+
"step": 395
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 6.67,
|
| 997 |
+
"grad_norm": 0.03436155617237091,
|
| 998 |
+
"learning_rate": 6.7e-05,
|
| 999 |
+
"loss": 0.0097,
|
| 1000 |
+
"step": 400
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 6.67,
|
| 1004 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1005 |
+
"eval_loss": 0.1434980034828186,
|
| 1006 |
+
"eval_runtime": 9.1677,
|
| 1007 |
+
"eval_samples_per_second": 13.199,
|
| 1008 |
+
"eval_steps_per_second": 1.745,
|
| 1009 |
+
"step": 400
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 6.75,
|
| 1013 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1014 |
+
"eval_loss": 0.14506025612354279,
|
| 1015 |
+
"eval_runtime": 8.7551,
|
| 1016 |
+
"eval_samples_per_second": 13.82,
|
| 1017 |
+
"eval_steps_per_second": 1.827,
|
| 1018 |
+
"step": 405
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 6.83,
|
| 1022 |
+
"grad_norm": 0.019292179495096207,
|
| 1023 |
+
"learning_rate": 6.366666666666668e-05,
|
| 1024 |
+
"loss": 0.0153,
|
| 1025 |
+
"step": 410
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 6.83,
|
| 1029 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1030 |
+
"eval_loss": 0.14391547441482544,
|
| 1031 |
+
"eval_runtime": 8.6401,
|
| 1032 |
+
"eval_samples_per_second": 14.004,
|
| 1033 |
+
"eval_steps_per_second": 1.852,
|
| 1034 |
+
"step": 410
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 6.92,
|
| 1038 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1039 |
+
"eval_loss": 0.1352916657924652,
|
| 1040 |
+
"eval_runtime": 8.9781,
|
| 1041 |
+
"eval_samples_per_second": 13.477,
|
| 1042 |
+
"eval_steps_per_second": 1.782,
|
| 1043 |
+
"step": 415
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 7.0,
|
| 1047 |
+
"grad_norm": 0.1764516532421112,
|
| 1048 |
+
"learning_rate": 6.033333333333334e-05,
|
| 1049 |
+
"loss": 0.0406,
|
| 1050 |
+
"step": 420
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 7.0,
|
| 1054 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1055 |
+
"eval_loss": 0.13157208263874054,
|
| 1056 |
+
"eval_runtime": 8.4013,
|
| 1057 |
+
"eval_samples_per_second": 14.402,
|
| 1058 |
+
"eval_steps_per_second": 1.904,
|
| 1059 |
+
"step": 420
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 7.08,
|
| 1063 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1064 |
+
"eval_loss": 0.13093091547489166,
|
| 1065 |
+
"eval_runtime": 9.2645,
|
| 1066 |
+
"eval_samples_per_second": 13.061,
|
| 1067 |
+
"eval_steps_per_second": 1.727,
|
| 1068 |
+
"step": 425
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 7.17,
|
| 1072 |
+
"grad_norm": 0.025451194494962692,
|
| 1073 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 1074 |
+
"loss": 0.0154,
|
| 1075 |
+
"step": 430
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 7.17,
|
| 1079 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1080 |
+
"eval_loss": 0.13050581514835358,
|
| 1081 |
+
"eval_runtime": 8.9669,
|
| 1082 |
+
"eval_samples_per_second": 13.494,
|
| 1083 |
+
"eval_steps_per_second": 1.784,
|
| 1084 |
+
"step": 430
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 7.25,
|
| 1088 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1089 |
+
"eval_loss": 0.13096679747104645,
|
| 1090 |
+
"eval_runtime": 8.8412,
|
| 1091 |
+
"eval_samples_per_second": 13.686,
|
| 1092 |
+
"eval_steps_per_second": 1.81,
|
| 1093 |
+
"step": 435
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 7.33,
|
| 1097 |
+
"grad_norm": 0.01770775578916073,
|
| 1098 |
+
"learning_rate": 5.3666666666666666e-05,
|
| 1099 |
+
"loss": 0.0209,
|
| 1100 |
+
"step": 440
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 7.33,
|
| 1104 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1105 |
+
"eval_loss": 0.13012637197971344,
|
| 1106 |
+
"eval_runtime": 8.8578,
|
| 1107 |
+
"eval_samples_per_second": 13.66,
|
| 1108 |
+
"eval_steps_per_second": 1.806,
|
| 1109 |
+
"step": 440
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 7.42,
|
| 1113 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1114 |
+
"eval_loss": 0.145931214094162,
|
| 1115 |
+
"eval_runtime": 8.6371,
|
| 1116 |
+
"eval_samples_per_second": 14.009,
|
| 1117 |
+
"eval_steps_per_second": 1.852,
|
| 1118 |
+
"step": 445
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 7.5,
|
| 1122 |
+
"grad_norm": 17.443572998046875,
|
| 1123 |
+
"learning_rate": 5.0333333333333335e-05,
|
| 1124 |
+
"loss": 0.0298,
|
| 1125 |
+
"step": 450
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 7.5,
|
| 1129 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1130 |
+
"eval_loss": 0.16629938781261444,
|
| 1131 |
+
"eval_runtime": 8.2752,
|
| 1132 |
+
"eval_samples_per_second": 14.622,
|
| 1133 |
+
"eval_steps_per_second": 1.933,
|
| 1134 |
+
"step": 450
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 7.58,
|
| 1138 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1139 |
+
"eval_loss": 0.15594066679477692,
|
| 1140 |
+
"eval_runtime": 8.806,
|
| 1141 |
+
"eval_samples_per_second": 13.741,
|
| 1142 |
+
"eval_steps_per_second": 1.817,
|
| 1143 |
+
"step": 455
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.67,
|
| 1147 |
+
"grad_norm": 0.030463455244898796,
|
| 1148 |
+
"learning_rate": 4.7e-05,
|
| 1149 |
+
"loss": 0.0052,
|
| 1150 |
+
"step": 460
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 7.67,
|
| 1154 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1155 |
+
"eval_loss": 0.15159618854522705,
|
| 1156 |
+
"eval_runtime": 8.9863,
|
| 1157 |
+
"eval_samples_per_second": 13.465,
|
| 1158 |
+
"eval_steps_per_second": 1.78,
|
| 1159 |
+
"step": 460
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 7.75,
|
| 1163 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1164 |
+
"eval_loss": 0.13964863121509552,
|
| 1165 |
+
"eval_runtime": 8.4469,
|
| 1166 |
+
"eval_samples_per_second": 14.325,
|
| 1167 |
+
"eval_steps_per_second": 1.894,
|
| 1168 |
+
"step": 465
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 7.83,
|
| 1172 |
+
"grad_norm": 0.11391649395227432,
|
| 1173 |
+
"learning_rate": 4.3666666666666666e-05,
|
| 1174 |
+
"loss": 0.0172,
|
| 1175 |
+
"step": 470
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 7.83,
|
| 1179 |
+
"eval_accuracy": 0.9586776859504132,
|
| 1180 |
+
"eval_loss": 0.13303633034229279,
|
| 1181 |
+
"eval_runtime": 8.9949,
|
| 1182 |
+
"eval_samples_per_second": 13.452,
|
| 1183 |
+
"eval_steps_per_second": 1.779,
|
| 1184 |
+
"step": 470
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 7.92,
|
| 1188 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1189 |
+
"eval_loss": 0.12355250120162964,
|
| 1190 |
+
"eval_runtime": 8.1109,
|
| 1191 |
+
"eval_samples_per_second": 14.918,
|
| 1192 |
+
"eval_steps_per_second": 1.973,
|
| 1193 |
+
"step": 475
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 8.0,
|
| 1197 |
+
"grad_norm": 0.059335533529520035,
|
| 1198 |
+
"learning_rate": 4.0333333333333336e-05,
|
| 1199 |
+
"loss": 0.0348,
|
| 1200 |
+
"step": 480
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 8.0,
|
| 1204 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1205 |
+
"eval_loss": 0.12098132073879242,
|
| 1206 |
+
"eval_runtime": 8.8837,
|
| 1207 |
+
"eval_samples_per_second": 13.621,
|
| 1208 |
+
"eval_steps_per_second": 1.801,
|
| 1209 |
+
"step": 480
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 8.08,
|
| 1213 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1214 |
+
"eval_loss": 0.11751802265644073,
|
| 1215 |
+
"eval_runtime": 9.8045,
|
| 1216 |
+
"eval_samples_per_second": 12.341,
|
| 1217 |
+
"eval_steps_per_second": 1.632,
|
| 1218 |
+
"step": 485
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 8.17,
|
| 1222 |
+
"grad_norm": 0.019469719380140305,
|
| 1223 |
+
"learning_rate": 3.7e-05,
|
| 1224 |
+
"loss": 0.0068,
|
| 1225 |
+
"step": 490
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 8.17,
|
| 1229 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1230 |
+
"eval_loss": 0.1185019463300705,
|
| 1231 |
+
"eval_runtime": 8.8757,
|
| 1232 |
+
"eval_samples_per_second": 13.633,
|
| 1233 |
+
"eval_steps_per_second": 1.803,
|
| 1234 |
+
"step": 490
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 8.25,
|
| 1238 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1239 |
+
"eval_loss": 0.12291049212217331,
|
| 1240 |
+
"eval_runtime": 8.9999,
|
| 1241 |
+
"eval_samples_per_second": 13.445,
|
| 1242 |
+
"eval_steps_per_second": 1.778,
|
| 1243 |
+
"step": 495
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 8.33,
|
| 1247 |
+
"grad_norm": 0.029128719121217728,
|
| 1248 |
+
"learning_rate": 3.366666666666667e-05,
|
| 1249 |
+
"loss": 0.0305,
|
| 1250 |
+
"step": 500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 8.33,
|
| 1254 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1255 |
+
"eval_loss": 0.12297818809747696,
|
| 1256 |
+
"eval_runtime": 9.0782,
|
| 1257 |
+
"eval_samples_per_second": 13.329,
|
| 1258 |
+
"eval_steps_per_second": 1.762,
|
| 1259 |
+
"step": 500
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 8.42,
|
| 1263 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1264 |
+
"eval_loss": 0.12048203498125076,
|
| 1265 |
+
"eval_runtime": 8.2377,
|
| 1266 |
+
"eval_samples_per_second": 14.688,
|
| 1267 |
+
"eval_steps_per_second": 1.942,
|
| 1268 |
+
"step": 505
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 8.5,
|
| 1272 |
+
"grad_norm": 0.01591232791543007,
|
| 1273 |
+
"learning_rate": 3.0333333333333337e-05,
|
| 1274 |
+
"loss": 0.0154,
|
| 1275 |
+
"step": 510
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 8.5,
|
| 1279 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1280 |
+
"eval_loss": 0.11965296417474747,
|
| 1281 |
+
"eval_runtime": 9.3646,
|
| 1282 |
+
"eval_samples_per_second": 12.921,
|
| 1283 |
+
"eval_steps_per_second": 1.709,
|
| 1284 |
+
"step": 510
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 8.58,
|
| 1288 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1289 |
+
"eval_loss": 0.12173377722501755,
|
| 1290 |
+
"eval_runtime": 9.3963,
|
| 1291 |
+
"eval_samples_per_second": 12.877,
|
| 1292 |
+
"eval_steps_per_second": 1.703,
|
| 1293 |
+
"step": 515
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 8.67,
|
| 1297 |
+
"grad_norm": 0.02258380502462387,
|
| 1298 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 1299 |
+
"loss": 0.0177,
|
| 1300 |
+
"step": 520
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 8.67,
|
| 1304 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1305 |
+
"eval_loss": 0.12387024611234665,
|
| 1306 |
+
"eval_runtime": 9.1961,
|
| 1307 |
+
"eval_samples_per_second": 13.158,
|
| 1308 |
+
"eval_steps_per_second": 1.74,
|
| 1309 |
+
"step": 520
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 8.75,
|
| 1313 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1314 |
+
"eval_loss": 0.12437601387500763,
|
| 1315 |
+
"eval_runtime": 9.196,
|
| 1316 |
+
"eval_samples_per_second": 13.158,
|
| 1317 |
+
"eval_steps_per_second": 1.74,
|
| 1318 |
+
"step": 525
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 8.83,
|
| 1322 |
+
"grad_norm": 0.020360412076115608,
|
| 1323 |
+
"learning_rate": 2.3666666666666668e-05,
|
| 1324 |
+
"loss": 0.0123,
|
| 1325 |
+
"step": 530
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 8.83,
|
| 1329 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1330 |
+
"eval_loss": 0.12708893418312073,
|
| 1331 |
+
"eval_runtime": 8.8063,
|
| 1332 |
+
"eval_samples_per_second": 13.74,
|
| 1333 |
+
"eval_steps_per_second": 1.817,
|
| 1334 |
+
"step": 530
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 8.92,
|
| 1338 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1339 |
+
"eval_loss": 0.1300380975008011,
|
| 1340 |
+
"eval_runtime": 8.9605,
|
| 1341 |
+
"eval_samples_per_second": 13.504,
|
| 1342 |
+
"eval_steps_per_second": 1.786,
|
| 1343 |
+
"step": 535
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 9.0,
|
| 1347 |
+
"grad_norm": 0.6857224702835083,
|
| 1348 |
+
"learning_rate": 2.0333333333333334e-05,
|
| 1349 |
+
"loss": 0.0154,
|
| 1350 |
+
"step": 540
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 9.0,
|
| 1354 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1355 |
+
"eval_loss": 0.13137011229991913,
|
| 1356 |
+
"eval_runtime": 8.8274,
|
| 1357 |
+
"eval_samples_per_second": 13.707,
|
| 1358 |
+
"eval_steps_per_second": 1.813,
|
| 1359 |
+
"step": 540
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 9.08,
|
| 1363 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1364 |
+
"eval_loss": 0.1295723170042038,
|
| 1365 |
+
"eval_runtime": 8.8666,
|
| 1366 |
+
"eval_samples_per_second": 13.647,
|
| 1367 |
+
"eval_steps_per_second": 1.805,
|
| 1368 |
+
"step": 545
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 9.17,
|
| 1372 |
+
"grad_norm": 0.4675326943397522,
|
| 1373 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 1374 |
+
"loss": 0.0331,
|
| 1375 |
+
"step": 550
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 9.17,
|
| 1379 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1380 |
+
"eval_loss": 0.12511087954044342,
|
| 1381 |
+
"eval_runtime": 9.3211,
|
| 1382 |
+
"eval_samples_per_second": 12.981,
|
| 1383 |
+
"eval_steps_per_second": 1.717,
|
| 1384 |
+
"step": 550
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 9.25,
|
| 1388 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1389 |
+
"eval_loss": 0.12691423296928406,
|
| 1390 |
+
"eval_runtime": 9.1389,
|
| 1391 |
+
"eval_samples_per_second": 13.24,
|
| 1392 |
+
"eval_steps_per_second": 1.751,
|
| 1393 |
+
"step": 555
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 9.33,
|
| 1397 |
+
"grad_norm": 0.02136796899139881,
|
| 1398 |
+
"learning_rate": 1.3666666666666666e-05,
|
| 1399 |
+
"loss": 0.0196,
|
| 1400 |
+
"step": 560
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 9.33,
|
| 1404 |
+
"eval_accuracy": 0.9752066115702479,
|
| 1405 |
+
"eval_loss": 0.12836582958698273,
|
| 1406 |
+
"eval_runtime": 8.6274,
|
| 1407 |
+
"eval_samples_per_second": 14.025,
|
| 1408 |
+
"eval_steps_per_second": 1.855,
|
| 1409 |
+
"step": 560
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 9.42,
|
| 1413 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1414 |
+
"eval_loss": 0.1298026293516159,
|
| 1415 |
+
"eval_runtime": 8.3157,
|
| 1416 |
+
"eval_samples_per_second": 14.551,
|
| 1417 |
+
"eval_steps_per_second": 1.924,
|
| 1418 |
+
"step": 565
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 9.5,
|
| 1422 |
+
"grad_norm": 0.014311849139630795,
|
| 1423 |
+
"learning_rate": 1.0333333333333333e-05,
|
| 1424 |
+
"loss": 0.0058,
|
| 1425 |
+
"step": 570
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 9.5,
|
| 1429 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1430 |
+
"eval_loss": 0.13134212791919708,
|
| 1431 |
+
"eval_runtime": 9.1247,
|
| 1432 |
+
"eval_samples_per_second": 13.261,
|
| 1433 |
+
"eval_steps_per_second": 1.753,
|
| 1434 |
+
"step": 570
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 9.58,
|
| 1438 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1439 |
+
"eval_loss": 0.13212385773658752,
|
| 1440 |
+
"eval_runtime": 8.6468,
|
| 1441 |
+
"eval_samples_per_second": 13.994,
|
| 1442 |
+
"eval_steps_per_second": 1.85,
|
| 1443 |
+
"step": 575
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 9.67,
|
| 1447 |
+
"grad_norm": 0.4894683361053467,
|
| 1448 |
+
"learning_rate": 7.000000000000001e-06,
|
| 1449 |
+
"loss": 0.012,
|
| 1450 |
+
"step": 580
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 9.67,
|
| 1454 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1455 |
+
"eval_loss": 0.1326775997877121,
|
| 1456 |
+
"eval_runtime": 9.2181,
|
| 1457 |
+
"eval_samples_per_second": 13.126,
|
| 1458 |
+
"eval_steps_per_second": 1.736,
|
| 1459 |
+
"step": 580
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 9.75,
|
| 1463 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1464 |
+
"eval_loss": 0.13264916837215424,
|
| 1465 |
+
"eval_runtime": 8.9299,
|
| 1466 |
+
"eval_samples_per_second": 13.55,
|
| 1467 |
+
"eval_steps_per_second": 1.792,
|
| 1468 |
+
"step": 585
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 9.83,
|
| 1472 |
+
"grad_norm": 0.015242637135088444,
|
| 1473 |
+
"learning_rate": 3.666666666666667e-06,
|
| 1474 |
+
"loss": 0.0081,
|
| 1475 |
+
"step": 590
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 9.83,
|
| 1479 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1480 |
+
"eval_loss": 0.13293592631816864,
|
| 1481 |
+
"eval_runtime": 9.2123,
|
| 1482 |
+
"eval_samples_per_second": 13.135,
|
| 1483 |
+
"eval_steps_per_second": 1.737,
|
| 1484 |
+
"step": 590
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 9.92,
|
| 1488 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1489 |
+
"eval_loss": 0.13364511728286743,
|
| 1490 |
+
"eval_runtime": 8.9976,
|
| 1491 |
+
"eval_samples_per_second": 13.448,
|
| 1492 |
+
"eval_steps_per_second": 1.778,
|
| 1493 |
+
"step": 595
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 10.0,
|
| 1497 |
+
"grad_norm": 0.014774391427636147,
|
| 1498 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 1499 |
+
"loss": 0.0083,
|
| 1500 |
+
"step": 600
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 10.0,
|
| 1504 |
+
"eval_accuracy": 0.9669421487603306,
|
| 1505 |
+
"eval_loss": 0.1337580680847168,
|
| 1506 |
+
"eval_runtime": 8.6721,
|
| 1507 |
+
"eval_samples_per_second": 13.953,
|
| 1508 |
+
"eval_steps_per_second": 1.845,
|
| 1509 |
+
"step": 600
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 10.0,
|
| 1513 |
+
"step": 600,
|
| 1514 |
+
"total_flos": 7.39286832673751e+17,
|
| 1515 |
+
"train_loss": 0.06743512197087208,
|
| 1516 |
+
"train_runtime": 3546.8991,
|
| 1517 |
+
"train_samples_per_second": 2.69,
|
| 1518 |
+
"train_steps_per_second": 0.169
|
| 1519 |
+
}
|
| 1520 |
+
],
|
| 1521 |
+
"logging_steps": 10,
|
| 1522 |
+
"max_steps": 600,
|
| 1523 |
+
"num_input_tokens_seen": 0,
|
| 1524 |
+
"num_train_epochs": 10,
|
| 1525 |
+
"save_steps": 5,
|
| 1526 |
+
"total_flos": 7.39286832673751e+17,
|
| 1527 |
+
"train_batch_size": 16,
|
| 1528 |
+
"trial_name": null,
|
| 1529 |
+
"trial_params": null
|
| 1530 |
+
}
|