jalaneunos
commited on
Commit
•
47cc262
1
Parent(s):
9113858
End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar18_16-33-47_e3106b2fe063/events.out.tfevents.1710780454.e3106b2fe063.997.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1534 -0
README.md
CHANGED
@@ -22,7 +22,7 @@ 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
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [WinKawaks/vit-tiny-patch16-224](https://huggingface.co/WinKawaks/vit-tiny-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 0.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.842911877394636
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [WinKawaks/vit-tiny-patch16-224](https://huggingface.co/WinKawaks/vit-tiny-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.3542
|
36 |
+
- Accuracy: 0.8429
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.842911877394636,
|
4 |
+
"eval_loss": 0.3542243540287018,
|
5 |
+
"eval_runtime": 5.8665,
|
6 |
+
"eval_samples_per_second": 489.386,
|
7 |
+
"eval_steps_per_second": 15.341,
|
8 |
+
"total_flos": 1.2892209231838003e+18,
|
9 |
+
"train_loss": 0.45288851414576614,
|
10 |
+
"train_runtime": 810.9556,
|
11 |
+
"train_samples_per_second": 318.599,
|
12 |
+
"train_steps_per_second": 2.491
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.842911877394636,
|
4 |
+
"eval_loss": 0.3542243540287018,
|
5 |
+
"eval_runtime": 5.8665,
|
6 |
+
"eval_samples_per_second": 489.386,
|
7 |
+
"eval_steps_per_second": 15.341
|
8 |
+
}
|
runs/Mar18_16-33-47_e3106b2fe063/events.out.tfevents.1710780454.e3106b2fe063.997.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2a277391455f7a0a3384c8dc36f81b4f7ce45d62948466c7922e98b7e6dd1be3
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"total_flos": 1.2892209231838003e+18,
|
4 |
+
"train_loss": 0.45288851414576614,
|
5 |
+
"train_runtime": 810.9556,
|
6 |
+
"train_samples_per_second": 318.599,
|
7 |
+
"train_steps_per_second": 2.491
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1534 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.842911877394636,
|
3 |
+
"best_model_checkpoint": "vit-tiny-patch16-224-winkawaks/checkpoint-1818",
|
4 |
+
"epoch": 10.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2020,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"grad_norm": 11.655478477478027,
|
14 |
+
"learning_rate": 2.4752475247524753e-06,
|
15 |
+
"loss": 0.8812,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.1,
|
20 |
+
"grad_norm": 6.654660224914551,
|
21 |
+
"learning_rate": 4.950495049504951e-06,
|
22 |
+
"loss": 0.7531,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.15,
|
27 |
+
"grad_norm": 4.010664939880371,
|
28 |
+
"learning_rate": 7.4257425742574256e-06,
|
29 |
+
"loss": 0.7312,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.2,
|
34 |
+
"grad_norm": 3.320401191711426,
|
35 |
+
"learning_rate": 9.900990099009901e-06,
|
36 |
+
"loss": 0.6753,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.25,
|
41 |
+
"grad_norm": 4.002676010131836,
|
42 |
+
"learning_rate": 1.2376237623762377e-05,
|
43 |
+
"loss": 0.6656,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.3,
|
48 |
+
"grad_norm": 5.1362690925598145,
|
49 |
+
"learning_rate": 1.4851485148514851e-05,
|
50 |
+
"loss": 0.636,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.35,
|
55 |
+
"grad_norm": 4.089080333709717,
|
56 |
+
"learning_rate": 1.7326732673267325e-05,
|
57 |
+
"loss": 0.6258,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.4,
|
62 |
+
"grad_norm": 2.782809257507324,
|
63 |
+
"learning_rate": 1.9801980198019803e-05,
|
64 |
+
"loss": 0.6074,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.45,
|
69 |
+
"grad_norm": 4.684001922607422,
|
70 |
+
"learning_rate": 2.227722772277228e-05,
|
71 |
+
"loss": 0.6114,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.5,
|
76 |
+
"grad_norm": 8.801894187927246,
|
77 |
+
"learning_rate": 2.4752475247524754e-05,
|
78 |
+
"loss": 0.5888,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.54,
|
83 |
+
"grad_norm": 10.810859680175781,
|
84 |
+
"learning_rate": 2.722772277227723e-05,
|
85 |
+
"loss": 0.5858,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.59,
|
90 |
+
"grad_norm": 4.189158916473389,
|
91 |
+
"learning_rate": 2.9702970297029702e-05,
|
92 |
+
"loss": 0.5849,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.64,
|
97 |
+
"grad_norm": 6.533910274505615,
|
98 |
+
"learning_rate": 3.217821782178218e-05,
|
99 |
+
"loss": 0.5811,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.69,
|
104 |
+
"grad_norm": 15.806050300598145,
|
105 |
+
"learning_rate": 3.465346534653465e-05,
|
106 |
+
"loss": 0.6041,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.74,
|
111 |
+
"grad_norm": 3.087372064590454,
|
112 |
+
"learning_rate": 3.712871287128713e-05,
|
113 |
+
"loss": 0.5402,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.79,
|
118 |
+
"grad_norm": 6.852709770202637,
|
119 |
+
"learning_rate": 3.9603960396039605e-05,
|
120 |
+
"loss": 0.5435,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.84,
|
125 |
+
"grad_norm": 3.9751667976379395,
|
126 |
+
"learning_rate": 4.207920792079208e-05,
|
127 |
+
"loss": 0.5826,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.89,
|
132 |
+
"grad_norm": 8.571554183959961,
|
133 |
+
"learning_rate": 4.455445544554456e-05,
|
134 |
+
"loss": 0.5399,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.94,
|
139 |
+
"grad_norm": 2.956425905227661,
|
140 |
+
"learning_rate": 4.702970297029703e-05,
|
141 |
+
"loss": 0.5627,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.99,
|
146 |
+
"grad_norm": 10.44449520111084,
|
147 |
+
"learning_rate": 4.950495049504951e-05,
|
148 |
+
"loss": 0.5345,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.0,
|
153 |
+
"eval_accuracy": 0.7770811563915012,
|
154 |
+
"eval_loss": 0.4689331650733948,
|
155 |
+
"eval_runtime": 5.627,
|
156 |
+
"eval_samples_per_second": 510.22,
|
157 |
+
"eval_steps_per_second": 15.994,
|
158 |
+
"step": 202
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.04,
|
162 |
+
"grad_norm": 5.914970397949219,
|
163 |
+
"learning_rate": 4.977997799779978e-05,
|
164 |
+
"loss": 0.533,
|
165 |
+
"step": 210
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 1.09,
|
169 |
+
"grad_norm": 7.073808193206787,
|
170 |
+
"learning_rate": 4.950495049504951e-05,
|
171 |
+
"loss": 0.517,
|
172 |
+
"step": 220
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 1.14,
|
176 |
+
"grad_norm": 3.3154642581939697,
|
177 |
+
"learning_rate": 4.9229922992299234e-05,
|
178 |
+
"loss": 0.5385,
|
179 |
+
"step": 230
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.19,
|
183 |
+
"grad_norm": 3.500061273574829,
|
184 |
+
"learning_rate": 4.895489548954896e-05,
|
185 |
+
"loss": 0.5345,
|
186 |
+
"step": 240
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.24,
|
190 |
+
"grad_norm": 8.62484359741211,
|
191 |
+
"learning_rate": 4.867986798679868e-05,
|
192 |
+
"loss": 0.5368,
|
193 |
+
"step": 250
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 1.29,
|
197 |
+
"grad_norm": 9.027387619018555,
|
198 |
+
"learning_rate": 4.8404840484048406e-05,
|
199 |
+
"loss": 0.5247,
|
200 |
+
"step": 260
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.34,
|
204 |
+
"grad_norm": 7.457286357879639,
|
205 |
+
"learning_rate": 4.812981298129813e-05,
|
206 |
+
"loss": 0.5044,
|
207 |
+
"step": 270
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 1.39,
|
211 |
+
"grad_norm": 7.2032790184021,
|
212 |
+
"learning_rate": 4.785478547854786e-05,
|
213 |
+
"loss": 0.5156,
|
214 |
+
"step": 280
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 1.44,
|
218 |
+
"grad_norm": 5.909176826477051,
|
219 |
+
"learning_rate": 4.7579757975797585e-05,
|
220 |
+
"loss": 0.5191,
|
221 |
+
"step": 290
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.49,
|
225 |
+
"grad_norm": 4.56859827041626,
|
226 |
+
"learning_rate": 4.730473047304731e-05,
|
227 |
+
"loss": 0.4891,
|
228 |
+
"step": 300
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 1.53,
|
232 |
+
"grad_norm": 5.181271553039551,
|
233 |
+
"learning_rate": 4.702970297029703e-05,
|
234 |
+
"loss": 0.5016,
|
235 |
+
"step": 310
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 1.58,
|
239 |
+
"grad_norm": 6.948479175567627,
|
240 |
+
"learning_rate": 4.675467546754676e-05,
|
241 |
+
"loss": 0.5008,
|
242 |
+
"step": 320
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.63,
|
246 |
+
"grad_norm": 7.155640125274658,
|
247 |
+
"learning_rate": 4.647964796479648e-05,
|
248 |
+
"loss": 0.5451,
|
249 |
+
"step": 330
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.68,
|
253 |
+
"grad_norm": 3.5051815509796143,
|
254 |
+
"learning_rate": 4.62046204620462e-05,
|
255 |
+
"loss": 0.4998,
|
256 |
+
"step": 340
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 1.73,
|
260 |
+
"grad_norm": 12.015359878540039,
|
261 |
+
"learning_rate": 4.592959295929593e-05,
|
262 |
+
"loss": 0.5302,
|
263 |
+
"step": 350
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.78,
|
267 |
+
"grad_norm": 5.719058990478516,
|
268 |
+
"learning_rate": 4.5654565456545655e-05,
|
269 |
+
"loss": 0.5052,
|
270 |
+
"step": 360
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 1.83,
|
274 |
+
"grad_norm": 6.292553424835205,
|
275 |
+
"learning_rate": 4.537953795379538e-05,
|
276 |
+
"loss": 0.5192,
|
277 |
+
"step": 370
|
278 |
+
},
|
279 |
+
{
|
280 |
+
"epoch": 1.88,
|
281 |
+
"grad_norm": 2.8755393028259277,
|
282 |
+
"learning_rate": 4.510451045104511e-05,
|
283 |
+
"loss": 0.4966,
|
284 |
+
"step": 380
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 1.93,
|
288 |
+
"grad_norm": 3.096843719482422,
|
289 |
+
"learning_rate": 4.4829482948294834e-05,
|
290 |
+
"loss": 0.5032,
|
291 |
+
"step": 390
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 1.98,
|
295 |
+
"grad_norm": 3.9766151905059814,
|
296 |
+
"learning_rate": 4.455445544554456e-05,
|
297 |
+
"loss": 0.4936,
|
298 |
+
"step": 400
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 2.0,
|
302 |
+
"eval_accuracy": 0.7485196795541623,
|
303 |
+
"eval_loss": 0.5022404789924622,
|
304 |
+
"eval_runtime": 5.6861,
|
305 |
+
"eval_samples_per_second": 504.916,
|
306 |
+
"eval_steps_per_second": 15.828,
|
307 |
+
"step": 404
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 2.03,
|
311 |
+
"grad_norm": 7.3236870765686035,
|
312 |
+
"learning_rate": 4.427942794279428e-05,
|
313 |
+
"loss": 0.4828,
|
314 |
+
"step": 410
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 2.08,
|
318 |
+
"grad_norm": 4.203598976135254,
|
319 |
+
"learning_rate": 4.4004400440044006e-05,
|
320 |
+
"loss": 0.5108,
|
321 |
+
"step": 420
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 2.13,
|
325 |
+
"grad_norm": 4.851642608642578,
|
326 |
+
"learning_rate": 4.372937293729373e-05,
|
327 |
+
"loss": 0.474,
|
328 |
+
"step": 430
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 2.18,
|
332 |
+
"grad_norm": 8.77710247039795,
|
333 |
+
"learning_rate": 4.345434543454346e-05,
|
334 |
+
"loss": 0.4902,
|
335 |
+
"step": 440
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 2.23,
|
339 |
+
"grad_norm": 6.436154365539551,
|
340 |
+
"learning_rate": 4.3179317931793185e-05,
|
341 |
+
"loss": 0.4838,
|
342 |
+
"step": 450
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 2.28,
|
346 |
+
"grad_norm": 4.809104919433594,
|
347 |
+
"learning_rate": 4.2904290429042904e-05,
|
348 |
+
"loss": 0.5064,
|
349 |
+
"step": 460
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 2.33,
|
353 |
+
"grad_norm": 4.125964641571045,
|
354 |
+
"learning_rate": 4.262926292629263e-05,
|
355 |
+
"loss": 0.4819,
|
356 |
+
"step": 470
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 2.38,
|
360 |
+
"grad_norm": 6.8852763175964355,
|
361 |
+
"learning_rate": 4.2354235423542356e-05,
|
362 |
+
"loss": 0.5011,
|
363 |
+
"step": 480
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 2.43,
|
367 |
+
"grad_norm": 3.3697924613952637,
|
368 |
+
"learning_rate": 4.207920792079208e-05,
|
369 |
+
"loss": 0.4676,
|
370 |
+
"step": 490
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 2.48,
|
374 |
+
"grad_norm": 5.354856967926025,
|
375 |
+
"learning_rate": 4.18041804180418e-05,
|
376 |
+
"loss": 0.507,
|
377 |
+
"step": 500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 2.52,
|
381 |
+
"grad_norm": 7.638093948364258,
|
382 |
+
"learning_rate": 4.152915291529153e-05,
|
383 |
+
"loss": 0.5046,
|
384 |
+
"step": 510
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 2.57,
|
388 |
+
"grad_norm": 4.444268226623535,
|
389 |
+
"learning_rate": 4.1254125412541255e-05,
|
390 |
+
"loss": 0.4635,
|
391 |
+
"step": 520
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 2.62,
|
395 |
+
"grad_norm": 8.037334442138672,
|
396 |
+
"learning_rate": 4.097909790979098e-05,
|
397 |
+
"loss": 0.4796,
|
398 |
+
"step": 530
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 2.67,
|
402 |
+
"grad_norm": 6.52307653427124,
|
403 |
+
"learning_rate": 4.070407040704071e-05,
|
404 |
+
"loss": 0.4668,
|
405 |
+
"step": 540
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 2.72,
|
409 |
+
"grad_norm": 7.490472316741943,
|
410 |
+
"learning_rate": 4.042904290429043e-05,
|
411 |
+
"loss": 0.4747,
|
412 |
+
"step": 550
|
413 |
+
},
|
414 |
+
{
|
415 |
+
"epoch": 2.77,
|
416 |
+
"grad_norm": 3.393075942993164,
|
417 |
+
"learning_rate": 4.015401540154016e-05,
|
418 |
+
"loss": 0.4667,
|
419 |
+
"step": 560
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 2.82,
|
423 |
+
"grad_norm": 4.454715251922607,
|
424 |
+
"learning_rate": 3.987898789878988e-05,
|
425 |
+
"loss": 0.4795,
|
426 |
+
"step": 570
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 2.87,
|
430 |
+
"grad_norm": 5.316873550415039,
|
431 |
+
"learning_rate": 3.9603960396039605e-05,
|
432 |
+
"loss": 0.499,
|
433 |
+
"step": 580
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 2.92,
|
437 |
+
"grad_norm": 2.8893849849700928,
|
438 |
+
"learning_rate": 3.932893289328933e-05,
|
439 |
+
"loss": 0.4913,
|
440 |
+
"step": 590
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 2.97,
|
444 |
+
"grad_norm": 2.6538729667663574,
|
445 |
+
"learning_rate": 3.905390539053906e-05,
|
446 |
+
"loss": 0.4911,
|
447 |
+
"step": 600
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 3.0,
|
451 |
+
"eval_accuracy": 0.82793451758969,
|
452 |
+
"eval_loss": 0.3886525630950928,
|
453 |
+
"eval_runtime": 5.5731,
|
454 |
+
"eval_samples_per_second": 515.151,
|
455 |
+
"eval_steps_per_second": 16.149,
|
456 |
+
"step": 606
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 3.02,
|
460 |
+
"grad_norm": 9.823522567749023,
|
461 |
+
"learning_rate": 3.877887788778878e-05,
|
462 |
+
"loss": 0.4731,
|
463 |
+
"step": 610
|
464 |
+
},
|
465 |
+
{
|
466 |
+
"epoch": 3.07,
|
467 |
+
"grad_norm": 5.899537086486816,
|
468 |
+
"learning_rate": 3.8503850385038503e-05,
|
469 |
+
"loss": 0.4621,
|
470 |
+
"step": 620
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 3.12,
|
474 |
+
"grad_norm": 4.964333534240723,
|
475 |
+
"learning_rate": 3.822882288228823e-05,
|
476 |
+
"loss": 0.4762,
|
477 |
+
"step": 630
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 3.17,
|
481 |
+
"grad_norm": 4.469970703125,
|
482 |
+
"learning_rate": 3.7953795379537956e-05,
|
483 |
+
"loss": 0.4774,
|
484 |
+
"step": 640
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 3.22,
|
488 |
+
"grad_norm": 3.393183469772339,
|
489 |
+
"learning_rate": 3.767876787678768e-05,
|
490 |
+
"loss": 0.4501,
|
491 |
+
"step": 650
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 3.27,
|
495 |
+
"grad_norm": 11.101873397827148,
|
496 |
+
"learning_rate": 3.74037403740374e-05,
|
497 |
+
"loss": 0.4596,
|
498 |
+
"step": 660
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 3.32,
|
502 |
+
"grad_norm": 4.9833831787109375,
|
503 |
+
"learning_rate": 3.712871287128713e-05,
|
504 |
+
"loss": 0.4516,
|
505 |
+
"step": 670
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 3.37,
|
509 |
+
"grad_norm": 5.423743724822998,
|
510 |
+
"learning_rate": 3.6853685368536854e-05,
|
511 |
+
"loss": 0.4887,
|
512 |
+
"step": 680
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 3.42,
|
516 |
+
"grad_norm": 4.94615364074707,
|
517 |
+
"learning_rate": 3.657865786578658e-05,
|
518 |
+
"loss": 0.4538,
|
519 |
+
"step": 690
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 3.47,
|
523 |
+
"grad_norm": 5.113562107086182,
|
524 |
+
"learning_rate": 3.6303630363036307e-05,
|
525 |
+
"loss": 0.4645,
|
526 |
+
"step": 700
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 3.51,
|
530 |
+
"grad_norm": 7.3234357833862305,
|
531 |
+
"learning_rate": 3.602860286028603e-05,
|
532 |
+
"loss": 0.4645,
|
533 |
+
"step": 710
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 3.56,
|
537 |
+
"grad_norm": 4.696035385131836,
|
538 |
+
"learning_rate": 3.575357535753576e-05,
|
539 |
+
"loss": 0.4584,
|
540 |
+
"step": 720
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 3.61,
|
544 |
+
"grad_norm": 3.724952459335327,
|
545 |
+
"learning_rate": 3.5478547854785485e-05,
|
546 |
+
"loss": 0.427,
|
547 |
+
"step": 730
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 3.66,
|
551 |
+
"grad_norm": 3.8439409732818604,
|
552 |
+
"learning_rate": 3.5203520352035205e-05,
|
553 |
+
"loss": 0.4554,
|
554 |
+
"step": 740
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 3.71,
|
558 |
+
"grad_norm": 3.4024925231933594,
|
559 |
+
"learning_rate": 3.492849284928493e-05,
|
560 |
+
"loss": 0.4413,
|
561 |
+
"step": 750
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 3.76,
|
565 |
+
"grad_norm": 8.046459197998047,
|
566 |
+
"learning_rate": 3.465346534653465e-05,
|
567 |
+
"loss": 0.4638,
|
568 |
+
"step": 760
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 3.81,
|
572 |
+
"grad_norm": 5.1486358642578125,
|
573 |
+
"learning_rate": 3.4378437843784377e-05,
|
574 |
+
"loss": 0.4852,
|
575 |
+
"step": 770
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 3.86,
|
579 |
+
"grad_norm": 3.408259868621826,
|
580 |
+
"learning_rate": 3.41034103410341e-05,
|
581 |
+
"loss": 0.4347,
|
582 |
+
"step": 780
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 3.91,
|
586 |
+
"grad_norm": 5.650546073913574,
|
587 |
+
"learning_rate": 3.382838283828383e-05,
|
588 |
+
"loss": 0.4294,
|
589 |
+
"step": 790
|
590 |
+
},
|
591 |
+
{
|
592 |
+
"epoch": 3.96,
|
593 |
+
"grad_norm": 3.8805394172668457,
|
594 |
+
"learning_rate": 3.3553355335533555e-05,
|
595 |
+
"loss": 0.4191,
|
596 |
+
"step": 800
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 4.0,
|
600 |
+
"eval_accuracy": 0.8098223615464994,
|
601 |
+
"eval_loss": 0.4121237099170685,
|
602 |
+
"eval_runtime": 5.4396,
|
603 |
+
"eval_samples_per_second": 527.797,
|
604 |
+
"eval_steps_per_second": 16.545,
|
605 |
+
"step": 808
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 4.01,
|
609 |
+
"grad_norm": 4.736340522766113,
|
610 |
+
"learning_rate": 3.327832783278328e-05,
|
611 |
+
"loss": 0.4824,
|
612 |
+
"step": 810
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 4.06,
|
616 |
+
"grad_norm": 4.468814373016357,
|
617 |
+
"learning_rate": 3.300330033003301e-05,
|
618 |
+
"loss": 0.468,
|
619 |
+
"step": 820
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 4.11,
|
623 |
+
"grad_norm": 3.311174154281616,
|
624 |
+
"learning_rate": 3.272827282728273e-05,
|
625 |
+
"loss": 0.442,
|
626 |
+
"step": 830
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 4.16,
|
630 |
+
"grad_norm": 6.016156196594238,
|
631 |
+
"learning_rate": 3.2453245324532453e-05,
|
632 |
+
"loss": 0.4748,
|
633 |
+
"step": 840
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 4.21,
|
637 |
+
"grad_norm": 6.195199966430664,
|
638 |
+
"learning_rate": 3.217821782178218e-05,
|
639 |
+
"loss": 0.4488,
|
640 |
+
"step": 850
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 4.26,
|
644 |
+
"grad_norm": 3.6519343852996826,
|
645 |
+
"learning_rate": 3.1903190319031906e-05,
|
646 |
+
"loss": 0.4285,
|
647 |
+
"step": 860
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 4.31,
|
651 |
+
"grad_norm": 5.461367130279541,
|
652 |
+
"learning_rate": 3.162816281628163e-05,
|
653 |
+
"loss": 0.4124,
|
654 |
+
"step": 870
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 4.36,
|
658 |
+
"grad_norm": 3.7743630409240723,
|
659 |
+
"learning_rate": 3.135313531353136e-05,
|
660 |
+
"loss": 0.4361,
|
661 |
+
"step": 880
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 4.41,
|
665 |
+
"grad_norm": 3.741748571395874,
|
666 |
+
"learning_rate": 3.1078107810781085e-05,
|
667 |
+
"loss": 0.4196,
|
668 |
+
"step": 890
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 4.46,
|
672 |
+
"grad_norm": 5.458765983581543,
|
673 |
+
"learning_rate": 3.0803080308030804e-05,
|
674 |
+
"loss": 0.4814,
|
675 |
+
"step": 900
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 4.5,
|
679 |
+
"grad_norm": 5.118380546569824,
|
680 |
+
"learning_rate": 3.052805280528053e-05,
|
681 |
+
"loss": 0.4441,
|
682 |
+
"step": 910
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 4.55,
|
686 |
+
"grad_norm": 4.428480625152588,
|
687 |
+
"learning_rate": 3.0253025302530253e-05,
|
688 |
+
"loss": 0.4217,
|
689 |
+
"step": 920
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 4.6,
|
693 |
+
"grad_norm": 3.2278711795806885,
|
694 |
+
"learning_rate": 2.9977997799779976e-05,
|
695 |
+
"loss": 0.4259,
|
696 |
+
"step": 930
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 4.65,
|
700 |
+
"grad_norm": 3.5108141899108887,
|
701 |
+
"learning_rate": 2.9702970297029702e-05,
|
702 |
+
"loss": 0.4507,
|
703 |
+
"step": 940
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 4.7,
|
707 |
+
"grad_norm": 3.02825665473938,
|
708 |
+
"learning_rate": 2.942794279427943e-05,
|
709 |
+
"loss": 0.4095,
|
710 |
+
"step": 950
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 4.75,
|
714 |
+
"grad_norm": 3.8650879859924316,
|
715 |
+
"learning_rate": 2.9152915291529155e-05,
|
716 |
+
"loss": 0.4251,
|
717 |
+
"step": 960
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 4.8,
|
721 |
+
"grad_norm": 5.335313320159912,
|
722 |
+
"learning_rate": 2.8877887788778878e-05,
|
723 |
+
"loss": 0.428,
|
724 |
+
"step": 970
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 4.85,
|
728 |
+
"grad_norm": 5.564798831939697,
|
729 |
+
"learning_rate": 2.8602860286028604e-05,
|
730 |
+
"loss": 0.4713,
|
731 |
+
"step": 980
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 4.9,
|
735 |
+
"grad_norm": 4.036501407623291,
|
736 |
+
"learning_rate": 2.832783278327833e-05,
|
737 |
+
"loss": 0.4152,
|
738 |
+
"step": 990
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 4.95,
|
742 |
+
"grad_norm": 4.295027732849121,
|
743 |
+
"learning_rate": 2.8052805280528056e-05,
|
744 |
+
"loss": 0.4209,
|
745 |
+
"step": 1000
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 5.0,
|
749 |
+
"grad_norm": 2.8911187648773193,
|
750 |
+
"learning_rate": 2.777777777777778e-05,
|
751 |
+
"loss": 0.4408,
|
752 |
+
"step": 1010
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 5.0,
|
756 |
+
"eval_accuracy": 0.8254963427377221,
|
757 |
+
"eval_loss": 0.38967224955558777,
|
758 |
+
"eval_runtime": 5.6692,
|
759 |
+
"eval_samples_per_second": 506.425,
|
760 |
+
"eval_steps_per_second": 15.875,
|
761 |
+
"step": 1010
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 5.05,
|
765 |
+
"grad_norm": 7.270760536193848,
|
766 |
+
"learning_rate": 2.7502750275027505e-05,
|
767 |
+
"loss": 0.4415,
|
768 |
+
"step": 1020
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 5.1,
|
772 |
+
"grad_norm": 2.75962495803833,
|
773 |
+
"learning_rate": 2.722772277227723e-05,
|
774 |
+
"loss": 0.4282,
|
775 |
+
"step": 1030
|
776 |
+
},
|
777 |
+
{
|
778 |
+
"epoch": 5.15,
|
779 |
+
"grad_norm": 3.6654067039489746,
|
780 |
+
"learning_rate": 2.6952695269526958e-05,
|
781 |
+
"loss": 0.4265,
|
782 |
+
"step": 1040
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 5.2,
|
786 |
+
"grad_norm": 5.461548805236816,
|
787 |
+
"learning_rate": 2.667766776677668e-05,
|
788 |
+
"loss": 0.4188,
|
789 |
+
"step": 1050
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 5.25,
|
793 |
+
"grad_norm": 3.629088878631592,
|
794 |
+
"learning_rate": 2.64026402640264e-05,
|
795 |
+
"loss": 0.438,
|
796 |
+
"step": 1060
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 5.3,
|
800 |
+
"grad_norm": 5.7036614418029785,
|
801 |
+
"learning_rate": 2.6127612761276126e-05,
|
802 |
+
"loss": 0.4073,
|
803 |
+
"step": 1070
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 5.35,
|
807 |
+
"grad_norm": 7.140379905700684,
|
808 |
+
"learning_rate": 2.5852585258525853e-05,
|
809 |
+
"loss": 0.4287,
|
810 |
+
"step": 1080
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 5.4,
|
814 |
+
"grad_norm": 3.2449722290039062,
|
815 |
+
"learning_rate": 2.557755775577558e-05,
|
816 |
+
"loss": 0.425,
|
817 |
+
"step": 1090
|
818 |
+
},
|
819 |
+
{
|
820 |
+
"epoch": 5.45,
|
821 |
+
"grad_norm": 4.488211631774902,
|
822 |
+
"learning_rate": 2.53025302530253e-05,
|
823 |
+
"loss": 0.4203,
|
824 |
+
"step": 1100
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 5.5,
|
828 |
+
"grad_norm": 3.7270421981811523,
|
829 |
+
"learning_rate": 2.5027502750275028e-05,
|
830 |
+
"loss": 0.4174,
|
831 |
+
"step": 1110
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 5.54,
|
835 |
+
"grad_norm": 4.254336357116699,
|
836 |
+
"learning_rate": 2.4752475247524754e-05,
|
837 |
+
"loss": 0.4533,
|
838 |
+
"step": 1120
|
839 |
+
},
|
840 |
+
{
|
841 |
+
"epoch": 5.59,
|
842 |
+
"grad_norm": 3.1208856105804443,
|
843 |
+
"learning_rate": 2.447744774477448e-05,
|
844 |
+
"loss": 0.4221,
|
845 |
+
"step": 1130
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 5.64,
|
849 |
+
"grad_norm": 6.634130001068115,
|
850 |
+
"learning_rate": 2.4202420242024203e-05,
|
851 |
+
"loss": 0.4377,
|
852 |
+
"step": 1140
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 5.69,
|
856 |
+
"grad_norm": 5.457045078277588,
|
857 |
+
"learning_rate": 2.392739273927393e-05,
|
858 |
+
"loss": 0.4459,
|
859 |
+
"step": 1150
|
860 |
+
},
|
861 |
+
{
|
862 |
+
"epoch": 5.74,
|
863 |
+
"grad_norm": 3.537189483642578,
|
864 |
+
"learning_rate": 2.3652365236523656e-05,
|
865 |
+
"loss": 0.4423,
|
866 |
+
"step": 1160
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 5.79,
|
870 |
+
"grad_norm": 5.883282661437988,
|
871 |
+
"learning_rate": 2.337733773377338e-05,
|
872 |
+
"loss": 0.4121,
|
873 |
+
"step": 1170
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 5.84,
|
877 |
+
"grad_norm": 4.0562744140625,
|
878 |
+
"learning_rate": 2.31023102310231e-05,
|
879 |
+
"loss": 0.4114,
|
880 |
+
"step": 1180
|
881 |
+
},
|
882 |
+
{
|
883 |
+
"epoch": 5.89,
|
884 |
+
"grad_norm": 5.0152974128723145,
|
885 |
+
"learning_rate": 2.2827282728272828e-05,
|
886 |
+
"loss": 0.4287,
|
887 |
+
"step": 1190
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 5.94,
|
891 |
+
"grad_norm": 4.672534465789795,
|
892 |
+
"learning_rate": 2.2552255225522554e-05,
|
893 |
+
"loss": 0.4124,
|
894 |
+
"step": 1200
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 5.99,
|
898 |
+
"grad_norm": 5.874300479888916,
|
899 |
+
"learning_rate": 2.227722772277228e-05,
|
900 |
+
"loss": 0.4134,
|
901 |
+
"step": 1210
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 6.0,
|
905 |
+
"eval_accuracy": 0.8331591779867642,
|
906 |
+
"eval_loss": 0.3714359402656555,
|
907 |
+
"eval_runtime": 5.6334,
|
908 |
+
"eval_samples_per_second": 509.636,
|
909 |
+
"eval_steps_per_second": 15.976,
|
910 |
+
"step": 1212
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 6.04,
|
914 |
+
"grad_norm": 3.3733067512512207,
|
915 |
+
"learning_rate": 2.2002200220022003e-05,
|
916 |
+
"loss": 0.404,
|
917 |
+
"step": 1220
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 6.09,
|
921 |
+
"grad_norm": 3.479614734649658,
|
922 |
+
"learning_rate": 2.172717271727173e-05,
|
923 |
+
"loss": 0.4102,
|
924 |
+
"step": 1230
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 6.14,
|
928 |
+
"grad_norm": 6.636502742767334,
|
929 |
+
"learning_rate": 2.1452145214521452e-05,
|
930 |
+
"loss": 0.4261,
|
931 |
+
"step": 1240
|
932 |
+
},
|
933 |
+
{
|
934 |
+
"epoch": 6.19,
|
935 |
+
"grad_norm": 3.6467645168304443,
|
936 |
+
"learning_rate": 2.1177117711771178e-05,
|
937 |
+
"loss": 0.4061,
|
938 |
+
"step": 1250
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 6.24,
|
942 |
+
"grad_norm": 3.435811996459961,
|
943 |
+
"learning_rate": 2.09020902090209e-05,
|
944 |
+
"loss": 0.4198,
|
945 |
+
"step": 1260
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 6.29,
|
949 |
+
"grad_norm": 5.085671424865723,
|
950 |
+
"learning_rate": 2.0627062706270627e-05,
|
951 |
+
"loss": 0.3963,
|
952 |
+
"step": 1270
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 6.34,
|
956 |
+
"grad_norm": 9.153255462646484,
|
957 |
+
"learning_rate": 2.0352035203520354e-05,
|
958 |
+
"loss": 0.4239,
|
959 |
+
"step": 1280
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 6.39,
|
963 |
+
"grad_norm": 5.1853251457214355,
|
964 |
+
"learning_rate": 2.007700770077008e-05,
|
965 |
+
"loss": 0.3908,
|
966 |
+
"step": 1290
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 6.44,
|
970 |
+
"grad_norm": 7.626086235046387,
|
971 |
+
"learning_rate": 1.9801980198019803e-05,
|
972 |
+
"loss": 0.4132,
|
973 |
+
"step": 1300
|
974 |
+
},
|
975 |
+
{
|
976 |
+
"epoch": 6.49,
|
977 |
+
"grad_norm": 5.670557022094727,
|
978 |
+
"learning_rate": 1.952695269526953e-05,
|
979 |
+
"loss": 0.4228,
|
980 |
+
"step": 1310
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 6.53,
|
984 |
+
"grad_norm": 5.062951564788818,
|
985 |
+
"learning_rate": 1.9251925192519252e-05,
|
986 |
+
"loss": 0.4385,
|
987 |
+
"step": 1320
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 6.58,
|
991 |
+
"grad_norm": 3.8000316619873047,
|
992 |
+
"learning_rate": 1.8976897689768978e-05,
|
993 |
+
"loss": 0.414,
|
994 |
+
"step": 1330
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 6.63,
|
998 |
+
"grad_norm": 3.444866895675659,
|
999 |
+
"learning_rate": 1.87018701870187e-05,
|
1000 |
+
"loss": 0.4246,
|
1001 |
+
"step": 1340
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 6.68,
|
1005 |
+
"grad_norm": 3.809499740600586,
|
1006 |
+
"learning_rate": 1.8426842684268427e-05,
|
1007 |
+
"loss": 0.4198,
|
1008 |
+
"step": 1350
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 6.73,
|
1012 |
+
"grad_norm": 4.515871047973633,
|
1013 |
+
"learning_rate": 1.8151815181518153e-05,
|
1014 |
+
"loss": 0.4051,
|
1015 |
+
"step": 1360
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 6.78,
|
1019 |
+
"grad_norm": 4.38671350479126,
|
1020 |
+
"learning_rate": 1.787678767876788e-05,
|
1021 |
+
"loss": 0.4153,
|
1022 |
+
"step": 1370
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 6.83,
|
1026 |
+
"grad_norm": 3.238205909729004,
|
1027 |
+
"learning_rate": 1.7601760176017602e-05,
|
1028 |
+
"loss": 0.3947,
|
1029 |
+
"step": 1380
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 6.88,
|
1033 |
+
"grad_norm": 3.455275058746338,
|
1034 |
+
"learning_rate": 1.7326732673267325e-05,
|
1035 |
+
"loss": 0.4018,
|
1036 |
+
"step": 1390
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 6.93,
|
1040 |
+
"grad_norm": 3.587259531021118,
|
1041 |
+
"learning_rate": 1.705170517051705e-05,
|
1042 |
+
"loss": 0.3983,
|
1043 |
+
"step": 1400
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 6.98,
|
1047 |
+
"grad_norm": 4.972535610198975,
|
1048 |
+
"learning_rate": 1.6776677667766778e-05,
|
1049 |
+
"loss": 0.4117,
|
1050 |
+
"step": 1410
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 7.0,
|
1054 |
+
"eval_accuracy": 0.8376872169975619,
|
1055 |
+
"eval_loss": 0.36851075291633606,
|
1056 |
+
"eval_runtime": 5.6001,
|
1057 |
+
"eval_samples_per_second": 512.666,
|
1058 |
+
"eval_steps_per_second": 16.071,
|
1059 |
+
"step": 1414
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 7.03,
|
1063 |
+
"grad_norm": 4.680665493011475,
|
1064 |
+
"learning_rate": 1.6501650165016504e-05,
|
1065 |
+
"loss": 0.3791,
|
1066 |
+
"step": 1420
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 7.08,
|
1070 |
+
"grad_norm": 4.995221138000488,
|
1071 |
+
"learning_rate": 1.6226622662266227e-05,
|
1072 |
+
"loss": 0.4119,
|
1073 |
+
"step": 1430
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 7.13,
|
1077 |
+
"grad_norm": 3.969187021255493,
|
1078 |
+
"learning_rate": 1.5951595159515953e-05,
|
1079 |
+
"loss": 0.3864,
|
1080 |
+
"step": 1440
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 7.18,
|
1084 |
+
"grad_norm": 3.4930386543273926,
|
1085 |
+
"learning_rate": 1.567656765676568e-05,
|
1086 |
+
"loss": 0.3914,
|
1087 |
+
"step": 1450
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 7.23,
|
1091 |
+
"grad_norm": 4.228696823120117,
|
1092 |
+
"learning_rate": 1.5401540154015402e-05,
|
1093 |
+
"loss": 0.4072,
|
1094 |
+
"step": 1460
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 7.28,
|
1098 |
+
"grad_norm": 3.5527050495147705,
|
1099 |
+
"learning_rate": 1.5126512651265127e-05,
|
1100 |
+
"loss": 0.3864,
|
1101 |
+
"step": 1470
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 7.33,
|
1105 |
+
"grad_norm": 3.4274041652679443,
|
1106 |
+
"learning_rate": 1.4851485148514851e-05,
|
1107 |
+
"loss": 0.3957,
|
1108 |
+
"step": 1480
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 7.38,
|
1112 |
+
"grad_norm": 4.775713920593262,
|
1113 |
+
"learning_rate": 1.4576457645764577e-05,
|
1114 |
+
"loss": 0.4034,
|
1115 |
+
"step": 1490
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 7.43,
|
1119 |
+
"grad_norm": 4.741795539855957,
|
1120 |
+
"learning_rate": 1.4301430143014302e-05,
|
1121 |
+
"loss": 0.4042,
|
1122 |
+
"step": 1500
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 7.48,
|
1126 |
+
"grad_norm": 6.42737340927124,
|
1127 |
+
"learning_rate": 1.4026402640264028e-05,
|
1128 |
+
"loss": 0.4197,
|
1129 |
+
"step": 1510
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 7.52,
|
1133 |
+
"grad_norm": 5.1894121170043945,
|
1134 |
+
"learning_rate": 1.3751375137513753e-05,
|
1135 |
+
"loss": 0.3956,
|
1136 |
+
"step": 1520
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 7.57,
|
1140 |
+
"grad_norm": 5.648491859436035,
|
1141 |
+
"learning_rate": 1.3476347634763479e-05,
|
1142 |
+
"loss": 0.4005,
|
1143 |
+
"step": 1530
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 7.62,
|
1147 |
+
"grad_norm": 4.315338611602783,
|
1148 |
+
"learning_rate": 1.32013201320132e-05,
|
1149 |
+
"loss": 0.3784,
|
1150 |
+
"step": 1540
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 7.67,
|
1154 |
+
"grad_norm": 4.007127285003662,
|
1155 |
+
"learning_rate": 1.2926292629262926e-05,
|
1156 |
+
"loss": 0.3948,
|
1157 |
+
"step": 1550
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 7.72,
|
1161 |
+
"grad_norm": 4.685734748840332,
|
1162 |
+
"learning_rate": 1.265126512651265e-05,
|
1163 |
+
"loss": 0.4165,
|
1164 |
+
"step": 1560
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 7.77,
|
1168 |
+
"grad_norm": 5.51500940322876,
|
1169 |
+
"learning_rate": 1.2376237623762377e-05,
|
1170 |
+
"loss": 0.3872,
|
1171 |
+
"step": 1570
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 7.82,
|
1175 |
+
"grad_norm": 3.208813428878784,
|
1176 |
+
"learning_rate": 1.2101210121012102e-05,
|
1177 |
+
"loss": 0.3886,
|
1178 |
+
"step": 1580
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 7.87,
|
1182 |
+
"grad_norm": 4.27180814743042,
|
1183 |
+
"learning_rate": 1.1826182618261828e-05,
|
1184 |
+
"loss": 0.3837,
|
1185 |
+
"step": 1590
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 7.92,
|
1189 |
+
"grad_norm": 9.008807182312012,
|
1190 |
+
"learning_rate": 1.155115511551155e-05,
|
1191 |
+
"loss": 0.4174,
|
1192 |
+
"step": 1600
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 7.97,
|
1196 |
+
"grad_norm": 3.7245564460754395,
|
1197 |
+
"learning_rate": 1.1276127612761277e-05,
|
1198 |
+
"loss": 0.3991,
|
1199 |
+
"step": 1610
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 8.0,
|
1203 |
+
"eval_accuracy": 0.8411703239289446,
|
1204 |
+
"eval_loss": 0.3601776659488678,
|
1205 |
+
"eval_runtime": 5.6002,
|
1206 |
+
"eval_samples_per_second": 512.663,
|
1207 |
+
"eval_steps_per_second": 16.071,
|
1208 |
+
"step": 1616
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 8.02,
|
1212 |
+
"grad_norm": 4.604074954986572,
|
1213 |
+
"learning_rate": 1.1001100110011001e-05,
|
1214 |
+
"loss": 0.4121,
|
1215 |
+
"step": 1620
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 8.07,
|
1219 |
+
"grad_norm": 6.1971611976623535,
|
1220 |
+
"learning_rate": 1.0726072607260726e-05,
|
1221 |
+
"loss": 0.3925,
|
1222 |
+
"step": 1630
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 8.12,
|
1226 |
+
"grad_norm": 3.589857816696167,
|
1227 |
+
"learning_rate": 1.045104510451045e-05,
|
1228 |
+
"loss": 0.4339,
|
1229 |
+
"step": 1640
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 8.17,
|
1233 |
+
"grad_norm": 3.007169246673584,
|
1234 |
+
"learning_rate": 1.0176017601760177e-05,
|
1235 |
+
"loss": 0.3841,
|
1236 |
+
"step": 1650
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 8.22,
|
1240 |
+
"grad_norm": 8.007019996643066,
|
1241 |
+
"learning_rate": 9.900990099009901e-06,
|
1242 |
+
"loss": 0.4005,
|
1243 |
+
"step": 1660
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 8.27,
|
1247 |
+
"grad_norm": 4.514450550079346,
|
1248 |
+
"learning_rate": 9.625962596259626e-06,
|
1249 |
+
"loss": 0.4072,
|
1250 |
+
"step": 1670
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 8.32,
|
1254 |
+
"grad_norm": 6.048130989074707,
|
1255 |
+
"learning_rate": 9.35093509350935e-06,
|
1256 |
+
"loss": 0.3756,
|
1257 |
+
"step": 1680
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 8.37,
|
1261 |
+
"grad_norm": 6.211109638214111,
|
1262 |
+
"learning_rate": 9.075907590759077e-06,
|
1263 |
+
"loss": 0.3953,
|
1264 |
+
"step": 1690
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 8.42,
|
1268 |
+
"grad_norm": 3.7817318439483643,
|
1269 |
+
"learning_rate": 8.800880088008801e-06,
|
1270 |
+
"loss": 0.3685,
|
1271 |
+
"step": 1700
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 8.47,
|
1275 |
+
"grad_norm": 5.379549980163574,
|
1276 |
+
"learning_rate": 8.525852585258526e-06,
|
1277 |
+
"loss": 0.3713,
|
1278 |
+
"step": 1710
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 8.51,
|
1282 |
+
"grad_norm": 4.123336315155029,
|
1283 |
+
"learning_rate": 8.250825082508252e-06,
|
1284 |
+
"loss": 0.3629,
|
1285 |
+
"step": 1720
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 8.56,
|
1289 |
+
"grad_norm": 4.12966251373291,
|
1290 |
+
"learning_rate": 7.975797579757976e-06,
|
1291 |
+
"loss": 0.3884,
|
1292 |
+
"step": 1730
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 8.61,
|
1296 |
+
"grad_norm": 4.9768967628479,
|
1297 |
+
"learning_rate": 7.700770077007701e-06,
|
1298 |
+
"loss": 0.4092,
|
1299 |
+
"step": 1740
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 8.66,
|
1303 |
+
"grad_norm": 5.166417598724365,
|
1304 |
+
"learning_rate": 7.4257425742574256e-06,
|
1305 |
+
"loss": 0.3891,
|
1306 |
+
"step": 1750
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 8.71,
|
1310 |
+
"grad_norm": 6.434201240539551,
|
1311 |
+
"learning_rate": 7.150715071507151e-06,
|
1312 |
+
"loss": 0.3592,
|
1313 |
+
"step": 1760
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 8.76,
|
1317 |
+
"grad_norm": 7.199638843536377,
|
1318 |
+
"learning_rate": 6.875687568756876e-06,
|
1319 |
+
"loss": 0.3909,
|
1320 |
+
"step": 1770
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 8.81,
|
1324 |
+
"grad_norm": 7.480076313018799,
|
1325 |
+
"learning_rate": 6.6006600660066e-06,
|
1326 |
+
"loss": 0.3895,
|
1327 |
+
"step": 1780
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 8.86,
|
1331 |
+
"grad_norm": 5.124411582946777,
|
1332 |
+
"learning_rate": 6.325632563256325e-06,
|
1333 |
+
"loss": 0.391,
|
1334 |
+
"step": 1790
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 8.91,
|
1338 |
+
"grad_norm": 9.751836776733398,
|
1339 |
+
"learning_rate": 6.050605060506051e-06,
|
1340 |
+
"loss": 0.3908,
|
1341 |
+
"step": 1800
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 8.96,
|
1345 |
+
"grad_norm": 4.501916408538818,
|
1346 |
+
"learning_rate": 5.775577557755775e-06,
|
1347 |
+
"loss": 0.3936,
|
1348 |
+
"step": 1810
|
1349 |
+
},
|
1350 |
+
{
|
1351 |
+
"epoch": 9.0,
|
1352 |
+
"eval_accuracy": 0.842911877394636,
|
1353 |
+
"eval_loss": 0.3542243540287018,
|
1354 |
+
"eval_runtime": 5.623,
|
1355 |
+
"eval_samples_per_second": 510.58,
|
1356 |
+
"eval_steps_per_second": 16.006,
|
1357 |
+
"step": 1818
|
1358 |
+
},
|
1359 |
+
{
|
1360 |
+
"epoch": 9.01,
|
1361 |
+
"grad_norm": 4.195034027099609,
|
1362 |
+
"learning_rate": 5.500550055005501e-06,
|
1363 |
+
"loss": 0.4147,
|
1364 |
+
"step": 1820
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 9.06,
|
1368 |
+
"grad_norm": 4.904818058013916,
|
1369 |
+
"learning_rate": 5.225522552255225e-06,
|
1370 |
+
"loss": 0.3822,
|
1371 |
+
"step": 1830
|
1372 |
+
},
|
1373 |
+
{
|
1374 |
+
"epoch": 9.11,
|
1375 |
+
"grad_norm": 4.105266094207764,
|
1376 |
+
"learning_rate": 4.950495049504951e-06,
|
1377 |
+
"loss": 0.401,
|
1378 |
+
"step": 1840
|
1379 |
+
},
|
1380 |
+
{
|
1381 |
+
"epoch": 9.16,
|
1382 |
+
"grad_norm": 4.364089488983154,
|
1383 |
+
"learning_rate": 4.675467546754675e-06,
|
1384 |
+
"loss": 0.398,
|
1385 |
+
"step": 1850
|
1386 |
+
},
|
1387 |
+
{
|
1388 |
+
"epoch": 9.21,
|
1389 |
+
"grad_norm": 5.043737411499023,
|
1390 |
+
"learning_rate": 4.400440044004401e-06,
|
1391 |
+
"loss": 0.4218,
|
1392 |
+
"step": 1860
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 9.26,
|
1396 |
+
"grad_norm": 4.617894172668457,
|
1397 |
+
"learning_rate": 4.125412541254126e-06,
|
1398 |
+
"loss": 0.4002,
|
1399 |
+
"step": 1870
|
1400 |
+
},
|
1401 |
+
{
|
1402 |
+
"epoch": 9.31,
|
1403 |
+
"grad_norm": 3.750427007675171,
|
1404 |
+
"learning_rate": 3.8503850385038505e-06,
|
1405 |
+
"loss": 0.4005,
|
1406 |
+
"step": 1880
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 9.36,
|
1410 |
+
"grad_norm": 4.708156585693359,
|
1411 |
+
"learning_rate": 3.5753575357535755e-06,
|
1412 |
+
"loss": 0.4067,
|
1413 |
+
"step": 1890
|
1414 |
+
},
|
1415 |
+
{
|
1416 |
+
"epoch": 9.41,
|
1417 |
+
"grad_norm": 4.125495910644531,
|
1418 |
+
"learning_rate": 3.3003300330033e-06,
|
1419 |
+
"loss": 0.3745,
|
1420 |
+
"step": 1900
|
1421 |
+
},
|
1422 |
+
{
|
1423 |
+
"epoch": 9.46,
|
1424 |
+
"grad_norm": 3.2616422176361084,
|
1425 |
+
"learning_rate": 3.0253025302530254e-06,
|
1426 |
+
"loss": 0.3914,
|
1427 |
+
"step": 1910
|
1428 |
+
},
|
1429 |
+
{
|
1430 |
+
"epoch": 9.5,
|
1431 |
+
"grad_norm": 3.8317136764526367,
|
1432 |
+
"learning_rate": 2.7502750275027504e-06,
|
1433 |
+
"loss": 0.383,
|
1434 |
+
"step": 1920
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 9.55,
|
1438 |
+
"grad_norm": 4.044767379760742,
|
1439 |
+
"learning_rate": 2.4752475247524753e-06,
|
1440 |
+
"loss": 0.3713,
|
1441 |
+
"step": 1930
|
1442 |
+
},
|
1443 |
+
{
|
1444 |
+
"epoch": 9.6,
|
1445 |
+
"grad_norm": 5.957603931427002,
|
1446 |
+
"learning_rate": 2.2002200220022003e-06,
|
1447 |
+
"loss": 0.3524,
|
1448 |
+
"step": 1940
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 9.65,
|
1452 |
+
"grad_norm": 3.8878390789031982,
|
1453 |
+
"learning_rate": 1.9251925192519253e-06,
|
1454 |
+
"loss": 0.3876,
|
1455 |
+
"step": 1950
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 9.7,
|
1459 |
+
"grad_norm": 4.049957752227783,
|
1460 |
+
"learning_rate": 1.65016501650165e-06,
|
1461 |
+
"loss": 0.3726,
|
1462 |
+
"step": 1960
|
1463 |
+
},
|
1464 |
+
{
|
1465 |
+
"epoch": 9.75,
|
1466 |
+
"grad_norm": 3.9568910598754883,
|
1467 |
+
"learning_rate": 1.3751375137513752e-06,
|
1468 |
+
"loss": 0.397,
|
1469 |
+
"step": 1970
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 9.8,
|
1473 |
+
"grad_norm": 5.049424171447754,
|
1474 |
+
"learning_rate": 1.1001100110011001e-06,
|
1475 |
+
"loss": 0.3771,
|
1476 |
+
"step": 1980
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 9.85,
|
1480 |
+
"grad_norm": 3.9506354331970215,
|
1481 |
+
"learning_rate": 8.25082508250825e-07,
|
1482 |
+
"loss": 0.3816,
|
1483 |
+
"step": 1990
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 9.9,
|
1487 |
+
"grad_norm": 4.543933391571045,
|
1488 |
+
"learning_rate": 5.500550055005501e-07,
|
1489 |
+
"loss": 0.3961,
|
1490 |
+
"step": 2000
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 9.95,
|
1494 |
+
"grad_norm": 4.267475605010986,
|
1495 |
+
"learning_rate": 2.7502750275027504e-07,
|
1496 |
+
"loss": 0.3612,
|
1497 |
+
"step": 2010
|
1498 |
+
},
|
1499 |
+
{
|
1500 |
+
"epoch": 10.0,
|
1501 |
+
"grad_norm": 7.206649303436279,
|
1502 |
+
"learning_rate": 0.0,
|
1503 |
+
"loss": 0.3422,
|
1504 |
+
"step": 2020
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 10.0,
|
1508 |
+
"eval_accuracy": 0.8397770811563915,
|
1509 |
+
"eval_loss": 0.35397040843963623,
|
1510 |
+
"eval_runtime": 5.6519,
|
1511 |
+
"eval_samples_per_second": 507.973,
|
1512 |
+
"eval_steps_per_second": 15.924,
|
1513 |
+
"step": 2020
|
1514 |
+
},
|
1515 |
+
{
|
1516 |
+
"epoch": 10.0,
|
1517 |
+
"step": 2020,
|
1518 |
+
"total_flos": 1.2892209231838003e+18,
|
1519 |
+
"train_loss": 0.45288851414576614,
|
1520 |
+
"train_runtime": 810.9556,
|
1521 |
+
"train_samples_per_second": 318.599,
|
1522 |
+
"train_steps_per_second": 2.491
|
1523 |
+
}
|
1524 |
+
],
|
1525 |
+
"logging_steps": 10,
|
1526 |
+
"max_steps": 2020,
|
1527 |
+
"num_input_tokens_seen": 0,
|
1528 |
+
"num_train_epochs": 10,
|
1529 |
+
"save_steps": 500,
|
1530 |
+
"total_flos": 1.2892209231838003e+18,
|
1531 |
+
"train_batch_size": 32,
|
1532 |
+
"trial_name": null,
|
1533 |
+
"trial_params": null
|
1534 |
+
}
|