🍻 cheers
Browse files- README.md +6 -5
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May09_14-30-54_ed3f40b4fac8/events.out.tfevents.1715268878.ed3f40b4fac8.34.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +3727 -0
README.md
CHANGED
@@ -2,6 +2,7 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: microsoft/swinv2-tiny-patch4-window8-256
|
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 |
# microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug
|
32 |
|
33 |
-
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) 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: microsoft/swinv2-tiny-patch4-window8-256
|
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: bengali_food_images
|
19 |
type: imagefolder
|
20 |
config: default
|
21 |
split: train
|
|
|
23 |
metrics:
|
24 |
- name: Accuracy
|
25 |
type: accuracy
|
26 |
+
value: 0.9456521739130435
|
27 |
---
|
28 |
|
29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
31 |
|
32 |
# microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug
|
33 |
|
34 |
+
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the bengali_food_images dataset.
|
35 |
It achieves the following results on the evaluation set:
|
36 |
+
- Loss: 0.2321
|
37 |
+
- Accuracy: 0.9457
|
38 |
|
39 |
## Model description
|
40 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9456521739130435,
|
4 |
+
"eval_loss": 0.2321375608444214,
|
5 |
+
"eval_runtime": 21.9373,
|
6 |
+
"eval_samples_per_second": 33.55,
|
7 |
+
"eval_steps_per_second": 4.194,
|
8 |
+
"total_flos": 1.870424802038661e+18,
|
9 |
+
"train_loss": 0.5094418521630693,
|
10 |
+
"train_runtime": 3760.9881,
|
11 |
+
"train_samples_per_second": 19.953,
|
12 |
+
"train_steps_per_second": 1.248
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9456521739130435,
|
4 |
+
"eval_loss": 0.2321375608444214,
|
5 |
+
"eval_runtime": 21.9373,
|
6 |
+
"eval_samples_per_second": 33.55,
|
7 |
+
"eval_steps_per_second": 4.194
|
8 |
+
}
|
runs/May09_14-30-54_ed3f40b4fac8/events.out.tfevents.1715268878.ed3f40b4fac8.34.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fc912e12f32a3672091165624bbf81548153c987aa1d6d9682bea375879ef575
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 1.870424802038661e+18,
|
4 |
+
"train_loss": 0.5094418521630693,
|
5 |
+
"train_runtime": 3760.9881,
|
6 |
+
"train_samples_per_second": 19.953,
|
7 |
+
"train_steps_per_second": 1.248
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3727 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.2321375608444214,
|
3 |
+
"best_model_checkpoint": "/kaggle/working/Model/microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug/checkpoint-4400",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 4692,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"grad_norm": NaN,
|
14 |
+
"learning_rate": 0.0002,
|
15 |
+
"loss": 3.2461,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"grad_norm": NaN,
|
21 |
+
"learning_rate": 0.0002,
|
22 |
+
"loss": 3.2031,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.03,
|
27 |
+
"grad_norm": 5.058732509613037,
|
28 |
+
"learning_rate": 0.0001999147485080989,
|
29 |
+
"loss": 3.2604,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.03,
|
34 |
+
"grad_norm": 2.7387874126434326,
|
35 |
+
"learning_rate": 0.00019948849104859336,
|
36 |
+
"loss": 3.1794,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.04,
|
41 |
+
"grad_norm": 2.7913503646850586,
|
42 |
+
"learning_rate": 0.0001990622335890878,
|
43 |
+
"loss": 2.9532,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.05,
|
48 |
+
"grad_norm": 4.803112030029297,
|
49 |
+
"learning_rate": 0.00019867860187553282,
|
50 |
+
"loss": 2.692,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.06,
|
55 |
+
"grad_norm": 5.479024887084961,
|
56 |
+
"learning_rate": 0.0001982523444160273,
|
57 |
+
"loss": 2.4673,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.07,
|
62 |
+
"grad_norm": 5.984738826751709,
|
63 |
+
"learning_rate": 0.00019782608695652175,
|
64 |
+
"loss": 2.2248,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.08,
|
69 |
+
"grad_norm": 8.306194305419922,
|
70 |
+
"learning_rate": 0.0001973998294970162,
|
71 |
+
"loss": 1.9695,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.09,
|
76 |
+
"grad_norm": 6.3569512367248535,
|
77 |
+
"learning_rate": 0.00019697357203751065,
|
78 |
+
"loss": 1.7162,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.09,
|
83 |
+
"eval_accuracy": 0.7078804347826086,
|
84 |
+
"eval_loss": 1.4225009679794312,
|
85 |
+
"eval_runtime": 26.8467,
|
86 |
+
"eval_samples_per_second": 27.415,
|
87 |
+
"eval_steps_per_second": 3.427,
|
88 |
+
"step": 100
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.09,
|
92 |
+
"grad_norm": 6.7053914070129395,
|
93 |
+
"learning_rate": 0.00019654731457800512,
|
94 |
+
"loss": 1.5791,
|
95 |
+
"step": 110
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.1,
|
99 |
+
"grad_norm": 6.75339412689209,
|
100 |
+
"learning_rate": 0.00019616368286445014,
|
101 |
+
"loss": 1.5743,
|
102 |
+
"step": 120
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 0.11,
|
106 |
+
"grad_norm": 6.368953704833984,
|
107 |
+
"learning_rate": 0.0001957374254049446,
|
108 |
+
"loss": 1.4221,
|
109 |
+
"step": 130
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.12,
|
113 |
+
"grad_norm": 7.507160663604736,
|
114 |
+
"learning_rate": 0.00019531116794543904,
|
115 |
+
"loss": 1.4591,
|
116 |
+
"step": 140
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.13,
|
120 |
+
"grad_norm": 6.467655181884766,
|
121 |
+
"learning_rate": 0.00019488491048593351,
|
122 |
+
"loss": 1.5203,
|
123 |
+
"step": 150
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.14,
|
127 |
+
"grad_norm": 7.42750883102417,
|
128 |
+
"learning_rate": 0.00019445865302642796,
|
129 |
+
"loss": 1.3534,
|
130 |
+
"step": 160
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.14,
|
134 |
+
"grad_norm": 8.539589881896973,
|
135 |
+
"learning_rate": 0.00019403239556692244,
|
136 |
+
"loss": 1.517,
|
137 |
+
"step": 170
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.15,
|
141 |
+
"grad_norm": 8.488027572631836,
|
142 |
+
"learning_rate": 0.0001936061381074169,
|
143 |
+
"loss": 1.3041,
|
144 |
+
"step": 180
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.16,
|
148 |
+
"grad_norm": 6.249123573303223,
|
149 |
+
"learning_rate": 0.00019317988064791134,
|
150 |
+
"loss": 1.178,
|
151 |
+
"step": 190
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 0.17,
|
155 |
+
"grad_norm": 8.70413589477539,
|
156 |
+
"learning_rate": 0.0001927536231884058,
|
157 |
+
"loss": 1.2286,
|
158 |
+
"step": 200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.17,
|
162 |
+
"eval_accuracy": 0.7934782608695652,
|
163 |
+
"eval_loss": 0.946118175983429,
|
164 |
+
"eval_runtime": 22.2277,
|
165 |
+
"eval_samples_per_second": 33.112,
|
166 |
+
"eval_steps_per_second": 4.139,
|
167 |
+
"step": 200
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 0.18,
|
171 |
+
"grad_norm": 6.648161888122559,
|
172 |
+
"learning_rate": 0.00019232736572890027,
|
173 |
+
"loss": 1.1328,
|
174 |
+
"step": 210
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.19,
|
178 |
+
"grad_norm": 6.3985419273376465,
|
179 |
+
"learning_rate": 0.00019190110826939472,
|
180 |
+
"loss": 1.3468,
|
181 |
+
"step": 220
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 0.2,
|
185 |
+
"grad_norm": 6.855421543121338,
|
186 |
+
"learning_rate": 0.0001914748508098892,
|
187 |
+
"loss": 1.121,
|
188 |
+
"step": 230
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.2,
|
192 |
+
"grad_norm": 7.411104202270508,
|
193 |
+
"learning_rate": 0.00019104859335038364,
|
194 |
+
"loss": 1.1668,
|
195 |
+
"step": 240
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.21,
|
199 |
+
"grad_norm": 6.608767986297607,
|
200 |
+
"learning_rate": 0.0001906223358908781,
|
201 |
+
"loss": 1.1276,
|
202 |
+
"step": 250
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 0.22,
|
206 |
+
"grad_norm": 7.618008613586426,
|
207 |
+
"learning_rate": 0.00019019607843137254,
|
208 |
+
"loss": 1.0934,
|
209 |
+
"step": 260
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.23,
|
213 |
+
"grad_norm": 8.76831340789795,
|
214 |
+
"learning_rate": 0.00018976982097186702,
|
215 |
+
"loss": 1.0502,
|
216 |
+
"step": 270
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.24,
|
220 |
+
"grad_norm": 6.6039323806762695,
|
221 |
+
"learning_rate": 0.00018934356351236147,
|
222 |
+
"loss": 0.9731,
|
223 |
+
"step": 280
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 0.25,
|
227 |
+
"grad_norm": 5.2799072265625,
|
228 |
+
"learning_rate": 0.00018891730605285594,
|
229 |
+
"loss": 0.9236,
|
230 |
+
"step": 290
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.26,
|
234 |
+
"grad_norm": 8.616353988647461,
|
235 |
+
"learning_rate": 0.0001884910485933504,
|
236 |
+
"loss": 1.0323,
|
237 |
+
"step": 300
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 0.26,
|
241 |
+
"eval_accuracy": 0.8355978260869565,
|
242 |
+
"eval_loss": 0.7366186380386353,
|
243 |
+
"eval_runtime": 22.3822,
|
244 |
+
"eval_samples_per_second": 32.883,
|
245 |
+
"eval_steps_per_second": 4.11,
|
246 |
+
"step": 300
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.26,
|
250 |
+
"grad_norm": 6.827239513397217,
|
251 |
+
"learning_rate": 0.00018806479113384484,
|
252 |
+
"loss": 1.0934,
|
253 |
+
"step": 310
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 0.27,
|
257 |
+
"grad_norm": 7.527755260467529,
|
258 |
+
"learning_rate": 0.0001876385336743393,
|
259 |
+
"loss": 1.0528,
|
260 |
+
"step": 320
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.28,
|
264 |
+
"grad_norm": 6.007165431976318,
|
265 |
+
"learning_rate": 0.00018721227621483377,
|
266 |
+
"loss": 0.8891,
|
267 |
+
"step": 330
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.29,
|
271 |
+
"grad_norm": 8.186945915222168,
|
272 |
+
"learning_rate": 0.00018678601875532822,
|
273 |
+
"loss": 0.965,
|
274 |
+
"step": 340
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 0.3,
|
278 |
+
"grad_norm": 6.866254806518555,
|
279 |
+
"learning_rate": 0.0001863597612958227,
|
280 |
+
"loss": 0.8948,
|
281 |
+
"step": 350
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 0.31,
|
285 |
+
"grad_norm": 6.580564975738525,
|
286 |
+
"learning_rate": 0.00018593350383631715,
|
287 |
+
"loss": 0.9438,
|
288 |
+
"step": 360
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 0.32,
|
292 |
+
"grad_norm": 5.725220680236816,
|
293 |
+
"learning_rate": 0.0001855072463768116,
|
294 |
+
"loss": 0.8196,
|
295 |
+
"step": 370
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 0.32,
|
299 |
+
"grad_norm": 7.180054664611816,
|
300 |
+
"learning_rate": 0.00018508098891730605,
|
301 |
+
"loss": 0.8645,
|
302 |
+
"step": 380
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.33,
|
306 |
+
"grad_norm": 5.196268081665039,
|
307 |
+
"learning_rate": 0.00018469735720375109,
|
308 |
+
"loss": 0.8335,
|
309 |
+
"step": 390
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.34,
|
313 |
+
"grad_norm": 5.930848121643066,
|
314 |
+
"learning_rate": 0.00018427109974424554,
|
315 |
+
"loss": 0.8678,
|
316 |
+
"step": 400
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.34,
|
320 |
+
"eval_accuracy": 0.8627717391304348,
|
321 |
+
"eval_loss": 0.6210773587226868,
|
322 |
+
"eval_runtime": 22.1531,
|
323 |
+
"eval_samples_per_second": 33.223,
|
324 |
+
"eval_steps_per_second": 4.153,
|
325 |
+
"step": 400
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 0.35,
|
329 |
+
"grad_norm": 7.423695087432861,
|
330 |
+
"learning_rate": 0.00018384484228473998,
|
331 |
+
"loss": 0.7715,
|
332 |
+
"step": 410
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.36,
|
336 |
+
"grad_norm": 4.49029016494751,
|
337 |
+
"learning_rate": 0.00018341858482523443,
|
338 |
+
"loss": 0.6917,
|
339 |
+
"step": 420
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 0.37,
|
343 |
+
"grad_norm": 4.553171157836914,
|
344 |
+
"learning_rate": 0.0001829923273657289,
|
345 |
+
"loss": 0.7546,
|
346 |
+
"step": 430
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 0.38,
|
350 |
+
"grad_norm": 5.762471675872803,
|
351 |
+
"learning_rate": 0.00018256606990622336,
|
352 |
+
"loss": 0.876,
|
353 |
+
"step": 440
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.38,
|
357 |
+
"grad_norm": 7.234831809997559,
|
358 |
+
"learning_rate": 0.00018213981244671784,
|
359 |
+
"loss": 0.819,
|
360 |
+
"step": 450
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.39,
|
364 |
+
"grad_norm": 6.18890380859375,
|
365 |
+
"learning_rate": 0.0001817135549872123,
|
366 |
+
"loss": 0.8027,
|
367 |
+
"step": 460
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 0.4,
|
371 |
+
"grad_norm": 5.732609272003174,
|
372 |
+
"learning_rate": 0.00018128729752770674,
|
373 |
+
"loss": 0.7294,
|
374 |
+
"step": 470
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.41,
|
378 |
+
"grad_norm": 6.72756290435791,
|
379 |
+
"learning_rate": 0.0001808610400682012,
|
380 |
+
"loss": 0.8894,
|
381 |
+
"step": 480
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 0.42,
|
385 |
+
"grad_norm": 4.72413444519043,
|
386 |
+
"learning_rate": 0.00018043478260869566,
|
387 |
+
"loss": 0.7052,
|
388 |
+
"step": 490
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 0.43,
|
392 |
+
"grad_norm": 3.8523027896881104,
|
393 |
+
"learning_rate": 0.0001800085251491901,
|
394 |
+
"loss": 0.7849,
|
395 |
+
"step": 500
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.43,
|
399 |
+
"eval_accuracy": 0.8654891304347826,
|
400 |
+
"eval_loss": 0.5353636741638184,
|
401 |
+
"eval_runtime": 22.0834,
|
402 |
+
"eval_samples_per_second": 33.328,
|
403 |
+
"eval_steps_per_second": 4.166,
|
404 |
+
"step": 500
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 0.43,
|
408 |
+
"grad_norm": 5.818722248077393,
|
409 |
+
"learning_rate": 0.0001795822676896846,
|
410 |
+
"loss": 0.6421,
|
411 |
+
"step": 510
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 0.44,
|
415 |
+
"grad_norm": 7.893730640411377,
|
416 |
+
"learning_rate": 0.00017915601023017904,
|
417 |
+
"loss": 0.7131,
|
418 |
+
"step": 520
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 0.45,
|
422 |
+
"grad_norm": 6.222229957580566,
|
423 |
+
"learning_rate": 0.0001787297527706735,
|
424 |
+
"loss": 0.642,
|
425 |
+
"step": 530
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 0.46,
|
429 |
+
"grad_norm": 6.1911139488220215,
|
430 |
+
"learning_rate": 0.00017830349531116794,
|
431 |
+
"loss": 0.7096,
|
432 |
+
"step": 540
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 0.47,
|
436 |
+
"grad_norm": 7.124819278717041,
|
437 |
+
"learning_rate": 0.00017787723785166242,
|
438 |
+
"loss": 0.6946,
|
439 |
+
"step": 550
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 0.48,
|
443 |
+
"grad_norm": 4.969889163970947,
|
444 |
+
"learning_rate": 0.00017745098039215687,
|
445 |
+
"loss": 0.6995,
|
446 |
+
"step": 560
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.49,
|
450 |
+
"grad_norm": 4.456979751586914,
|
451 |
+
"learning_rate": 0.00017702472293265134,
|
452 |
+
"loss": 0.6533,
|
453 |
+
"step": 570
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 0.49,
|
457 |
+
"grad_norm": 7.239711284637451,
|
458 |
+
"learning_rate": 0.0001765984654731458,
|
459 |
+
"loss": 0.755,
|
460 |
+
"step": 580
|
461 |
+
},
|
462 |
+
{
|
463 |
+
"epoch": 0.5,
|
464 |
+
"grad_norm": 4.915600776672363,
|
465 |
+
"learning_rate": 0.00017617220801364024,
|
466 |
+
"loss": 0.6239,
|
467 |
+
"step": 590
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 0.51,
|
471 |
+
"grad_norm": 4.89735221862793,
|
472 |
+
"learning_rate": 0.0001757459505541347,
|
473 |
+
"loss": 0.7105,
|
474 |
+
"step": 600
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 0.51,
|
478 |
+
"eval_accuracy": 0.8899456521739131,
|
479 |
+
"eval_loss": 0.47926023602485657,
|
480 |
+
"eval_runtime": 21.8438,
|
481 |
+
"eval_samples_per_second": 33.694,
|
482 |
+
"eval_steps_per_second": 4.212,
|
483 |
+
"step": 600
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 0.52,
|
487 |
+
"grad_norm": 4.159371376037598,
|
488 |
+
"learning_rate": 0.00017531969309462917,
|
489 |
+
"loss": 0.5948,
|
490 |
+
"step": 610
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 0.53,
|
494 |
+
"grad_norm": 4.996119976043701,
|
495 |
+
"learning_rate": 0.00017489343563512362,
|
496 |
+
"loss": 0.6705,
|
497 |
+
"step": 620
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 0.54,
|
501 |
+
"grad_norm": 5.600012302398682,
|
502 |
+
"learning_rate": 0.0001744671781756181,
|
503 |
+
"loss": 0.6742,
|
504 |
+
"step": 630
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 0.55,
|
508 |
+
"grad_norm": 8.147639274597168,
|
509 |
+
"learning_rate": 0.00017404092071611254,
|
510 |
+
"loss": 0.7171,
|
511 |
+
"step": 640
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 0.55,
|
515 |
+
"grad_norm": 4.924251556396484,
|
516 |
+
"learning_rate": 0.000173614663256607,
|
517 |
+
"loss": 0.6655,
|
518 |
+
"step": 650
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 0.56,
|
522 |
+
"grad_norm": 4.968270301818848,
|
523 |
+
"learning_rate": 0.00017318840579710144,
|
524 |
+
"loss": 0.641,
|
525 |
+
"step": 660
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 0.57,
|
529 |
+
"grad_norm": 6.600828647613525,
|
530 |
+
"learning_rate": 0.00017276214833759592,
|
531 |
+
"loss": 0.6549,
|
532 |
+
"step": 670
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.58,
|
536 |
+
"grad_norm": 5.408564567565918,
|
537 |
+
"learning_rate": 0.00017233589087809037,
|
538 |
+
"loss": 0.6431,
|
539 |
+
"step": 680
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.59,
|
543 |
+
"grad_norm": 6.0379228591918945,
|
544 |
+
"learning_rate": 0.00017190963341858485,
|
545 |
+
"loss": 0.6226,
|
546 |
+
"step": 690
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 0.6,
|
550 |
+
"grad_norm": 6.015142917633057,
|
551 |
+
"learning_rate": 0.0001714833759590793,
|
552 |
+
"loss": 0.6198,
|
553 |
+
"step": 700
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 0.6,
|
557 |
+
"eval_accuracy": 0.9089673913043478,
|
558 |
+
"eval_loss": 0.43188050389289856,
|
559 |
+
"eval_runtime": 21.8402,
|
560 |
+
"eval_samples_per_second": 33.699,
|
561 |
+
"eval_steps_per_second": 4.212,
|
562 |
+
"step": 700
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.61,
|
566 |
+
"grad_norm": 5.850244045257568,
|
567 |
+
"learning_rate": 0.00017105711849957375,
|
568 |
+
"loss": 0.5817,
|
569 |
+
"step": 710
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.61,
|
573 |
+
"grad_norm": 3.7645974159240723,
|
574 |
+
"learning_rate": 0.0001706308610400682,
|
575 |
+
"loss": 0.6146,
|
576 |
+
"step": 720
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.62,
|
580 |
+
"grad_norm": 4.481291770935059,
|
581 |
+
"learning_rate": 0.00017020460358056267,
|
582 |
+
"loss": 0.6362,
|
583 |
+
"step": 730
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.63,
|
587 |
+
"grad_norm": 3.906785488128662,
|
588 |
+
"learning_rate": 0.00016977834612105712,
|
589 |
+
"loss": 0.603,
|
590 |
+
"step": 740
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.64,
|
594 |
+
"grad_norm": 6.198018550872803,
|
595 |
+
"learning_rate": 0.0001693520886615516,
|
596 |
+
"loss": 0.547,
|
597 |
+
"step": 750
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.65,
|
601 |
+
"grad_norm": 3.281864881515503,
|
602 |
+
"learning_rate": 0.00016892583120204605,
|
603 |
+
"loss": 0.5979,
|
604 |
+
"step": 760
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.66,
|
608 |
+
"grad_norm": 3.2829701900482178,
|
609 |
+
"learning_rate": 0.0001684995737425405,
|
610 |
+
"loss": 0.6047,
|
611 |
+
"step": 770
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.66,
|
615 |
+
"grad_norm": 4.538403511047363,
|
616 |
+
"learning_rate": 0.00016807331628303495,
|
617 |
+
"loss": 0.6037,
|
618 |
+
"step": 780
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.67,
|
622 |
+
"grad_norm": 3.8807320594787598,
|
623 |
+
"learning_rate": 0.00016764705882352942,
|
624 |
+
"loss": 0.4783,
|
625 |
+
"step": 790
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.68,
|
629 |
+
"grad_norm": 5.805025100708008,
|
630 |
+
"learning_rate": 0.00016722080136402387,
|
631 |
+
"loss": 0.6276,
|
632 |
+
"step": 800
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.68,
|
636 |
+
"eval_accuracy": 0.8980978260869565,
|
637 |
+
"eval_loss": 0.4021691083908081,
|
638 |
+
"eval_runtime": 22.1281,
|
639 |
+
"eval_samples_per_second": 33.261,
|
640 |
+
"eval_steps_per_second": 4.158,
|
641 |
+
"step": 800
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 0.69,
|
645 |
+
"grad_norm": 3.921274185180664,
|
646 |
+
"learning_rate": 0.00016679454390451835,
|
647 |
+
"loss": 0.6383,
|
648 |
+
"step": 810
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 0.7,
|
652 |
+
"grad_norm": 8.53153133392334,
|
653 |
+
"learning_rate": 0.0001663682864450128,
|
654 |
+
"loss": 0.6362,
|
655 |
+
"step": 820
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 0.71,
|
659 |
+
"grad_norm": 3.754373073577881,
|
660 |
+
"learning_rate": 0.00016594202898550725,
|
661 |
+
"loss": 0.502,
|
662 |
+
"step": 830
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.72,
|
666 |
+
"grad_norm": 5.654985427856445,
|
667 |
+
"learning_rate": 0.0001655157715260017,
|
668 |
+
"loss": 0.6056,
|
669 |
+
"step": 840
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 0.72,
|
673 |
+
"grad_norm": 4.316379070281982,
|
674 |
+
"learning_rate": 0.00016508951406649618,
|
675 |
+
"loss": 0.5892,
|
676 |
+
"step": 850
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 0.73,
|
680 |
+
"grad_norm": 5.630128860473633,
|
681 |
+
"learning_rate": 0.00016466325660699063,
|
682 |
+
"loss": 0.6279,
|
683 |
+
"step": 860
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 0.74,
|
687 |
+
"grad_norm": 5.5036797523498535,
|
688 |
+
"learning_rate": 0.0001642369991474851,
|
689 |
+
"loss": 0.5664,
|
690 |
+
"step": 870
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 0.75,
|
694 |
+
"grad_norm": 3.7804858684539795,
|
695 |
+
"learning_rate": 0.00016381074168797955,
|
696 |
+
"loss": 0.6377,
|
697 |
+
"step": 880
|
698 |
+
},
|
699 |
+
{
|
700 |
+
"epoch": 0.76,
|
701 |
+
"grad_norm": 3.757089138031006,
|
702 |
+
"learning_rate": 0.000163384484228474,
|
703 |
+
"loss": 0.6601,
|
704 |
+
"step": 890
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 0.77,
|
708 |
+
"grad_norm": 5.010238170623779,
|
709 |
+
"learning_rate": 0.00016295822676896845,
|
710 |
+
"loss": 0.5411,
|
711 |
+
"step": 900
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 0.77,
|
715 |
+
"eval_accuracy": 0.9116847826086957,
|
716 |
+
"eval_loss": 0.38156262040138245,
|
717 |
+
"eval_runtime": 22.137,
|
718 |
+
"eval_samples_per_second": 33.248,
|
719 |
+
"eval_steps_per_second": 4.156,
|
720 |
+
"step": 900
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 0.78,
|
724 |
+
"grad_norm": 4.0843987464904785,
|
725 |
+
"learning_rate": 0.00016253196930946293,
|
726 |
+
"loss": 0.5379,
|
727 |
+
"step": 910
|
728 |
+
},
|
729 |
+
{
|
730 |
+
"epoch": 0.78,
|
731 |
+
"grad_norm": 2.8297243118286133,
|
732 |
+
"learning_rate": 0.00016210571184995738,
|
733 |
+
"loss": 0.4184,
|
734 |
+
"step": 920
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 0.79,
|
738 |
+
"grad_norm": 5.259457111358643,
|
739 |
+
"learning_rate": 0.00016167945439045185,
|
740 |
+
"loss": 0.4892,
|
741 |
+
"step": 930
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 0.8,
|
745 |
+
"grad_norm": 6.436057090759277,
|
746 |
+
"learning_rate": 0.0001612531969309463,
|
747 |
+
"loss": 0.5708,
|
748 |
+
"step": 940
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 0.81,
|
752 |
+
"grad_norm": 5.9696736335754395,
|
753 |
+
"learning_rate": 0.00016082693947144075,
|
754 |
+
"loss": 0.5395,
|
755 |
+
"step": 950
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 0.82,
|
759 |
+
"grad_norm": 5.614426136016846,
|
760 |
+
"learning_rate": 0.0001604006820119352,
|
761 |
+
"loss": 0.573,
|
762 |
+
"step": 960
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 0.83,
|
766 |
+
"grad_norm": 3.9581282138824463,
|
767 |
+
"learning_rate": 0.00015997442455242968,
|
768 |
+
"loss": 0.5813,
|
769 |
+
"step": 970
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.84,
|
773 |
+
"grad_norm": 4.684549808502197,
|
774 |
+
"learning_rate": 0.00015954816709292413,
|
775 |
+
"loss": 0.5533,
|
776 |
+
"step": 980
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 0.84,
|
780 |
+
"grad_norm": 5.443897247314453,
|
781 |
+
"learning_rate": 0.0001591219096334186,
|
782 |
+
"loss": 0.5539,
|
783 |
+
"step": 990
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 0.85,
|
787 |
+
"grad_norm": 4.034463882446289,
|
788 |
+
"learning_rate": 0.00015869565217391306,
|
789 |
+
"loss": 0.4984,
|
790 |
+
"step": 1000
|
791 |
+
},
|
792 |
+
{
|
793 |
+
"epoch": 0.85,
|
794 |
+
"eval_accuracy": 0.9021739130434783,
|
795 |
+
"eval_loss": 0.38239946961402893,
|
796 |
+
"eval_runtime": 21.9989,
|
797 |
+
"eval_samples_per_second": 33.456,
|
798 |
+
"eval_steps_per_second": 4.182,
|
799 |
+
"step": 1000
|
800 |
+
},
|
801 |
+
{
|
802 |
+
"epoch": 0.86,
|
803 |
+
"grad_norm": 3.4797463417053223,
|
804 |
+
"learning_rate": 0.0001582693947144075,
|
805 |
+
"loss": 0.5063,
|
806 |
+
"step": 1010
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 0.87,
|
810 |
+
"grad_norm": 4.582981586456299,
|
811 |
+
"learning_rate": 0.00015784313725490196,
|
812 |
+
"loss": 0.5875,
|
813 |
+
"step": 1020
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 0.88,
|
817 |
+
"grad_norm": 4.804072856903076,
|
818 |
+
"learning_rate": 0.00015741687979539643,
|
819 |
+
"loss": 0.5107,
|
820 |
+
"step": 1030
|
821 |
+
},
|
822 |
+
{
|
823 |
+
"epoch": 0.89,
|
824 |
+
"grad_norm": 4.848588943481445,
|
825 |
+
"learning_rate": 0.00015699062233589088,
|
826 |
+
"loss": 0.5255,
|
827 |
+
"step": 1040
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 0.9,
|
831 |
+
"grad_norm": 3.0660464763641357,
|
832 |
+
"learning_rate": 0.00015656436487638536,
|
833 |
+
"loss": 0.4437,
|
834 |
+
"step": 1050
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 0.9,
|
838 |
+
"grad_norm": 5.267394065856934,
|
839 |
+
"learning_rate": 0.0001561381074168798,
|
840 |
+
"loss": 0.5443,
|
841 |
+
"step": 1060
|
842 |
+
},
|
843 |
+
{
|
844 |
+
"epoch": 0.91,
|
845 |
+
"grad_norm": 5.676567077636719,
|
846 |
+
"learning_rate": 0.00015571184995737426,
|
847 |
+
"loss": 0.5238,
|
848 |
+
"step": 1070
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.92,
|
852 |
+
"grad_norm": 4.262234210968018,
|
853 |
+
"learning_rate": 0.0001552855924978687,
|
854 |
+
"loss": 0.5148,
|
855 |
+
"step": 1080
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 0.93,
|
859 |
+
"grad_norm": 3.7277231216430664,
|
860 |
+
"learning_rate": 0.00015485933503836318,
|
861 |
+
"loss": 0.4952,
|
862 |
+
"step": 1090
|
863 |
+
},
|
864 |
+
{
|
865 |
+
"epoch": 0.94,
|
866 |
+
"grad_norm": 4.57068395614624,
|
867 |
+
"learning_rate": 0.00015443307757885763,
|
868 |
+
"loss": 0.5665,
|
869 |
+
"step": 1100
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 0.94,
|
873 |
+
"eval_accuracy": 0.9211956521739131,
|
874 |
+
"eval_loss": 0.34602978825569153,
|
875 |
+
"eval_runtime": 22.2733,
|
876 |
+
"eval_samples_per_second": 33.044,
|
877 |
+
"eval_steps_per_second": 4.131,
|
878 |
+
"step": 1100
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.95,
|
882 |
+
"grad_norm": 3.960221767425537,
|
883 |
+
"learning_rate": 0.0001540068201193521,
|
884 |
+
"loss": 0.4565,
|
885 |
+
"step": 1110
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 0.95,
|
889 |
+
"grad_norm": 3.6928794384002686,
|
890 |
+
"learning_rate": 0.00015358056265984656,
|
891 |
+
"loss": 0.4239,
|
892 |
+
"step": 1120
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 0.96,
|
896 |
+
"grad_norm": 2.560225009918213,
|
897 |
+
"learning_rate": 0.000153154305200341,
|
898 |
+
"loss": 0.5735,
|
899 |
+
"step": 1130
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 0.97,
|
903 |
+
"grad_norm": 4.45659065246582,
|
904 |
+
"learning_rate": 0.00015272804774083546,
|
905 |
+
"loss": 0.4299,
|
906 |
+
"step": 1140
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 0.98,
|
910 |
+
"grad_norm": 5.458113670349121,
|
911 |
+
"learning_rate": 0.00015230179028132994,
|
912 |
+
"loss": 0.4727,
|
913 |
+
"step": 1150
|
914 |
+
},
|
915 |
+
{
|
916 |
+
"epoch": 0.99,
|
917 |
+
"grad_norm": 4.833206653594971,
|
918 |
+
"learning_rate": 0.0001518755328218244,
|
919 |
+
"loss": 0.4722,
|
920 |
+
"step": 1160
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 1.0,
|
924 |
+
"grad_norm": 4.604361057281494,
|
925 |
+
"learning_rate": 0.00015144927536231886,
|
926 |
+
"loss": 0.5225,
|
927 |
+
"step": 1170
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 1.01,
|
931 |
+
"grad_norm": 2.7065069675445557,
|
932 |
+
"learning_rate": 0.0001510230179028133,
|
933 |
+
"loss": 0.4548,
|
934 |
+
"step": 1180
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 1.01,
|
938 |
+
"grad_norm": 4.440161228179932,
|
939 |
+
"learning_rate": 0.00015059676044330776,
|
940 |
+
"loss": 0.5144,
|
941 |
+
"step": 1190
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 1.02,
|
945 |
+
"grad_norm": 4.055997371673584,
|
946 |
+
"learning_rate": 0.0001501705029838022,
|
947 |
+
"loss": 0.5741,
|
948 |
+
"step": 1200
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 1.02,
|
952 |
+
"eval_accuracy": 0.9157608695652174,
|
953 |
+
"eval_loss": 0.3336072862148285,
|
954 |
+
"eval_runtime": 21.7333,
|
955 |
+
"eval_samples_per_second": 33.865,
|
956 |
+
"eval_steps_per_second": 4.233,
|
957 |
+
"step": 1200
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 1.03,
|
961 |
+
"grad_norm": 3.851508617401123,
|
962 |
+
"learning_rate": 0.0001497442455242967,
|
963 |
+
"loss": 0.5196,
|
964 |
+
"step": 1210
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 1.04,
|
968 |
+
"grad_norm": 6.436770915985107,
|
969 |
+
"learning_rate": 0.00014931798806479114,
|
970 |
+
"loss": 0.5433,
|
971 |
+
"step": 1220
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 1.05,
|
975 |
+
"grad_norm": 4.73512601852417,
|
976 |
+
"learning_rate": 0.00014889173060528562,
|
977 |
+
"loss": 0.466,
|
978 |
+
"step": 1230
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 1.06,
|
982 |
+
"grad_norm": 6.008419036865234,
|
983 |
+
"learning_rate": 0.00014846547314578007,
|
984 |
+
"loss": 0.4656,
|
985 |
+
"step": 1240
|
986 |
+
},
|
987 |
+
{
|
988 |
+
"epoch": 1.07,
|
989 |
+
"grad_norm": 4.851034164428711,
|
990 |
+
"learning_rate": 0.00014803921568627451,
|
991 |
+
"loss": 0.4271,
|
992 |
+
"step": 1250
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 1.07,
|
996 |
+
"grad_norm": 4.398035526275635,
|
997 |
+
"learning_rate": 0.00014761295822676896,
|
998 |
+
"loss": 0.4296,
|
999 |
+
"step": 1260
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 1.08,
|
1003 |
+
"grad_norm": 4.0159454345703125,
|
1004 |
+
"learning_rate": 0.00014718670076726344,
|
1005 |
+
"loss": 0.4648,
|
1006 |
+
"step": 1270
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 1.09,
|
1010 |
+
"grad_norm": 5.387024879455566,
|
1011 |
+
"learning_rate": 0.0001467604433077579,
|
1012 |
+
"loss": 0.4818,
|
1013 |
+
"step": 1280
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 1.1,
|
1017 |
+
"grad_norm": 4.121237277984619,
|
1018 |
+
"learning_rate": 0.00014633418584825237,
|
1019 |
+
"loss": 0.4996,
|
1020 |
+
"step": 1290
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 1.11,
|
1024 |
+
"grad_norm": 2.9251198768615723,
|
1025 |
+
"learning_rate": 0.00014590792838874682,
|
1026 |
+
"loss": 0.4039,
|
1027 |
+
"step": 1300
|
1028 |
+
},
|
1029 |
+
{
|
1030 |
+
"epoch": 1.11,
|
1031 |
+
"eval_accuracy": 0.9130434782608695,
|
1032 |
+
"eval_loss": 0.32037827372550964,
|
1033 |
+
"eval_runtime": 22.2814,
|
1034 |
+
"eval_samples_per_second": 33.032,
|
1035 |
+
"eval_steps_per_second": 4.129,
|
1036 |
+
"step": 1300
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 1.12,
|
1040 |
+
"grad_norm": 5.795650482177734,
|
1041 |
+
"learning_rate": 0.00014548167092924127,
|
1042 |
+
"loss": 0.3762,
|
1043 |
+
"step": 1310
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 1.13,
|
1047 |
+
"grad_norm": 3.4435791969299316,
|
1048 |
+
"learning_rate": 0.00014505541346973572,
|
1049 |
+
"loss": 0.4195,
|
1050 |
+
"step": 1320
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 1.13,
|
1054 |
+
"grad_norm": 5.9583563804626465,
|
1055 |
+
"learning_rate": 0.0001446291560102302,
|
1056 |
+
"loss": 0.4561,
|
1057 |
+
"step": 1330
|
1058 |
+
},
|
1059 |
+
{
|
1060 |
+
"epoch": 1.14,
|
1061 |
+
"grad_norm": 4.873754501342773,
|
1062 |
+
"learning_rate": 0.00014420289855072464,
|
1063 |
+
"loss": 0.4756,
|
1064 |
+
"step": 1340
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 1.15,
|
1068 |
+
"grad_norm": 2.2037878036499023,
|
1069 |
+
"learning_rate": 0.00014377664109121912,
|
1070 |
+
"loss": 0.4494,
|
1071 |
+
"step": 1350
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 1.16,
|
1075 |
+
"grad_norm": 3.80642032623291,
|
1076 |
+
"learning_rate": 0.00014335038363171357,
|
1077 |
+
"loss": 0.454,
|
1078 |
+
"step": 1360
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 1.17,
|
1082 |
+
"grad_norm": 3.6881439685821533,
|
1083 |
+
"learning_rate": 0.00014292412617220802,
|
1084 |
+
"loss": 0.5275,
|
1085 |
+
"step": 1370
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.18,
|
1089 |
+
"grad_norm": 3.6306328773498535,
|
1090 |
+
"learning_rate": 0.00014249786871270247,
|
1091 |
+
"loss": 0.3859,
|
1092 |
+
"step": 1380
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 1.18,
|
1096 |
+
"grad_norm": 2.9194376468658447,
|
1097 |
+
"learning_rate": 0.00014207161125319695,
|
1098 |
+
"loss": 0.3732,
|
1099 |
+
"step": 1390
|
1100 |
+
},
|
1101 |
+
{
|
1102 |
+
"epoch": 1.19,
|
1103 |
+
"grad_norm": 6.103586673736572,
|
1104 |
+
"learning_rate": 0.0001416453537936914,
|
1105 |
+
"loss": 0.4347,
|
1106 |
+
"step": 1400
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 1.19,
|
1110 |
+
"eval_accuracy": 0.9307065217391305,
|
1111 |
+
"eval_loss": 0.3037649989128113,
|
1112 |
+
"eval_runtime": 21.9092,
|
1113 |
+
"eval_samples_per_second": 33.593,
|
1114 |
+
"eval_steps_per_second": 4.199,
|
1115 |
+
"step": 1400
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 1.2,
|
1119 |
+
"grad_norm": 3.9999098777770996,
|
1120 |
+
"learning_rate": 0.00014121909633418587,
|
1121 |
+
"loss": 0.33,
|
1122 |
+
"step": 1410
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 1.21,
|
1126 |
+
"grad_norm": 4.35922384262085,
|
1127 |
+
"learning_rate": 0.00014079283887468032,
|
1128 |
+
"loss": 0.4913,
|
1129 |
+
"step": 1420
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 1.22,
|
1133 |
+
"grad_norm": 5.2613019943237305,
|
1134 |
+
"learning_rate": 0.00014036658141517477,
|
1135 |
+
"loss": 0.4595,
|
1136 |
+
"step": 1430
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.23,
|
1140 |
+
"grad_norm": 4.525338172912598,
|
1141 |
+
"learning_rate": 0.00013994032395566922,
|
1142 |
+
"loss": 0.4338,
|
1143 |
+
"step": 1440
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 1.24,
|
1147 |
+
"grad_norm": 3.5422630310058594,
|
1148 |
+
"learning_rate": 0.0001395140664961637,
|
1149 |
+
"loss": 0.3084,
|
1150 |
+
"step": 1450
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 1.24,
|
1154 |
+
"grad_norm": 4.331848621368408,
|
1155 |
+
"learning_rate": 0.00013908780903665815,
|
1156 |
+
"loss": 0.4736,
|
1157 |
+
"step": 1460
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 1.25,
|
1161 |
+
"grad_norm": 3.7015225887298584,
|
1162 |
+
"learning_rate": 0.00013866155157715262,
|
1163 |
+
"loss": 0.4691,
|
1164 |
+
"step": 1470
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 1.26,
|
1168 |
+
"grad_norm": 3.8468117713928223,
|
1169 |
+
"learning_rate": 0.00013823529411764707,
|
1170 |
+
"loss": 0.3247,
|
1171 |
+
"step": 1480
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 1.27,
|
1175 |
+
"grad_norm": 4.560232639312744,
|
1176 |
+
"learning_rate": 0.00013780903665814152,
|
1177 |
+
"loss": 0.4162,
|
1178 |
+
"step": 1490
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 1.28,
|
1182 |
+
"grad_norm": 4.275567531585693,
|
1183 |
+
"learning_rate": 0.00013738277919863597,
|
1184 |
+
"loss": 0.3639,
|
1185 |
+
"step": 1500
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 1.28,
|
1189 |
+
"eval_accuracy": 0.9252717391304348,
|
1190 |
+
"eval_loss": 0.2954687476158142,
|
1191 |
+
"eval_runtime": 21.3813,
|
1192 |
+
"eval_samples_per_second": 34.423,
|
1193 |
+
"eval_steps_per_second": 4.303,
|
1194 |
+
"step": 1500
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 1.29,
|
1198 |
+
"grad_norm": 4.1551513671875,
|
1199 |
+
"learning_rate": 0.00013695652173913045,
|
1200 |
+
"loss": 0.4241,
|
1201 |
+
"step": 1510
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 1.3,
|
1205 |
+
"grad_norm": 3.172189712524414,
|
1206 |
+
"learning_rate": 0.0001365302642796249,
|
1207 |
+
"loss": 0.5576,
|
1208 |
+
"step": 1520
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 1.3,
|
1212 |
+
"grad_norm": 4.926159858703613,
|
1213 |
+
"learning_rate": 0.00013610400682011938,
|
1214 |
+
"loss": 0.3687,
|
1215 |
+
"step": 1530
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 1.31,
|
1219 |
+
"grad_norm": 5.438605785369873,
|
1220 |
+
"learning_rate": 0.00013567774936061383,
|
1221 |
+
"loss": 0.4823,
|
1222 |
+
"step": 1540
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 1.32,
|
1226 |
+
"grad_norm": 4.588318347930908,
|
1227 |
+
"learning_rate": 0.00013525149190110828,
|
1228 |
+
"loss": 0.3816,
|
1229 |
+
"step": 1550
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.33,
|
1233 |
+
"grad_norm": 3.6473512649536133,
|
1234 |
+
"learning_rate": 0.00013482523444160273,
|
1235 |
+
"loss": 0.4009,
|
1236 |
+
"step": 1560
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 1.34,
|
1240 |
+
"grad_norm": 1.983258605003357,
|
1241 |
+
"learning_rate": 0.0001343989769820972,
|
1242 |
+
"loss": 0.412,
|
1243 |
+
"step": 1570
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 1.35,
|
1247 |
+
"grad_norm": 4.585762023925781,
|
1248 |
+
"learning_rate": 0.00013397271952259165,
|
1249 |
+
"loss": 0.4427,
|
1250 |
+
"step": 1580
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.36,
|
1254 |
+
"grad_norm": 3.362663984298706,
|
1255 |
+
"learning_rate": 0.00013354646206308613,
|
1256 |
+
"loss": 0.3946,
|
1257 |
+
"step": 1590
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 1.36,
|
1261 |
+
"grad_norm": 2.671499729156494,
|
1262 |
+
"learning_rate": 0.00013312020460358058,
|
1263 |
+
"loss": 0.4282,
|
1264 |
+
"step": 1600
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 1.36,
|
1268 |
+
"eval_accuracy": 0.9293478260869565,
|
1269 |
+
"eval_loss": 0.29482707381248474,
|
1270 |
+
"eval_runtime": 22.1321,
|
1271 |
+
"eval_samples_per_second": 33.255,
|
1272 |
+
"eval_steps_per_second": 4.157,
|
1273 |
+
"step": 1600
|
1274 |
+
},
|
1275 |
+
{
|
1276 |
+
"epoch": 1.37,
|
1277 |
+
"grad_norm": 4.811617374420166,
|
1278 |
+
"learning_rate": 0.00013269394714407503,
|
1279 |
+
"loss": 0.4589,
|
1280 |
+
"step": 1610
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.38,
|
1284 |
+
"grad_norm": 3.9043726921081543,
|
1285 |
+
"learning_rate": 0.00013226768968456948,
|
1286 |
+
"loss": 0.4281,
|
1287 |
+
"step": 1620
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 1.39,
|
1291 |
+
"grad_norm": 3.5376808643341064,
|
1292 |
+
"learning_rate": 0.00013184143222506395,
|
1293 |
+
"loss": 0.3975,
|
1294 |
+
"step": 1630
|
1295 |
+
},
|
1296 |
+
{
|
1297 |
+
"epoch": 1.4,
|
1298 |
+
"grad_norm": 4.834260940551758,
|
1299 |
+
"learning_rate": 0.0001314151747655584,
|
1300 |
+
"loss": 0.4094,
|
1301 |
+
"step": 1640
|
1302 |
+
},
|
1303 |
+
{
|
1304 |
+
"epoch": 1.41,
|
1305 |
+
"grad_norm": 4.845081806182861,
|
1306 |
+
"learning_rate": 0.00013098891730605288,
|
1307 |
+
"loss": 0.3806,
|
1308 |
+
"step": 1650
|
1309 |
+
},
|
1310 |
+
{
|
1311 |
+
"epoch": 1.42,
|
1312 |
+
"grad_norm": 4.995219707489014,
|
1313 |
+
"learning_rate": 0.00013056265984654733,
|
1314 |
+
"loss": 0.4762,
|
1315 |
+
"step": 1660
|
1316 |
+
},
|
1317 |
+
{
|
1318 |
+
"epoch": 1.42,
|
1319 |
+
"grad_norm": 4.277749061584473,
|
1320 |
+
"learning_rate": 0.00013013640238704178,
|
1321 |
+
"loss": 0.4923,
|
1322 |
+
"step": 1670
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 1.43,
|
1326 |
+
"grad_norm": 5.35748815536499,
|
1327 |
+
"learning_rate": 0.00012971014492753623,
|
1328 |
+
"loss": 0.3878,
|
1329 |
+
"step": 1680
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.44,
|
1333 |
+
"grad_norm": 3.3792693614959717,
|
1334 |
+
"learning_rate": 0.0001292838874680307,
|
1335 |
+
"loss": 0.369,
|
1336 |
+
"step": 1690
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.45,
|
1340 |
+
"grad_norm": 4.180275917053223,
|
1341 |
+
"learning_rate": 0.00012885763000852516,
|
1342 |
+
"loss": 0.4375,
|
1343 |
+
"step": 1700
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 1.45,
|
1347 |
+
"eval_accuracy": 0.9211956521739131,
|
1348 |
+
"eval_loss": 0.2868107855319977,
|
1349 |
+
"eval_runtime": 22.1903,
|
1350 |
+
"eval_samples_per_second": 33.168,
|
1351 |
+
"eval_steps_per_second": 4.146,
|
1352 |
+
"step": 1700
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 1.46,
|
1356 |
+
"grad_norm": 2.387194871902466,
|
1357 |
+
"learning_rate": 0.00012843137254901963,
|
1358 |
+
"loss": 0.4346,
|
1359 |
+
"step": 1710
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 1.47,
|
1363 |
+
"grad_norm": 1.87982177734375,
|
1364 |
+
"learning_rate": 0.00012800511508951408,
|
1365 |
+
"loss": 0.3212,
|
1366 |
+
"step": 1720
|
1367 |
+
},
|
1368 |
+
{
|
1369 |
+
"epoch": 1.47,
|
1370 |
+
"grad_norm": 2.5354747772216797,
|
1371 |
+
"learning_rate": 0.00012757885763000853,
|
1372 |
+
"loss": 0.3389,
|
1373 |
+
"step": 1730
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 1.48,
|
1377 |
+
"grad_norm": 3.975893259048462,
|
1378 |
+
"learning_rate": 0.00012715260017050298,
|
1379 |
+
"loss": 0.422,
|
1380 |
+
"step": 1740
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 1.49,
|
1384 |
+
"grad_norm": 5.26716423034668,
|
1385 |
+
"learning_rate": 0.00012672634271099746,
|
1386 |
+
"loss": 0.356,
|
1387 |
+
"step": 1750
|
1388 |
+
},
|
1389 |
+
{
|
1390 |
+
"epoch": 1.5,
|
1391 |
+
"grad_norm": 3.5218489170074463,
|
1392 |
+
"learning_rate": 0.0001263000852514919,
|
1393 |
+
"loss": 0.5009,
|
1394 |
+
"step": 1760
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 1.51,
|
1398 |
+
"grad_norm": 3.965714931488037,
|
1399 |
+
"learning_rate": 0.00012587382779198638,
|
1400 |
+
"loss": 0.4176,
|
1401 |
+
"step": 1770
|
1402 |
+
},
|
1403 |
+
{
|
1404 |
+
"epoch": 1.52,
|
1405 |
+
"grad_norm": 4.640725612640381,
|
1406 |
+
"learning_rate": 0.00012544757033248083,
|
1407 |
+
"loss": 0.4312,
|
1408 |
+
"step": 1780
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 1.53,
|
1412 |
+
"grad_norm": 3.008267879486084,
|
1413 |
+
"learning_rate": 0.00012502131287297528,
|
1414 |
+
"loss": 0.4443,
|
1415 |
+
"step": 1790
|
1416 |
+
},
|
1417 |
+
{
|
1418 |
+
"epoch": 1.53,
|
1419 |
+
"grad_norm": 2.9221649169921875,
|
1420 |
+
"learning_rate": 0.00012459505541346973,
|
1421 |
+
"loss": 0.3063,
|
1422 |
+
"step": 1800
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 1.53,
|
1426 |
+
"eval_accuracy": 0.9334239130434783,
|
1427 |
+
"eval_loss": 0.28607919812202454,
|
1428 |
+
"eval_runtime": 21.4966,
|
1429 |
+
"eval_samples_per_second": 34.238,
|
1430 |
+
"eval_steps_per_second": 4.28,
|
1431 |
+
"step": 1800
|
1432 |
+
},
|
1433 |
+
{
|
1434 |
+
"epoch": 1.54,
|
1435 |
+
"grad_norm": 4.676169395446777,
|
1436 |
+
"learning_rate": 0.0001241687979539642,
|
1437 |
+
"loss": 0.4685,
|
1438 |
+
"step": 1810
|
1439 |
+
},
|
1440 |
+
{
|
1441 |
+
"epoch": 1.55,
|
1442 |
+
"grad_norm": 4.044462203979492,
|
1443 |
+
"learning_rate": 0.00012374254049445866,
|
1444 |
+
"loss": 0.4094,
|
1445 |
+
"step": 1820
|
1446 |
+
},
|
1447 |
+
{
|
1448 |
+
"epoch": 1.56,
|
1449 |
+
"grad_norm": 3.891371726989746,
|
1450 |
+
"learning_rate": 0.00012331628303495314,
|
1451 |
+
"loss": 0.4552,
|
1452 |
+
"step": 1830
|
1453 |
+
},
|
1454 |
+
{
|
1455 |
+
"epoch": 1.57,
|
1456 |
+
"grad_norm": 5.24341344833374,
|
1457 |
+
"learning_rate": 0.0001228900255754476,
|
1458 |
+
"loss": 0.2682,
|
1459 |
+
"step": 1840
|
1460 |
+
},
|
1461 |
+
{
|
1462 |
+
"epoch": 1.58,
|
1463 |
+
"grad_norm": 5.291731834411621,
|
1464 |
+
"learning_rate": 0.00012246376811594204,
|
1465 |
+
"loss": 0.4047,
|
1466 |
+
"step": 1850
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.59,
|
1470 |
+
"grad_norm": 2.1534364223480225,
|
1471 |
+
"learning_rate": 0.0001220375106564365,
|
1472 |
+
"loss": 0.2985,
|
1473 |
+
"step": 1860
|
1474 |
+
},
|
1475 |
+
{
|
1476 |
+
"epoch": 1.59,
|
1477 |
+
"grad_norm": 4.051937580108643,
|
1478 |
+
"learning_rate": 0.00012161125319693096,
|
1479 |
+
"loss": 0.3872,
|
1480 |
+
"step": 1870
|
1481 |
+
},
|
1482 |
+
{
|
1483 |
+
"epoch": 1.6,
|
1484 |
+
"grad_norm": 4.822917461395264,
|
1485 |
+
"learning_rate": 0.00012118499573742541,
|
1486 |
+
"loss": 0.3532,
|
1487 |
+
"step": 1880
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 1.61,
|
1491 |
+
"grad_norm": 2.520695209503174,
|
1492 |
+
"learning_rate": 0.00012075873827791986,
|
1493 |
+
"loss": 0.3878,
|
1494 |
+
"step": 1890
|
1495 |
+
},
|
1496 |
+
{
|
1497 |
+
"epoch": 1.62,
|
1498 |
+
"grad_norm": 1.7465733289718628,
|
1499 |
+
"learning_rate": 0.00012033248081841433,
|
1500 |
+
"loss": 0.3549,
|
1501 |
+
"step": 1900
|
1502 |
+
},
|
1503 |
+
{
|
1504 |
+
"epoch": 1.62,
|
1505 |
+
"eval_accuracy": 0.9293478260869565,
|
1506 |
+
"eval_loss": 0.28257426619529724,
|
1507 |
+
"eval_runtime": 22.1114,
|
1508 |
+
"eval_samples_per_second": 33.286,
|
1509 |
+
"eval_steps_per_second": 4.161,
|
1510 |
+
"step": 1900
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 1.63,
|
1514 |
+
"grad_norm": 4.117008686065674,
|
1515 |
+
"learning_rate": 0.00011990622335890877,
|
1516 |
+
"loss": 0.345,
|
1517 |
+
"step": 1910
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 1.64,
|
1521 |
+
"grad_norm": 2.7594146728515625,
|
1522 |
+
"learning_rate": 0.00011947996589940325,
|
1523 |
+
"loss": 0.3515,
|
1524 |
+
"step": 1920
|
1525 |
+
},
|
1526 |
+
{
|
1527 |
+
"epoch": 1.65,
|
1528 |
+
"grad_norm": 3.63336181640625,
|
1529 |
+
"learning_rate": 0.00011905370843989769,
|
1530 |
+
"loss": 0.4509,
|
1531 |
+
"step": 1930
|
1532 |
+
},
|
1533 |
+
{
|
1534 |
+
"epoch": 1.65,
|
1535 |
+
"grad_norm": 2.090906858444214,
|
1536 |
+
"learning_rate": 0.00011862745098039216,
|
1537 |
+
"loss": 0.4249,
|
1538 |
+
"step": 1940
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 1.66,
|
1542 |
+
"grad_norm": 7.205491542816162,
|
1543 |
+
"learning_rate": 0.00011820119352088661,
|
1544 |
+
"loss": 0.3791,
|
1545 |
+
"step": 1950
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 1.67,
|
1549 |
+
"grad_norm": 4.450414657592773,
|
1550 |
+
"learning_rate": 0.00011777493606138108,
|
1551 |
+
"loss": 0.3818,
|
1552 |
+
"step": 1960
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 1.68,
|
1556 |
+
"grad_norm": 3.7030208110809326,
|
1557 |
+
"learning_rate": 0.00011734867860187553,
|
1558 |
+
"loss": 0.3881,
|
1559 |
+
"step": 1970
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 1.69,
|
1563 |
+
"grad_norm": 3.2162046432495117,
|
1564 |
+
"learning_rate": 0.00011692242114237,
|
1565 |
+
"loss": 0.4438,
|
1566 |
+
"step": 1980
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 1.7,
|
1570 |
+
"grad_norm": 3.964151382446289,
|
1571 |
+
"learning_rate": 0.00011649616368286444,
|
1572 |
+
"loss": 0.3491,
|
1573 |
+
"step": 1990
|
1574 |
+
},
|
1575 |
+
{
|
1576 |
+
"epoch": 1.71,
|
1577 |
+
"grad_norm": 6.592412948608398,
|
1578 |
+
"learning_rate": 0.00011606990622335892,
|
1579 |
+
"loss": 0.4326,
|
1580 |
+
"step": 2000
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 1.71,
|
1584 |
+
"eval_accuracy": 0.9347826086956522,
|
1585 |
+
"eval_loss": 0.26979920268058777,
|
1586 |
+
"eval_runtime": 21.9026,
|
1587 |
+
"eval_samples_per_second": 33.603,
|
1588 |
+
"eval_steps_per_second": 4.2,
|
1589 |
+
"step": 2000
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 1.71,
|
1593 |
+
"grad_norm": 3.8649954795837402,
|
1594 |
+
"learning_rate": 0.00011564364876385337,
|
1595 |
+
"loss": 0.4652,
|
1596 |
+
"step": 2010
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.72,
|
1600 |
+
"grad_norm": 4.012170791625977,
|
1601 |
+
"learning_rate": 0.00011521739130434783,
|
1602 |
+
"loss": 0.4145,
|
1603 |
+
"step": 2020
|
1604 |
+
},
|
1605 |
+
{
|
1606 |
+
"epoch": 1.73,
|
1607 |
+
"grad_norm": 3.8874316215515137,
|
1608 |
+
"learning_rate": 0.00011479113384484228,
|
1609 |
+
"loss": 0.3241,
|
1610 |
+
"step": 2030
|
1611 |
+
},
|
1612 |
+
{
|
1613 |
+
"epoch": 1.74,
|
1614 |
+
"grad_norm": 3.7281954288482666,
|
1615 |
+
"learning_rate": 0.00011436487638533676,
|
1616 |
+
"loss": 0.3844,
|
1617 |
+
"step": 2040
|
1618 |
+
},
|
1619 |
+
{
|
1620 |
+
"epoch": 1.75,
|
1621 |
+
"grad_norm": 3.496342182159424,
|
1622 |
+
"learning_rate": 0.00011393861892583119,
|
1623 |
+
"loss": 0.4773,
|
1624 |
+
"step": 2050
|
1625 |
+
},
|
1626 |
+
{
|
1627 |
+
"epoch": 1.76,
|
1628 |
+
"grad_norm": 2.408202886581421,
|
1629 |
+
"learning_rate": 0.00011351236146632567,
|
1630 |
+
"loss": 0.44,
|
1631 |
+
"step": 2060
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 1.76,
|
1635 |
+
"grad_norm": 1.211472749710083,
|
1636 |
+
"learning_rate": 0.00011308610400682012,
|
1637 |
+
"loss": 0.2852,
|
1638 |
+
"step": 2070
|
1639 |
+
},
|
1640 |
+
{
|
1641 |
+
"epoch": 1.77,
|
1642 |
+
"grad_norm": 3.826094627380371,
|
1643 |
+
"learning_rate": 0.00011265984654731458,
|
1644 |
+
"loss": 0.4446,
|
1645 |
+
"step": 2080
|
1646 |
+
},
|
1647 |
+
{
|
1648 |
+
"epoch": 1.78,
|
1649 |
+
"grad_norm": 3.4480748176574707,
|
1650 |
+
"learning_rate": 0.00011223358908780903,
|
1651 |
+
"loss": 0.4111,
|
1652 |
+
"step": 2090
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 1.79,
|
1656 |
+
"grad_norm": 4.725152492523193,
|
1657 |
+
"learning_rate": 0.00011180733162830351,
|
1658 |
+
"loss": 0.3697,
|
1659 |
+
"step": 2100
|
1660 |
+
},
|
1661 |
+
{
|
1662 |
+
"epoch": 1.79,
|
1663 |
+
"eval_accuracy": 0.9279891304347826,
|
1664 |
+
"eval_loss": 0.2602107524871826,
|
1665 |
+
"eval_runtime": 21.9521,
|
1666 |
+
"eval_samples_per_second": 33.528,
|
1667 |
+
"eval_steps_per_second": 4.191,
|
1668 |
+
"step": 2100
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 1.8,
|
1672 |
+
"grad_norm": 3.1105542182922363,
|
1673 |
+
"learning_rate": 0.00011138107416879794,
|
1674 |
+
"loss": 0.3935,
|
1675 |
+
"step": 2110
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 1.81,
|
1679 |
+
"grad_norm": 4.203391075134277,
|
1680 |
+
"learning_rate": 0.00011095481670929242,
|
1681 |
+
"loss": 0.4265,
|
1682 |
+
"step": 2120
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 1.82,
|
1686 |
+
"grad_norm": 3.371384382247925,
|
1687 |
+
"learning_rate": 0.00011052855924978687,
|
1688 |
+
"loss": 0.3914,
|
1689 |
+
"step": 2130
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 1.82,
|
1693 |
+
"grad_norm": 4.558941841125488,
|
1694 |
+
"learning_rate": 0.00011010230179028133,
|
1695 |
+
"loss": 0.4196,
|
1696 |
+
"step": 2140
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 1.83,
|
1700 |
+
"grad_norm": 2.8182742595672607,
|
1701 |
+
"learning_rate": 0.00010967604433077578,
|
1702 |
+
"loss": 0.3749,
|
1703 |
+
"step": 2150
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 1.84,
|
1707 |
+
"grad_norm": 5.837876319885254,
|
1708 |
+
"learning_rate": 0.00010924978687127026,
|
1709 |
+
"loss": 0.3435,
|
1710 |
+
"step": 2160
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 1.85,
|
1714 |
+
"grad_norm": 4.261623859405518,
|
1715 |
+
"learning_rate": 0.0001088235294117647,
|
1716 |
+
"loss": 0.377,
|
1717 |
+
"step": 2170
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 1.86,
|
1721 |
+
"grad_norm": 3.3105015754699707,
|
1722 |
+
"learning_rate": 0.00010839727195225917,
|
1723 |
+
"loss": 0.3968,
|
1724 |
+
"step": 2180
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 1.87,
|
1728 |
+
"grad_norm": 4.884303092956543,
|
1729 |
+
"learning_rate": 0.00010797101449275362,
|
1730 |
+
"loss": 0.3476,
|
1731 |
+
"step": 2190
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 1.88,
|
1735 |
+
"grad_norm": 2.950490951538086,
|
1736 |
+
"learning_rate": 0.00010754475703324809,
|
1737 |
+
"loss": 0.3155,
|
1738 |
+
"step": 2200
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 1.88,
|
1742 |
+
"eval_accuracy": 0.936141304347826,
|
1743 |
+
"eval_loss": 0.2523466646671295,
|
1744 |
+
"eval_runtime": 21.8751,
|
1745 |
+
"eval_samples_per_second": 33.646,
|
1746 |
+
"eval_steps_per_second": 4.206,
|
1747 |
+
"step": 2200
|
1748 |
+
},
|
1749 |
+
{
|
1750 |
+
"epoch": 1.88,
|
1751 |
+
"grad_norm": 2.758512258529663,
|
1752 |
+
"learning_rate": 0.00010711849957374254,
|
1753 |
+
"loss": 0.2808,
|
1754 |
+
"step": 2210
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 1.89,
|
1758 |
+
"grad_norm": 3.4424259662628174,
|
1759 |
+
"learning_rate": 0.00010669224211423701,
|
1760 |
+
"loss": 0.3434,
|
1761 |
+
"step": 2220
|
1762 |
+
},
|
1763 |
+
{
|
1764 |
+
"epoch": 1.9,
|
1765 |
+
"grad_norm": 3.334716558456421,
|
1766 |
+
"learning_rate": 0.00010626598465473145,
|
1767 |
+
"loss": 0.4246,
|
1768 |
+
"step": 2230
|
1769 |
+
},
|
1770 |
+
{
|
1771 |
+
"epoch": 1.91,
|
1772 |
+
"grad_norm": 3.309183120727539,
|
1773 |
+
"learning_rate": 0.00010583972719522593,
|
1774 |
+
"loss": 0.3535,
|
1775 |
+
"step": 2240
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 1.92,
|
1779 |
+
"grad_norm": 3.0825111865997314,
|
1780 |
+
"learning_rate": 0.00010541346973572037,
|
1781 |
+
"loss": 0.4027,
|
1782 |
+
"step": 2250
|
1783 |
+
},
|
1784 |
+
{
|
1785 |
+
"epoch": 1.93,
|
1786 |
+
"grad_norm": 4.07969856262207,
|
1787 |
+
"learning_rate": 0.00010498721227621484,
|
1788 |
+
"loss": 0.4758,
|
1789 |
+
"step": 2260
|
1790 |
+
},
|
1791 |
+
{
|
1792 |
+
"epoch": 1.94,
|
1793 |
+
"grad_norm": 3.2373547554016113,
|
1794 |
+
"learning_rate": 0.00010456095481670929,
|
1795 |
+
"loss": 0.4168,
|
1796 |
+
"step": 2270
|
1797 |
+
},
|
1798 |
+
{
|
1799 |
+
"epoch": 1.94,
|
1800 |
+
"grad_norm": 3.9943110942840576,
|
1801 |
+
"learning_rate": 0.00010413469735720376,
|
1802 |
+
"loss": 0.3319,
|
1803 |
+
"step": 2280
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 1.95,
|
1807 |
+
"grad_norm": 4.852569103240967,
|
1808 |
+
"learning_rate": 0.0001037084398976982,
|
1809 |
+
"loss": 0.3544,
|
1810 |
+
"step": 2290
|
1811 |
+
},
|
1812 |
+
{
|
1813 |
+
"epoch": 1.96,
|
1814 |
+
"grad_norm": 2.613586902618408,
|
1815 |
+
"learning_rate": 0.00010328218243819268,
|
1816 |
+
"loss": 0.3348,
|
1817 |
+
"step": 2300
|
1818 |
+
},
|
1819 |
+
{
|
1820 |
+
"epoch": 1.96,
|
1821 |
+
"eval_accuracy": 0.9470108695652174,
|
1822 |
+
"eval_loss": 0.2506195604801178,
|
1823 |
+
"eval_runtime": 22.0934,
|
1824 |
+
"eval_samples_per_second": 33.313,
|
1825 |
+
"eval_steps_per_second": 4.164,
|
1826 |
+
"step": 2300
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 1.97,
|
1830 |
+
"grad_norm": 3.5085036754608154,
|
1831 |
+
"learning_rate": 0.00010285592497868713,
|
1832 |
+
"loss": 0.3908,
|
1833 |
+
"step": 2310
|
1834 |
+
},
|
1835 |
+
{
|
1836 |
+
"epoch": 1.98,
|
1837 |
+
"grad_norm": 2.5577197074890137,
|
1838 |
+
"learning_rate": 0.00010242966751918159,
|
1839 |
+
"loss": 0.3995,
|
1840 |
+
"step": 2320
|
1841 |
+
},
|
1842 |
+
{
|
1843 |
+
"epoch": 1.99,
|
1844 |
+
"grad_norm": 3.079270124435425,
|
1845 |
+
"learning_rate": 0.00010200341005967604,
|
1846 |
+
"loss": 0.2755,
|
1847 |
+
"step": 2330
|
1848 |
+
},
|
1849 |
+
{
|
1850 |
+
"epoch": 1.99,
|
1851 |
+
"grad_norm": 2.5068159103393555,
|
1852 |
+
"learning_rate": 0.00010157715260017052,
|
1853 |
+
"loss": 0.4502,
|
1854 |
+
"step": 2340
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 2.0,
|
1858 |
+
"grad_norm": 4.730781555175781,
|
1859 |
+
"learning_rate": 0.00010115089514066495,
|
1860 |
+
"loss": 0.3456,
|
1861 |
+
"step": 2350
|
1862 |
+
},
|
1863 |
+
{
|
1864 |
+
"epoch": 2.01,
|
1865 |
+
"grad_norm": 3.066279888153076,
|
1866 |
+
"learning_rate": 0.00010072463768115943,
|
1867 |
+
"loss": 0.3106,
|
1868 |
+
"step": 2360
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 2.02,
|
1872 |
+
"grad_norm": 5.494949817657471,
|
1873 |
+
"learning_rate": 0.00010029838022165388,
|
1874 |
+
"loss": 0.4091,
|
1875 |
+
"step": 2370
|
1876 |
+
},
|
1877 |
+
{
|
1878 |
+
"epoch": 2.03,
|
1879 |
+
"grad_norm": 4.616675853729248,
|
1880 |
+
"learning_rate": 9.987212276214834e-05,
|
1881 |
+
"loss": 0.349,
|
1882 |
+
"step": 2380
|
1883 |
+
},
|
1884 |
+
{
|
1885 |
+
"epoch": 2.04,
|
1886 |
+
"grad_norm": 3.0624680519104004,
|
1887 |
+
"learning_rate": 9.94458653026428e-05,
|
1888 |
+
"loss": 0.3441,
|
1889 |
+
"step": 2390
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 2.05,
|
1893 |
+
"grad_norm": 3.0316741466522217,
|
1894 |
+
"learning_rate": 9.901960784313727e-05,
|
1895 |
+
"loss": 0.3854,
|
1896 |
+
"step": 2400
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 2.05,
|
1900 |
+
"eval_accuracy": 0.9320652173913043,
|
1901 |
+
"eval_loss": 0.25646623969078064,
|
1902 |
+
"eval_runtime": 21.873,
|
1903 |
+
"eval_samples_per_second": 33.649,
|
1904 |
+
"eval_steps_per_second": 4.206,
|
1905 |
+
"step": 2400
|
1906 |
+
},
|
1907 |
+
{
|
1908 |
+
"epoch": 2.05,
|
1909 |
+
"grad_norm": 3.854029893875122,
|
1910 |
+
"learning_rate": 9.859335038363172e-05,
|
1911 |
+
"loss": 0.3514,
|
1912 |
+
"step": 2410
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 2.06,
|
1916 |
+
"grad_norm": 4.106563568115234,
|
1917 |
+
"learning_rate": 9.816709292412618e-05,
|
1918 |
+
"loss": 0.3358,
|
1919 |
+
"step": 2420
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 2.07,
|
1923 |
+
"grad_norm": 3.2283833026885986,
|
1924 |
+
"learning_rate": 9.774083546462064e-05,
|
1925 |
+
"loss": 0.4073,
|
1926 |
+
"step": 2430
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 2.08,
|
1930 |
+
"grad_norm": 3.7362446784973145,
|
1931 |
+
"learning_rate": 9.73145780051151e-05,
|
1932 |
+
"loss": 0.3858,
|
1933 |
+
"step": 2440
|
1934 |
+
},
|
1935 |
+
{
|
1936 |
+
"epoch": 2.09,
|
1937 |
+
"grad_norm": 2.262639284133911,
|
1938 |
+
"learning_rate": 9.688832054560956e-05,
|
1939 |
+
"loss": 0.3603,
|
1940 |
+
"step": 2450
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 2.1,
|
1944 |
+
"grad_norm": 3.2846786975860596,
|
1945 |
+
"learning_rate": 9.646206308610402e-05,
|
1946 |
+
"loss": 0.3249,
|
1947 |
+
"step": 2460
|
1948 |
+
},
|
1949 |
+
{
|
1950 |
+
"epoch": 2.11,
|
1951 |
+
"grad_norm": 3.0911221504211426,
|
1952 |
+
"learning_rate": 9.603580562659847e-05,
|
1953 |
+
"loss": 0.3276,
|
1954 |
+
"step": 2470
|
1955 |
+
},
|
1956 |
+
{
|
1957 |
+
"epoch": 2.11,
|
1958 |
+
"grad_norm": 3.0129778385162354,
|
1959 |
+
"learning_rate": 9.560954816709293e-05,
|
1960 |
+
"loss": 0.3321,
|
1961 |
+
"step": 2480
|
1962 |
+
},
|
1963 |
+
{
|
1964 |
+
"epoch": 2.12,
|
1965 |
+
"grad_norm": 1.167391061782837,
|
1966 |
+
"learning_rate": 9.51832907075874e-05,
|
1967 |
+
"loss": 0.3583,
|
1968 |
+
"step": 2490
|
1969 |
+
},
|
1970 |
+
{
|
1971 |
+
"epoch": 2.13,
|
1972 |
+
"grad_norm": 3.6704165935516357,
|
1973 |
+
"learning_rate": 9.475703324808185e-05,
|
1974 |
+
"loss": 0.3951,
|
1975 |
+
"step": 2500
|
1976 |
+
},
|
1977 |
+
{
|
1978 |
+
"epoch": 2.13,
|
1979 |
+
"eval_accuracy": 0.9402173913043478,
|
1980 |
+
"eval_loss": 0.24821878969669342,
|
1981 |
+
"eval_runtime": 22.0497,
|
1982 |
+
"eval_samples_per_second": 33.379,
|
1983 |
+
"eval_steps_per_second": 4.172,
|
1984 |
+
"step": 2500
|
1985 |
+
},
|
1986 |
+
{
|
1987 |
+
"epoch": 2.14,
|
1988 |
+
"grad_norm": 4.554658889770508,
|
1989 |
+
"learning_rate": 9.433077578857631e-05,
|
1990 |
+
"loss": 0.4089,
|
1991 |
+
"step": 2510
|
1992 |
+
},
|
1993 |
+
{
|
1994 |
+
"epoch": 2.15,
|
1995 |
+
"grad_norm": 1.6168781518936157,
|
1996 |
+
"learning_rate": 9.390451832907077e-05,
|
1997 |
+
"loss": 0.381,
|
1998 |
+
"step": 2520
|
1999 |
+
},
|
2000 |
+
{
|
2001 |
+
"epoch": 2.16,
|
2002 |
+
"grad_norm": 3.4385123252868652,
|
2003 |
+
"learning_rate": 9.347826086956522e-05,
|
2004 |
+
"loss": 0.3826,
|
2005 |
+
"step": 2530
|
2006 |
+
},
|
2007 |
+
{
|
2008 |
+
"epoch": 2.17,
|
2009 |
+
"grad_norm": 2.3573708534240723,
|
2010 |
+
"learning_rate": 9.305200341005969e-05,
|
2011 |
+
"loss": 0.4197,
|
2012 |
+
"step": 2540
|
2013 |
+
},
|
2014 |
+
{
|
2015 |
+
"epoch": 2.17,
|
2016 |
+
"grad_norm": 3.4988811016082764,
|
2017 |
+
"learning_rate": 9.262574595055415e-05,
|
2018 |
+
"loss": 0.3862,
|
2019 |
+
"step": 2550
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 2.18,
|
2023 |
+
"grad_norm": 2.5783936977386475,
|
2024 |
+
"learning_rate": 9.21994884910486e-05,
|
2025 |
+
"loss": 0.4243,
|
2026 |
+
"step": 2560
|
2027 |
+
},
|
2028 |
+
{
|
2029 |
+
"epoch": 2.19,
|
2030 |
+
"grad_norm": 4.15625,
|
2031 |
+
"learning_rate": 9.177323103154306e-05,
|
2032 |
+
"loss": 0.4088,
|
2033 |
+
"step": 2570
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 2.2,
|
2037 |
+
"grad_norm": 1.7744431495666504,
|
2038 |
+
"learning_rate": 9.134697357203753e-05,
|
2039 |
+
"loss": 0.2933,
|
2040 |
+
"step": 2580
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 2.21,
|
2044 |
+
"grad_norm": 3.8392601013183594,
|
2045 |
+
"learning_rate": 9.092071611253197e-05,
|
2046 |
+
"loss": 0.3433,
|
2047 |
+
"step": 2590
|
2048 |
+
},
|
2049 |
+
{
|
2050 |
+
"epoch": 2.22,
|
2051 |
+
"grad_norm": 1.8646149635314941,
|
2052 |
+
"learning_rate": 9.049445865302644e-05,
|
2053 |
+
"loss": 0.3531,
|
2054 |
+
"step": 2600
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 2.22,
|
2058 |
+
"eval_accuracy": 0.9402173913043478,
|
2059 |
+
"eval_loss": 0.24554158747196198,
|
2060 |
+
"eval_runtime": 22.0249,
|
2061 |
+
"eval_samples_per_second": 33.417,
|
2062 |
+
"eval_steps_per_second": 4.177,
|
2063 |
+
"step": 2600
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 2.23,
|
2067 |
+
"grad_norm": 2.4835286140441895,
|
2068 |
+
"learning_rate": 9.00682011935209e-05,
|
2069 |
+
"loss": 0.3631,
|
2070 |
+
"step": 2610
|
2071 |
+
},
|
2072 |
+
{
|
2073 |
+
"epoch": 2.23,
|
2074 |
+
"grad_norm": 4.589507102966309,
|
2075 |
+
"learning_rate": 8.964194373401535e-05,
|
2076 |
+
"loss": 0.2747,
|
2077 |
+
"step": 2620
|
2078 |
+
},
|
2079 |
+
{
|
2080 |
+
"epoch": 2.24,
|
2081 |
+
"grad_norm": 4.478066444396973,
|
2082 |
+
"learning_rate": 8.921568627450981e-05,
|
2083 |
+
"loss": 0.352,
|
2084 |
+
"step": 2630
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 2.25,
|
2088 |
+
"grad_norm": 3.647719383239746,
|
2089 |
+
"learning_rate": 8.878942881500428e-05,
|
2090 |
+
"loss": 0.3804,
|
2091 |
+
"step": 2640
|
2092 |
+
},
|
2093 |
+
{
|
2094 |
+
"epoch": 2.26,
|
2095 |
+
"grad_norm": 3.516533613204956,
|
2096 |
+
"learning_rate": 8.836317135549873e-05,
|
2097 |
+
"loss": 0.3542,
|
2098 |
+
"step": 2650
|
2099 |
+
},
|
2100 |
+
{
|
2101 |
+
"epoch": 2.27,
|
2102 |
+
"grad_norm": 4.30831241607666,
|
2103 |
+
"learning_rate": 8.793691389599319e-05,
|
2104 |
+
"loss": 0.4834,
|
2105 |
+
"step": 2660
|
2106 |
+
},
|
2107 |
+
{
|
2108 |
+
"epoch": 2.28,
|
2109 |
+
"grad_norm": 3.8890135288238525,
|
2110 |
+
"learning_rate": 8.751065643648765e-05,
|
2111 |
+
"loss": 0.4297,
|
2112 |
+
"step": 2670
|
2113 |
+
},
|
2114 |
+
{
|
2115 |
+
"epoch": 2.28,
|
2116 |
+
"grad_norm": 2.649815559387207,
|
2117 |
+
"learning_rate": 8.70843989769821e-05,
|
2118 |
+
"loss": 0.4457,
|
2119 |
+
"step": 2680
|
2120 |
+
},
|
2121 |
+
{
|
2122 |
+
"epoch": 2.29,
|
2123 |
+
"grad_norm": 2.874537467956543,
|
2124 |
+
"learning_rate": 8.665814151747657e-05,
|
2125 |
+
"loss": 0.3606,
|
2126 |
+
"step": 2690
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 2.3,
|
2130 |
+
"grad_norm": 3.6767563819885254,
|
2131 |
+
"learning_rate": 8.623188405797103e-05,
|
2132 |
+
"loss": 0.3643,
|
2133 |
+
"step": 2700
|
2134 |
+
},
|
2135 |
+
{
|
2136 |
+
"epoch": 2.3,
|
2137 |
+
"eval_accuracy": 0.9375,
|
2138 |
+
"eval_loss": 0.25128769874572754,
|
2139 |
+
"eval_runtime": 22.204,
|
2140 |
+
"eval_samples_per_second": 33.147,
|
2141 |
+
"eval_steps_per_second": 4.143,
|
2142 |
+
"step": 2700
|
2143 |
+
},
|
2144 |
+
{
|
2145 |
+
"epoch": 2.31,
|
2146 |
+
"grad_norm": 2.8548221588134766,
|
2147 |
+
"learning_rate": 8.580562659846548e-05,
|
2148 |
+
"loss": 0.4496,
|
2149 |
+
"step": 2710
|
2150 |
+
},
|
2151 |
+
{
|
2152 |
+
"epoch": 2.32,
|
2153 |
+
"grad_norm": 3.32646107673645,
|
2154 |
+
"learning_rate": 8.537936913895993e-05,
|
2155 |
+
"loss": 0.4037,
|
2156 |
+
"step": 2720
|
2157 |
+
},
|
2158 |
+
{
|
2159 |
+
"epoch": 2.33,
|
2160 |
+
"grad_norm": 4.917088508605957,
|
2161 |
+
"learning_rate": 8.495311167945439e-05,
|
2162 |
+
"loss": 0.3816,
|
2163 |
+
"step": 2730
|
2164 |
+
},
|
2165 |
+
{
|
2166 |
+
"epoch": 2.34,
|
2167 |
+
"grad_norm": 4.138692378997803,
|
2168 |
+
"learning_rate": 8.452685421994884e-05,
|
2169 |
+
"loss": 0.4166,
|
2170 |
+
"step": 2740
|
2171 |
+
},
|
2172 |
+
{
|
2173 |
+
"epoch": 2.34,
|
2174 |
+
"grad_norm": 2.5747594833374023,
|
2175 |
+
"learning_rate": 8.41005967604433e-05,
|
2176 |
+
"loss": 0.3947,
|
2177 |
+
"step": 2750
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 2.35,
|
2181 |
+
"grad_norm": 3.0434372425079346,
|
2182 |
+
"learning_rate": 8.367433930093777e-05,
|
2183 |
+
"loss": 0.2987,
|
2184 |
+
"step": 2760
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 2.36,
|
2188 |
+
"grad_norm": 4.922779083251953,
|
2189 |
+
"learning_rate": 8.324808184143222e-05,
|
2190 |
+
"loss": 0.3555,
|
2191 |
+
"step": 2770
|
2192 |
+
},
|
2193 |
+
{
|
2194 |
+
"epoch": 2.37,
|
2195 |
+
"grad_norm": 3.4687201976776123,
|
2196 |
+
"learning_rate": 8.282182438192668e-05,
|
2197 |
+
"loss": 0.3274,
|
2198 |
+
"step": 2780
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 2.38,
|
2202 |
+
"grad_norm": 3.288496494293213,
|
2203 |
+
"learning_rate": 8.239556692242114e-05,
|
2204 |
+
"loss": 0.3321,
|
2205 |
+
"step": 2790
|
2206 |
+
},
|
2207 |
+
{
|
2208 |
+
"epoch": 2.39,
|
2209 |
+
"grad_norm": 3.5918378829956055,
|
2210 |
+
"learning_rate": 8.19693094629156e-05,
|
2211 |
+
"loss": 0.3393,
|
2212 |
+
"step": 2800
|
2213 |
+
},
|
2214 |
+
{
|
2215 |
+
"epoch": 2.39,
|
2216 |
+
"eval_accuracy": 0.9429347826086957,
|
2217 |
+
"eval_loss": 0.24919526278972626,
|
2218 |
+
"eval_runtime": 21.7658,
|
2219 |
+
"eval_samples_per_second": 33.814,
|
2220 |
+
"eval_steps_per_second": 4.227,
|
2221 |
+
"step": 2800
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 2.4,
|
2225 |
+
"grad_norm": 2.7565417289733887,
|
2226 |
+
"learning_rate": 8.154305200341006e-05,
|
2227 |
+
"loss": 0.3414,
|
2228 |
+
"step": 2810
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 2.4,
|
2232 |
+
"grad_norm": 3.525897979736328,
|
2233 |
+
"learning_rate": 8.111679454390452e-05,
|
2234 |
+
"loss": 0.4407,
|
2235 |
+
"step": 2820
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 2.41,
|
2239 |
+
"grad_norm": 2.774212121963501,
|
2240 |
+
"learning_rate": 8.069053708439897e-05,
|
2241 |
+
"loss": 0.3403,
|
2242 |
+
"step": 2830
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 2.42,
|
2246 |
+
"grad_norm": 3.12599515914917,
|
2247 |
+
"learning_rate": 8.026427962489343e-05,
|
2248 |
+
"loss": 0.3312,
|
2249 |
+
"step": 2840
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 2.43,
|
2253 |
+
"grad_norm": 5.231620788574219,
|
2254 |
+
"learning_rate": 7.98380221653879e-05,
|
2255 |
+
"loss": 0.3958,
|
2256 |
+
"step": 2850
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 2.44,
|
2260 |
+
"grad_norm": 3.423818588256836,
|
2261 |
+
"learning_rate": 7.941176470588235e-05,
|
2262 |
+
"loss": 0.3513,
|
2263 |
+
"step": 2860
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 2.45,
|
2267 |
+
"grad_norm": 2.218860387802124,
|
2268 |
+
"learning_rate": 7.898550724637681e-05,
|
2269 |
+
"loss": 0.3747,
|
2270 |
+
"step": 2870
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 2.46,
|
2274 |
+
"grad_norm": 4.602892875671387,
|
2275 |
+
"learning_rate": 7.855924978687127e-05,
|
2276 |
+
"loss": 0.4701,
|
2277 |
+
"step": 2880
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 2.46,
|
2281 |
+
"grad_norm": 3.3462207317352295,
|
2282 |
+
"learning_rate": 7.813299232736572e-05,
|
2283 |
+
"loss": 0.3967,
|
2284 |
+
"step": 2890
|
2285 |
+
},
|
2286 |
+
{
|
2287 |
+
"epoch": 2.47,
|
2288 |
+
"grad_norm": 2.690305233001709,
|
2289 |
+
"learning_rate": 7.770673486786019e-05,
|
2290 |
+
"loss": 0.3635,
|
2291 |
+
"step": 2900
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 2.47,
|
2295 |
+
"eval_accuracy": 0.9402173913043478,
|
2296 |
+
"eval_loss": 0.2394125610589981,
|
2297 |
+
"eval_runtime": 21.6739,
|
2298 |
+
"eval_samples_per_second": 33.958,
|
2299 |
+
"eval_steps_per_second": 4.245,
|
2300 |
+
"step": 2900
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 2.48,
|
2304 |
+
"grad_norm": 3.6983702182769775,
|
2305 |
+
"learning_rate": 7.728047740835465e-05,
|
2306 |
+
"loss": 0.3699,
|
2307 |
+
"step": 2910
|
2308 |
+
},
|
2309 |
+
{
|
2310 |
+
"epoch": 2.49,
|
2311 |
+
"grad_norm": 3.6911110877990723,
|
2312 |
+
"learning_rate": 7.68542199488491e-05,
|
2313 |
+
"loss": 0.4104,
|
2314 |
+
"step": 2920
|
2315 |
+
},
|
2316 |
+
{
|
2317 |
+
"epoch": 2.5,
|
2318 |
+
"grad_norm": 2.3493428230285645,
|
2319 |
+
"learning_rate": 7.642796248934356e-05,
|
2320 |
+
"loss": 0.323,
|
2321 |
+
"step": 2930
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 2.51,
|
2325 |
+
"grad_norm": 3.3958208560943604,
|
2326 |
+
"learning_rate": 7.600170502983802e-05,
|
2327 |
+
"loss": 0.3458,
|
2328 |
+
"step": 2940
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 2.51,
|
2332 |
+
"grad_norm": 3.3971476554870605,
|
2333 |
+
"learning_rate": 7.557544757033247e-05,
|
2334 |
+
"loss": 0.3147,
|
2335 |
+
"step": 2950
|
2336 |
+
},
|
2337 |
+
{
|
2338 |
+
"epoch": 2.52,
|
2339 |
+
"grad_norm": 3.6205222606658936,
|
2340 |
+
"learning_rate": 7.514919011082694e-05,
|
2341 |
+
"loss": 0.4776,
|
2342 |
+
"step": 2960
|
2343 |
+
},
|
2344 |
+
{
|
2345 |
+
"epoch": 2.53,
|
2346 |
+
"grad_norm": 2.697317123413086,
|
2347 |
+
"learning_rate": 7.47229326513214e-05,
|
2348 |
+
"loss": 0.3806,
|
2349 |
+
"step": 2970
|
2350 |
+
},
|
2351 |
+
{
|
2352 |
+
"epoch": 2.54,
|
2353 |
+
"grad_norm": 4.160821437835693,
|
2354 |
+
"learning_rate": 7.429667519181585e-05,
|
2355 |
+
"loss": 0.4149,
|
2356 |
+
"step": 2980
|
2357 |
+
},
|
2358 |
+
{
|
2359 |
+
"epoch": 2.55,
|
2360 |
+
"grad_norm": 4.857287406921387,
|
2361 |
+
"learning_rate": 7.387041773231031e-05,
|
2362 |
+
"loss": 0.2937,
|
2363 |
+
"step": 2990
|
2364 |
+
},
|
2365 |
+
{
|
2366 |
+
"epoch": 2.56,
|
2367 |
+
"grad_norm": 3.1837852001190186,
|
2368 |
+
"learning_rate": 7.344416027280478e-05,
|
2369 |
+
"loss": 0.3624,
|
2370 |
+
"step": 3000
|
2371 |
+
},
|
2372 |
+
{
|
2373 |
+
"epoch": 2.56,
|
2374 |
+
"eval_accuracy": 0.938858695652174,
|
2375 |
+
"eval_loss": 0.24252061545848846,
|
2376 |
+
"eval_runtime": 22.0444,
|
2377 |
+
"eval_samples_per_second": 33.387,
|
2378 |
+
"eval_steps_per_second": 4.173,
|
2379 |
+
"step": 3000
|
2380 |
+
},
|
2381 |
+
{
|
2382 |
+
"epoch": 2.57,
|
2383 |
+
"grad_norm": 4.424566745758057,
|
2384 |
+
"learning_rate": 7.301790281329923e-05,
|
2385 |
+
"loss": 0.2947,
|
2386 |
+
"step": 3010
|
2387 |
+
},
|
2388 |
+
{
|
2389 |
+
"epoch": 2.57,
|
2390 |
+
"grad_norm": 4.352055549621582,
|
2391 |
+
"learning_rate": 7.259164535379369e-05,
|
2392 |
+
"loss": 0.408,
|
2393 |
+
"step": 3020
|
2394 |
+
},
|
2395 |
+
{
|
2396 |
+
"epoch": 2.58,
|
2397 |
+
"grad_norm": 4.074793338775635,
|
2398 |
+
"learning_rate": 7.216538789428815e-05,
|
2399 |
+
"loss": 0.3518,
|
2400 |
+
"step": 3030
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 2.59,
|
2404 |
+
"grad_norm": 2.518249750137329,
|
2405 |
+
"learning_rate": 7.17391304347826e-05,
|
2406 |
+
"loss": 0.3168,
|
2407 |
+
"step": 3040
|
2408 |
+
},
|
2409 |
+
{
|
2410 |
+
"epoch": 2.6,
|
2411 |
+
"grad_norm": 3.670358180999756,
|
2412 |
+
"learning_rate": 7.131287297527707e-05,
|
2413 |
+
"loss": 0.3239,
|
2414 |
+
"step": 3050
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 2.61,
|
2418 |
+
"grad_norm": 3.765688180923462,
|
2419 |
+
"learning_rate": 7.088661551577153e-05,
|
2420 |
+
"loss": 0.3946,
|
2421 |
+
"step": 3060
|
2422 |
+
},
|
2423 |
+
{
|
2424 |
+
"epoch": 2.62,
|
2425 |
+
"grad_norm": 3.1018218994140625,
|
2426 |
+
"learning_rate": 7.046035805626598e-05,
|
2427 |
+
"loss": 0.3853,
|
2428 |
+
"step": 3070
|
2429 |
+
},
|
2430 |
+
{
|
2431 |
+
"epoch": 2.63,
|
2432 |
+
"grad_norm": 1.9965764284133911,
|
2433 |
+
"learning_rate": 7.003410059676044e-05,
|
2434 |
+
"loss": 0.3506,
|
2435 |
+
"step": 3080
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 2.63,
|
2439 |
+
"grad_norm": 4.730674743652344,
|
2440 |
+
"learning_rate": 6.96078431372549e-05,
|
2441 |
+
"loss": 0.3642,
|
2442 |
+
"step": 3090
|
2443 |
+
},
|
2444 |
+
{
|
2445 |
+
"epoch": 2.64,
|
2446 |
+
"grad_norm": 2.4963278770446777,
|
2447 |
+
"learning_rate": 6.918158567774935e-05,
|
2448 |
+
"loss": 0.3608,
|
2449 |
+
"step": 3100
|
2450 |
+
},
|
2451 |
+
{
|
2452 |
+
"epoch": 2.64,
|
2453 |
+
"eval_accuracy": 0.9456521739130435,
|
2454 |
+
"eval_loss": 0.2389669418334961,
|
2455 |
+
"eval_runtime": 22.1191,
|
2456 |
+
"eval_samples_per_second": 33.274,
|
2457 |
+
"eval_steps_per_second": 4.159,
|
2458 |
+
"step": 3100
|
2459 |
+
},
|
2460 |
+
{
|
2461 |
+
"epoch": 2.65,
|
2462 |
+
"grad_norm": 2.049652576446533,
|
2463 |
+
"learning_rate": 6.875532821824382e-05,
|
2464 |
+
"loss": 0.374,
|
2465 |
+
"step": 3110
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 2.66,
|
2469 |
+
"grad_norm": 3.511080026626587,
|
2470 |
+
"learning_rate": 6.832907075873828e-05,
|
2471 |
+
"loss": 0.3842,
|
2472 |
+
"step": 3120
|
2473 |
+
},
|
2474 |
+
{
|
2475 |
+
"epoch": 2.67,
|
2476 |
+
"grad_norm": 4.446420192718506,
|
2477 |
+
"learning_rate": 6.790281329923273e-05,
|
2478 |
+
"loss": 0.288,
|
2479 |
+
"step": 3130
|
2480 |
+
},
|
2481 |
+
{
|
2482 |
+
"epoch": 2.68,
|
2483 |
+
"grad_norm": 3.2927162647247314,
|
2484 |
+
"learning_rate": 6.74765558397272e-05,
|
2485 |
+
"loss": 0.2999,
|
2486 |
+
"step": 3140
|
2487 |
+
},
|
2488 |
+
{
|
2489 |
+
"epoch": 2.69,
|
2490 |
+
"grad_norm": 3.2927627563476562,
|
2491 |
+
"learning_rate": 6.705029838022166e-05,
|
2492 |
+
"loss": 0.2792,
|
2493 |
+
"step": 3150
|
2494 |
+
},
|
2495 |
+
{
|
2496 |
+
"epoch": 2.69,
|
2497 |
+
"grad_norm": 4.26877498626709,
|
2498 |
+
"learning_rate": 6.66240409207161e-05,
|
2499 |
+
"loss": 0.3553,
|
2500 |
+
"step": 3160
|
2501 |
+
},
|
2502 |
+
{
|
2503 |
+
"epoch": 2.7,
|
2504 |
+
"grad_norm": 2.3921847343444824,
|
2505 |
+
"learning_rate": 6.619778346121057e-05,
|
2506 |
+
"loss": 0.3472,
|
2507 |
+
"step": 3170
|
2508 |
+
},
|
2509 |
+
{
|
2510 |
+
"epoch": 2.71,
|
2511 |
+
"grad_norm": 2.2627205848693848,
|
2512 |
+
"learning_rate": 6.577152600170503e-05,
|
2513 |
+
"loss": 0.3367,
|
2514 |
+
"step": 3180
|
2515 |
+
},
|
2516 |
+
{
|
2517 |
+
"epoch": 2.72,
|
2518 |
+
"grad_norm": 4.409063816070557,
|
2519 |
+
"learning_rate": 6.534526854219948e-05,
|
2520 |
+
"loss": 0.4007,
|
2521 |
+
"step": 3190
|
2522 |
+
},
|
2523 |
+
{
|
2524 |
+
"epoch": 2.73,
|
2525 |
+
"grad_norm": 2.80055570602417,
|
2526 |
+
"learning_rate": 6.491901108269395e-05,
|
2527 |
+
"loss": 0.3215,
|
2528 |
+
"step": 3200
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 2.73,
|
2532 |
+
"eval_accuracy": 0.9320652173913043,
|
2533 |
+
"eval_loss": 0.2482815384864807,
|
2534 |
+
"eval_runtime": 22.058,
|
2535 |
+
"eval_samples_per_second": 33.367,
|
2536 |
+
"eval_steps_per_second": 4.171,
|
2537 |
+
"step": 3200
|
2538 |
+
},
|
2539 |
+
{
|
2540 |
+
"epoch": 2.74,
|
2541 |
+
"grad_norm": 2.1410982608795166,
|
2542 |
+
"learning_rate": 6.449275362318841e-05,
|
2543 |
+
"loss": 0.3288,
|
2544 |
+
"step": 3210
|
2545 |
+
},
|
2546 |
+
{
|
2547 |
+
"epoch": 2.75,
|
2548 |
+
"grad_norm": 1.486505389213562,
|
2549 |
+
"learning_rate": 6.406649616368286e-05,
|
2550 |
+
"loss": 0.3833,
|
2551 |
+
"step": 3220
|
2552 |
+
},
|
2553 |
+
{
|
2554 |
+
"epoch": 2.75,
|
2555 |
+
"grad_norm": 2.9427695274353027,
|
2556 |
+
"learning_rate": 6.364023870417732e-05,
|
2557 |
+
"loss": 0.3629,
|
2558 |
+
"step": 3230
|
2559 |
+
},
|
2560 |
+
{
|
2561 |
+
"epoch": 2.76,
|
2562 |
+
"grad_norm": 6.134848594665527,
|
2563 |
+
"learning_rate": 6.321398124467179e-05,
|
2564 |
+
"loss": 0.3732,
|
2565 |
+
"step": 3240
|
2566 |
+
},
|
2567 |
+
{
|
2568 |
+
"epoch": 2.77,
|
2569 |
+
"grad_norm": 2.851616144180298,
|
2570 |
+
"learning_rate": 6.278772378516623e-05,
|
2571 |
+
"loss": 0.3866,
|
2572 |
+
"step": 3250
|
2573 |
+
},
|
2574 |
+
{
|
2575 |
+
"epoch": 2.78,
|
2576 |
+
"grad_norm": 4.025167465209961,
|
2577 |
+
"learning_rate": 6.23614663256607e-05,
|
2578 |
+
"loss": 0.3245,
|
2579 |
+
"step": 3260
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 2.79,
|
2583 |
+
"grad_norm": 6.220050811767578,
|
2584 |
+
"learning_rate": 6.193520886615516e-05,
|
2585 |
+
"loss": 0.3233,
|
2586 |
+
"step": 3270
|
2587 |
+
},
|
2588 |
+
{
|
2589 |
+
"epoch": 2.8,
|
2590 |
+
"grad_norm": 2.960378408432007,
|
2591 |
+
"learning_rate": 6.150895140664961e-05,
|
2592 |
+
"loss": 0.3752,
|
2593 |
+
"step": 3280
|
2594 |
+
},
|
2595 |
+
{
|
2596 |
+
"epoch": 2.8,
|
2597 |
+
"grad_norm": 3.8177359104156494,
|
2598 |
+
"learning_rate": 6.108269394714407e-05,
|
2599 |
+
"loss": 0.3581,
|
2600 |
+
"step": 3290
|
2601 |
+
},
|
2602 |
+
{
|
2603 |
+
"epoch": 2.81,
|
2604 |
+
"grad_norm": 3.3736770153045654,
|
2605 |
+
"learning_rate": 6.065643648763854e-05,
|
2606 |
+
"loss": 0.2971,
|
2607 |
+
"step": 3300
|
2608 |
+
},
|
2609 |
+
{
|
2610 |
+
"epoch": 2.81,
|
2611 |
+
"eval_accuracy": 0.9402173913043478,
|
2612 |
+
"eval_loss": 0.2455403059720993,
|
2613 |
+
"eval_runtime": 22.1061,
|
2614 |
+
"eval_samples_per_second": 33.294,
|
2615 |
+
"eval_steps_per_second": 4.162,
|
2616 |
+
"step": 3300
|
2617 |
+
},
|
2618 |
+
{
|
2619 |
+
"epoch": 2.82,
|
2620 |
+
"grad_norm": 5.297087669372559,
|
2621 |
+
"learning_rate": 6.0230179028132994e-05,
|
2622 |
+
"loss": 0.3563,
|
2623 |
+
"step": 3310
|
2624 |
+
},
|
2625 |
+
{
|
2626 |
+
"epoch": 2.83,
|
2627 |
+
"grad_norm": 2.4093313217163086,
|
2628 |
+
"learning_rate": 5.980392156862745e-05,
|
2629 |
+
"loss": 0.2741,
|
2630 |
+
"step": 3320
|
2631 |
+
},
|
2632 |
+
{
|
2633 |
+
"epoch": 2.84,
|
2634 |
+
"grad_norm": 2.5030243396759033,
|
2635 |
+
"learning_rate": 5.9377664109121913e-05,
|
2636 |
+
"loss": 0.37,
|
2637 |
+
"step": 3330
|
2638 |
+
},
|
2639 |
+
{
|
2640 |
+
"epoch": 2.85,
|
2641 |
+
"grad_norm": 3.5400922298431396,
|
2642 |
+
"learning_rate": 5.895140664961637e-05,
|
2643 |
+
"loss": 0.3389,
|
2644 |
+
"step": 3340
|
2645 |
+
},
|
2646 |
+
{
|
2647 |
+
"epoch": 2.86,
|
2648 |
+
"grad_norm": 2.54128098487854,
|
2649 |
+
"learning_rate": 5.8525149190110826e-05,
|
2650 |
+
"loss": 0.3459,
|
2651 |
+
"step": 3350
|
2652 |
+
},
|
2653 |
+
{
|
2654 |
+
"epoch": 2.86,
|
2655 |
+
"grad_norm": 3.479625701904297,
|
2656 |
+
"learning_rate": 5.809889173060529e-05,
|
2657 |
+
"loss": 0.4011,
|
2658 |
+
"step": 3360
|
2659 |
+
},
|
2660 |
+
{
|
2661 |
+
"epoch": 2.87,
|
2662 |
+
"grad_norm": 2.196369171142578,
|
2663 |
+
"learning_rate": 5.7672634271099746e-05,
|
2664 |
+
"loss": 0.4212,
|
2665 |
+
"step": 3370
|
2666 |
+
},
|
2667 |
+
{
|
2668 |
+
"epoch": 2.88,
|
2669 |
+
"grad_norm": 4.405612468719482,
|
2670 |
+
"learning_rate": 5.72463768115942e-05,
|
2671 |
+
"loss": 0.4886,
|
2672 |
+
"step": 3380
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 2.89,
|
2676 |
+
"grad_norm": 4.257590293884277,
|
2677 |
+
"learning_rate": 5.6820119352088666e-05,
|
2678 |
+
"loss": 0.3434,
|
2679 |
+
"step": 3390
|
2680 |
+
},
|
2681 |
+
{
|
2682 |
+
"epoch": 2.9,
|
2683 |
+
"grad_norm": 5.413020610809326,
|
2684 |
+
"learning_rate": 5.639386189258312e-05,
|
2685 |
+
"loss": 0.3838,
|
2686 |
+
"step": 3400
|
2687 |
+
},
|
2688 |
+
{
|
2689 |
+
"epoch": 2.9,
|
2690 |
+
"eval_accuracy": 0.9470108695652174,
|
2691 |
+
"eval_loss": 0.23631170392036438,
|
2692 |
+
"eval_runtime": 22.1897,
|
2693 |
+
"eval_samples_per_second": 33.169,
|
2694 |
+
"eval_steps_per_second": 4.146,
|
2695 |
+
"step": 3400
|
2696 |
+
},
|
2697 |
+
{
|
2698 |
+
"epoch": 2.91,
|
2699 |
+
"grad_norm": 2.7590572834014893,
|
2700 |
+
"learning_rate": 5.596760443307758e-05,
|
2701 |
+
"loss": 0.3481,
|
2702 |
+
"step": 3410
|
2703 |
+
},
|
2704 |
+
{
|
2705 |
+
"epoch": 2.92,
|
2706 |
+
"grad_norm": 3.4297289848327637,
|
2707 |
+
"learning_rate": 5.554134697357204e-05,
|
2708 |
+
"loss": 0.3478,
|
2709 |
+
"step": 3420
|
2710 |
+
},
|
2711 |
+
{
|
2712 |
+
"epoch": 2.92,
|
2713 |
+
"grad_norm": 3.9188895225524902,
|
2714 |
+
"learning_rate": 5.51150895140665e-05,
|
2715 |
+
"loss": 0.4562,
|
2716 |
+
"step": 3430
|
2717 |
+
},
|
2718 |
+
{
|
2719 |
+
"epoch": 2.93,
|
2720 |
+
"grad_norm": 2.648061513900757,
|
2721 |
+
"learning_rate": 5.4688832054560955e-05,
|
2722 |
+
"loss": 0.2639,
|
2723 |
+
"step": 3440
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 2.94,
|
2727 |
+
"grad_norm": 3.416527509689331,
|
2728 |
+
"learning_rate": 5.426257459505542e-05,
|
2729 |
+
"loss": 0.3647,
|
2730 |
+
"step": 3450
|
2731 |
+
},
|
2732 |
+
{
|
2733 |
+
"epoch": 2.95,
|
2734 |
+
"grad_norm": 2.152862310409546,
|
2735 |
+
"learning_rate": 5.3836317135549874e-05,
|
2736 |
+
"loss": 0.3866,
|
2737 |
+
"step": 3460
|
2738 |
+
},
|
2739 |
+
{
|
2740 |
+
"epoch": 2.96,
|
2741 |
+
"grad_norm": 3.239790439605713,
|
2742 |
+
"learning_rate": 5.341005967604433e-05,
|
2743 |
+
"loss": 0.3115,
|
2744 |
+
"step": 3470
|
2745 |
+
},
|
2746 |
+
{
|
2747 |
+
"epoch": 2.97,
|
2748 |
+
"grad_norm": 4.865437030792236,
|
2749 |
+
"learning_rate": 5.2983802216538794e-05,
|
2750 |
+
"loss": 0.4108,
|
2751 |
+
"step": 3480
|
2752 |
+
},
|
2753 |
+
{
|
2754 |
+
"epoch": 2.98,
|
2755 |
+
"grad_norm": 2.4955708980560303,
|
2756 |
+
"learning_rate": 5.255754475703325e-05,
|
2757 |
+
"loss": 0.3349,
|
2758 |
+
"step": 3490
|
2759 |
+
},
|
2760 |
+
{
|
2761 |
+
"epoch": 2.98,
|
2762 |
+
"grad_norm": 3.560739755630493,
|
2763 |
+
"learning_rate": 5.213128729752771e-05,
|
2764 |
+
"loss": 0.3036,
|
2765 |
+
"step": 3500
|
2766 |
+
},
|
2767 |
+
{
|
2768 |
+
"epoch": 2.98,
|
2769 |
+
"eval_accuracy": 0.9402173913043478,
|
2770 |
+
"eval_loss": 0.2422226220369339,
|
2771 |
+
"eval_runtime": 22.1146,
|
2772 |
+
"eval_samples_per_second": 33.281,
|
2773 |
+
"eval_steps_per_second": 4.16,
|
2774 |
+
"step": 3500
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 2.99,
|
2778 |
+
"grad_norm": 2.6449167728424072,
|
2779 |
+
"learning_rate": 5.170502983802217e-05,
|
2780 |
+
"loss": 0.3553,
|
2781 |
+
"step": 3510
|
2782 |
+
},
|
2783 |
+
{
|
2784 |
+
"epoch": 3.0,
|
2785 |
+
"grad_norm": 3.7996511459350586,
|
2786 |
+
"learning_rate": 5.1278772378516626e-05,
|
2787 |
+
"loss": 0.2752,
|
2788 |
+
"step": 3520
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 3.01,
|
2792 |
+
"grad_norm": 2.710324764251709,
|
2793 |
+
"learning_rate": 5.085251491901108e-05,
|
2794 |
+
"loss": 0.3815,
|
2795 |
+
"step": 3530
|
2796 |
+
},
|
2797 |
+
{
|
2798 |
+
"epoch": 3.02,
|
2799 |
+
"grad_norm": 3.8157262802124023,
|
2800 |
+
"learning_rate": 5.0426257459505546e-05,
|
2801 |
+
"loss": 0.3643,
|
2802 |
+
"step": 3540
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 3.03,
|
2806 |
+
"grad_norm": 3.8065927028656006,
|
2807 |
+
"learning_rate": 5e-05,
|
2808 |
+
"loss": 0.3013,
|
2809 |
+
"step": 3550
|
2810 |
+
},
|
2811 |
+
{
|
2812 |
+
"epoch": 3.03,
|
2813 |
+
"grad_norm": 3.4787912368774414,
|
2814 |
+
"learning_rate": 4.957374254049446e-05,
|
2815 |
+
"loss": 0.3826,
|
2816 |
+
"step": 3560
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 3.04,
|
2820 |
+
"grad_norm": 3.8217809200286865,
|
2821 |
+
"learning_rate": 4.914748508098892e-05,
|
2822 |
+
"loss": 0.3892,
|
2823 |
+
"step": 3570
|
2824 |
+
},
|
2825 |
+
{
|
2826 |
+
"epoch": 3.05,
|
2827 |
+
"grad_norm": 3.754225015640259,
|
2828 |
+
"learning_rate": 4.872122762148338e-05,
|
2829 |
+
"loss": 0.258,
|
2830 |
+
"step": 3580
|
2831 |
+
},
|
2832 |
+
{
|
2833 |
+
"epoch": 3.06,
|
2834 |
+
"grad_norm": 3.32442045211792,
|
2835 |
+
"learning_rate": 4.8294970161977835e-05,
|
2836 |
+
"loss": 0.2541,
|
2837 |
+
"step": 3590
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 3.07,
|
2841 |
+
"grad_norm": 3.7520785331726074,
|
2842 |
+
"learning_rate": 4.78687127024723e-05,
|
2843 |
+
"loss": 0.401,
|
2844 |
+
"step": 3600
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 3.07,
|
2848 |
+
"eval_accuracy": 0.9429347826086957,
|
2849 |
+
"eval_loss": 0.23978637158870697,
|
2850 |
+
"eval_runtime": 21.9597,
|
2851 |
+
"eval_samples_per_second": 33.516,
|
2852 |
+
"eval_steps_per_second": 4.189,
|
2853 |
+
"step": 3600
|
2854 |
+
},
|
2855 |
+
{
|
2856 |
+
"epoch": 3.08,
|
2857 |
+
"grad_norm": 2.6882004737854004,
|
2858 |
+
"learning_rate": 4.7442455242966755e-05,
|
2859 |
+
"loss": 0.4118,
|
2860 |
+
"step": 3610
|
2861 |
+
},
|
2862 |
+
{
|
2863 |
+
"epoch": 3.09,
|
2864 |
+
"grad_norm": 2.4365506172180176,
|
2865 |
+
"learning_rate": 4.701619778346121e-05,
|
2866 |
+
"loss": 0.3586,
|
2867 |
+
"step": 3620
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 3.09,
|
2871 |
+
"grad_norm": 1.2782219648361206,
|
2872 |
+
"learning_rate": 4.6589940323955674e-05,
|
2873 |
+
"loss": 0.3192,
|
2874 |
+
"step": 3630
|
2875 |
+
},
|
2876 |
+
{
|
2877 |
+
"epoch": 3.1,
|
2878 |
+
"grad_norm": 4.13897180557251,
|
2879 |
+
"learning_rate": 4.616368286445013e-05,
|
2880 |
+
"loss": 0.3101,
|
2881 |
+
"step": 3640
|
2882 |
+
},
|
2883 |
+
{
|
2884 |
+
"epoch": 3.11,
|
2885 |
+
"grad_norm": 1.7676732540130615,
|
2886 |
+
"learning_rate": 4.573742540494459e-05,
|
2887 |
+
"loss": 0.3854,
|
2888 |
+
"step": 3650
|
2889 |
+
},
|
2890 |
+
{
|
2891 |
+
"epoch": 3.12,
|
2892 |
+
"grad_norm": 3.285656213760376,
|
2893 |
+
"learning_rate": 4.531116794543905e-05,
|
2894 |
+
"loss": 0.4299,
|
2895 |
+
"step": 3660
|
2896 |
+
},
|
2897 |
+
{
|
2898 |
+
"epoch": 3.13,
|
2899 |
+
"grad_norm": 2.670168876647949,
|
2900 |
+
"learning_rate": 4.488491048593351e-05,
|
2901 |
+
"loss": 0.2919,
|
2902 |
+
"step": 3670
|
2903 |
+
},
|
2904 |
+
{
|
2905 |
+
"epoch": 3.14,
|
2906 |
+
"grad_norm": 3.4926791191101074,
|
2907 |
+
"learning_rate": 4.445865302642796e-05,
|
2908 |
+
"loss": 0.356,
|
2909 |
+
"step": 3680
|
2910 |
+
},
|
2911 |
+
{
|
2912 |
+
"epoch": 3.15,
|
2913 |
+
"grad_norm": 1.534806728363037,
|
2914 |
+
"learning_rate": 4.4032395566922426e-05,
|
2915 |
+
"loss": 0.3157,
|
2916 |
+
"step": 3690
|
2917 |
+
},
|
2918 |
+
{
|
2919 |
+
"epoch": 3.15,
|
2920 |
+
"grad_norm": 2.911130905151367,
|
2921 |
+
"learning_rate": 4.360613810741688e-05,
|
2922 |
+
"loss": 0.3458,
|
2923 |
+
"step": 3700
|
2924 |
+
},
|
2925 |
+
{
|
2926 |
+
"epoch": 3.15,
|
2927 |
+
"eval_accuracy": 0.9429347826086957,
|
2928 |
+
"eval_loss": 0.2517460286617279,
|
2929 |
+
"eval_runtime": 22.0673,
|
2930 |
+
"eval_samples_per_second": 33.353,
|
2931 |
+
"eval_steps_per_second": 4.169,
|
2932 |
+
"step": 3700
|
2933 |
+
},
|
2934 |
+
{
|
2935 |
+
"epoch": 3.16,
|
2936 |
+
"grad_norm": 3.810058116912842,
|
2937 |
+
"learning_rate": 4.317988064791134e-05,
|
2938 |
+
"loss": 0.351,
|
2939 |
+
"step": 3710
|
2940 |
+
},
|
2941 |
+
{
|
2942 |
+
"epoch": 3.17,
|
2943 |
+
"grad_norm": 3.799814462661743,
|
2944 |
+
"learning_rate": 4.27536231884058e-05,
|
2945 |
+
"loss": 0.4715,
|
2946 |
+
"step": 3720
|
2947 |
+
},
|
2948 |
+
{
|
2949 |
+
"epoch": 3.18,
|
2950 |
+
"grad_norm": 2.787813663482666,
|
2951 |
+
"learning_rate": 4.232736572890026e-05,
|
2952 |
+
"loss": 0.3037,
|
2953 |
+
"step": 3730
|
2954 |
+
},
|
2955 |
+
{
|
2956 |
+
"epoch": 3.19,
|
2957 |
+
"grad_norm": 3.3836419582366943,
|
2958 |
+
"learning_rate": 4.1901108269394715e-05,
|
2959 |
+
"loss": 0.3279,
|
2960 |
+
"step": 3740
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 3.2,
|
2964 |
+
"grad_norm": 3.981520652770996,
|
2965 |
+
"learning_rate": 4.147485080988918e-05,
|
2966 |
+
"loss": 0.4078,
|
2967 |
+
"step": 3750
|
2968 |
+
},
|
2969 |
+
{
|
2970 |
+
"epoch": 3.21,
|
2971 |
+
"grad_norm": 2.715733051300049,
|
2972 |
+
"learning_rate": 4.1048593350383635e-05,
|
2973 |
+
"loss": 0.3375,
|
2974 |
+
"step": 3760
|
2975 |
+
},
|
2976 |
+
{
|
2977 |
+
"epoch": 3.21,
|
2978 |
+
"grad_norm": 3.2656917572021484,
|
2979 |
+
"learning_rate": 4.062233589087809e-05,
|
2980 |
+
"loss": 0.3842,
|
2981 |
+
"step": 3770
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 3.22,
|
2985 |
+
"grad_norm": 4.852041721343994,
|
2986 |
+
"learning_rate": 4.0196078431372555e-05,
|
2987 |
+
"loss": 0.3119,
|
2988 |
+
"step": 3780
|
2989 |
+
},
|
2990 |
+
{
|
2991 |
+
"epoch": 3.23,
|
2992 |
+
"grad_norm": 3.6040267944335938,
|
2993 |
+
"learning_rate": 3.976982097186701e-05,
|
2994 |
+
"loss": 0.3623,
|
2995 |
+
"step": 3790
|
2996 |
+
},
|
2997 |
+
{
|
2998 |
+
"epoch": 3.24,
|
2999 |
+
"grad_norm": 2.597151279449463,
|
3000 |
+
"learning_rate": 3.934356351236147e-05,
|
3001 |
+
"loss": 0.2908,
|
3002 |
+
"step": 3800
|
3003 |
+
},
|
3004 |
+
{
|
3005 |
+
"epoch": 3.24,
|
3006 |
+
"eval_accuracy": 0.9456521739130435,
|
3007 |
+
"eval_loss": 0.24226675927639008,
|
3008 |
+
"eval_runtime": 21.9179,
|
3009 |
+
"eval_samples_per_second": 33.58,
|
3010 |
+
"eval_steps_per_second": 4.197,
|
3011 |
+
"step": 3800
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 3.25,
|
3015 |
+
"grad_norm": 4.236294269561768,
|
3016 |
+
"learning_rate": 3.891730605285593e-05,
|
3017 |
+
"loss": 0.3942,
|
3018 |
+
"step": 3810
|
3019 |
+
},
|
3020 |
+
{
|
3021 |
+
"epoch": 3.26,
|
3022 |
+
"grad_norm": 1.8580868244171143,
|
3023 |
+
"learning_rate": 3.849104859335039e-05,
|
3024 |
+
"loss": 0.2885,
|
3025 |
+
"step": 3820
|
3026 |
+
},
|
3027 |
+
{
|
3028 |
+
"epoch": 3.27,
|
3029 |
+
"grad_norm": 2.5142822265625,
|
3030 |
+
"learning_rate": 3.8064791133844843e-05,
|
3031 |
+
"loss": 0.322,
|
3032 |
+
"step": 3830
|
3033 |
+
},
|
3034 |
+
{
|
3035 |
+
"epoch": 3.27,
|
3036 |
+
"grad_norm": 3.2371296882629395,
|
3037 |
+
"learning_rate": 3.763853367433931e-05,
|
3038 |
+
"loss": 0.2543,
|
3039 |
+
"step": 3840
|
3040 |
+
},
|
3041 |
+
{
|
3042 |
+
"epoch": 3.28,
|
3043 |
+
"grad_norm": 3.4256720542907715,
|
3044 |
+
"learning_rate": 3.721227621483376e-05,
|
3045 |
+
"loss": 0.3408,
|
3046 |
+
"step": 3850
|
3047 |
+
},
|
3048 |
+
{
|
3049 |
+
"epoch": 3.29,
|
3050 |
+
"grad_norm": 2.5072007179260254,
|
3051 |
+
"learning_rate": 3.678601875532822e-05,
|
3052 |
+
"loss": 0.3094,
|
3053 |
+
"step": 3860
|
3054 |
+
},
|
3055 |
+
{
|
3056 |
+
"epoch": 3.3,
|
3057 |
+
"grad_norm": 3.4915339946746826,
|
3058 |
+
"learning_rate": 3.635976129582268e-05,
|
3059 |
+
"loss": 0.4899,
|
3060 |
+
"step": 3870
|
3061 |
+
},
|
3062 |
+
{
|
3063 |
+
"epoch": 3.31,
|
3064 |
+
"grad_norm": 2.5262367725372314,
|
3065 |
+
"learning_rate": 3.593350383631714e-05,
|
3066 |
+
"loss": 0.377,
|
3067 |
+
"step": 3880
|
3068 |
+
},
|
3069 |
+
{
|
3070 |
+
"epoch": 3.32,
|
3071 |
+
"grad_norm": 2.8197975158691406,
|
3072 |
+
"learning_rate": 3.5507246376811596e-05,
|
3073 |
+
"loss": 0.4031,
|
3074 |
+
"step": 3890
|
3075 |
+
},
|
3076 |
+
{
|
3077 |
+
"epoch": 3.32,
|
3078 |
+
"grad_norm": 1.553293228149414,
|
3079 |
+
"learning_rate": 3.508098891730606e-05,
|
3080 |
+
"loss": 0.3016,
|
3081 |
+
"step": 3900
|
3082 |
+
},
|
3083 |
+
{
|
3084 |
+
"epoch": 3.32,
|
3085 |
+
"eval_accuracy": 0.9442934782608695,
|
3086 |
+
"eval_loss": 0.24018557369709015,
|
3087 |
+
"eval_runtime": 21.8606,
|
3088 |
+
"eval_samples_per_second": 33.668,
|
3089 |
+
"eval_steps_per_second": 4.208,
|
3090 |
+
"step": 3900
|
3091 |
+
},
|
3092 |
+
{
|
3093 |
+
"epoch": 3.33,
|
3094 |
+
"grad_norm": 4.228199481964111,
|
3095 |
+
"learning_rate": 3.4654731457800515e-05,
|
3096 |
+
"loss": 0.3767,
|
3097 |
+
"step": 3910
|
3098 |
+
},
|
3099 |
+
{
|
3100 |
+
"epoch": 3.34,
|
3101 |
+
"grad_norm": 3.0265350341796875,
|
3102 |
+
"learning_rate": 3.422847399829497e-05,
|
3103 |
+
"loss": 0.3735,
|
3104 |
+
"step": 3920
|
3105 |
+
},
|
3106 |
+
{
|
3107 |
+
"epoch": 3.35,
|
3108 |
+
"grad_norm": 5.04302453994751,
|
3109 |
+
"learning_rate": 3.3802216538789435e-05,
|
3110 |
+
"loss": 0.3223,
|
3111 |
+
"step": 3930
|
3112 |
+
},
|
3113 |
+
{
|
3114 |
+
"epoch": 3.36,
|
3115 |
+
"grad_norm": 3.406611919403076,
|
3116 |
+
"learning_rate": 3.337595907928389e-05,
|
3117 |
+
"loss": 0.3978,
|
3118 |
+
"step": 3940
|
3119 |
+
},
|
3120 |
+
{
|
3121 |
+
"epoch": 3.37,
|
3122 |
+
"grad_norm": 3.3483784198760986,
|
3123 |
+
"learning_rate": 3.294970161977835e-05,
|
3124 |
+
"loss": 0.2903,
|
3125 |
+
"step": 3950
|
3126 |
+
},
|
3127 |
+
{
|
3128 |
+
"epoch": 3.38,
|
3129 |
+
"grad_norm": 4.2682905197143555,
|
3130 |
+
"learning_rate": 3.252344416027281e-05,
|
3131 |
+
"loss": 0.3356,
|
3132 |
+
"step": 3960
|
3133 |
+
},
|
3134 |
+
{
|
3135 |
+
"epoch": 3.38,
|
3136 |
+
"grad_norm": 1.5132865905761719,
|
3137 |
+
"learning_rate": 3.209718670076726e-05,
|
3138 |
+
"loss": 0.3522,
|
3139 |
+
"step": 3970
|
3140 |
+
},
|
3141 |
+
{
|
3142 |
+
"epoch": 3.39,
|
3143 |
+
"grad_norm": 2.6350772380828857,
|
3144 |
+
"learning_rate": 3.1670929241261724e-05,
|
3145 |
+
"loss": 0.3271,
|
3146 |
+
"step": 3980
|
3147 |
+
},
|
3148 |
+
{
|
3149 |
+
"epoch": 3.4,
|
3150 |
+
"grad_norm": 2.6944427490234375,
|
3151 |
+
"learning_rate": 3.124467178175618e-05,
|
3152 |
+
"loss": 0.2867,
|
3153 |
+
"step": 3990
|
3154 |
+
},
|
3155 |
+
{
|
3156 |
+
"epoch": 3.41,
|
3157 |
+
"grad_norm": 2.3761823177337646,
|
3158 |
+
"learning_rate": 3.081841432225064e-05,
|
3159 |
+
"loss": 0.2961,
|
3160 |
+
"step": 4000
|
3161 |
+
},
|
3162 |
+
{
|
3163 |
+
"epoch": 3.41,
|
3164 |
+
"eval_accuracy": 0.9456521739130435,
|
3165 |
+
"eval_loss": 0.2413594275712967,
|
3166 |
+
"eval_runtime": 22.1329,
|
3167 |
+
"eval_samples_per_second": 33.254,
|
3168 |
+
"eval_steps_per_second": 4.157,
|
3169 |
+
"step": 4000
|
3170 |
+
},
|
3171 |
+
{
|
3172 |
+
"epoch": 3.42,
|
3173 |
+
"grad_norm": 3.0776336193084717,
|
3174 |
+
"learning_rate": 3.0392156862745097e-05,
|
3175 |
+
"loss": 0.333,
|
3176 |
+
"step": 4010
|
3177 |
+
},
|
3178 |
+
{
|
3179 |
+
"epoch": 3.43,
|
3180 |
+
"grad_norm": 2.033477783203125,
|
3181 |
+
"learning_rate": 2.9965899403239556e-05,
|
3182 |
+
"loss": 0.3263,
|
3183 |
+
"step": 4020
|
3184 |
+
},
|
3185 |
+
{
|
3186 |
+
"epoch": 3.44,
|
3187 |
+
"grad_norm": 4.718287467956543,
|
3188 |
+
"learning_rate": 2.9539641943734013e-05,
|
3189 |
+
"loss": 0.3373,
|
3190 |
+
"step": 4030
|
3191 |
+
},
|
3192 |
+
{
|
3193 |
+
"epoch": 3.44,
|
3194 |
+
"grad_norm": 5.81247615814209,
|
3195 |
+
"learning_rate": 2.9113384484228473e-05,
|
3196 |
+
"loss": 0.3212,
|
3197 |
+
"step": 4040
|
3198 |
+
},
|
3199 |
+
{
|
3200 |
+
"epoch": 3.45,
|
3201 |
+
"grad_norm": 4.8096723556518555,
|
3202 |
+
"learning_rate": 2.8687127024722932e-05,
|
3203 |
+
"loss": 0.3458,
|
3204 |
+
"step": 4050
|
3205 |
+
},
|
3206 |
+
{
|
3207 |
+
"epoch": 3.46,
|
3208 |
+
"grad_norm": 3.833155632019043,
|
3209 |
+
"learning_rate": 2.826086956521739e-05,
|
3210 |
+
"loss": 0.3212,
|
3211 |
+
"step": 4060
|
3212 |
+
},
|
3213 |
+
{
|
3214 |
+
"epoch": 3.47,
|
3215 |
+
"grad_norm": 2.942125082015991,
|
3216 |
+
"learning_rate": 2.783461210571185e-05,
|
3217 |
+
"loss": 0.2945,
|
3218 |
+
"step": 4070
|
3219 |
+
},
|
3220 |
+
{
|
3221 |
+
"epoch": 3.48,
|
3222 |
+
"grad_norm": 5.171932697296143,
|
3223 |
+
"learning_rate": 2.740835464620631e-05,
|
3224 |
+
"loss": 0.3338,
|
3225 |
+
"step": 4080
|
3226 |
+
},
|
3227 |
+
{
|
3228 |
+
"epoch": 3.49,
|
3229 |
+
"grad_norm": 4.480559825897217,
|
3230 |
+
"learning_rate": 2.6982097186700765e-05,
|
3231 |
+
"loss": 0.3755,
|
3232 |
+
"step": 4090
|
3233 |
+
},
|
3234 |
+
{
|
3235 |
+
"epoch": 3.5,
|
3236 |
+
"grad_norm": 3.524714708328247,
|
3237 |
+
"learning_rate": 2.6555839727195225e-05,
|
3238 |
+
"loss": 0.3822,
|
3239 |
+
"step": 4100
|
3240 |
+
},
|
3241 |
+
{
|
3242 |
+
"epoch": 3.5,
|
3243 |
+
"eval_accuracy": 0.9415760869565217,
|
3244 |
+
"eval_loss": 0.2412695288658142,
|
3245 |
+
"eval_runtime": 22.1034,
|
3246 |
+
"eval_samples_per_second": 33.298,
|
3247 |
+
"eval_steps_per_second": 4.162,
|
3248 |
+
"step": 4100
|
3249 |
+
},
|
3250 |
+
{
|
3251 |
+
"epoch": 3.5,
|
3252 |
+
"grad_norm": 2.857635736465454,
|
3253 |
+
"learning_rate": 2.6129582267689685e-05,
|
3254 |
+
"loss": 0.3709,
|
3255 |
+
"step": 4110
|
3256 |
+
},
|
3257 |
+
{
|
3258 |
+
"epoch": 3.51,
|
3259 |
+
"grad_norm": 3.15679931640625,
|
3260 |
+
"learning_rate": 2.5703324808184144e-05,
|
3261 |
+
"loss": 0.2529,
|
3262 |
+
"step": 4120
|
3263 |
+
},
|
3264 |
+
{
|
3265 |
+
"epoch": 3.52,
|
3266 |
+
"grad_norm": 2.1586387157440186,
|
3267 |
+
"learning_rate": 2.52770673486786e-05,
|
3268 |
+
"loss": 0.2946,
|
3269 |
+
"step": 4130
|
3270 |
+
},
|
3271 |
+
{
|
3272 |
+
"epoch": 3.53,
|
3273 |
+
"grad_norm": 3.978802442550659,
|
3274 |
+
"learning_rate": 2.4850809889173064e-05,
|
3275 |
+
"loss": 0.3867,
|
3276 |
+
"step": 4140
|
3277 |
+
},
|
3278 |
+
{
|
3279 |
+
"epoch": 3.54,
|
3280 |
+
"grad_norm": 4.345022201538086,
|
3281 |
+
"learning_rate": 2.442455242966752e-05,
|
3282 |
+
"loss": 0.3664,
|
3283 |
+
"step": 4150
|
3284 |
+
},
|
3285 |
+
{
|
3286 |
+
"epoch": 3.55,
|
3287 |
+
"grad_norm": 3.7032830715179443,
|
3288 |
+
"learning_rate": 2.399829497016198e-05,
|
3289 |
+
"loss": 0.3485,
|
3290 |
+
"step": 4160
|
3291 |
+
},
|
3292 |
+
{
|
3293 |
+
"epoch": 3.55,
|
3294 |
+
"grad_norm": 3.59366774559021,
|
3295 |
+
"learning_rate": 2.357203751065644e-05,
|
3296 |
+
"loss": 0.3255,
|
3297 |
+
"step": 4170
|
3298 |
+
},
|
3299 |
+
{
|
3300 |
+
"epoch": 3.56,
|
3301 |
+
"grad_norm": 5.068453311920166,
|
3302 |
+
"learning_rate": 2.3145780051150897e-05,
|
3303 |
+
"loss": 0.3121,
|
3304 |
+
"step": 4180
|
3305 |
+
},
|
3306 |
+
{
|
3307 |
+
"epoch": 3.57,
|
3308 |
+
"grad_norm": 5.033252239227295,
|
3309 |
+
"learning_rate": 2.2719522591645353e-05,
|
3310 |
+
"loss": 0.368,
|
3311 |
+
"step": 4190
|
3312 |
+
},
|
3313 |
+
{
|
3314 |
+
"epoch": 3.58,
|
3315 |
+
"grad_norm": 2.29129958152771,
|
3316 |
+
"learning_rate": 2.2293265132139813e-05,
|
3317 |
+
"loss": 0.2596,
|
3318 |
+
"step": 4200
|
3319 |
+
},
|
3320 |
+
{
|
3321 |
+
"epoch": 3.58,
|
3322 |
+
"eval_accuracy": 0.9456521739130435,
|
3323 |
+
"eval_loss": 0.23559102416038513,
|
3324 |
+
"eval_runtime": 22.0843,
|
3325 |
+
"eval_samples_per_second": 33.327,
|
3326 |
+
"eval_steps_per_second": 4.166,
|
3327 |
+
"step": 4200
|
3328 |
+
},
|
3329 |
+
{
|
3330 |
+
"epoch": 3.59,
|
3331 |
+
"grad_norm": 3.8676092624664307,
|
3332 |
+
"learning_rate": 2.1867007672634273e-05,
|
3333 |
+
"loss": 0.3853,
|
3334 |
+
"step": 4210
|
3335 |
+
},
|
3336 |
+
{
|
3337 |
+
"epoch": 3.6,
|
3338 |
+
"grad_norm": 2.257540702819824,
|
3339 |
+
"learning_rate": 2.144075021312873e-05,
|
3340 |
+
"loss": 0.4,
|
3341 |
+
"step": 4220
|
3342 |
+
},
|
3343 |
+
{
|
3344 |
+
"epoch": 3.61,
|
3345 |
+
"grad_norm": 4.077911853790283,
|
3346 |
+
"learning_rate": 2.101449275362319e-05,
|
3347 |
+
"loss": 0.3717,
|
3348 |
+
"step": 4230
|
3349 |
+
},
|
3350 |
+
{
|
3351 |
+
"epoch": 3.61,
|
3352 |
+
"grad_norm": 3.7997450828552246,
|
3353 |
+
"learning_rate": 2.058823529411765e-05,
|
3354 |
+
"loss": 0.3895,
|
3355 |
+
"step": 4240
|
3356 |
+
},
|
3357 |
+
{
|
3358 |
+
"epoch": 3.62,
|
3359 |
+
"grad_norm": 2.1893258094787598,
|
3360 |
+
"learning_rate": 2.0161977834612105e-05,
|
3361 |
+
"loss": 0.3243,
|
3362 |
+
"step": 4250
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 3.63,
|
3366 |
+
"grad_norm": 2.298306941986084,
|
3367 |
+
"learning_rate": 1.9735720375106565e-05,
|
3368 |
+
"loss": 0.3166,
|
3369 |
+
"step": 4260
|
3370 |
+
},
|
3371 |
+
{
|
3372 |
+
"epoch": 3.64,
|
3373 |
+
"grad_norm": 2.121025562286377,
|
3374 |
+
"learning_rate": 1.9309462915601025e-05,
|
3375 |
+
"loss": 0.2771,
|
3376 |
+
"step": 4270
|
3377 |
+
},
|
3378 |
+
{
|
3379 |
+
"epoch": 3.65,
|
3380 |
+
"grad_norm": 1.8228780031204224,
|
3381 |
+
"learning_rate": 1.888320545609548e-05,
|
3382 |
+
"loss": 0.2858,
|
3383 |
+
"step": 4280
|
3384 |
+
},
|
3385 |
+
{
|
3386 |
+
"epoch": 3.66,
|
3387 |
+
"grad_norm": 4.4333977699279785,
|
3388 |
+
"learning_rate": 1.845694799658994e-05,
|
3389 |
+
"loss": 0.3386,
|
3390 |
+
"step": 4290
|
3391 |
+
},
|
3392 |
+
{
|
3393 |
+
"epoch": 3.67,
|
3394 |
+
"grad_norm": 3.167982578277588,
|
3395 |
+
"learning_rate": 1.80306905370844e-05,
|
3396 |
+
"loss": 0.3064,
|
3397 |
+
"step": 4300
|
3398 |
+
},
|
3399 |
+
{
|
3400 |
+
"epoch": 3.67,
|
3401 |
+
"eval_accuracy": 0.9497282608695652,
|
3402 |
+
"eval_loss": 0.23243670165538788,
|
3403 |
+
"eval_runtime": 22.1685,
|
3404 |
+
"eval_samples_per_second": 33.2,
|
3405 |
+
"eval_steps_per_second": 4.15,
|
3406 |
+
"step": 4300
|
3407 |
+
},
|
3408 |
+
{
|
3409 |
+
"epoch": 3.67,
|
3410 |
+
"grad_norm": 3.177164316177368,
|
3411 |
+
"learning_rate": 1.7604433077578857e-05,
|
3412 |
+
"loss": 0.2947,
|
3413 |
+
"step": 4310
|
3414 |
+
},
|
3415 |
+
{
|
3416 |
+
"epoch": 3.68,
|
3417 |
+
"grad_norm": 4.344371318817139,
|
3418 |
+
"learning_rate": 1.7178175618073317e-05,
|
3419 |
+
"loss": 0.3845,
|
3420 |
+
"step": 4320
|
3421 |
+
},
|
3422 |
+
{
|
3423 |
+
"epoch": 3.69,
|
3424 |
+
"grad_norm": 2.364387035369873,
|
3425 |
+
"learning_rate": 1.6751918158567777e-05,
|
3426 |
+
"loss": 0.3276,
|
3427 |
+
"step": 4330
|
3428 |
+
},
|
3429 |
+
{
|
3430 |
+
"epoch": 3.7,
|
3431 |
+
"grad_norm": 4.086526870727539,
|
3432 |
+
"learning_rate": 1.6325660699062233e-05,
|
3433 |
+
"loss": 0.3708,
|
3434 |
+
"step": 4340
|
3435 |
+
},
|
3436 |
+
{
|
3437 |
+
"epoch": 3.71,
|
3438 |
+
"grad_norm": 4.65876579284668,
|
3439 |
+
"learning_rate": 1.5899403239556693e-05,
|
3440 |
+
"loss": 0.3258,
|
3441 |
+
"step": 4350
|
3442 |
+
},
|
3443 |
+
{
|
3444 |
+
"epoch": 3.72,
|
3445 |
+
"grad_norm": 4.176472187042236,
|
3446 |
+
"learning_rate": 1.5473145780051153e-05,
|
3447 |
+
"loss": 0.2992,
|
3448 |
+
"step": 4360
|
3449 |
+
},
|
3450 |
+
{
|
3451 |
+
"epoch": 3.73,
|
3452 |
+
"grad_norm": 3.989961624145508,
|
3453 |
+
"learning_rate": 1.504688832054561e-05,
|
3454 |
+
"loss": 0.3359,
|
3455 |
+
"step": 4370
|
3456 |
+
},
|
3457 |
+
{
|
3458 |
+
"epoch": 3.73,
|
3459 |
+
"grad_norm": 2.337566614151001,
|
3460 |
+
"learning_rate": 1.462063086104007e-05,
|
3461 |
+
"loss": 0.3066,
|
3462 |
+
"step": 4380
|
3463 |
+
},
|
3464 |
+
{
|
3465 |
+
"epoch": 3.74,
|
3466 |
+
"grad_norm": 2.9135842323303223,
|
3467 |
+
"learning_rate": 1.4194373401534527e-05,
|
3468 |
+
"loss": 0.2812,
|
3469 |
+
"step": 4390
|
3470 |
+
},
|
3471 |
+
{
|
3472 |
+
"epoch": 3.75,
|
3473 |
+
"grad_norm": 4.203680038452148,
|
3474 |
+
"learning_rate": 1.3768115942028985e-05,
|
3475 |
+
"loss": 0.3059,
|
3476 |
+
"step": 4400
|
3477 |
+
},
|
3478 |
+
{
|
3479 |
+
"epoch": 3.75,
|
3480 |
+
"eval_accuracy": 0.9456521739130435,
|
3481 |
+
"eval_loss": 0.2321375608444214,
|
3482 |
+
"eval_runtime": 21.9466,
|
3483 |
+
"eval_samples_per_second": 33.536,
|
3484 |
+
"eval_steps_per_second": 4.192,
|
3485 |
+
"step": 4400
|
3486 |
+
},
|
3487 |
+
{
|
3488 |
+
"epoch": 3.76,
|
3489 |
+
"grad_norm": 4.143373489379883,
|
3490 |
+
"learning_rate": 1.3341858482523445e-05,
|
3491 |
+
"loss": 0.354,
|
3492 |
+
"step": 4410
|
3493 |
+
},
|
3494 |
+
{
|
3495 |
+
"epoch": 3.77,
|
3496 |
+
"grad_norm": 2.7329490184783936,
|
3497 |
+
"learning_rate": 1.2915601023017903e-05,
|
3498 |
+
"loss": 0.3319,
|
3499 |
+
"step": 4420
|
3500 |
+
},
|
3501 |
+
{
|
3502 |
+
"epoch": 3.78,
|
3503 |
+
"grad_norm": 3.1333062648773193,
|
3504 |
+
"learning_rate": 1.2489343563512362e-05,
|
3505 |
+
"loss": 0.3421,
|
3506 |
+
"step": 4430
|
3507 |
+
},
|
3508 |
+
{
|
3509 |
+
"epoch": 3.79,
|
3510 |
+
"grad_norm": 3.976710319519043,
|
3511 |
+
"learning_rate": 1.2063086104006821e-05,
|
3512 |
+
"loss": 0.3263,
|
3513 |
+
"step": 4440
|
3514 |
+
},
|
3515 |
+
{
|
3516 |
+
"epoch": 3.79,
|
3517 |
+
"grad_norm": 1.6640021800994873,
|
3518 |
+
"learning_rate": 1.163682864450128e-05,
|
3519 |
+
"loss": 0.3224,
|
3520 |
+
"step": 4450
|
3521 |
+
},
|
3522 |
+
{
|
3523 |
+
"epoch": 3.8,
|
3524 |
+
"grad_norm": 2.7301018238067627,
|
3525 |
+
"learning_rate": 1.121057118499574e-05,
|
3526 |
+
"loss": 0.3694,
|
3527 |
+
"step": 4460
|
3528 |
+
},
|
3529 |
+
{
|
3530 |
+
"epoch": 3.81,
|
3531 |
+
"grad_norm": 2.9358885288238525,
|
3532 |
+
"learning_rate": 1.0784313725490197e-05,
|
3533 |
+
"loss": 0.3103,
|
3534 |
+
"step": 4470
|
3535 |
+
},
|
3536 |
+
{
|
3537 |
+
"epoch": 3.82,
|
3538 |
+
"grad_norm": 4.6855387687683105,
|
3539 |
+
"learning_rate": 1.0400682011935209e-05,
|
3540 |
+
"loss": 0.3927,
|
3541 |
+
"step": 4480
|
3542 |
+
},
|
3543 |
+
{
|
3544 |
+
"epoch": 3.83,
|
3545 |
+
"grad_norm": 2.2591495513916016,
|
3546 |
+
"learning_rate": 9.974424552429668e-06,
|
3547 |
+
"loss": 0.4017,
|
3548 |
+
"step": 4490
|
3549 |
+
},
|
3550 |
+
{
|
3551 |
+
"epoch": 3.84,
|
3552 |
+
"grad_norm": 2.375493049621582,
|
3553 |
+
"learning_rate": 9.548167092924126e-06,
|
3554 |
+
"loss": 0.42,
|
3555 |
+
"step": 4500
|
3556 |
+
},
|
3557 |
+
{
|
3558 |
+
"epoch": 3.84,
|
3559 |
+
"eval_accuracy": 0.9402173913043478,
|
3560 |
+
"eval_loss": 0.25556182861328125,
|
3561 |
+
"eval_runtime": 22.225,
|
3562 |
+
"eval_samples_per_second": 33.116,
|
3563 |
+
"eval_steps_per_second": 4.139,
|
3564 |
+
"step": 4500
|
3565 |
+
},
|
3566 |
+
{
|
3567 |
+
"epoch": 3.84,
|
3568 |
+
"grad_norm": 2.5054142475128174,
|
3569 |
+
"learning_rate": 9.121909633418585e-06,
|
3570 |
+
"loss": 0.3716,
|
3571 |
+
"step": 4510
|
3572 |
+
},
|
3573 |
+
{
|
3574 |
+
"epoch": 3.85,
|
3575 |
+
"grad_norm": 3.321662425994873,
|
3576 |
+
"learning_rate": 8.695652173913044e-06,
|
3577 |
+
"loss": 0.3185,
|
3578 |
+
"step": 4520
|
3579 |
+
},
|
3580 |
+
{
|
3581 |
+
"epoch": 3.86,
|
3582 |
+
"grad_norm": 2.8269574642181396,
|
3583 |
+
"learning_rate": 8.269394714407503e-06,
|
3584 |
+
"loss": 0.2985,
|
3585 |
+
"step": 4530
|
3586 |
+
},
|
3587 |
+
{
|
3588 |
+
"epoch": 3.87,
|
3589 |
+
"grad_norm": 3.426715612411499,
|
3590 |
+
"learning_rate": 7.84313725490196e-06,
|
3591 |
+
"loss": 0.3068,
|
3592 |
+
"step": 4540
|
3593 |
+
},
|
3594 |
+
{
|
3595 |
+
"epoch": 3.88,
|
3596 |
+
"grad_norm": 6.179994583129883,
|
3597 |
+
"learning_rate": 7.41687979539642e-06,
|
3598 |
+
"loss": 0.411,
|
3599 |
+
"step": 4550
|
3600 |
+
},
|
3601 |
+
{
|
3602 |
+
"epoch": 3.89,
|
3603 |
+
"grad_norm": 3.369870662689209,
|
3604 |
+
"learning_rate": 6.990622335890879e-06,
|
3605 |
+
"loss": 0.3128,
|
3606 |
+
"step": 4560
|
3607 |
+
},
|
3608 |
+
{
|
3609 |
+
"epoch": 3.9,
|
3610 |
+
"grad_norm": 4.271053791046143,
|
3611 |
+
"learning_rate": 6.564364876385337e-06,
|
3612 |
+
"loss": 0.3924,
|
3613 |
+
"step": 4570
|
3614 |
+
},
|
3615 |
+
{
|
3616 |
+
"epoch": 3.9,
|
3617 |
+
"grad_norm": 3.3079607486724854,
|
3618 |
+
"learning_rate": 6.138107416879796e-06,
|
3619 |
+
"loss": 0.34,
|
3620 |
+
"step": 4580
|
3621 |
+
},
|
3622 |
+
{
|
3623 |
+
"epoch": 3.91,
|
3624 |
+
"grad_norm": 6.019475936889648,
|
3625 |
+
"learning_rate": 5.711849957374255e-06,
|
3626 |
+
"loss": 0.3955,
|
3627 |
+
"step": 4590
|
3628 |
+
},
|
3629 |
+
{
|
3630 |
+
"epoch": 3.92,
|
3631 |
+
"grad_norm": 3.133575916290283,
|
3632 |
+
"learning_rate": 5.285592497868714e-06,
|
3633 |
+
"loss": 0.2959,
|
3634 |
+
"step": 4600
|
3635 |
+
},
|
3636 |
+
{
|
3637 |
+
"epoch": 3.92,
|
3638 |
+
"eval_accuracy": 0.9415760869565217,
|
3639 |
+
"eval_loss": 0.24908488988876343,
|
3640 |
+
"eval_runtime": 22.0463,
|
3641 |
+
"eval_samples_per_second": 33.384,
|
3642 |
+
"eval_steps_per_second": 4.173,
|
3643 |
+
"step": 4600
|
3644 |
+
},
|
3645 |
+
{
|
3646 |
+
"epoch": 3.93,
|
3647 |
+
"grad_norm": 4.321446895599365,
|
3648 |
+
"learning_rate": 4.859335038363171e-06,
|
3649 |
+
"loss": 0.4081,
|
3650 |
+
"step": 4610
|
3651 |
+
},
|
3652 |
+
{
|
3653 |
+
"epoch": 3.94,
|
3654 |
+
"grad_norm": 4.398526191711426,
|
3655 |
+
"learning_rate": 4.43307757885763e-06,
|
3656 |
+
"loss": 0.289,
|
3657 |
+
"step": 4620
|
3658 |
+
},
|
3659 |
+
{
|
3660 |
+
"epoch": 3.95,
|
3661 |
+
"grad_norm": 3.48781681060791,
|
3662 |
+
"learning_rate": 4.006820119352089e-06,
|
3663 |
+
"loss": 0.311,
|
3664 |
+
"step": 4630
|
3665 |
+
},
|
3666 |
+
{
|
3667 |
+
"epoch": 3.96,
|
3668 |
+
"grad_norm": 3.4008705615997314,
|
3669 |
+
"learning_rate": 3.5805626598465474e-06,
|
3670 |
+
"loss": 0.3046,
|
3671 |
+
"step": 4640
|
3672 |
+
},
|
3673 |
+
{
|
3674 |
+
"epoch": 3.96,
|
3675 |
+
"grad_norm": 3.372878074645996,
|
3676 |
+
"learning_rate": 3.154305200341006e-06,
|
3677 |
+
"loss": 0.4128,
|
3678 |
+
"step": 4650
|
3679 |
+
},
|
3680 |
+
{
|
3681 |
+
"epoch": 3.97,
|
3682 |
+
"grad_norm": 4.755558967590332,
|
3683 |
+
"learning_rate": 2.728047740835465e-06,
|
3684 |
+
"loss": 0.3619,
|
3685 |
+
"step": 4660
|
3686 |
+
},
|
3687 |
+
{
|
3688 |
+
"epoch": 3.98,
|
3689 |
+
"grad_norm": 2.060793161392212,
|
3690 |
+
"learning_rate": 2.3017902813299235e-06,
|
3691 |
+
"loss": 0.2876,
|
3692 |
+
"step": 4670
|
3693 |
+
},
|
3694 |
+
{
|
3695 |
+
"epoch": 3.99,
|
3696 |
+
"grad_norm": 4.377244472503662,
|
3697 |
+
"learning_rate": 1.875532821824382e-06,
|
3698 |
+
"loss": 0.3278,
|
3699 |
+
"step": 4680
|
3700 |
+
},
|
3701 |
+
{
|
3702 |
+
"epoch": 4.0,
|
3703 |
+
"grad_norm": 3.747706651687622,
|
3704 |
+
"learning_rate": 1.4492753623188406e-06,
|
3705 |
+
"loss": 0.3186,
|
3706 |
+
"step": 4690
|
3707 |
+
},
|
3708 |
+
{
|
3709 |
+
"epoch": 4.0,
|
3710 |
+
"step": 4692,
|
3711 |
+
"total_flos": 1.870424802038661e+18,
|
3712 |
+
"train_loss": 0.5094418521630693,
|
3713 |
+
"train_runtime": 3760.9881,
|
3714 |
+
"train_samples_per_second": 19.953,
|
3715 |
+
"train_steps_per_second": 1.248
|
3716 |
+
}
|
3717 |
+
],
|
3718 |
+
"logging_steps": 10,
|
3719 |
+
"max_steps": 4692,
|
3720 |
+
"num_input_tokens_seen": 0,
|
3721 |
+
"num_train_epochs": 4,
|
3722 |
+
"save_steps": 100,
|
3723 |
+
"total_flos": 1.870424802038661e+18,
|
3724 |
+
"train_batch_size": 16,
|
3725 |
+
"trial_name": null,
|
3726 |
+
"trial_params": null
|
3727 |
+
}
|