🍻 cheers
Browse files- README.md +6 -5
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr15_19-54-29_beac81f7e97f/events.out.tfevents.1713214056.beac81f7e97f.34.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1408 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 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 |
# Action_model
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the
|
| 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: google/vit-base-patch16-224-in21k
|
| 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: action_class
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.799047619047619
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# Action_model
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the action_class dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.6551
|
| 37 |
+
- Accuracy: 0.7990
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.799047619047619,
|
| 4 |
+
"eval_loss": 0.6551458239555359,
|
| 5 |
+
"eval_runtime": 15.0447,
|
| 6 |
+
"eval_samples_per_second": 69.792,
|
| 7 |
+
"eval_steps_per_second": 8.774,
|
| 8 |
+
"total_flos": 7.776878731479245e+18,
|
| 9 |
+
"train_loss": 0.2955047018209081,
|
| 10 |
+
"train_runtime": 3022.5495,
|
| 11 |
+
"train_samples_per_second": 33.2,
|
| 12 |
+
"train_steps_per_second": 1.039
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.799047619047619,
|
| 4 |
+
"eval_loss": 0.6551458239555359,
|
| 5 |
+
"eval_runtime": 15.0447,
|
| 6 |
+
"eval_samples_per_second": 69.792,
|
| 7 |
+
"eval_steps_per_second": 8.774
|
| 8 |
+
}
|
runs/Apr15_19-54-29_beac81f7e97f/events.out.tfevents.1713214056.beac81f7e97f.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34aba966447902b2ea0c8fbd33e28307e507b2476396c3ba78ac52c0e8786503
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.776878731479245e+18,
|
| 4 |
+
"train_loss": 0.2955047018209081,
|
| 5 |
+
"train_runtime": 3022.5495,
|
| 6 |
+
"train_samples_per_second": 33.2,
|
| 7 |
+
"train_steps_per_second": 1.039
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6551458239555359,
|
| 3 |
+
"best_model_checkpoint": "Action_model/checkpoint-600",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 3140,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"grad_norm": 1.6790106296539307,
|
| 14 |
+
"learning_rate": 9.936305732484077e-05,
|
| 15 |
+
"loss": 2.1638,
|
| 16 |
+
"step": 20
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.13,
|
| 20 |
+
"grad_norm": 1.8781601190567017,
|
| 21 |
+
"learning_rate": 9.872611464968153e-05,
|
| 22 |
+
"loss": 1.8702,
|
| 23 |
+
"step": 40
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.19,
|
| 27 |
+
"grad_norm": 2.154752016067505,
|
| 28 |
+
"learning_rate": 9.80891719745223e-05,
|
| 29 |
+
"loss": 1.496,
|
| 30 |
+
"step": 60
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.25,
|
| 34 |
+
"grad_norm": 2.715329885482788,
|
| 35 |
+
"learning_rate": 9.745222929936307e-05,
|
| 36 |
+
"loss": 1.2633,
|
| 37 |
+
"step": 80
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.32,
|
| 41 |
+
"grad_norm": 2.8564321994781494,
|
| 42 |
+
"learning_rate": 9.681528662420382e-05,
|
| 43 |
+
"loss": 1.1382,
|
| 44 |
+
"step": 100
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.32,
|
| 48 |
+
"eval_accuracy": 0.7676190476190476,
|
| 49 |
+
"eval_loss": 1.0001901388168335,
|
| 50 |
+
"eval_runtime": 19.3002,
|
| 51 |
+
"eval_samples_per_second": 54.404,
|
| 52 |
+
"eval_steps_per_second": 6.839,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.38,
|
| 57 |
+
"grad_norm": 2.2369894981384277,
|
| 58 |
+
"learning_rate": 9.617834394904459e-05,
|
| 59 |
+
"loss": 1.0084,
|
| 60 |
+
"step": 120
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.45,
|
| 64 |
+
"grad_norm": 3.4310965538024902,
|
| 65 |
+
"learning_rate": 9.554140127388536e-05,
|
| 66 |
+
"loss": 0.9195,
|
| 67 |
+
"step": 140
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.51,
|
| 71 |
+
"grad_norm": 3.0611016750335693,
|
| 72 |
+
"learning_rate": 9.490445859872612e-05,
|
| 73 |
+
"loss": 0.9319,
|
| 74 |
+
"step": 160
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.57,
|
| 78 |
+
"grad_norm": 4.078617095947266,
|
| 79 |
+
"learning_rate": 9.426751592356689e-05,
|
| 80 |
+
"loss": 0.7825,
|
| 81 |
+
"step": 180
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.64,
|
| 85 |
+
"grad_norm": 2.7804274559020996,
|
| 86 |
+
"learning_rate": 9.363057324840766e-05,
|
| 87 |
+
"loss": 0.782,
|
| 88 |
+
"step": 200
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.64,
|
| 92 |
+
"eval_accuracy": 0.7676190476190476,
|
| 93 |
+
"eval_loss": 0.7673064470291138,
|
| 94 |
+
"eval_runtime": 15.1028,
|
| 95 |
+
"eval_samples_per_second": 69.523,
|
| 96 |
+
"eval_steps_per_second": 8.74,
|
| 97 |
+
"step": 200
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.7,
|
| 101 |
+
"grad_norm": 1.4780933856964111,
|
| 102 |
+
"learning_rate": 9.299363057324841e-05,
|
| 103 |
+
"loss": 0.6899,
|
| 104 |
+
"step": 220
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.76,
|
| 108 |
+
"grad_norm": 2.280031681060791,
|
| 109 |
+
"learning_rate": 9.238853503184714e-05,
|
| 110 |
+
"loss": 0.6333,
|
| 111 |
+
"step": 240
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.83,
|
| 115 |
+
"grad_norm": 2.9191763401031494,
|
| 116 |
+
"learning_rate": 9.17515923566879e-05,
|
| 117 |
+
"loss": 0.7143,
|
| 118 |
+
"step": 260
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.89,
|
| 122 |
+
"grad_norm": 2.002995729446411,
|
| 123 |
+
"learning_rate": 9.111464968152866e-05,
|
| 124 |
+
"loss": 0.6578,
|
| 125 |
+
"step": 280
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.96,
|
| 129 |
+
"grad_norm": 3.061410427093506,
|
| 130 |
+
"learning_rate": 9.047770700636943e-05,
|
| 131 |
+
"loss": 0.6289,
|
| 132 |
+
"step": 300
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.96,
|
| 136 |
+
"eval_accuracy": 0.7866666666666666,
|
| 137 |
+
"eval_loss": 0.70728999376297,
|
| 138 |
+
"eval_runtime": 15.1443,
|
| 139 |
+
"eval_samples_per_second": 69.333,
|
| 140 |
+
"eval_steps_per_second": 8.716,
|
| 141 |
+
"step": 300
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.02,
|
| 145 |
+
"grad_norm": 3.693359375,
|
| 146 |
+
"learning_rate": 8.984076433121019e-05,
|
| 147 |
+
"loss": 0.5495,
|
| 148 |
+
"step": 320
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.08,
|
| 152 |
+
"grad_norm": 3.992403507232666,
|
| 153 |
+
"learning_rate": 8.920382165605096e-05,
|
| 154 |
+
"loss": 0.5563,
|
| 155 |
+
"step": 340
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.15,
|
| 159 |
+
"grad_norm": 2.6919829845428467,
|
| 160 |
+
"learning_rate": 8.856687898089173e-05,
|
| 161 |
+
"loss": 0.5681,
|
| 162 |
+
"step": 360
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.21,
|
| 166 |
+
"grad_norm": 3.0533978939056396,
|
| 167 |
+
"learning_rate": 8.796178343949045e-05,
|
| 168 |
+
"loss": 0.5286,
|
| 169 |
+
"step": 380
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.27,
|
| 173 |
+
"grad_norm": 2.9173035621643066,
|
| 174 |
+
"learning_rate": 8.732484076433122e-05,
|
| 175 |
+
"loss": 0.5028,
|
| 176 |
+
"step": 400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.27,
|
| 180 |
+
"eval_accuracy": 0.7685714285714286,
|
| 181 |
+
"eval_loss": 0.7260778546333313,
|
| 182 |
+
"eval_runtime": 17.5337,
|
| 183 |
+
"eval_samples_per_second": 59.885,
|
| 184 |
+
"eval_steps_per_second": 7.528,
|
| 185 |
+
"step": 400
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.34,
|
| 189 |
+
"grad_norm": 1.7928365468978882,
|
| 190 |
+
"learning_rate": 8.668789808917198e-05,
|
| 191 |
+
"loss": 0.4379,
|
| 192 |
+
"step": 420
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.4,
|
| 196 |
+
"grad_norm": 4.260186195373535,
|
| 197 |
+
"learning_rate": 8.605095541401275e-05,
|
| 198 |
+
"loss": 0.48,
|
| 199 |
+
"step": 440
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.46,
|
| 203 |
+
"grad_norm": 2.8024277687072754,
|
| 204 |
+
"learning_rate": 8.541401273885352e-05,
|
| 205 |
+
"loss": 0.3689,
|
| 206 |
+
"step": 460
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.53,
|
| 210 |
+
"grad_norm": 4.045362949371338,
|
| 211 |
+
"learning_rate": 8.477707006369427e-05,
|
| 212 |
+
"loss": 0.3922,
|
| 213 |
+
"step": 480
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.59,
|
| 217 |
+
"grad_norm": 1.6472926139831543,
|
| 218 |
+
"learning_rate": 8.414012738853504e-05,
|
| 219 |
+
"loss": 0.4746,
|
| 220 |
+
"step": 500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.59,
|
| 224 |
+
"eval_accuracy": 0.7619047619047619,
|
| 225 |
+
"eval_loss": 0.7463707327842712,
|
| 226 |
+
"eval_runtime": 15.5218,
|
| 227 |
+
"eval_samples_per_second": 67.647,
|
| 228 |
+
"eval_steps_per_second": 8.504,
|
| 229 |
+
"step": 500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.66,
|
| 233 |
+
"grad_norm": 4.001276969909668,
|
| 234 |
+
"learning_rate": 8.350318471337581e-05,
|
| 235 |
+
"loss": 0.4123,
|
| 236 |
+
"step": 520
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.72,
|
| 240 |
+
"grad_norm": 4.151864528656006,
|
| 241 |
+
"learning_rate": 8.286624203821657e-05,
|
| 242 |
+
"loss": 0.4626,
|
| 243 |
+
"step": 540
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.78,
|
| 247 |
+
"grad_norm": 3.43729567527771,
|
| 248 |
+
"learning_rate": 8.222929936305733e-05,
|
| 249 |
+
"loss": 0.4279,
|
| 250 |
+
"step": 560
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.85,
|
| 254 |
+
"grad_norm": 3.8993654251098633,
|
| 255 |
+
"learning_rate": 8.159235668789809e-05,
|
| 256 |
+
"loss": 0.4288,
|
| 257 |
+
"step": 580
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.91,
|
| 261 |
+
"grad_norm": 3.0417702198028564,
|
| 262 |
+
"learning_rate": 8.095541401273886e-05,
|
| 263 |
+
"loss": 0.4298,
|
| 264 |
+
"step": 600
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.91,
|
| 268 |
+
"eval_accuracy": 0.799047619047619,
|
| 269 |
+
"eval_loss": 0.6551458239555359,
|
| 270 |
+
"eval_runtime": 15.0487,
|
| 271 |
+
"eval_samples_per_second": 69.773,
|
| 272 |
+
"eval_steps_per_second": 8.771,
|
| 273 |
+
"step": 600
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.97,
|
| 277 |
+
"grad_norm": 4.088045597076416,
|
| 278 |
+
"learning_rate": 8.031847133757963e-05,
|
| 279 |
+
"loss": 0.404,
|
| 280 |
+
"step": 620
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.04,
|
| 284 |
+
"grad_norm": 2.1314449310302734,
|
| 285 |
+
"learning_rate": 7.968152866242038e-05,
|
| 286 |
+
"loss": 0.4096,
|
| 287 |
+
"step": 640
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 2.1,
|
| 291 |
+
"grad_norm": 5.116842746734619,
|
| 292 |
+
"learning_rate": 7.904458598726115e-05,
|
| 293 |
+
"loss": 0.3332,
|
| 294 |
+
"step": 660
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.17,
|
| 298 |
+
"grad_norm": 4.045914649963379,
|
| 299 |
+
"learning_rate": 7.840764331210192e-05,
|
| 300 |
+
"loss": 0.2961,
|
| 301 |
+
"step": 680
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 2.23,
|
| 305 |
+
"grad_norm": 1.9349473714828491,
|
| 306 |
+
"learning_rate": 7.777070063694268e-05,
|
| 307 |
+
"loss": 0.3488,
|
| 308 |
+
"step": 700
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.23,
|
| 312 |
+
"eval_accuracy": 0.7733333333333333,
|
| 313 |
+
"eval_loss": 0.7358552813529968,
|
| 314 |
+
"eval_runtime": 15.4659,
|
| 315 |
+
"eval_samples_per_second": 67.891,
|
| 316 |
+
"eval_steps_per_second": 8.535,
|
| 317 |
+
"step": 700
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.29,
|
| 321 |
+
"grad_norm": 4.3049845695495605,
|
| 322 |
+
"learning_rate": 7.713375796178345e-05,
|
| 323 |
+
"loss": 0.3078,
|
| 324 |
+
"step": 720
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.36,
|
| 328 |
+
"grad_norm": 2.4363415241241455,
|
| 329 |
+
"learning_rate": 7.649681528662422e-05,
|
| 330 |
+
"loss": 0.4005,
|
| 331 |
+
"step": 740
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.42,
|
| 335 |
+
"grad_norm": 3.518944501876831,
|
| 336 |
+
"learning_rate": 7.585987261146497e-05,
|
| 337 |
+
"loss": 0.3428,
|
| 338 |
+
"step": 760
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.48,
|
| 342 |
+
"grad_norm": 3.8007328510284424,
|
| 343 |
+
"learning_rate": 7.522292993630574e-05,
|
| 344 |
+
"loss": 0.3471,
|
| 345 |
+
"step": 780
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.55,
|
| 349 |
+
"grad_norm": 1.3215352296829224,
|
| 350 |
+
"learning_rate": 7.45859872611465e-05,
|
| 351 |
+
"loss": 0.266,
|
| 352 |
+
"step": 800
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.55,
|
| 356 |
+
"eval_accuracy": 0.7514285714285714,
|
| 357 |
+
"eval_loss": 0.829559862613678,
|
| 358 |
+
"eval_runtime": 15.0559,
|
| 359 |
+
"eval_samples_per_second": 69.74,
|
| 360 |
+
"eval_steps_per_second": 8.767,
|
| 361 |
+
"step": 800
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.61,
|
| 365 |
+
"grad_norm": 1.9990004301071167,
|
| 366 |
+
"learning_rate": 7.394904458598727e-05,
|
| 367 |
+
"loss": 0.2918,
|
| 368 |
+
"step": 820
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.68,
|
| 372 |
+
"grad_norm": 5.201882362365723,
|
| 373 |
+
"learning_rate": 7.331210191082802e-05,
|
| 374 |
+
"loss": 0.3175,
|
| 375 |
+
"step": 840
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.74,
|
| 379 |
+
"grad_norm": 4.187939167022705,
|
| 380 |
+
"learning_rate": 7.267515923566879e-05,
|
| 381 |
+
"loss": 0.304,
|
| 382 |
+
"step": 860
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.8,
|
| 386 |
+
"grad_norm": 2.6038854122161865,
|
| 387 |
+
"learning_rate": 7.203821656050955e-05,
|
| 388 |
+
"loss": 0.3119,
|
| 389 |
+
"step": 880
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.87,
|
| 393 |
+
"grad_norm": 3.0886316299438477,
|
| 394 |
+
"learning_rate": 7.140127388535032e-05,
|
| 395 |
+
"loss": 0.3651,
|
| 396 |
+
"step": 900
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.87,
|
| 400 |
+
"eval_accuracy": 0.7304761904761905,
|
| 401 |
+
"eval_loss": 0.8660680651664734,
|
| 402 |
+
"eval_runtime": 15.7752,
|
| 403 |
+
"eval_samples_per_second": 66.56,
|
| 404 |
+
"eval_steps_per_second": 8.368,
|
| 405 |
+
"step": 900
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.93,
|
| 409 |
+
"grad_norm": 3.3455846309661865,
|
| 410 |
+
"learning_rate": 7.076433121019108e-05,
|
| 411 |
+
"loss": 0.2739,
|
| 412 |
+
"step": 920
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 2.99,
|
| 416 |
+
"grad_norm": 3.0882346630096436,
|
| 417 |
+
"learning_rate": 7.012738853503184e-05,
|
| 418 |
+
"loss": 0.2327,
|
| 419 |
+
"step": 940
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.06,
|
| 423 |
+
"grad_norm": 4.43536901473999,
|
| 424 |
+
"learning_rate": 6.949044585987261e-05,
|
| 425 |
+
"loss": 0.2699,
|
| 426 |
+
"step": 960
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 3.12,
|
| 430 |
+
"grad_norm": 3.8387985229492188,
|
| 431 |
+
"learning_rate": 6.885350318471338e-05,
|
| 432 |
+
"loss": 0.295,
|
| 433 |
+
"step": 980
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 3.18,
|
| 437 |
+
"grad_norm": 2.8090808391571045,
|
| 438 |
+
"learning_rate": 6.821656050955413e-05,
|
| 439 |
+
"loss": 0.2796,
|
| 440 |
+
"step": 1000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 3.18,
|
| 444 |
+
"eval_accuracy": 0.7866666666666666,
|
| 445 |
+
"eval_loss": 0.7188078165054321,
|
| 446 |
+
"eval_runtime": 15.2769,
|
| 447 |
+
"eval_samples_per_second": 68.731,
|
| 448 |
+
"eval_steps_per_second": 8.641,
|
| 449 |
+
"step": 1000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.25,
|
| 453 |
+
"grad_norm": 2.789705991744995,
|
| 454 |
+
"learning_rate": 6.75796178343949e-05,
|
| 455 |
+
"loss": 0.234,
|
| 456 |
+
"step": 1020
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 3.31,
|
| 460 |
+
"grad_norm": 4.02871561050415,
|
| 461 |
+
"learning_rate": 6.694267515923567e-05,
|
| 462 |
+
"loss": 0.2282,
|
| 463 |
+
"step": 1040
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.38,
|
| 467 |
+
"grad_norm": 2.786869764328003,
|
| 468 |
+
"learning_rate": 6.630573248407643e-05,
|
| 469 |
+
"loss": 0.3052,
|
| 470 |
+
"step": 1060
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.44,
|
| 474 |
+
"grad_norm": 3.5847015380859375,
|
| 475 |
+
"learning_rate": 6.56687898089172e-05,
|
| 476 |
+
"loss": 0.2343,
|
| 477 |
+
"step": 1080
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.5,
|
| 481 |
+
"grad_norm": 2.2771642208099365,
|
| 482 |
+
"learning_rate": 6.503184713375797e-05,
|
| 483 |
+
"loss": 0.2703,
|
| 484 |
+
"step": 1100
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 3.5,
|
| 488 |
+
"eval_accuracy": 0.7476190476190476,
|
| 489 |
+
"eval_loss": 0.8421508073806763,
|
| 490 |
+
"eval_runtime": 15.1919,
|
| 491 |
+
"eval_samples_per_second": 69.116,
|
| 492 |
+
"eval_steps_per_second": 8.689,
|
| 493 |
+
"step": 1100
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 3.57,
|
| 497 |
+
"grad_norm": 3.3567535877227783,
|
| 498 |
+
"learning_rate": 6.439490445859872e-05,
|
| 499 |
+
"loss": 0.2429,
|
| 500 |
+
"step": 1120
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 3.63,
|
| 504 |
+
"grad_norm": 4.021266937255859,
|
| 505 |
+
"learning_rate": 6.375796178343949e-05,
|
| 506 |
+
"loss": 0.2875,
|
| 507 |
+
"step": 1140
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.69,
|
| 511 |
+
"grad_norm": 1.5646867752075195,
|
| 512 |
+
"learning_rate": 6.312101910828026e-05,
|
| 513 |
+
"loss": 0.2355,
|
| 514 |
+
"step": 1160
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 3.76,
|
| 518 |
+
"grad_norm": 0.8209928274154663,
|
| 519 |
+
"learning_rate": 6.248407643312102e-05,
|
| 520 |
+
"loss": 0.2367,
|
| 521 |
+
"step": 1180
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 3.82,
|
| 525 |
+
"grad_norm": 5.591761589050293,
|
| 526 |
+
"learning_rate": 6.184713375796178e-05,
|
| 527 |
+
"loss": 0.2608,
|
| 528 |
+
"step": 1200
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 3.82,
|
| 532 |
+
"eval_accuracy": 0.7723809523809524,
|
| 533 |
+
"eval_loss": 0.8207409381866455,
|
| 534 |
+
"eval_runtime": 15.3778,
|
| 535 |
+
"eval_samples_per_second": 68.28,
|
| 536 |
+
"eval_steps_per_second": 8.584,
|
| 537 |
+
"step": 1200
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.89,
|
| 541 |
+
"grad_norm": 2.2955307960510254,
|
| 542 |
+
"learning_rate": 6.121019108280255e-05,
|
| 543 |
+
"loss": 0.2174,
|
| 544 |
+
"step": 1220
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 3.95,
|
| 548 |
+
"grad_norm": 4.40664005279541,
|
| 549 |
+
"learning_rate": 6.057324840764331e-05,
|
| 550 |
+
"loss": 0.2168,
|
| 551 |
+
"step": 1240
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 4.01,
|
| 555 |
+
"grad_norm": 4.821913719177246,
|
| 556 |
+
"learning_rate": 5.993630573248408e-05,
|
| 557 |
+
"loss": 0.251,
|
| 558 |
+
"step": 1260
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 4.08,
|
| 562 |
+
"grad_norm": 6.526182174682617,
|
| 563 |
+
"learning_rate": 5.929936305732484e-05,
|
| 564 |
+
"loss": 0.2424,
|
| 565 |
+
"step": 1280
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 4.14,
|
| 569 |
+
"grad_norm": 1.996484398841858,
|
| 570 |
+
"learning_rate": 5.86624203821656e-05,
|
| 571 |
+
"loss": 0.251,
|
| 572 |
+
"step": 1300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 4.14,
|
| 576 |
+
"eval_accuracy": 0.7266666666666667,
|
| 577 |
+
"eval_loss": 1.0251611471176147,
|
| 578 |
+
"eval_runtime": 14.903,
|
| 579 |
+
"eval_samples_per_second": 70.455,
|
| 580 |
+
"eval_steps_per_second": 8.857,
|
| 581 |
+
"step": 1300
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 4.2,
|
| 585 |
+
"grad_norm": 2.93237566947937,
|
| 586 |
+
"learning_rate": 5.802547770700637e-05,
|
| 587 |
+
"loss": 0.1727,
|
| 588 |
+
"step": 1320
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 4.27,
|
| 592 |
+
"grad_norm": 3.0160069465637207,
|
| 593 |
+
"learning_rate": 5.7388535031847135e-05,
|
| 594 |
+
"loss": 0.2049,
|
| 595 |
+
"step": 1340
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.33,
|
| 599 |
+
"grad_norm": 2.878361701965332,
|
| 600 |
+
"learning_rate": 5.67515923566879e-05,
|
| 601 |
+
"loss": 0.2221,
|
| 602 |
+
"step": 1360
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 4.39,
|
| 606 |
+
"grad_norm": 3.4867329597473145,
|
| 607 |
+
"learning_rate": 5.6114649681528666e-05,
|
| 608 |
+
"loss": 0.2459,
|
| 609 |
+
"step": 1380
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.46,
|
| 613 |
+
"grad_norm": 4.972071170806885,
|
| 614 |
+
"learning_rate": 5.547770700636943e-05,
|
| 615 |
+
"loss": 0.2085,
|
| 616 |
+
"step": 1400
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 4.46,
|
| 620 |
+
"eval_accuracy": 0.7171428571428572,
|
| 621 |
+
"eval_loss": 1.0474802255630493,
|
| 622 |
+
"eval_runtime": 15.353,
|
| 623 |
+
"eval_samples_per_second": 68.391,
|
| 624 |
+
"eval_steps_per_second": 8.598,
|
| 625 |
+
"step": 1400
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 4.52,
|
| 629 |
+
"grad_norm": 3.7226266860961914,
|
| 630 |
+
"learning_rate": 5.484076433121019e-05,
|
| 631 |
+
"loss": 0.22,
|
| 632 |
+
"step": 1420
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 4.59,
|
| 636 |
+
"grad_norm": 3.0898613929748535,
|
| 637 |
+
"learning_rate": 5.420382165605096e-05,
|
| 638 |
+
"loss": 0.206,
|
| 639 |
+
"step": 1440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 4.65,
|
| 643 |
+
"grad_norm": 5.401129722595215,
|
| 644 |
+
"learning_rate": 5.356687898089172e-05,
|
| 645 |
+
"loss": 0.2215,
|
| 646 |
+
"step": 1460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 4.71,
|
| 650 |
+
"grad_norm": 3.430591344833374,
|
| 651 |
+
"learning_rate": 5.2929936305732485e-05,
|
| 652 |
+
"loss": 0.1883,
|
| 653 |
+
"step": 1480
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 4.78,
|
| 657 |
+
"grad_norm": 0.33961915969848633,
|
| 658 |
+
"learning_rate": 5.229299363057325e-05,
|
| 659 |
+
"loss": 0.1715,
|
| 660 |
+
"step": 1500
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 4.78,
|
| 664 |
+
"eval_accuracy": 0.7495238095238095,
|
| 665 |
+
"eval_loss": 0.8852301836013794,
|
| 666 |
+
"eval_runtime": 15.0006,
|
| 667 |
+
"eval_samples_per_second": 69.997,
|
| 668 |
+
"eval_steps_per_second": 8.8,
|
| 669 |
+
"step": 1500
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.84,
|
| 673 |
+
"grad_norm": 3.065584182739258,
|
| 674 |
+
"learning_rate": 5.1656050955414016e-05,
|
| 675 |
+
"loss": 0.1687,
|
| 676 |
+
"step": 1520
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 4.9,
|
| 680 |
+
"grad_norm": 4.120232582092285,
|
| 681 |
+
"learning_rate": 5.101910828025478e-05,
|
| 682 |
+
"loss": 0.214,
|
| 683 |
+
"step": 1540
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 4.97,
|
| 687 |
+
"grad_norm": 0.5130987763404846,
|
| 688 |
+
"learning_rate": 5.038216560509554e-05,
|
| 689 |
+
"loss": 0.1631,
|
| 690 |
+
"step": 1560
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 5.03,
|
| 694 |
+
"grad_norm": 4.085451126098633,
|
| 695 |
+
"learning_rate": 4.974522292993631e-05,
|
| 696 |
+
"loss": 0.1907,
|
| 697 |
+
"step": 1580
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 5.1,
|
| 701 |
+
"grad_norm": 3.028500556945801,
|
| 702 |
+
"learning_rate": 4.910828025477707e-05,
|
| 703 |
+
"loss": 0.2051,
|
| 704 |
+
"step": 1600
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 5.1,
|
| 708 |
+
"eval_accuracy": 0.7790476190476191,
|
| 709 |
+
"eval_loss": 0.8164414763450623,
|
| 710 |
+
"eval_runtime": 15.3889,
|
| 711 |
+
"eval_samples_per_second": 68.231,
|
| 712 |
+
"eval_steps_per_second": 8.578,
|
| 713 |
+
"step": 1600
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 5.16,
|
| 717 |
+
"grad_norm": 2.5563900470733643,
|
| 718 |
+
"learning_rate": 4.8471337579617835e-05,
|
| 719 |
+
"loss": 0.187,
|
| 720 |
+
"step": 1620
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 5.22,
|
| 724 |
+
"grad_norm": 3.853022813796997,
|
| 725 |
+
"learning_rate": 4.7834394904458604e-05,
|
| 726 |
+
"loss": 0.2186,
|
| 727 |
+
"step": 1640
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 5.29,
|
| 731 |
+
"grad_norm": 6.835115909576416,
|
| 732 |
+
"learning_rate": 4.7197452229299366e-05,
|
| 733 |
+
"loss": 0.1717,
|
| 734 |
+
"step": 1660
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 5.35,
|
| 738 |
+
"grad_norm": 3.7477526664733887,
|
| 739 |
+
"learning_rate": 4.656050955414013e-05,
|
| 740 |
+
"loss": 0.2352,
|
| 741 |
+
"step": 1680
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 5.41,
|
| 745 |
+
"grad_norm": 3.5091373920440674,
|
| 746 |
+
"learning_rate": 4.59235668789809e-05,
|
| 747 |
+
"loss": 0.1481,
|
| 748 |
+
"step": 1700
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 5.41,
|
| 752 |
+
"eval_accuracy": 0.7628571428571429,
|
| 753 |
+
"eval_loss": 0.8825291991233826,
|
| 754 |
+
"eval_runtime": 15.2694,
|
| 755 |
+
"eval_samples_per_second": 68.765,
|
| 756 |
+
"eval_steps_per_second": 8.645,
|
| 757 |
+
"step": 1700
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 5.48,
|
| 761 |
+
"grad_norm": 4.236053943634033,
|
| 762 |
+
"learning_rate": 4.528662420382166e-05,
|
| 763 |
+
"loss": 0.1482,
|
| 764 |
+
"step": 1720
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 5.54,
|
| 768 |
+
"grad_norm": 3.335090160369873,
|
| 769 |
+
"learning_rate": 4.464968152866242e-05,
|
| 770 |
+
"loss": 0.1399,
|
| 771 |
+
"step": 1740
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 5.61,
|
| 775 |
+
"grad_norm": 2.238372802734375,
|
| 776 |
+
"learning_rate": 4.4012738853503185e-05,
|
| 777 |
+
"loss": 0.1664,
|
| 778 |
+
"step": 1760
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 5.67,
|
| 782 |
+
"grad_norm": 1.2184184789657593,
|
| 783 |
+
"learning_rate": 4.3375796178343954e-05,
|
| 784 |
+
"loss": 0.179,
|
| 785 |
+
"step": 1780
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 5.73,
|
| 789 |
+
"grad_norm": 7.519371509552002,
|
| 790 |
+
"learning_rate": 4.2738853503184716e-05,
|
| 791 |
+
"loss": 0.177,
|
| 792 |
+
"step": 1800
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 5.73,
|
| 796 |
+
"eval_accuracy": 0.7866666666666666,
|
| 797 |
+
"eval_loss": 0.8622841238975525,
|
| 798 |
+
"eval_runtime": 16.3495,
|
| 799 |
+
"eval_samples_per_second": 64.222,
|
| 800 |
+
"eval_steps_per_second": 8.074,
|
| 801 |
+
"step": 1800
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 5.8,
|
| 805 |
+
"grad_norm": 3.128350019454956,
|
| 806 |
+
"learning_rate": 4.210191082802548e-05,
|
| 807 |
+
"loss": 0.1943,
|
| 808 |
+
"step": 1820
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 5.86,
|
| 812 |
+
"grad_norm": 3.5789175033569336,
|
| 813 |
+
"learning_rate": 4.146496815286625e-05,
|
| 814 |
+
"loss": 0.1431,
|
| 815 |
+
"step": 1840
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 5.92,
|
| 819 |
+
"grad_norm": 5.671403408050537,
|
| 820 |
+
"learning_rate": 4.082802547770701e-05,
|
| 821 |
+
"loss": 0.1761,
|
| 822 |
+
"step": 1860
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 5.99,
|
| 826 |
+
"grad_norm": 2.8949170112609863,
|
| 827 |
+
"learning_rate": 4.019108280254777e-05,
|
| 828 |
+
"loss": 0.1639,
|
| 829 |
+
"step": 1880
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 6.05,
|
| 833 |
+
"grad_norm": 6.811347961425781,
|
| 834 |
+
"learning_rate": 3.955414012738854e-05,
|
| 835 |
+
"loss": 0.1607,
|
| 836 |
+
"step": 1900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 6.05,
|
| 840 |
+
"eval_accuracy": 0.7609523809523809,
|
| 841 |
+
"eval_loss": 0.948749303817749,
|
| 842 |
+
"eval_runtime": 15.6299,
|
| 843 |
+
"eval_samples_per_second": 67.179,
|
| 844 |
+
"eval_steps_per_second": 8.445,
|
| 845 |
+
"step": 1900
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 6.11,
|
| 849 |
+
"grad_norm": 3.2780776023864746,
|
| 850 |
+
"learning_rate": 3.8917197452229304e-05,
|
| 851 |
+
"loss": 0.2392,
|
| 852 |
+
"step": 1920
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 6.18,
|
| 856 |
+
"grad_norm": 1.425671935081482,
|
| 857 |
+
"learning_rate": 3.8280254777070066e-05,
|
| 858 |
+
"loss": 0.1988,
|
| 859 |
+
"step": 1940
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 6.24,
|
| 863 |
+
"grad_norm": 4.055123329162598,
|
| 864 |
+
"learning_rate": 3.7643312101910836e-05,
|
| 865 |
+
"loss": 0.1563,
|
| 866 |
+
"step": 1960
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 6.31,
|
| 870 |
+
"grad_norm": 5.6207356452941895,
|
| 871 |
+
"learning_rate": 3.700636942675159e-05,
|
| 872 |
+
"loss": 0.1364,
|
| 873 |
+
"step": 1980
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 6.37,
|
| 877 |
+
"grad_norm": 0.6465654373168945,
|
| 878 |
+
"learning_rate": 3.6369426751592353e-05,
|
| 879 |
+
"loss": 0.1273,
|
| 880 |
+
"step": 2000
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 6.37,
|
| 884 |
+
"eval_accuracy": 0.7733333333333333,
|
| 885 |
+
"eval_loss": 0.8984624743461609,
|
| 886 |
+
"eval_runtime": 15.237,
|
| 887 |
+
"eval_samples_per_second": 68.911,
|
| 888 |
+
"eval_steps_per_second": 8.663,
|
| 889 |
+
"step": 2000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 6.43,
|
| 893 |
+
"grad_norm": 3.3592026233673096,
|
| 894 |
+
"learning_rate": 3.573248407643312e-05,
|
| 895 |
+
"loss": 0.1903,
|
| 896 |
+
"step": 2020
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 6.5,
|
| 900 |
+
"grad_norm": 2.187608480453491,
|
| 901 |
+
"learning_rate": 3.5095541401273885e-05,
|
| 902 |
+
"loss": 0.1684,
|
| 903 |
+
"step": 2040
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 6.56,
|
| 907 |
+
"grad_norm": 2.657270908355713,
|
| 908 |
+
"learning_rate": 3.445859872611465e-05,
|
| 909 |
+
"loss": 0.1619,
|
| 910 |
+
"step": 2060
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 6.62,
|
| 914 |
+
"grad_norm": 2.2679970264434814,
|
| 915 |
+
"learning_rate": 3.3821656050955416e-05,
|
| 916 |
+
"loss": 0.1556,
|
| 917 |
+
"step": 2080
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 6.69,
|
| 921 |
+
"grad_norm": 1.9460710287094116,
|
| 922 |
+
"learning_rate": 3.318471337579618e-05,
|
| 923 |
+
"loss": 0.1609,
|
| 924 |
+
"step": 2100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 6.69,
|
| 928 |
+
"eval_accuracy": 0.7504761904761905,
|
| 929 |
+
"eval_loss": 0.9624072313308716,
|
| 930 |
+
"eval_runtime": 15.076,
|
| 931 |
+
"eval_samples_per_second": 69.647,
|
| 932 |
+
"eval_steps_per_second": 8.756,
|
| 933 |
+
"step": 2100
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 6.75,
|
| 937 |
+
"grad_norm": 2.056673526763916,
|
| 938 |
+
"learning_rate": 3.254777070063694e-05,
|
| 939 |
+
"loss": 0.1779,
|
| 940 |
+
"step": 2120
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 6.82,
|
| 944 |
+
"grad_norm": 2.4007034301757812,
|
| 945 |
+
"learning_rate": 3.191082802547771e-05,
|
| 946 |
+
"loss": 0.1359,
|
| 947 |
+
"step": 2140
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 6.88,
|
| 951 |
+
"grad_norm": 6.746215343475342,
|
| 952 |
+
"learning_rate": 3.127388535031847e-05,
|
| 953 |
+
"loss": 0.1653,
|
| 954 |
+
"step": 2160
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 6.94,
|
| 958 |
+
"grad_norm": 3.8807878494262695,
|
| 959 |
+
"learning_rate": 3.0636942675159235e-05,
|
| 960 |
+
"loss": 0.1434,
|
| 961 |
+
"step": 2180
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 7.01,
|
| 965 |
+
"grad_norm": 4.5821990966796875,
|
| 966 |
+
"learning_rate": 3e-05,
|
| 967 |
+
"loss": 0.1583,
|
| 968 |
+
"step": 2200
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 7.01,
|
| 972 |
+
"eval_accuracy": 0.7780952380952381,
|
| 973 |
+
"eval_loss": 0.9015449285507202,
|
| 974 |
+
"eval_runtime": 15.3184,
|
| 975 |
+
"eval_samples_per_second": 68.545,
|
| 976 |
+
"eval_steps_per_second": 8.617,
|
| 977 |
+
"step": 2200
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 7.07,
|
| 981 |
+
"grad_norm": 3.755427837371826,
|
| 982 |
+
"learning_rate": 2.9363057324840763e-05,
|
| 983 |
+
"loss": 0.1251,
|
| 984 |
+
"step": 2220
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 7.13,
|
| 988 |
+
"grad_norm": 2.709980010986328,
|
| 989 |
+
"learning_rate": 2.872611464968153e-05,
|
| 990 |
+
"loss": 0.1233,
|
| 991 |
+
"step": 2240
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 7.2,
|
| 995 |
+
"grad_norm": 2.7730648517608643,
|
| 996 |
+
"learning_rate": 2.8089171974522295e-05,
|
| 997 |
+
"loss": 0.1019,
|
| 998 |
+
"step": 2260
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 7.26,
|
| 1002 |
+
"grad_norm": 4.75289249420166,
|
| 1003 |
+
"learning_rate": 2.7452229299363057e-05,
|
| 1004 |
+
"loss": 0.1862,
|
| 1005 |
+
"step": 2280
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 7.32,
|
| 1009 |
+
"grad_norm": 2.336014747619629,
|
| 1010 |
+
"learning_rate": 2.6815286624203823e-05,
|
| 1011 |
+
"loss": 0.1178,
|
| 1012 |
+
"step": 2300
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 7.32,
|
| 1016 |
+
"eval_accuracy": 0.7761904761904762,
|
| 1017 |
+
"eval_loss": 0.9142788648605347,
|
| 1018 |
+
"eval_runtime": 15.5959,
|
| 1019 |
+
"eval_samples_per_second": 67.325,
|
| 1020 |
+
"eval_steps_per_second": 8.464,
|
| 1021 |
+
"step": 2300
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 7.39,
|
| 1025 |
+
"grad_norm": 5.943305015563965,
|
| 1026 |
+
"learning_rate": 2.617834394904459e-05,
|
| 1027 |
+
"loss": 0.146,
|
| 1028 |
+
"step": 2320
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 7.45,
|
| 1032 |
+
"grad_norm": 1.1991711854934692,
|
| 1033 |
+
"learning_rate": 2.554140127388535e-05,
|
| 1034 |
+
"loss": 0.1185,
|
| 1035 |
+
"step": 2340
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 7.52,
|
| 1039 |
+
"grad_norm": 1.5695414543151855,
|
| 1040 |
+
"learning_rate": 2.4904458598726117e-05,
|
| 1041 |
+
"loss": 0.1437,
|
| 1042 |
+
"step": 2360
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 7.58,
|
| 1046 |
+
"grad_norm": 0.3648098111152649,
|
| 1047 |
+
"learning_rate": 2.426751592356688e-05,
|
| 1048 |
+
"loss": 0.1028,
|
| 1049 |
+
"step": 2380
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 7.64,
|
| 1053 |
+
"grad_norm": 2.0215370655059814,
|
| 1054 |
+
"learning_rate": 2.3630573248407645e-05,
|
| 1055 |
+
"loss": 0.1175,
|
| 1056 |
+
"step": 2400
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 7.64,
|
| 1060 |
+
"eval_accuracy": 0.7590476190476191,
|
| 1061 |
+
"eval_loss": 0.9670929908752441,
|
| 1062 |
+
"eval_runtime": 18.4333,
|
| 1063 |
+
"eval_samples_per_second": 56.962,
|
| 1064 |
+
"eval_steps_per_second": 7.161,
|
| 1065 |
+
"step": 2400
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 7.71,
|
| 1069 |
+
"grad_norm": 5.889522075653076,
|
| 1070 |
+
"learning_rate": 2.299363057324841e-05,
|
| 1071 |
+
"loss": 0.152,
|
| 1072 |
+
"step": 2420
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 7.77,
|
| 1076 |
+
"grad_norm": 3.53730845451355,
|
| 1077 |
+
"learning_rate": 2.2356687898089173e-05,
|
| 1078 |
+
"loss": 0.165,
|
| 1079 |
+
"step": 2440
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 7.83,
|
| 1083 |
+
"grad_norm": 2.178981065750122,
|
| 1084 |
+
"learning_rate": 2.171974522292994e-05,
|
| 1085 |
+
"loss": 0.1128,
|
| 1086 |
+
"step": 2460
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 7.9,
|
| 1090 |
+
"grad_norm": 0.19914887845516205,
|
| 1091 |
+
"learning_rate": 2.1114649681528666e-05,
|
| 1092 |
+
"loss": 0.1307,
|
| 1093 |
+
"step": 2480
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 7.96,
|
| 1097 |
+
"grad_norm": 2.177159070968628,
|
| 1098 |
+
"learning_rate": 2.0477707006369428e-05,
|
| 1099 |
+
"loss": 0.1257,
|
| 1100 |
+
"step": 2500
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 7.96,
|
| 1104 |
+
"eval_accuracy": 0.7838095238095238,
|
| 1105 |
+
"eval_loss": 0.8925411701202393,
|
| 1106 |
+
"eval_runtime": 17.5418,
|
| 1107 |
+
"eval_samples_per_second": 59.857,
|
| 1108 |
+
"eval_steps_per_second": 7.525,
|
| 1109 |
+
"step": 2500
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 8.03,
|
| 1113 |
+
"grad_norm": 2.6388609409332275,
|
| 1114 |
+
"learning_rate": 1.9840764331210194e-05,
|
| 1115 |
+
"loss": 0.1199,
|
| 1116 |
+
"step": 2520
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 8.09,
|
| 1120 |
+
"grad_norm": 4.0329155921936035,
|
| 1121 |
+
"learning_rate": 1.9203821656050956e-05,
|
| 1122 |
+
"loss": 0.1158,
|
| 1123 |
+
"step": 2540
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 8.15,
|
| 1127 |
+
"grad_norm": 2.214768409729004,
|
| 1128 |
+
"learning_rate": 1.856687898089172e-05,
|
| 1129 |
+
"loss": 0.148,
|
| 1130 |
+
"step": 2560
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 8.22,
|
| 1134 |
+
"grad_norm": 3.5198891162872314,
|
| 1135 |
+
"learning_rate": 1.7929936305732484e-05,
|
| 1136 |
+
"loss": 0.1107,
|
| 1137 |
+
"step": 2580
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 8.28,
|
| 1141 |
+
"grad_norm": 1.0330649614334106,
|
| 1142 |
+
"learning_rate": 1.7292993630573247e-05,
|
| 1143 |
+
"loss": 0.0939,
|
| 1144 |
+
"step": 2600
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 8.28,
|
| 1148 |
+
"eval_accuracy": 0.7704761904761904,
|
| 1149 |
+
"eval_loss": 0.9257068634033203,
|
| 1150 |
+
"eval_runtime": 15.0117,
|
| 1151 |
+
"eval_samples_per_second": 69.945,
|
| 1152 |
+
"eval_steps_per_second": 8.793,
|
| 1153 |
+
"step": 2600
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 8.34,
|
| 1157 |
+
"grad_norm": 1.8858942985534668,
|
| 1158 |
+
"learning_rate": 1.6656050955414012e-05,
|
| 1159 |
+
"loss": 0.1063,
|
| 1160 |
+
"step": 2620
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 8.41,
|
| 1164 |
+
"grad_norm": 2.4143009185791016,
|
| 1165 |
+
"learning_rate": 1.6019108280254778e-05,
|
| 1166 |
+
"loss": 0.1624,
|
| 1167 |
+
"step": 2640
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 8.47,
|
| 1171 |
+
"grad_norm": 0.2723791003227234,
|
| 1172 |
+
"learning_rate": 1.538216560509554e-05,
|
| 1173 |
+
"loss": 0.1109,
|
| 1174 |
+
"step": 2660
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 8.54,
|
| 1178 |
+
"grad_norm": 3.380007266998291,
|
| 1179 |
+
"learning_rate": 1.4745222929936306e-05,
|
| 1180 |
+
"loss": 0.1375,
|
| 1181 |
+
"step": 2680
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 8.6,
|
| 1185 |
+
"grad_norm": 1.9235315322875977,
|
| 1186 |
+
"learning_rate": 1.410828025477707e-05,
|
| 1187 |
+
"loss": 0.1238,
|
| 1188 |
+
"step": 2700
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 8.6,
|
| 1192 |
+
"eval_accuracy": 0.7647619047619048,
|
| 1193 |
+
"eval_loss": 0.9797086715698242,
|
| 1194 |
+
"eval_runtime": 15.6554,
|
| 1195 |
+
"eval_samples_per_second": 67.07,
|
| 1196 |
+
"eval_steps_per_second": 8.432,
|
| 1197 |
+
"step": 2700
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 8.66,
|
| 1201 |
+
"grad_norm": 2.037036180496216,
|
| 1202 |
+
"learning_rate": 1.3471337579617834e-05,
|
| 1203 |
+
"loss": 0.1381,
|
| 1204 |
+
"step": 2720
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 8.73,
|
| 1208 |
+
"grad_norm": 1.407027006149292,
|
| 1209 |
+
"learning_rate": 1.2834394904458598e-05,
|
| 1210 |
+
"loss": 0.0994,
|
| 1211 |
+
"step": 2740
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 8.79,
|
| 1215 |
+
"grad_norm": 1.2764071226119995,
|
| 1216 |
+
"learning_rate": 1.2197452229299364e-05,
|
| 1217 |
+
"loss": 0.1033,
|
| 1218 |
+
"step": 2760
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 8.85,
|
| 1222 |
+
"grad_norm": 2.9256701469421387,
|
| 1223 |
+
"learning_rate": 1.1560509554140128e-05,
|
| 1224 |
+
"loss": 0.0982,
|
| 1225 |
+
"step": 2780
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 8.92,
|
| 1229 |
+
"grad_norm": 5.594937324523926,
|
| 1230 |
+
"learning_rate": 1.0923566878980892e-05,
|
| 1231 |
+
"loss": 0.1219,
|
| 1232 |
+
"step": 2800
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 8.92,
|
| 1236 |
+
"eval_accuracy": 0.7723809523809524,
|
| 1237 |
+
"eval_loss": 0.939895510673523,
|
| 1238 |
+
"eval_runtime": 14.8747,
|
| 1239 |
+
"eval_samples_per_second": 70.589,
|
| 1240 |
+
"eval_steps_per_second": 8.874,
|
| 1241 |
+
"step": 2800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 8.98,
|
| 1245 |
+
"grad_norm": 3.3030786514282227,
|
| 1246 |
+
"learning_rate": 1.0286624203821656e-05,
|
| 1247 |
+
"loss": 0.0996,
|
| 1248 |
+
"step": 2820
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 9.04,
|
| 1252 |
+
"grad_norm": 1.5488649606704712,
|
| 1253 |
+
"learning_rate": 9.649681528662422e-06,
|
| 1254 |
+
"loss": 0.1477,
|
| 1255 |
+
"step": 2840
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 9.11,
|
| 1259 |
+
"grad_norm": 4.593501091003418,
|
| 1260 |
+
"learning_rate": 9.012738853503185e-06,
|
| 1261 |
+
"loss": 0.1142,
|
| 1262 |
+
"step": 2860
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 9.17,
|
| 1266 |
+
"grad_norm": 6.411059379577637,
|
| 1267 |
+
"learning_rate": 8.375796178343949e-06,
|
| 1268 |
+
"loss": 0.1249,
|
| 1269 |
+
"step": 2880
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 9.24,
|
| 1273 |
+
"grad_norm": 2.7087924480438232,
|
| 1274 |
+
"learning_rate": 7.738853503184713e-06,
|
| 1275 |
+
"loss": 0.0985,
|
| 1276 |
+
"step": 2900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 9.24,
|
| 1280 |
+
"eval_accuracy": 0.7647619047619048,
|
| 1281 |
+
"eval_loss": 0.9940046072006226,
|
| 1282 |
+
"eval_runtime": 15.2749,
|
| 1283 |
+
"eval_samples_per_second": 68.74,
|
| 1284 |
+
"eval_steps_per_second": 8.642,
|
| 1285 |
+
"step": 2900
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 9.3,
|
| 1289 |
+
"grad_norm": 5.092051029205322,
|
| 1290 |
+
"learning_rate": 7.1019108280254775e-06,
|
| 1291 |
+
"loss": 0.1043,
|
| 1292 |
+
"step": 2920
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 9.36,
|
| 1296 |
+
"grad_norm": 10.890628814697266,
|
| 1297 |
+
"learning_rate": 6.464968152866242e-06,
|
| 1298 |
+
"loss": 0.1174,
|
| 1299 |
+
"step": 2940
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 9.43,
|
| 1303 |
+
"grad_norm": 3.2413973808288574,
|
| 1304 |
+
"learning_rate": 5.8280254777070065e-06,
|
| 1305 |
+
"loss": 0.105,
|
| 1306 |
+
"step": 2960
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 9.49,
|
| 1310 |
+
"grad_norm": 3.838075876235962,
|
| 1311 |
+
"learning_rate": 5.191082802547771e-06,
|
| 1312 |
+
"loss": 0.0857,
|
| 1313 |
+
"step": 2980
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 9.55,
|
| 1317 |
+
"grad_norm": 2.082455635070801,
|
| 1318 |
+
"learning_rate": 4.5541401273885346e-06,
|
| 1319 |
+
"loss": 0.1069,
|
| 1320 |
+
"step": 3000
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 9.55,
|
| 1324 |
+
"eval_accuracy": 0.7742857142857142,
|
| 1325 |
+
"eval_loss": 0.9392004013061523,
|
| 1326 |
+
"eval_runtime": 15.3798,
|
| 1327 |
+
"eval_samples_per_second": 68.271,
|
| 1328 |
+
"eval_steps_per_second": 8.583,
|
| 1329 |
+
"step": 3000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 9.62,
|
| 1333 |
+
"grad_norm": 0.4810134768486023,
|
| 1334 |
+
"learning_rate": 3.9171974522292995e-06,
|
| 1335 |
+
"loss": 0.0748,
|
| 1336 |
+
"step": 3020
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 9.68,
|
| 1340 |
+
"grad_norm": 1.6180094480514526,
|
| 1341 |
+
"learning_rate": 3.280254777070064e-06,
|
| 1342 |
+
"loss": 0.1049,
|
| 1343 |
+
"step": 3040
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 9.75,
|
| 1347 |
+
"grad_norm": 0.6423608660697937,
|
| 1348 |
+
"learning_rate": 2.6433121019108284e-06,
|
| 1349 |
+
"loss": 0.126,
|
| 1350 |
+
"step": 3060
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 9.81,
|
| 1354 |
+
"grad_norm": 1.139814853668213,
|
| 1355 |
+
"learning_rate": 2.0063694267515925e-06,
|
| 1356 |
+
"loss": 0.0813,
|
| 1357 |
+
"step": 3080
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 9.87,
|
| 1361 |
+
"grad_norm": 0.03859005495905876,
|
| 1362 |
+
"learning_rate": 1.3694267515923567e-06,
|
| 1363 |
+
"loss": 0.0589,
|
| 1364 |
+
"step": 3100
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 9.87,
|
| 1368 |
+
"eval_accuracy": 0.78,
|
| 1369 |
+
"eval_loss": 0.9408173561096191,
|
| 1370 |
+
"eval_runtime": 15.1635,
|
| 1371 |
+
"eval_samples_per_second": 69.245,
|
| 1372 |
+
"eval_steps_per_second": 8.705,
|
| 1373 |
+
"step": 3100
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 9.94,
|
| 1377 |
+
"grad_norm": 2.202677011489868,
|
| 1378 |
+
"learning_rate": 7.324840764331211e-07,
|
| 1379 |
+
"loss": 0.0856,
|
| 1380 |
+
"step": 3120
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 10.0,
|
| 1384 |
+
"grad_norm": 3.140180826187134,
|
| 1385 |
+
"learning_rate": 9.554140127388536e-08,
|
| 1386 |
+
"loss": 0.0997,
|
| 1387 |
+
"step": 3140
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 10.0,
|
| 1391 |
+
"step": 3140,
|
| 1392 |
+
"total_flos": 7.776878731479245e+18,
|
| 1393 |
+
"train_loss": 0.2955047018209081,
|
| 1394 |
+
"train_runtime": 3022.5495,
|
| 1395 |
+
"train_samples_per_second": 33.2,
|
| 1396 |
+
"train_steps_per_second": 1.039
|
| 1397 |
+
}
|
| 1398 |
+
],
|
| 1399 |
+
"logging_steps": 20,
|
| 1400 |
+
"max_steps": 3140,
|
| 1401 |
+
"num_input_tokens_seen": 0,
|
| 1402 |
+
"num_train_epochs": 10,
|
| 1403 |
+
"save_steps": 100,
|
| 1404 |
+
"total_flos": 7.776878731479245e+18,
|
| 1405 |
+
"train_batch_size": 32,
|
| 1406 |
+
"trial_name": null,
|
| 1407 |
+
"trial_params": null
|
| 1408 |
+
}
|