🍻 cheers
Browse files- README.md +6 -5
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Sep04_16-25-43_eec317420151/events.out.tfevents.1725468448.eec317420151.967.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1269 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: dima806/deepfake_vs_real_image_detection
|
| 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: train
|
|
@@ -22,7 +23,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# realFake-img
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [dima806/deepfake_vs_real_image_detection](https://huggingface.co/dima806/deepfake_vs_real_image_detection) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: dima806/deepfake_vs_real_image_detection
|
| 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: ai_real_images
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.8518181818181818
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# realFake-img
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [dima806/deepfake_vs_real_image_detection](https://huggingface.co/dima806/deepfake_vs_real_image_detection) on the ai_real_images dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.3329
|
| 37 |
+
- Accuracy: 0.8518
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8518181818181818,
|
| 4 |
+
"eval_loss": 0.33288896083831787,
|
| 5 |
+
"eval_runtime": 24.2544,
|
| 6 |
+
"eval_samples_per_second": 45.353,
|
| 7 |
+
"eval_steps_per_second": 5.69,
|
| 8 |
+
"total_flos": 1.9301704773202575e+18,
|
| 9 |
+
"train_loss": 0.266804637053074,
|
| 10 |
+
"train_runtime": 1157.865,
|
| 11 |
+
"train_samples_per_second": 21.512,
|
| 12 |
+
"train_steps_per_second": 1.347
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8518181818181818,
|
| 4 |
+
"eval_loss": 0.33288896083831787,
|
| 5 |
+
"eval_runtime": 24.2544,
|
| 6 |
+
"eval_samples_per_second": 45.353,
|
| 7 |
+
"eval_steps_per_second": 5.69
|
| 8 |
+
}
|
runs/Sep04_16-25-43_eec317420151/events.out.tfevents.1725468448.eec317420151.967.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76e1315ff7a25a92d56b47cb210475c69e8adb0848ba80e1e2402fe09dffa799
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.9301704773202575e+18,
|
| 4 |
+
"train_loss": 0.266804637053074,
|
| 5 |
+
"train_runtime": 1157.865,
|
| 6 |
+
"train_samples_per_second": 21.512,
|
| 7 |
+
"train_steps_per_second": 1.347
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1269 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.33288896083831787,
|
| 3 |
+
"best_model_checkpoint": "realFake-img/checkpoint-700",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1560,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02564102564102564,
|
| 13 |
+
"grad_norm": 2.928205728530884,
|
| 14 |
+
"learning_rate": 0.00019871794871794874,
|
| 15 |
+
"loss": 1.1209,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05128205128205128,
|
| 20 |
+
"grad_norm": 3.1282591819763184,
|
| 21 |
+
"learning_rate": 0.00019743589743589744,
|
| 22 |
+
"loss": 0.6435,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07692307692307693,
|
| 27 |
+
"grad_norm": 2.2042582035064697,
|
| 28 |
+
"learning_rate": 0.00019615384615384615,
|
| 29 |
+
"loss": 0.6513,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10256410256410256,
|
| 34 |
+
"grad_norm": 4.91236686706543,
|
| 35 |
+
"learning_rate": 0.00019487179487179487,
|
| 36 |
+
"loss": 0.7536,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1282051282051282,
|
| 41 |
+
"grad_norm": 2.019882917404175,
|
| 42 |
+
"learning_rate": 0.0001935897435897436,
|
| 43 |
+
"loss": 0.6197,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15384615384615385,
|
| 48 |
+
"grad_norm": 2.4161789417266846,
|
| 49 |
+
"learning_rate": 0.00019230769230769233,
|
| 50 |
+
"loss": 0.5531,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1794871794871795,
|
| 55 |
+
"grad_norm": 2.189767360687256,
|
| 56 |
+
"learning_rate": 0.00019102564102564104,
|
| 57 |
+
"loss": 0.4985,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.20512820512820512,
|
| 62 |
+
"grad_norm": 3.2690813541412354,
|
| 63 |
+
"learning_rate": 0.00018974358974358974,
|
| 64 |
+
"loss": 0.5499,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23076923076923078,
|
| 69 |
+
"grad_norm": 1.6541521549224854,
|
| 70 |
+
"learning_rate": 0.00018846153846153847,
|
| 71 |
+
"loss": 0.5365,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2564102564102564,
|
| 76 |
+
"grad_norm": 1.9506237506866455,
|
| 77 |
+
"learning_rate": 0.0001871794871794872,
|
| 78 |
+
"loss": 0.4892,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2564102564102564,
|
| 83 |
+
"eval_accuracy": 0.7227272727272728,
|
| 84 |
+
"eval_loss": 0.5756350159645081,
|
| 85 |
+
"eval_runtime": 24.0859,
|
| 86 |
+
"eval_samples_per_second": 45.67,
|
| 87 |
+
"eval_steps_per_second": 5.73,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.28205128205128205,
|
| 92 |
+
"grad_norm": 3.7501442432403564,
|
| 93 |
+
"learning_rate": 0.0001858974358974359,
|
| 94 |
+
"loss": 0.4306,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.3076923076923077,
|
| 99 |
+
"grad_norm": 2.692314386367798,
|
| 100 |
+
"learning_rate": 0.00018461538461538463,
|
| 101 |
+
"loss": 0.5122,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.3333333333333333,
|
| 106 |
+
"grad_norm": 2.5989458560943604,
|
| 107 |
+
"learning_rate": 0.00018333333333333334,
|
| 108 |
+
"loss": 0.4974,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.358974358974359,
|
| 113 |
+
"grad_norm": 1.8959237337112427,
|
| 114 |
+
"learning_rate": 0.00018205128205128207,
|
| 115 |
+
"loss": 0.4464,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.38461538461538464,
|
| 120 |
+
"grad_norm": 1.9950543642044067,
|
| 121 |
+
"learning_rate": 0.00018076923076923077,
|
| 122 |
+
"loss": 0.4366,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.41025641025641024,
|
| 127 |
+
"grad_norm": 1.4334654808044434,
|
| 128 |
+
"learning_rate": 0.0001794871794871795,
|
| 129 |
+
"loss": 0.5248,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.4358974358974359,
|
| 134 |
+
"grad_norm": 1.7422298192977905,
|
| 135 |
+
"learning_rate": 0.00017820512820512823,
|
| 136 |
+
"loss": 0.4314,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.46153846153846156,
|
| 141 |
+
"grad_norm": 1.0945038795471191,
|
| 142 |
+
"learning_rate": 0.00017692307692307693,
|
| 143 |
+
"loss": 0.3464,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.48717948717948717,
|
| 148 |
+
"grad_norm": 2.3801605701446533,
|
| 149 |
+
"learning_rate": 0.00017564102564102566,
|
| 150 |
+
"loss": 0.4059,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5128205128205128,
|
| 155 |
+
"grad_norm": 1.6624411344528198,
|
| 156 |
+
"learning_rate": 0.00017435897435897436,
|
| 157 |
+
"loss": 0.683,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.5128205128205128,
|
| 162 |
+
"eval_accuracy": 0.6372727272727273,
|
| 163 |
+
"eval_loss": 0.6742109656333923,
|
| 164 |
+
"eval_runtime": 24.1183,
|
| 165 |
+
"eval_samples_per_second": 45.609,
|
| 166 |
+
"eval_steps_per_second": 5.722,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.5384615384615384,
|
| 171 |
+
"grad_norm": 3.1722240447998047,
|
| 172 |
+
"learning_rate": 0.0001730769230769231,
|
| 173 |
+
"loss": 0.4548,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.5641025641025641,
|
| 178 |
+
"grad_norm": 2.0260214805603027,
|
| 179 |
+
"learning_rate": 0.0001717948717948718,
|
| 180 |
+
"loss": 0.4612,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.5897435897435898,
|
| 185 |
+
"grad_norm": 2.1661155223846436,
|
| 186 |
+
"learning_rate": 0.00017051282051282053,
|
| 187 |
+
"loss": 0.3824,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.6153846153846154,
|
| 192 |
+
"grad_norm": 2.2094335556030273,
|
| 193 |
+
"learning_rate": 0.00016923076923076923,
|
| 194 |
+
"loss": 0.3815,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.6410256410256411,
|
| 199 |
+
"grad_norm": 2.571754217147827,
|
| 200 |
+
"learning_rate": 0.00016794871794871796,
|
| 201 |
+
"loss": 0.3743,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.6666666666666666,
|
| 206 |
+
"grad_norm": 1.545766830444336,
|
| 207 |
+
"learning_rate": 0.0001666666666666667,
|
| 208 |
+
"loss": 0.3708,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.6923076923076923,
|
| 213 |
+
"grad_norm": 2.742072343826294,
|
| 214 |
+
"learning_rate": 0.0001653846153846154,
|
| 215 |
+
"loss": 0.3729,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.717948717948718,
|
| 220 |
+
"grad_norm": 2.677527904510498,
|
| 221 |
+
"learning_rate": 0.0001641025641025641,
|
| 222 |
+
"loss": 0.3555,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.7435897435897436,
|
| 227 |
+
"grad_norm": 1.6417704820632935,
|
| 228 |
+
"learning_rate": 0.00016282051282051282,
|
| 229 |
+
"loss": 0.4212,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.7692307692307693,
|
| 234 |
+
"grad_norm": 2.6961071491241455,
|
| 235 |
+
"learning_rate": 0.00016153846153846155,
|
| 236 |
+
"loss": 0.3737,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.7692307692307693,
|
| 241 |
+
"eval_accuracy": 0.7554545454545455,
|
| 242 |
+
"eval_loss": 0.5462190508842468,
|
| 243 |
+
"eval_runtime": 24.1467,
|
| 244 |
+
"eval_samples_per_second": 45.555,
|
| 245 |
+
"eval_steps_per_second": 5.715,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.7948717948717948,
|
| 250 |
+
"grad_norm": 3.5049915313720703,
|
| 251 |
+
"learning_rate": 0.00016025641025641028,
|
| 252 |
+
"loss": 0.3995,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.8205128205128205,
|
| 257 |
+
"grad_norm": 2.599503517150879,
|
| 258 |
+
"learning_rate": 0.00015897435897435896,
|
| 259 |
+
"loss": 0.3805,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.8461538461538461,
|
| 264 |
+
"grad_norm": 2.3924107551574707,
|
| 265 |
+
"learning_rate": 0.0001576923076923077,
|
| 266 |
+
"loss": 0.3586,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.8717948717948718,
|
| 271 |
+
"grad_norm": 3.0169456005096436,
|
| 272 |
+
"learning_rate": 0.00015641025641025642,
|
| 273 |
+
"loss": 0.3964,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.8974358974358975,
|
| 278 |
+
"grad_norm": 3.907792091369629,
|
| 279 |
+
"learning_rate": 0.00015512820512820515,
|
| 280 |
+
"loss": 0.3915,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.9230769230769231,
|
| 285 |
+
"grad_norm": 1.2597954273223877,
|
| 286 |
+
"learning_rate": 0.00015384615384615385,
|
| 287 |
+
"loss": 0.3471,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.9487179487179487,
|
| 292 |
+
"grad_norm": 4.323169231414795,
|
| 293 |
+
"learning_rate": 0.00015256410256410255,
|
| 294 |
+
"loss": 0.2476,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.9743589743589743,
|
| 299 |
+
"grad_norm": 3.7260568141937256,
|
| 300 |
+
"learning_rate": 0.00015128205128205128,
|
| 301 |
+
"loss": 0.41,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.0,
|
| 306 |
+
"grad_norm": 0.22700244188308716,
|
| 307 |
+
"learning_rate": 0.00015000000000000001,
|
| 308 |
+
"loss": 0.3773,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.0256410256410255,
|
| 313 |
+
"grad_norm": 1.8512117862701416,
|
| 314 |
+
"learning_rate": 0.00014871794871794872,
|
| 315 |
+
"loss": 0.3554,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.0256410256410255,
|
| 320 |
+
"eval_accuracy": 0.8009090909090909,
|
| 321 |
+
"eval_loss": 0.4354061186313629,
|
| 322 |
+
"eval_runtime": 24.0034,
|
| 323 |
+
"eval_samples_per_second": 45.827,
|
| 324 |
+
"eval_steps_per_second": 5.749,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.0512820512820513,
|
| 329 |
+
"grad_norm": 1.9039109945297241,
|
| 330 |
+
"learning_rate": 0.00014743589743589745,
|
| 331 |
+
"loss": 0.3033,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.0769230769230769,
|
| 336 |
+
"grad_norm": 2.7765700817108154,
|
| 337 |
+
"learning_rate": 0.00014615384615384615,
|
| 338 |
+
"loss": 0.2594,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.1025641025641026,
|
| 343 |
+
"grad_norm": 2.247612953186035,
|
| 344 |
+
"learning_rate": 0.00014487179487179488,
|
| 345 |
+
"loss": 0.3434,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.1282051282051282,
|
| 350 |
+
"grad_norm": 1.161192536354065,
|
| 351 |
+
"learning_rate": 0.0001435897435897436,
|
| 352 |
+
"loss": 0.2653,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.1538461538461537,
|
| 357 |
+
"grad_norm": 2.9420008659362793,
|
| 358 |
+
"learning_rate": 0.0001423076923076923,
|
| 359 |
+
"loss": 0.3184,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.1794871794871795,
|
| 364 |
+
"grad_norm": 2.359160900115967,
|
| 365 |
+
"learning_rate": 0.00014102564102564104,
|
| 366 |
+
"loss": 0.4049,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.205128205128205,
|
| 371 |
+
"grad_norm": 1.5929157733917236,
|
| 372 |
+
"learning_rate": 0.00013974358974358974,
|
| 373 |
+
"loss": 0.371,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.2307692307692308,
|
| 378 |
+
"grad_norm": 3.8561315536499023,
|
| 379 |
+
"learning_rate": 0.00013846153846153847,
|
| 380 |
+
"loss": 0.2864,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.2564102564102564,
|
| 385 |
+
"grad_norm": 2.111147403717041,
|
| 386 |
+
"learning_rate": 0.00013717948717948718,
|
| 387 |
+
"loss": 0.3173,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.282051282051282,
|
| 392 |
+
"grad_norm": 1.6825300455093384,
|
| 393 |
+
"learning_rate": 0.0001358974358974359,
|
| 394 |
+
"loss": 0.2368,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.282051282051282,
|
| 399 |
+
"eval_accuracy": 0.8309090909090909,
|
| 400 |
+
"eval_loss": 0.4046396017074585,
|
| 401 |
+
"eval_runtime": 24.1395,
|
| 402 |
+
"eval_samples_per_second": 45.568,
|
| 403 |
+
"eval_steps_per_second": 5.717,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.3076923076923077,
|
| 408 |
+
"grad_norm": 1.7883163690567017,
|
| 409 |
+
"learning_rate": 0.00013461538461538464,
|
| 410 |
+
"loss": 0.2616,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.3333333333333333,
|
| 415 |
+
"grad_norm": 3.3474502563476562,
|
| 416 |
+
"learning_rate": 0.00013333333333333334,
|
| 417 |
+
"loss": 0.292,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.358974358974359,
|
| 422 |
+
"grad_norm": 1.9872941970825195,
|
| 423 |
+
"learning_rate": 0.00013205128205128204,
|
| 424 |
+
"loss": 0.3284,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.3846153846153846,
|
| 429 |
+
"grad_norm": 1.508928894996643,
|
| 430 |
+
"learning_rate": 0.00013076923076923077,
|
| 431 |
+
"loss": 0.3132,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 1.4102564102564101,
|
| 436 |
+
"grad_norm": 2.3678171634674072,
|
| 437 |
+
"learning_rate": 0.0001294871794871795,
|
| 438 |
+
"loss": 0.2698,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.435897435897436,
|
| 443 |
+
"grad_norm": 3.961099147796631,
|
| 444 |
+
"learning_rate": 0.00012820512820512823,
|
| 445 |
+
"loss": 0.2522,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.4615384615384617,
|
| 450 |
+
"grad_norm": 1.7161500453948975,
|
| 451 |
+
"learning_rate": 0.00012692307692307693,
|
| 452 |
+
"loss": 0.3514,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.4871794871794872,
|
| 457 |
+
"grad_norm": 2.0210063457489014,
|
| 458 |
+
"learning_rate": 0.00012564102564102564,
|
| 459 |
+
"loss": 0.2064,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.5128205128205128,
|
| 464 |
+
"grad_norm": 1.9867080450057983,
|
| 465 |
+
"learning_rate": 0.00012435897435897437,
|
| 466 |
+
"loss": 0.3796,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.5384615384615383,
|
| 471 |
+
"grad_norm": 5.6288981437683105,
|
| 472 |
+
"learning_rate": 0.0001230769230769231,
|
| 473 |
+
"loss": 0.3696,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.5384615384615383,
|
| 478 |
+
"eval_accuracy": 0.7809090909090909,
|
| 479 |
+
"eval_loss": 0.5547047257423401,
|
| 480 |
+
"eval_runtime": 22.8194,
|
| 481 |
+
"eval_samples_per_second": 48.205,
|
| 482 |
+
"eval_steps_per_second": 6.047,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.564102564102564,
|
| 487 |
+
"grad_norm": 2.3821280002593994,
|
| 488 |
+
"learning_rate": 0.00012179487179487179,
|
| 489 |
+
"loss": 0.169,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.5897435897435899,
|
| 494 |
+
"grad_norm": 2.7712907791137695,
|
| 495 |
+
"learning_rate": 0.00012051282051282052,
|
| 496 |
+
"loss": 0.294,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.6153846153846154,
|
| 501 |
+
"grad_norm": 2.803150177001953,
|
| 502 |
+
"learning_rate": 0.00011923076923076923,
|
| 503 |
+
"loss": 0.3544,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.641025641025641,
|
| 508 |
+
"grad_norm": 2.660898447036743,
|
| 509 |
+
"learning_rate": 0.00011794871794871796,
|
| 510 |
+
"loss": 0.2811,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.6666666666666665,
|
| 515 |
+
"grad_norm": 1.4722263813018799,
|
| 516 |
+
"learning_rate": 0.00011666666666666668,
|
| 517 |
+
"loss": 0.26,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.6923076923076923,
|
| 522 |
+
"grad_norm": 2.1373977661132812,
|
| 523 |
+
"learning_rate": 0.00011538461538461538,
|
| 524 |
+
"loss": 0.2282,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.717948717948718,
|
| 529 |
+
"grad_norm": 4.172289848327637,
|
| 530 |
+
"learning_rate": 0.0001141025641025641,
|
| 531 |
+
"loss": 0.2673,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.7435897435897436,
|
| 536 |
+
"grad_norm": 1.547676682472229,
|
| 537 |
+
"learning_rate": 0.00011282051282051283,
|
| 538 |
+
"loss": 0.234,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 1.7692307692307692,
|
| 543 |
+
"grad_norm": 1.9279000759124756,
|
| 544 |
+
"learning_rate": 0.00011153846153846154,
|
| 545 |
+
"loss": 0.4459,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.7948717948717947,
|
| 550 |
+
"grad_norm": 2.7669806480407715,
|
| 551 |
+
"learning_rate": 0.00011025641025641027,
|
| 552 |
+
"loss": 0.2824,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.7948717948717947,
|
| 557 |
+
"eval_accuracy": 0.8518181818181818,
|
| 558 |
+
"eval_loss": 0.33288896083831787,
|
| 559 |
+
"eval_runtime": 23.8646,
|
| 560 |
+
"eval_samples_per_second": 46.093,
|
| 561 |
+
"eval_steps_per_second": 5.783,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.8205128205128205,
|
| 566 |
+
"grad_norm": 1.626448154449463,
|
| 567 |
+
"learning_rate": 0.00010897435897435896,
|
| 568 |
+
"loss": 0.2789,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.8461538461538463,
|
| 573 |
+
"grad_norm": 2.5008246898651123,
|
| 574 |
+
"learning_rate": 0.0001076923076923077,
|
| 575 |
+
"loss": 0.2939,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.8717948717948718,
|
| 580 |
+
"grad_norm": 1.4484879970550537,
|
| 581 |
+
"learning_rate": 0.00010641025641025641,
|
| 582 |
+
"loss": 0.3107,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.8974358974358974,
|
| 587 |
+
"grad_norm": 2.9797451496124268,
|
| 588 |
+
"learning_rate": 0.00010512820512820514,
|
| 589 |
+
"loss": 0.1989,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.9230769230769231,
|
| 594 |
+
"grad_norm": 2.551682710647583,
|
| 595 |
+
"learning_rate": 0.00010384615384615386,
|
| 596 |
+
"loss": 0.3133,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.9487179487179487,
|
| 601 |
+
"grad_norm": 3.318741798400879,
|
| 602 |
+
"learning_rate": 0.00010256410256410256,
|
| 603 |
+
"loss": 0.2384,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.9743589743589745,
|
| 608 |
+
"grad_norm": 0.9309015274047852,
|
| 609 |
+
"learning_rate": 0.00010128205128205129,
|
| 610 |
+
"loss": 0.2314,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.0,
|
| 615 |
+
"grad_norm": 3.10331392288208,
|
| 616 |
+
"learning_rate": 0.0001,
|
| 617 |
+
"loss": 0.2341,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.0256410256410255,
|
| 622 |
+
"grad_norm": 2.6792232990264893,
|
| 623 |
+
"learning_rate": 9.871794871794872e-05,
|
| 624 |
+
"loss": 0.1841,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.051282051282051,
|
| 629 |
+
"grad_norm": 2.074448585510254,
|
| 630 |
+
"learning_rate": 9.743589743589744e-05,
|
| 631 |
+
"loss": 0.2366,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.051282051282051,
|
| 636 |
+
"eval_accuracy": 0.8254545454545454,
|
| 637 |
+
"eval_loss": 0.45823609828948975,
|
| 638 |
+
"eval_runtime": 24.0195,
|
| 639 |
+
"eval_samples_per_second": 45.796,
|
| 640 |
+
"eval_steps_per_second": 5.745,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.076923076923077,
|
| 645 |
+
"grad_norm": 0.9467771649360657,
|
| 646 |
+
"learning_rate": 9.615384615384617e-05,
|
| 647 |
+
"loss": 0.1895,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.1025641025641026,
|
| 652 |
+
"grad_norm": 3.1332082748413086,
|
| 653 |
+
"learning_rate": 9.487179487179487e-05,
|
| 654 |
+
"loss": 0.2665,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.128205128205128,
|
| 659 |
+
"grad_norm": 3.9276282787323,
|
| 660 |
+
"learning_rate": 9.35897435897436e-05,
|
| 661 |
+
"loss": 0.2388,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.1538461538461537,
|
| 666 |
+
"grad_norm": 2.7033755779266357,
|
| 667 |
+
"learning_rate": 9.230769230769232e-05,
|
| 668 |
+
"loss": 0.1917,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.1794871794871793,
|
| 673 |
+
"grad_norm": 1.5250920057296753,
|
| 674 |
+
"learning_rate": 9.102564102564103e-05,
|
| 675 |
+
"loss": 0.3245,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.2051282051282053,
|
| 680 |
+
"grad_norm": 1.5101457834243774,
|
| 681 |
+
"learning_rate": 8.974358974358975e-05,
|
| 682 |
+
"loss": 0.1377,
|
| 683 |
+
"step": 860
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.230769230769231,
|
| 687 |
+
"grad_norm": 0.743198573589325,
|
| 688 |
+
"learning_rate": 8.846153846153847e-05,
|
| 689 |
+
"loss": 0.2243,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.2564102564102564,
|
| 694 |
+
"grad_norm": 5.429717540740967,
|
| 695 |
+
"learning_rate": 8.717948717948718e-05,
|
| 696 |
+
"loss": 0.2683,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.282051282051282,
|
| 701 |
+
"grad_norm": 2.3276283740997314,
|
| 702 |
+
"learning_rate": 8.58974358974359e-05,
|
| 703 |
+
"loss": 0.2083,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.3076923076923075,
|
| 708 |
+
"grad_norm": 1.3464454412460327,
|
| 709 |
+
"learning_rate": 8.461538461538461e-05,
|
| 710 |
+
"loss": 0.2212,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.3076923076923075,
|
| 715 |
+
"eval_accuracy": 0.8254545454545454,
|
| 716 |
+
"eval_loss": 0.4885379374027252,
|
| 717 |
+
"eval_runtime": 24.2062,
|
| 718 |
+
"eval_samples_per_second": 45.443,
|
| 719 |
+
"eval_steps_per_second": 5.701,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.3333333333333335,
|
| 724 |
+
"grad_norm": 2.0947823524475098,
|
| 725 |
+
"learning_rate": 8.333333333333334e-05,
|
| 726 |
+
"loss": 0.1469,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.358974358974359,
|
| 731 |
+
"grad_norm": 2.0833053588867188,
|
| 732 |
+
"learning_rate": 8.205128205128205e-05,
|
| 733 |
+
"loss": 0.1637,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.3846153846153846,
|
| 738 |
+
"grad_norm": 3.5598833560943604,
|
| 739 |
+
"learning_rate": 8.076923076923078e-05,
|
| 740 |
+
"loss": 0.2127,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 2.41025641025641,
|
| 745 |
+
"grad_norm": 5.30457878112793,
|
| 746 |
+
"learning_rate": 7.948717948717948e-05,
|
| 747 |
+
"loss": 0.1561,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.435897435897436,
|
| 752 |
+
"grad_norm": 2.163148880004883,
|
| 753 |
+
"learning_rate": 7.820512820512821e-05,
|
| 754 |
+
"loss": 0.1539,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.4615384615384617,
|
| 759 |
+
"grad_norm": 4.815582752227783,
|
| 760 |
+
"learning_rate": 7.692307692307693e-05,
|
| 761 |
+
"loss": 0.1802,
|
| 762 |
+
"step": 960
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.4871794871794872,
|
| 766 |
+
"grad_norm": 1.3422257900238037,
|
| 767 |
+
"learning_rate": 7.564102564102564e-05,
|
| 768 |
+
"loss": 0.2059,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.5128205128205128,
|
| 773 |
+
"grad_norm": 2.441047430038452,
|
| 774 |
+
"learning_rate": 7.435897435897436e-05,
|
| 775 |
+
"loss": 0.2389,
|
| 776 |
+
"step": 980
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.5384615384615383,
|
| 780 |
+
"grad_norm": 3.386566162109375,
|
| 781 |
+
"learning_rate": 7.307692307692307e-05,
|
| 782 |
+
"loss": 0.1882,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.564102564102564,
|
| 787 |
+
"grad_norm": 3.123497247695923,
|
| 788 |
+
"learning_rate": 7.17948717948718e-05,
|
| 789 |
+
"loss": 0.2031,
|
| 790 |
+
"step": 1000
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.564102564102564,
|
| 794 |
+
"eval_accuracy": 0.8563636363636363,
|
| 795 |
+
"eval_loss": 0.42818111181259155,
|
| 796 |
+
"eval_runtime": 24.1992,
|
| 797 |
+
"eval_samples_per_second": 45.456,
|
| 798 |
+
"eval_steps_per_second": 5.703,
|
| 799 |
+
"step": 1000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.58974358974359,
|
| 803 |
+
"grad_norm": 2.518524646759033,
|
| 804 |
+
"learning_rate": 7.051282051282052e-05,
|
| 805 |
+
"loss": 0.2551,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.6153846153846154,
|
| 810 |
+
"grad_norm": 2.376194715499878,
|
| 811 |
+
"learning_rate": 6.923076923076924e-05,
|
| 812 |
+
"loss": 0.1853,
|
| 813 |
+
"step": 1020
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.641025641025641,
|
| 817 |
+
"grad_norm": 0.8307498097419739,
|
| 818 |
+
"learning_rate": 6.794871794871795e-05,
|
| 819 |
+
"loss": 0.1643,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.6666666666666665,
|
| 824 |
+
"grad_norm": 2.050661087036133,
|
| 825 |
+
"learning_rate": 6.666666666666667e-05,
|
| 826 |
+
"loss": 0.105,
|
| 827 |
+
"step": 1040
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.6923076923076925,
|
| 831 |
+
"grad_norm": 2.984266996383667,
|
| 832 |
+
"learning_rate": 6.538461538461539e-05,
|
| 833 |
+
"loss": 0.1774,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.717948717948718,
|
| 838 |
+
"grad_norm": 3.933162212371826,
|
| 839 |
+
"learning_rate": 6.410256410256412e-05,
|
| 840 |
+
"loss": 0.1079,
|
| 841 |
+
"step": 1060
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.7435897435897436,
|
| 845 |
+
"grad_norm": 4.650693893432617,
|
| 846 |
+
"learning_rate": 6.282051282051282e-05,
|
| 847 |
+
"loss": 0.1542,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.769230769230769,
|
| 852 |
+
"grad_norm": 2.796116828918457,
|
| 853 |
+
"learning_rate": 6.153846153846155e-05,
|
| 854 |
+
"loss": 0.4463,
|
| 855 |
+
"step": 1080
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.7948717948717947,
|
| 859 |
+
"grad_norm": 3.1699883937835693,
|
| 860 |
+
"learning_rate": 6.025641025641026e-05,
|
| 861 |
+
"loss": 0.1348,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 2.8205128205128203,
|
| 866 |
+
"grad_norm": 2.0635199546813965,
|
| 867 |
+
"learning_rate": 5.897435897435898e-05,
|
| 868 |
+
"loss": 0.1717,
|
| 869 |
+
"step": 1100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 2.8205128205128203,
|
| 873 |
+
"eval_accuracy": 0.85,
|
| 874 |
+
"eval_loss": 0.4373130798339844,
|
| 875 |
+
"eval_runtime": 23.6976,
|
| 876 |
+
"eval_samples_per_second": 46.418,
|
| 877 |
+
"eval_steps_per_second": 5.823,
|
| 878 |
+
"step": 1100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 2.8461538461538463,
|
| 882 |
+
"grad_norm": 3.7501204013824463,
|
| 883 |
+
"learning_rate": 5.769230769230769e-05,
|
| 884 |
+
"loss": 0.1505,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.871794871794872,
|
| 889 |
+
"grad_norm": 0.866908609867096,
|
| 890 |
+
"learning_rate": 5.6410256410256414e-05,
|
| 891 |
+
"loss": 0.131,
|
| 892 |
+
"step": 1120
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 2.8974358974358974,
|
| 896 |
+
"grad_norm": 2.7631490230560303,
|
| 897 |
+
"learning_rate": 5.512820512820514e-05,
|
| 898 |
+
"loss": 0.1067,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 2.9230769230769234,
|
| 903 |
+
"grad_norm": 0.8835192918777466,
|
| 904 |
+
"learning_rate": 5.384615384615385e-05,
|
| 905 |
+
"loss": 0.2449,
|
| 906 |
+
"step": 1140
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 2.948717948717949,
|
| 910 |
+
"grad_norm": 0.17269015312194824,
|
| 911 |
+
"learning_rate": 5.256410256410257e-05,
|
| 912 |
+
"loss": 0.1235,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 2.9743589743589745,
|
| 917 |
+
"grad_norm": 2.5380775928497314,
|
| 918 |
+
"learning_rate": 5.128205128205128e-05,
|
| 919 |
+
"loss": 0.1214,
|
| 920 |
+
"step": 1160
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 3.0,
|
| 924 |
+
"grad_norm": 9.240225791931152,
|
| 925 |
+
"learning_rate": 5e-05,
|
| 926 |
+
"loss": 0.2372,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 3.0256410256410255,
|
| 931 |
+
"grad_norm": 0.454428106546402,
|
| 932 |
+
"learning_rate": 4.871794871794872e-05,
|
| 933 |
+
"loss": 0.1121,
|
| 934 |
+
"step": 1180
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 3.051282051282051,
|
| 938 |
+
"grad_norm": 3.3110735416412354,
|
| 939 |
+
"learning_rate": 4.7435897435897435e-05,
|
| 940 |
+
"loss": 0.1121,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 3.076923076923077,
|
| 945 |
+
"grad_norm": 0.4833953380584717,
|
| 946 |
+
"learning_rate": 4.615384615384616e-05,
|
| 947 |
+
"loss": 0.1303,
|
| 948 |
+
"step": 1200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 3.076923076923077,
|
| 952 |
+
"eval_accuracy": 0.8718181818181818,
|
| 953 |
+
"eval_loss": 0.36585894227027893,
|
| 954 |
+
"eval_runtime": 24.2959,
|
| 955 |
+
"eval_samples_per_second": 45.275,
|
| 956 |
+
"eval_steps_per_second": 5.68,
|
| 957 |
+
"step": 1200
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 3.1025641025641026,
|
| 961 |
+
"grad_norm": 0.08100098371505737,
|
| 962 |
+
"learning_rate": 4.4871794871794874e-05,
|
| 963 |
+
"loss": 0.0911,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 3.128205128205128,
|
| 968 |
+
"grad_norm": 0.30585813522338867,
|
| 969 |
+
"learning_rate": 4.358974358974359e-05,
|
| 970 |
+
"loss": 0.0834,
|
| 971 |
+
"step": 1220
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.1538461538461537,
|
| 975 |
+
"grad_norm": 4.129181385040283,
|
| 976 |
+
"learning_rate": 4.230769230769231e-05,
|
| 977 |
+
"loss": 0.1144,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 3.1794871794871793,
|
| 982 |
+
"grad_norm": 0.367727667093277,
|
| 983 |
+
"learning_rate": 4.1025641025641023e-05,
|
| 984 |
+
"loss": 0.0808,
|
| 985 |
+
"step": 1240
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.2051282051282053,
|
| 989 |
+
"grad_norm": 0.10303868353366852,
|
| 990 |
+
"learning_rate": 3.974358974358974e-05,
|
| 991 |
+
"loss": 0.1758,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.230769230769231,
|
| 996 |
+
"grad_norm": 2.300645589828491,
|
| 997 |
+
"learning_rate": 3.846153846153846e-05,
|
| 998 |
+
"loss": 0.227,
|
| 999 |
+
"step": 1260
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 3.2564102564102564,
|
| 1003 |
+
"grad_norm": 1.345780372619629,
|
| 1004 |
+
"learning_rate": 3.717948717948718e-05,
|
| 1005 |
+
"loss": 0.1345,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 3.282051282051282,
|
| 1010 |
+
"grad_norm": 2.5391829013824463,
|
| 1011 |
+
"learning_rate": 3.58974358974359e-05,
|
| 1012 |
+
"loss": 0.0496,
|
| 1013 |
+
"step": 1280
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 3.3076923076923075,
|
| 1017 |
+
"grad_norm": 0.31912463903427124,
|
| 1018 |
+
"learning_rate": 3.461538461538462e-05,
|
| 1019 |
+
"loss": 0.1165,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 3.3333333333333335,
|
| 1024 |
+
"grad_norm": 0.5431106686592102,
|
| 1025 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 1026 |
+
"loss": 0.0889,
|
| 1027 |
+
"step": 1300
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 3.3333333333333335,
|
| 1031 |
+
"eval_accuracy": 0.8736363636363637,
|
| 1032 |
+
"eval_loss": 0.3662668764591217,
|
| 1033 |
+
"eval_runtime": 23.4444,
|
| 1034 |
+
"eval_samples_per_second": 46.92,
|
| 1035 |
+
"eval_steps_per_second": 5.886,
|
| 1036 |
+
"step": 1300
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 3.358974358974359,
|
| 1040 |
+
"grad_norm": 2.443268299102783,
|
| 1041 |
+
"learning_rate": 3.205128205128206e-05,
|
| 1042 |
+
"loss": 0.1256,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 3.3846153846153846,
|
| 1047 |
+
"grad_norm": 2.0804026126861572,
|
| 1048 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 1049 |
+
"loss": 0.0973,
|
| 1050 |
+
"step": 1320
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 3.41025641025641,
|
| 1054 |
+
"grad_norm": 10.397607803344727,
|
| 1055 |
+
"learning_rate": 2.948717948717949e-05,
|
| 1056 |
+
"loss": 0.1183,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 3.435897435897436,
|
| 1061 |
+
"grad_norm": 3.746250867843628,
|
| 1062 |
+
"learning_rate": 2.8205128205128207e-05,
|
| 1063 |
+
"loss": 0.046,
|
| 1064 |
+
"step": 1340
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.4615384615384617,
|
| 1068 |
+
"grad_norm": 0.7118757367134094,
|
| 1069 |
+
"learning_rate": 2.6923076923076923e-05,
|
| 1070 |
+
"loss": 0.1611,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.4871794871794872,
|
| 1075 |
+
"grad_norm": 0.34771645069122314,
|
| 1076 |
+
"learning_rate": 2.564102564102564e-05,
|
| 1077 |
+
"loss": 0.1974,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 3.5128205128205128,
|
| 1082 |
+
"grad_norm": 6.590170860290527,
|
| 1083 |
+
"learning_rate": 2.435897435897436e-05,
|
| 1084 |
+
"loss": 0.1392,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 3.5384615384615383,
|
| 1089 |
+
"grad_norm": 3.6979663372039795,
|
| 1090 |
+
"learning_rate": 2.307692307692308e-05,
|
| 1091 |
+
"loss": 0.1153,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 3.564102564102564,
|
| 1096 |
+
"grad_norm": 0.12197946012020111,
|
| 1097 |
+
"learning_rate": 2.1794871794871795e-05,
|
| 1098 |
+
"loss": 0.1027,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 3.58974358974359,
|
| 1103 |
+
"grad_norm": 2.5246639251708984,
|
| 1104 |
+
"learning_rate": 2.0512820512820512e-05,
|
| 1105 |
+
"loss": 0.1157,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.58974358974359,
|
| 1110 |
+
"eval_accuracy": 0.8436363636363636,
|
| 1111 |
+
"eval_loss": 0.4587700366973877,
|
| 1112 |
+
"eval_runtime": 22.8201,
|
| 1113 |
+
"eval_samples_per_second": 48.203,
|
| 1114 |
+
"eval_steps_per_second": 6.047,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.6153846153846154,
|
| 1119 |
+
"grad_norm": 0.37446674704551697,
|
| 1120 |
+
"learning_rate": 1.923076923076923e-05,
|
| 1121 |
+
"loss": 0.0839,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.641025641025641,
|
| 1126 |
+
"grad_norm": 0.7361642718315125,
|
| 1127 |
+
"learning_rate": 1.794871794871795e-05,
|
| 1128 |
+
"loss": 0.0541,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.6666666666666665,
|
| 1133 |
+
"grad_norm": 0.11162062734365463,
|
| 1134 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 1135 |
+
"loss": 0.1791,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.6923076923076925,
|
| 1140 |
+
"grad_norm": 3.2151377201080322,
|
| 1141 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 1142 |
+
"loss": 0.0597,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.717948717948718,
|
| 1147 |
+
"grad_norm": 0.853471040725708,
|
| 1148 |
+
"learning_rate": 1.4102564102564104e-05,
|
| 1149 |
+
"loss": 0.1246,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.7435897435897436,
|
| 1154 |
+
"grad_norm": 0.2989501953125,
|
| 1155 |
+
"learning_rate": 1.282051282051282e-05,
|
| 1156 |
+
"loss": 0.0747,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.769230769230769,
|
| 1161 |
+
"grad_norm": 0.4194205403327942,
|
| 1162 |
+
"learning_rate": 1.153846153846154e-05,
|
| 1163 |
+
"loss": 0.078,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.7948717948717947,
|
| 1168 |
+
"grad_norm": 0.2623525857925415,
|
| 1169 |
+
"learning_rate": 1.0256410256410256e-05,
|
| 1170 |
+
"loss": 0.064,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.8205128205128203,
|
| 1175 |
+
"grad_norm": 1.1962109804153442,
|
| 1176 |
+
"learning_rate": 8.974358974358976e-06,
|
| 1177 |
+
"loss": 0.0955,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.8461538461538463,
|
| 1182 |
+
"grad_norm": 2.009432792663574,
|
| 1183 |
+
"learning_rate": 7.692307692307694e-06,
|
| 1184 |
+
"loss": 0.1215,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.8461538461538463,
|
| 1189 |
+
"eval_accuracy": 0.8654545454545455,
|
| 1190 |
+
"eval_loss": 0.43503817915916443,
|
| 1191 |
+
"eval_runtime": 23.6622,
|
| 1192 |
+
"eval_samples_per_second": 46.488,
|
| 1193 |
+
"eval_steps_per_second": 5.832,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 3.871794871794872,
|
| 1198 |
+
"grad_norm": 3.284787178039551,
|
| 1199 |
+
"learning_rate": 6.41025641025641e-06,
|
| 1200 |
+
"loss": 0.0614,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 3.8974358974358974,
|
| 1205 |
+
"grad_norm": 0.1390266716480255,
|
| 1206 |
+
"learning_rate": 5.128205128205128e-06,
|
| 1207 |
+
"loss": 0.0795,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 3.9230769230769234,
|
| 1212 |
+
"grad_norm": 3.4633984565734863,
|
| 1213 |
+
"learning_rate": 3.846153846153847e-06,
|
| 1214 |
+
"loss": 0.1268,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 3.948717948717949,
|
| 1219 |
+
"grad_norm": 3.78682804107666,
|
| 1220 |
+
"learning_rate": 2.564102564102564e-06,
|
| 1221 |
+
"loss": 0.1049,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 3.9743589743589745,
|
| 1226 |
+
"grad_norm": 3.6551170349121094,
|
| 1227 |
+
"learning_rate": 1.282051282051282e-06,
|
| 1228 |
+
"loss": 0.0924,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 4.0,
|
| 1233 |
+
"grad_norm": 0.1017698347568512,
|
| 1234 |
+
"learning_rate": 0.0,
|
| 1235 |
+
"loss": 0.1392,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 4.0,
|
| 1240 |
+
"step": 1560,
|
| 1241 |
+
"total_flos": 1.9301704773202575e+18,
|
| 1242 |
+
"train_loss": 0.266804637053074,
|
| 1243 |
+
"train_runtime": 1157.865,
|
| 1244 |
+
"train_samples_per_second": 21.512,
|
| 1245 |
+
"train_steps_per_second": 1.347
|
| 1246 |
+
}
|
| 1247 |
+
],
|
| 1248 |
+
"logging_steps": 10,
|
| 1249 |
+
"max_steps": 1560,
|
| 1250 |
+
"num_input_tokens_seen": 0,
|
| 1251 |
+
"num_train_epochs": 4,
|
| 1252 |
+
"save_steps": 100,
|
| 1253 |
+
"stateful_callbacks": {
|
| 1254 |
+
"TrainerControl": {
|
| 1255 |
+
"args": {
|
| 1256 |
+
"should_epoch_stop": false,
|
| 1257 |
+
"should_evaluate": false,
|
| 1258 |
+
"should_log": false,
|
| 1259 |
+
"should_save": true,
|
| 1260 |
+
"should_training_stop": true
|
| 1261 |
+
},
|
| 1262 |
+
"attributes": {}
|
| 1263 |
+
}
|
| 1264 |
+
},
|
| 1265 |
+
"total_flos": 1.9301704773202575e+18,
|
| 1266 |
+
"train_batch_size": 16,
|
| 1267 |
+
"trial_name": null,
|
| 1268 |
+
"trial_params": null
|
| 1269 |
+
}
|