End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Jan15_16-58-11_c583982b4f3d/events.out.tfevents.1736962575.c583982b4f3d.215.11 +3 -0
- train_results.json +8 -0
- trainer_state.json +1299 -0
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [facebook/vit-msn-small](https://huggingface.co/facebook/vit-msn-small) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss:
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.9602224123182207
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [facebook/vit-msn-small](https://huggingface.co/facebook/vit-msn-small) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2917
|
| 37 |
+
- Accuracy: 0.9602
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"eval_accuracy": 0.9602224123182207,
|
| 4 |
+
"eval_loss": 0.29173824191093445,
|
| 5 |
+
"eval_runtime": 10.0668,
|
| 6 |
+
"eval_samples_per_second": 232.248,
|
| 7 |
+
"eval_steps_per_second": 3.675,
|
| 8 |
+
"total_flos": 2.9138957540265e+18,
|
| 9 |
+
"train_loss": 0.273075803120931,
|
| 10 |
+
"train_runtime": 2186.3924,
|
| 11 |
+
"train_samples_per_second": 73.774,
|
| 12 |
+
"train_steps_per_second": 0.274
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"eval_accuracy": 0.9602224123182207,
|
| 4 |
+
"eval_loss": 0.29173824191093445,
|
| 5 |
+
"eval_runtime": 10.0668,
|
| 6 |
+
"eval_samples_per_second": 232.248,
|
| 7 |
+
"eval_steps_per_second": 3.675
|
| 8 |
+
}
|
runs/Jan15_16-58-11_c583982b4f3d/events.out.tfevents.1736962575.c583982b4f3d.215.11
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:997472b4d7e0357c2639cf0764ce01c5c6c9abdccd787c95a43fd12787bd1be8
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"total_flos": 2.9138957540265e+18,
|
| 4 |
+
"train_loss": 0.273075803120931,
|
| 5 |
+
"train_runtime": 2186.3924,
|
| 6 |
+
"train_samples_per_second": 73.774,
|
| 7 |
+
"train_steps_per_second": 0.274
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1299 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9602224123182207,
|
| 3 |
+
"best_model_checkpoint": "vit-msn-small-lateral_flow_ivalidation_green/checkpoint-32",
|
| 4 |
+
"epoch": 92.3076923076923,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9230769230769231,
|
| 13 |
+
"eval_accuracy": 0.9238665526090676,
|
| 14 |
+
"eval_loss": 0.5034469366073608,
|
| 15 |
+
"eval_runtime": 9.6118,
|
| 16 |
+
"eval_samples_per_second": 243.243,
|
| 17 |
+
"eval_steps_per_second": 3.849,
|
| 18 |
+
"step": 6
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.5384615384615383,
|
| 22 |
+
"grad_norm": 13.861084938049316,
|
| 23 |
+
"learning_rate": 8.333333333333334e-06,
|
| 24 |
+
"loss": 0.704,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_accuracy": 0.9409751924721984,
|
| 30 |
+
"eval_loss": 0.45238828659057617,
|
| 31 |
+
"eval_runtime": 9.8069,
|
| 32 |
+
"eval_samples_per_second": 238.404,
|
| 33 |
+
"eval_steps_per_second": 3.773,
|
| 34 |
+
"step": 13
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.9230769230769234,
|
| 38 |
+
"eval_accuracy": 0.9593669803250642,
|
| 39 |
+
"eval_loss": 0.26334497332572937,
|
| 40 |
+
"eval_runtime": 9.5224,
|
| 41 |
+
"eval_samples_per_second": 245.527,
|
| 42 |
+
"eval_steps_per_second": 3.886,
|
| 43 |
+
"step": 19
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.076923076923077,
|
| 47 |
+
"grad_norm": 5.300178527832031,
|
| 48 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 49 |
+
"loss": 0.505,
|
| 50 |
+
"step": 20
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.8092386655260907,
|
| 55 |
+
"eval_loss": 0.4748455882072449,
|
| 56 |
+
"eval_runtime": 9.8908,
|
| 57 |
+
"eval_samples_per_second": 236.381,
|
| 58 |
+
"eval_steps_per_second": 3.741,
|
| 59 |
+
"step": 26
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.615384615384615,
|
| 63 |
+
"grad_norm": 14.291367530822754,
|
| 64 |
+
"learning_rate": 2.5e-05,
|
| 65 |
+
"loss": 0.4456,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.923076923076923,
|
| 70 |
+
"eval_accuracy": 0.9602224123182207,
|
| 71 |
+
"eval_loss": 0.29173824191093445,
|
| 72 |
+
"eval_runtime": 10.0407,
|
| 73 |
+
"eval_samples_per_second": 232.852,
|
| 74 |
+
"eval_steps_per_second": 3.685,
|
| 75 |
+
"step": 32
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 6.0,
|
| 79 |
+
"eval_accuracy": 0.9221556886227545,
|
| 80 |
+
"eval_loss": 0.2621428668498993,
|
| 81 |
+
"eval_runtime": 9.9631,
|
| 82 |
+
"eval_samples_per_second": 234.666,
|
| 83 |
+
"eval_steps_per_second": 3.714,
|
| 84 |
+
"step": 39
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.153846153846154,
|
| 88 |
+
"grad_norm": 8.197896957397461,
|
| 89 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 90 |
+
"loss": 0.3908,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.923076923076923,
|
| 95 |
+
"eval_accuracy": 0.8190761334473909,
|
| 96 |
+
"eval_loss": 0.4519110918045044,
|
| 97 |
+
"eval_runtime": 9.949,
|
| 98 |
+
"eval_samples_per_second": 234.998,
|
| 99 |
+
"eval_steps_per_second": 3.719,
|
| 100 |
+
"step": 45
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 7.6923076923076925,
|
| 104 |
+
"grad_norm": 7.590042591094971,
|
| 105 |
+
"learning_rate": 4.166666666666667e-05,
|
| 106 |
+
"loss": 0.3628,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 8.0,
|
| 111 |
+
"eval_accuracy": 0.8622754491017964,
|
| 112 |
+
"eval_loss": 0.4092731475830078,
|
| 113 |
+
"eval_runtime": 10.096,
|
| 114 |
+
"eval_samples_per_second": 231.578,
|
| 115 |
+
"eval_steps_per_second": 3.665,
|
| 116 |
+
"step": 52
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 8.923076923076923,
|
| 120 |
+
"eval_accuracy": 0.935414884516681,
|
| 121 |
+
"eval_loss": 0.2705248296260834,
|
| 122 |
+
"eval_runtime": 10.1632,
|
| 123 |
+
"eval_samples_per_second": 230.046,
|
| 124 |
+
"eval_steps_per_second": 3.641,
|
| 125 |
+
"step": 58
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 9.23076923076923,
|
| 129 |
+
"grad_norm": 10.925556182861328,
|
| 130 |
+
"learning_rate": 5e-05,
|
| 131 |
+
"loss": 0.372,
|
| 132 |
+
"step": 60
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 10.0,
|
| 136 |
+
"eval_accuracy": 0.8545765611633875,
|
| 137 |
+
"eval_loss": 0.41367459297180176,
|
| 138 |
+
"eval_runtime": 10.0011,
|
| 139 |
+
"eval_samples_per_second": 233.774,
|
| 140 |
+
"eval_steps_per_second": 3.7,
|
| 141 |
+
"step": 65
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.76923076923077,
|
| 145 |
+
"grad_norm": 8.855840682983398,
|
| 146 |
+
"learning_rate": 4.9074074074074075e-05,
|
| 147 |
+
"loss": 0.36,
|
| 148 |
+
"step": 70
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.923076923076923,
|
| 152 |
+
"eval_accuracy": 0.8815226689478186,
|
| 153 |
+
"eval_loss": 0.34931161999702454,
|
| 154 |
+
"eval_runtime": 9.9523,
|
| 155 |
+
"eval_samples_per_second": 234.921,
|
| 156 |
+
"eval_steps_per_second": 3.718,
|
| 157 |
+
"step": 71
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 12.0,
|
| 161 |
+
"eval_accuracy": 0.9456800684345594,
|
| 162 |
+
"eval_loss": 0.21901701390743256,
|
| 163 |
+
"eval_runtime": 9.6913,
|
| 164 |
+
"eval_samples_per_second": 241.248,
|
| 165 |
+
"eval_steps_per_second": 3.818,
|
| 166 |
+
"step": 78
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.307692307692308,
|
| 170 |
+
"grad_norm": 7.141057014465332,
|
| 171 |
+
"learning_rate": 4.814814814814815e-05,
|
| 172 |
+
"loss": 0.36,
|
| 173 |
+
"step": 80
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 12.923076923076923,
|
| 177 |
+
"eval_accuracy": 0.9033361847733106,
|
| 178 |
+
"eval_loss": 0.3190420866012573,
|
| 179 |
+
"eval_runtime": 9.7798,
|
| 180 |
+
"eval_samples_per_second": 239.064,
|
| 181 |
+
"eval_steps_per_second": 3.783,
|
| 182 |
+
"step": 84
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.846153846153847,
|
| 186 |
+
"grad_norm": 11.677677154541016,
|
| 187 |
+
"learning_rate": 4.722222222222222e-05,
|
| 188 |
+
"loss": 0.3363,
|
| 189 |
+
"step": 90
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 14.0,
|
| 193 |
+
"eval_accuracy": 0.894781864841745,
|
| 194 |
+
"eval_loss": 0.337951123714447,
|
| 195 |
+
"eval_runtime": 9.7204,
|
| 196 |
+
"eval_samples_per_second": 240.525,
|
| 197 |
+
"eval_steps_per_second": 3.806,
|
| 198 |
+
"step": 91
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 14.923076923076923,
|
| 202 |
+
"eval_accuracy": 0.8982035928143712,
|
| 203 |
+
"eval_loss": 0.3342379331588745,
|
| 204 |
+
"eval_runtime": 9.9394,
|
| 205 |
+
"eval_samples_per_second": 235.226,
|
| 206 |
+
"eval_steps_per_second": 3.723,
|
| 207 |
+
"step": 97
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.384615384615385,
|
| 211 |
+
"grad_norm": 3.7460684776306152,
|
| 212 |
+
"learning_rate": 4.62962962962963e-05,
|
| 213 |
+
"loss": 0.327,
|
| 214 |
+
"step": 100
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.0,
|
| 218 |
+
"eval_accuracy": 0.8327630453378957,
|
| 219 |
+
"eval_loss": 0.4211990237236023,
|
| 220 |
+
"eval_runtime": 9.8418,
|
| 221 |
+
"eval_samples_per_second": 237.559,
|
| 222 |
+
"eval_steps_per_second": 3.759,
|
| 223 |
+
"step": 104
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.923076923076923,
|
| 227 |
+
"grad_norm": 12.34093952178955,
|
| 228 |
+
"learning_rate": 4.5370370370370374e-05,
|
| 229 |
+
"loss": 0.3257,
|
| 230 |
+
"step": 110
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 16.923076923076923,
|
| 234 |
+
"eval_accuracy": 0.7844311377245509,
|
| 235 |
+
"eval_loss": 0.5167170763015747,
|
| 236 |
+
"eval_runtime": 9.8234,
|
| 237 |
+
"eval_samples_per_second": 238.004,
|
| 238 |
+
"eval_steps_per_second": 3.767,
|
| 239 |
+
"step": 110
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 18.0,
|
| 243 |
+
"eval_accuracy": 0.7275449101796407,
|
| 244 |
+
"eval_loss": 0.5847879648208618,
|
| 245 |
+
"eval_runtime": 9.927,
|
| 246 |
+
"eval_samples_per_second": 235.518,
|
| 247 |
+
"eval_steps_per_second": 3.727,
|
| 248 |
+
"step": 117
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.46153846153846,
|
| 252 |
+
"grad_norm": 5.46023416519165,
|
| 253 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 254 |
+
"loss": 0.3175,
|
| 255 |
+
"step": 120
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 18.923076923076923,
|
| 259 |
+
"eval_accuracy": 0.8336184773310522,
|
| 260 |
+
"eval_loss": 0.4090871810913086,
|
| 261 |
+
"eval_runtime": 10.1427,
|
| 262 |
+
"eval_samples_per_second": 230.511,
|
| 263 |
+
"eval_steps_per_second": 3.648,
|
| 264 |
+
"step": 123
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 20.0,
|
| 268 |
+
"grad_norm": 9.724111557006836,
|
| 269 |
+
"learning_rate": 4.351851851851852e-05,
|
| 270 |
+
"loss": 0.3377,
|
| 271 |
+
"step": 130
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 20.0,
|
| 275 |
+
"eval_accuracy": 0.9161676646706587,
|
| 276 |
+
"eval_loss": 0.28380292654037476,
|
| 277 |
+
"eval_runtime": 10.1341,
|
| 278 |
+
"eval_samples_per_second": 230.706,
|
| 279 |
+
"eval_steps_per_second": 3.651,
|
| 280 |
+
"step": 130
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.923076923076923,
|
| 284 |
+
"eval_accuracy": 0.7262617621899059,
|
| 285 |
+
"eval_loss": 0.6106137633323669,
|
| 286 |
+
"eval_runtime": 9.9506,
|
| 287 |
+
"eval_samples_per_second": 234.962,
|
| 288 |
+
"eval_steps_per_second": 3.718,
|
| 289 |
+
"step": 136
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 21.53846153846154,
|
| 293 |
+
"grad_norm": 5.13586950302124,
|
| 294 |
+
"learning_rate": 4.259259259259259e-05,
|
| 295 |
+
"loss": 0.3129,
|
| 296 |
+
"step": 140
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 22.0,
|
| 300 |
+
"eval_accuracy": 0.7164242942686057,
|
| 301 |
+
"eval_loss": 0.6294828057289124,
|
| 302 |
+
"eval_runtime": 9.9909,
|
| 303 |
+
"eval_samples_per_second": 234.013,
|
| 304 |
+
"eval_steps_per_second": 3.703,
|
| 305 |
+
"step": 143
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 22.923076923076923,
|
| 309 |
+
"eval_accuracy": 0.5932420872540634,
|
| 310 |
+
"eval_loss": 0.7897723913192749,
|
| 311 |
+
"eval_runtime": 9.9526,
|
| 312 |
+
"eval_samples_per_second": 234.912,
|
| 313 |
+
"eval_steps_per_second": 3.718,
|
| 314 |
+
"step": 149
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 23.076923076923077,
|
| 318 |
+
"grad_norm": 15.414055824279785,
|
| 319 |
+
"learning_rate": 4.166666666666667e-05,
|
| 320 |
+
"loss": 0.3138,
|
| 321 |
+
"step": 150
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 24.0,
|
| 325 |
+
"eval_accuracy": 0.4846022241231822,
|
| 326 |
+
"eval_loss": 0.9407968521118164,
|
| 327 |
+
"eval_runtime": 9.7597,
|
| 328 |
+
"eval_samples_per_second": 239.556,
|
| 329 |
+
"eval_steps_per_second": 3.791,
|
| 330 |
+
"step": 156
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.615384615384617,
|
| 334 |
+
"grad_norm": 3.9247374534606934,
|
| 335 |
+
"learning_rate": 4.074074074074074e-05,
|
| 336 |
+
"loss": 0.3106,
|
| 337 |
+
"step": 160
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 24.923076923076923,
|
| 341 |
+
"eval_accuracy": 0.8832335329341318,
|
| 342 |
+
"eval_loss": 0.34852102398872375,
|
| 343 |
+
"eval_runtime": 9.8091,
|
| 344 |
+
"eval_samples_per_second": 238.35,
|
| 345 |
+
"eval_steps_per_second": 3.772,
|
| 346 |
+
"step": 162
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 26.0,
|
| 350 |
+
"eval_accuracy": 0.7865697177074422,
|
| 351 |
+
"eval_loss": 0.5201271176338196,
|
| 352 |
+
"eval_runtime": 9.9386,
|
| 353 |
+
"eval_samples_per_second": 235.245,
|
| 354 |
+
"eval_steps_per_second": 3.723,
|
| 355 |
+
"step": 169
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 26.153846153846153,
|
| 359 |
+
"grad_norm": 5.763908386230469,
|
| 360 |
+
"learning_rate": 3.981481481481482e-05,
|
| 361 |
+
"loss": 0.3157,
|
| 362 |
+
"step": 170
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 26.923076923076923,
|
| 366 |
+
"eval_accuracy": 0.6672369546621043,
|
| 367 |
+
"eval_loss": 0.72103351354599,
|
| 368 |
+
"eval_runtime": 9.9351,
|
| 369 |
+
"eval_samples_per_second": 235.327,
|
| 370 |
+
"eval_steps_per_second": 3.724,
|
| 371 |
+
"step": 175
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 27.692307692307693,
|
| 375 |
+
"grad_norm": 5.021239757537842,
|
| 376 |
+
"learning_rate": 3.888888888888889e-05,
|
| 377 |
+
"loss": 0.2896,
|
| 378 |
+
"step": 180
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 28.0,
|
| 382 |
+
"eval_accuracy": 0.6330196749358425,
|
| 383 |
+
"eval_loss": 0.7980794906616211,
|
| 384 |
+
"eval_runtime": 10.2592,
|
| 385 |
+
"eval_samples_per_second": 227.892,
|
| 386 |
+
"eval_steps_per_second": 3.607,
|
| 387 |
+
"step": 182
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.923076923076923,
|
| 391 |
+
"eval_accuracy": 0.6428571428571429,
|
| 392 |
+
"eval_loss": 0.7667437791824341,
|
| 393 |
+
"eval_runtime": 9.9249,
|
| 394 |
+
"eval_samples_per_second": 235.57,
|
| 395 |
+
"eval_steps_per_second": 3.728,
|
| 396 |
+
"step": 188
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 29.23076923076923,
|
| 400 |
+
"grad_norm": 8.59469985961914,
|
| 401 |
+
"learning_rate": 3.7962962962962964e-05,
|
| 402 |
+
"loss": 0.2867,
|
| 403 |
+
"step": 190
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 30.0,
|
| 407 |
+
"eval_accuracy": 0.6544054747647562,
|
| 408 |
+
"eval_loss": 0.7686835527420044,
|
| 409 |
+
"eval_runtime": 10.0746,
|
| 410 |
+
"eval_samples_per_second": 232.069,
|
| 411 |
+
"eval_steps_per_second": 3.673,
|
| 412 |
+
"step": 195
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 30.76923076923077,
|
| 416 |
+
"grad_norm": 5.601044654846191,
|
| 417 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 418 |
+
"loss": 0.2786,
|
| 419 |
+
"step": 200
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 30.923076923076923,
|
| 423 |
+
"eval_accuracy": 0.5209580838323353,
|
| 424 |
+
"eval_loss": 1.1714286804199219,
|
| 425 |
+
"eval_runtime": 9.9004,
|
| 426 |
+
"eval_samples_per_second": 236.151,
|
| 427 |
+
"eval_steps_per_second": 3.737,
|
| 428 |
+
"step": 201
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 32.0,
|
| 432 |
+
"eval_accuracy": 0.42728828058169377,
|
| 433 |
+
"eval_loss": 1.1744341850280762,
|
| 434 |
+
"eval_runtime": 9.8292,
|
| 435 |
+
"eval_samples_per_second": 237.862,
|
| 436 |
+
"eval_steps_per_second": 3.764,
|
| 437 |
+
"step": 208
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 32.30769230769231,
|
| 441 |
+
"grad_norm": 4.0939507484436035,
|
| 442 |
+
"learning_rate": 3.611111111111111e-05,
|
| 443 |
+
"loss": 0.2823,
|
| 444 |
+
"step": 210
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 32.92307692307692,
|
| 448 |
+
"eval_accuracy": 0.5444824636441403,
|
| 449 |
+
"eval_loss": 0.9260274767875671,
|
| 450 |
+
"eval_runtime": 9.8098,
|
| 451 |
+
"eval_samples_per_second": 238.334,
|
| 452 |
+
"eval_steps_per_second": 3.772,
|
| 453 |
+
"step": 214
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.84615384615385,
|
| 457 |
+
"grad_norm": 5.409052848815918,
|
| 458 |
+
"learning_rate": 3.518518518518519e-05,
|
| 459 |
+
"loss": 0.2864,
|
| 460 |
+
"step": 220
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 34.0,
|
| 464 |
+
"eval_accuracy": 0.6920444824636441,
|
| 465 |
+
"eval_loss": 0.7139692902565002,
|
| 466 |
+
"eval_runtime": 9.9272,
|
| 467 |
+
"eval_samples_per_second": 235.514,
|
| 468 |
+
"eval_steps_per_second": 3.727,
|
| 469 |
+
"step": 221
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 34.92307692307692,
|
| 473 |
+
"eval_accuracy": 0.7331052181351583,
|
| 474 |
+
"eval_loss": 0.6098384857177734,
|
| 475 |
+
"eval_runtime": 9.8226,
|
| 476 |
+
"eval_samples_per_second": 238.024,
|
| 477 |
+
"eval_steps_per_second": 3.767,
|
| 478 |
+
"step": 227
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 35.38461538461539,
|
| 482 |
+
"grad_norm": 3.7521326541900635,
|
| 483 |
+
"learning_rate": 3.425925925925926e-05,
|
| 484 |
+
"loss": 0.2707,
|
| 485 |
+
"step": 230
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 36.0,
|
| 489 |
+
"eval_accuracy": 0.6783575705731394,
|
| 490 |
+
"eval_loss": 0.6992803812026978,
|
| 491 |
+
"eval_runtime": 9.8614,
|
| 492 |
+
"eval_samples_per_second": 237.087,
|
| 493 |
+
"eval_steps_per_second": 3.752,
|
| 494 |
+
"step": 234
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.92307692307692,
|
| 498 |
+
"grad_norm": 5.1208319664001465,
|
| 499 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 500 |
+
"loss": 0.2921,
|
| 501 |
+
"step": 240
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.92307692307692,
|
| 505 |
+
"eval_accuracy": 0.6176218990590248,
|
| 506 |
+
"eval_loss": 0.8719092607498169,
|
| 507 |
+
"eval_runtime": 10.0768,
|
| 508 |
+
"eval_samples_per_second": 232.019,
|
| 509 |
+
"eval_steps_per_second": 3.672,
|
| 510 |
+
"step": 240
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 38.0,
|
| 514 |
+
"eval_accuracy": 0.6060735671514115,
|
| 515 |
+
"eval_loss": 0.8336823582649231,
|
| 516 |
+
"eval_runtime": 9.8195,
|
| 517 |
+
"eval_samples_per_second": 238.098,
|
| 518 |
+
"eval_steps_per_second": 3.768,
|
| 519 |
+
"step": 247
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 38.46153846153846,
|
| 523 |
+
"grad_norm": 6.667710304260254,
|
| 524 |
+
"learning_rate": 3.240740740740741e-05,
|
| 525 |
+
"loss": 0.2849,
|
| 526 |
+
"step": 250
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 38.92307692307692,
|
| 530 |
+
"eval_accuracy": 0.825491873396065,
|
| 531 |
+
"eval_loss": 0.4395623505115509,
|
| 532 |
+
"eval_runtime": 9.6605,
|
| 533 |
+
"eval_samples_per_second": 242.015,
|
| 534 |
+
"eval_steps_per_second": 3.83,
|
| 535 |
+
"step": 253
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 40.0,
|
| 539 |
+
"grad_norm": 5.09765625,
|
| 540 |
+
"learning_rate": 3.148148148148148e-05,
|
| 541 |
+
"loss": 0.2657,
|
| 542 |
+
"step": 260
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 40.0,
|
| 546 |
+
"eval_accuracy": 0.501710863986313,
|
| 547 |
+
"eval_loss": 1.0981603860855103,
|
| 548 |
+
"eval_runtime": 9.9306,
|
| 549 |
+
"eval_samples_per_second": 235.433,
|
| 550 |
+
"eval_steps_per_second": 3.726,
|
| 551 |
+
"step": 260
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 40.92307692307692,
|
| 555 |
+
"eval_accuracy": 0.5175363558597091,
|
| 556 |
+
"eval_loss": 1.093379259109497,
|
| 557 |
+
"eval_runtime": 9.7012,
|
| 558 |
+
"eval_samples_per_second": 241.001,
|
| 559 |
+
"eval_steps_per_second": 3.814,
|
| 560 |
+
"step": 266
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 41.53846153846154,
|
| 564 |
+
"grad_norm": 3.8766767978668213,
|
| 565 |
+
"learning_rate": 3.055555555555556e-05,
|
| 566 |
+
"loss": 0.2659,
|
| 567 |
+
"step": 270
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 42.0,
|
| 571 |
+
"eval_accuracy": 0.636869118905047,
|
| 572 |
+
"eval_loss": 0.8629169464111328,
|
| 573 |
+
"eval_runtime": 9.9076,
|
| 574 |
+
"eval_samples_per_second": 235.98,
|
| 575 |
+
"eval_steps_per_second": 3.734,
|
| 576 |
+
"step": 273
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 42.92307692307692,
|
| 580 |
+
"eval_accuracy": 0.41402908468776733,
|
| 581 |
+
"eval_loss": 1.4602264165878296,
|
| 582 |
+
"eval_runtime": 9.7024,
|
| 583 |
+
"eval_samples_per_second": 240.972,
|
| 584 |
+
"eval_steps_per_second": 3.813,
|
| 585 |
+
"step": 279
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 43.07692307692308,
|
| 589 |
+
"grad_norm": 7.324178695678711,
|
| 590 |
+
"learning_rate": 2.962962962962963e-05,
|
| 591 |
+
"loss": 0.2645,
|
| 592 |
+
"step": 280
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 44.0,
|
| 596 |
+
"eval_accuracy": 0.3421727972626176,
|
| 597 |
+
"eval_loss": 1.9095213413238525,
|
| 598 |
+
"eval_runtime": 10.1083,
|
| 599 |
+
"eval_samples_per_second": 231.294,
|
| 600 |
+
"eval_steps_per_second": 3.66,
|
| 601 |
+
"step": 286
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 44.61538461538461,
|
| 605 |
+
"grad_norm": 3.6502673625946045,
|
| 606 |
+
"learning_rate": 2.8703703703703706e-05,
|
| 607 |
+
"loss": 0.2424,
|
| 608 |
+
"step": 290
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 44.92307692307692,
|
| 612 |
+
"eval_accuracy": 0.43969204448246363,
|
| 613 |
+
"eval_loss": 1.2180449962615967,
|
| 614 |
+
"eval_runtime": 9.8511,
|
| 615 |
+
"eval_samples_per_second": 237.334,
|
| 616 |
+
"eval_steps_per_second": 3.756,
|
| 617 |
+
"step": 292
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 46.0,
|
| 621 |
+
"eval_accuracy": 0.6424294268605646,
|
| 622 |
+
"eval_loss": 0.7686424255371094,
|
| 623 |
+
"eval_runtime": 10.0098,
|
| 624 |
+
"eval_samples_per_second": 233.572,
|
| 625 |
+
"eval_steps_per_second": 3.696,
|
| 626 |
+
"step": 299
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 46.15384615384615,
|
| 630 |
+
"grad_norm": 4.265622138977051,
|
| 631 |
+
"learning_rate": 2.777777777777778e-05,
|
| 632 |
+
"loss": 0.2495,
|
| 633 |
+
"step": 300
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 46.92307692307692,
|
| 637 |
+
"eval_accuracy": 0.5795551753635586,
|
| 638 |
+
"eval_loss": 0.9899386763572693,
|
| 639 |
+
"eval_runtime": 9.9941,
|
| 640 |
+
"eval_samples_per_second": 233.939,
|
| 641 |
+
"eval_steps_per_second": 3.702,
|
| 642 |
+
"step": 305
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 47.69230769230769,
|
| 646 |
+
"grad_norm": 2.909935474395752,
|
| 647 |
+
"learning_rate": 2.6851851851851855e-05,
|
| 648 |
+
"loss": 0.2454,
|
| 649 |
+
"step": 310
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 48.0,
|
| 653 |
+
"eval_accuracy": 0.553464499572284,
|
| 654 |
+
"eval_loss": 1.0290616750717163,
|
| 655 |
+
"eval_runtime": 10.0955,
|
| 656 |
+
"eval_samples_per_second": 231.587,
|
| 657 |
+
"eval_steps_per_second": 3.665,
|
| 658 |
+
"step": 312
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 48.92307692307692,
|
| 662 |
+
"eval_accuracy": 0.6822070145423439,
|
| 663 |
+
"eval_loss": 0.7534288167953491,
|
| 664 |
+
"eval_runtime": 10.0774,
|
| 665 |
+
"eval_samples_per_second": 232.004,
|
| 666 |
+
"eval_steps_per_second": 3.672,
|
| 667 |
+
"step": 318
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 49.23076923076923,
|
| 671 |
+
"grad_norm": 6.560550212860107,
|
| 672 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 673 |
+
"loss": 0.2473,
|
| 674 |
+
"step": 320
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 50.0,
|
| 678 |
+
"eval_accuracy": 0.7091531223267751,
|
| 679 |
+
"eval_loss": 0.6591421961784363,
|
| 680 |
+
"eval_runtime": 10.0255,
|
| 681 |
+
"eval_samples_per_second": 233.204,
|
| 682 |
+
"eval_steps_per_second": 3.691,
|
| 683 |
+
"step": 325
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 50.76923076923077,
|
| 687 |
+
"grad_norm": 10.7844820022583,
|
| 688 |
+
"learning_rate": 2.5e-05,
|
| 689 |
+
"loss": 0.2716,
|
| 690 |
+
"step": 330
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 50.92307692307692,
|
| 694 |
+
"eval_accuracy": 0.7455089820359282,
|
| 695 |
+
"eval_loss": 0.58400559425354,
|
| 696 |
+
"eval_runtime": 9.8847,
|
| 697 |
+
"eval_samples_per_second": 236.527,
|
| 698 |
+
"eval_steps_per_second": 3.743,
|
| 699 |
+
"step": 331
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 52.0,
|
| 703 |
+
"eval_accuracy": 0.47647562018819506,
|
| 704 |
+
"eval_loss": 1.2430182695388794,
|
| 705 |
+
"eval_runtime": 9.8852,
|
| 706 |
+
"eval_samples_per_second": 236.515,
|
| 707 |
+
"eval_steps_per_second": 3.743,
|
| 708 |
+
"step": 338
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 52.30769230769231,
|
| 712 |
+
"grad_norm": 7.81274938583374,
|
| 713 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 714 |
+
"loss": 0.234,
|
| 715 |
+
"step": 340
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 52.92307692307692,
|
| 719 |
+
"eval_accuracy": 0.5145423438836613,
|
| 720 |
+
"eval_loss": 1.299268126487732,
|
| 721 |
+
"eval_runtime": 10.1543,
|
| 722 |
+
"eval_samples_per_second": 230.247,
|
| 723 |
+
"eval_steps_per_second": 3.644,
|
| 724 |
+
"step": 344
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 53.84615384615385,
|
| 728 |
+
"grad_norm": 8.652073860168457,
|
| 729 |
+
"learning_rate": 2.314814814814815e-05,
|
| 730 |
+
"loss": 0.2482,
|
| 731 |
+
"step": 350
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 54.0,
|
| 735 |
+
"eval_accuracy": 0.7172797262617622,
|
| 736 |
+
"eval_loss": 0.6042024493217468,
|
| 737 |
+
"eval_runtime": 10.1638,
|
| 738 |
+
"eval_samples_per_second": 230.033,
|
| 739 |
+
"eval_steps_per_second": 3.64,
|
| 740 |
+
"step": 351
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 54.92307692307692,
|
| 744 |
+
"eval_accuracy": 0.6026518391787853,
|
| 745 |
+
"eval_loss": 0.8891559839248657,
|
| 746 |
+
"eval_runtime": 9.9638,
|
| 747 |
+
"eval_samples_per_second": 234.65,
|
| 748 |
+
"eval_steps_per_second": 3.713,
|
| 749 |
+
"step": 357
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 55.38461538461539,
|
| 753 |
+
"grad_norm": 3.8207547664642334,
|
| 754 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 755 |
+
"loss": 0.2339,
|
| 756 |
+
"step": 360
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 56.0,
|
| 760 |
+
"eval_accuracy": 0.316082121471343,
|
| 761 |
+
"eval_loss": 1.8545583486557007,
|
| 762 |
+
"eval_runtime": 9.9886,
|
| 763 |
+
"eval_samples_per_second": 234.066,
|
| 764 |
+
"eval_steps_per_second": 3.704,
|
| 765 |
+
"step": 364
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 56.92307692307692,
|
| 769 |
+
"grad_norm": 4.086548328399658,
|
| 770 |
+
"learning_rate": 2.1296296296296296e-05,
|
| 771 |
+
"loss": 0.2461,
|
| 772 |
+
"step": 370
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 56.92307692307692,
|
| 776 |
+
"eval_accuracy": 0.5359281437125748,
|
| 777 |
+
"eval_loss": 1.0858689546585083,
|
| 778 |
+
"eval_runtime": 9.9977,
|
| 779 |
+
"eval_samples_per_second": 233.854,
|
| 780 |
+
"eval_steps_per_second": 3.701,
|
| 781 |
+
"step": 370
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 58.0,
|
| 785 |
+
"eval_accuracy": 0.6176218990590248,
|
| 786 |
+
"eval_loss": 0.8690257668495178,
|
| 787 |
+
"eval_runtime": 10.0183,
|
| 788 |
+
"eval_samples_per_second": 233.374,
|
| 789 |
+
"eval_steps_per_second": 3.693,
|
| 790 |
+
"step": 377
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 58.46153846153846,
|
| 794 |
+
"grad_norm": 4.676217079162598,
|
| 795 |
+
"learning_rate": 2.037037037037037e-05,
|
| 796 |
+
"loss": 0.2395,
|
| 797 |
+
"step": 380
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 58.92307692307692,
|
| 801 |
+
"eval_accuracy": 0.6693755346449958,
|
| 802 |
+
"eval_loss": 0.7557449340820312,
|
| 803 |
+
"eval_runtime": 9.868,
|
| 804 |
+
"eval_samples_per_second": 236.928,
|
| 805 |
+
"eval_steps_per_second": 3.75,
|
| 806 |
+
"step": 383
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 60.0,
|
| 810 |
+
"grad_norm": 7.573139667510986,
|
| 811 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 812 |
+
"loss": 0.2159,
|
| 813 |
+
"step": 390
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 60.0,
|
| 817 |
+
"eval_accuracy": 0.5701454234388366,
|
| 818 |
+
"eval_loss": 1.053432822227478,
|
| 819 |
+
"eval_runtime": 10.2535,
|
| 820 |
+
"eval_samples_per_second": 228.02,
|
| 821 |
+
"eval_steps_per_second": 3.609,
|
| 822 |
+
"step": 390
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 60.92307692307692,
|
| 826 |
+
"eval_accuracy": 0.5812660393498716,
|
| 827 |
+
"eval_loss": 0.9855865240097046,
|
| 828 |
+
"eval_runtime": 10.0776,
|
| 829 |
+
"eval_samples_per_second": 231.999,
|
| 830 |
+
"eval_steps_per_second": 3.672,
|
| 831 |
+
"step": 396
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 61.53846153846154,
|
| 835 |
+
"grad_norm": 4.489895820617676,
|
| 836 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 837 |
+
"loss": 0.2309,
|
| 838 |
+
"step": 400
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 62.0,
|
| 842 |
+
"eval_accuracy": 0.5500427715996579,
|
| 843 |
+
"eval_loss": 0.9999585151672363,
|
| 844 |
+
"eval_runtime": 9.7878,
|
| 845 |
+
"eval_samples_per_second": 238.869,
|
| 846 |
+
"eval_steps_per_second": 3.78,
|
| 847 |
+
"step": 403
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 62.92307692307692,
|
| 851 |
+
"eval_accuracy": 0.5179640718562875,
|
| 852 |
+
"eval_loss": 1.1939594745635986,
|
| 853 |
+
"eval_runtime": 9.8311,
|
| 854 |
+
"eval_samples_per_second": 237.818,
|
| 855 |
+
"eval_steps_per_second": 3.764,
|
| 856 |
+
"step": 409
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 63.07692307692308,
|
| 860 |
+
"grad_norm": 6.18975830078125,
|
| 861 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 862 |
+
"loss": 0.2117,
|
| 863 |
+
"step": 410
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 64.0,
|
| 867 |
+
"eval_accuracy": 0.5153977758768178,
|
| 868 |
+
"eval_loss": 1.1580592393875122,
|
| 869 |
+
"eval_runtime": 10.0265,
|
| 870 |
+
"eval_samples_per_second": 233.182,
|
| 871 |
+
"eval_steps_per_second": 3.69,
|
| 872 |
+
"step": 416
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 64.61538461538461,
|
| 876 |
+
"grad_norm": 4.720950603485107,
|
| 877 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 878 |
+
"loss": 0.2307,
|
| 879 |
+
"step": 420
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 64.92307692307692,
|
| 883 |
+
"eval_accuracy": 0.5337895637296834,
|
| 884 |
+
"eval_loss": 0.9987441897392273,
|
| 885 |
+
"eval_runtime": 10.0605,
|
| 886 |
+
"eval_samples_per_second": 232.395,
|
| 887 |
+
"eval_steps_per_second": 3.678,
|
| 888 |
+
"step": 422
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 66.0,
|
| 892 |
+
"eval_accuracy": 0.5414884516680923,
|
| 893 |
+
"eval_loss": 1.084990382194519,
|
| 894 |
+
"eval_runtime": 9.9921,
|
| 895 |
+
"eval_samples_per_second": 233.986,
|
| 896 |
+
"eval_steps_per_second": 3.703,
|
| 897 |
+
"step": 429
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 66.15384615384616,
|
| 901 |
+
"grad_norm": 5.033910274505615,
|
| 902 |
+
"learning_rate": 1.574074074074074e-05,
|
| 903 |
+
"loss": 0.2068,
|
| 904 |
+
"step": 430
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 66.92307692307692,
|
| 908 |
+
"eval_accuracy": 0.6013686911890505,
|
| 909 |
+
"eval_loss": 0.942755401134491,
|
| 910 |
+
"eval_runtime": 9.7471,
|
| 911 |
+
"eval_samples_per_second": 239.866,
|
| 912 |
+
"eval_steps_per_second": 3.796,
|
| 913 |
+
"step": 435
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 67.6923076923077,
|
| 917 |
+
"grad_norm": 5.34838342666626,
|
| 918 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 919 |
+
"loss": 0.2126,
|
| 920 |
+
"step": 440
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 68.0,
|
| 924 |
+
"eval_accuracy": 0.5115483319076134,
|
| 925 |
+
"eval_loss": 1.237959861755371,
|
| 926 |
+
"eval_runtime": 9.9382,
|
| 927 |
+
"eval_samples_per_second": 235.253,
|
| 928 |
+
"eval_steps_per_second": 3.723,
|
| 929 |
+
"step": 442
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 68.92307692307692,
|
| 933 |
+
"eval_accuracy": 0.5859709153122327,
|
| 934 |
+
"eval_loss": 0.9992711544036865,
|
| 935 |
+
"eval_runtime": 10.002,
|
| 936 |
+
"eval_samples_per_second": 233.754,
|
| 937 |
+
"eval_steps_per_second": 3.699,
|
| 938 |
+
"step": 448
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 69.23076923076923,
|
| 942 |
+
"grad_norm": 3.1523709297180176,
|
| 943 |
+
"learning_rate": 1.388888888888889e-05,
|
| 944 |
+
"loss": 0.2176,
|
| 945 |
+
"step": 450
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 70.0,
|
| 949 |
+
"eval_accuracy": 0.5021385799828914,
|
| 950 |
+
"eval_loss": 1.190958023071289,
|
| 951 |
+
"eval_runtime": 9.7096,
|
| 952 |
+
"eval_samples_per_second": 240.793,
|
| 953 |
+
"eval_steps_per_second": 3.811,
|
| 954 |
+
"step": 455
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 70.76923076923077,
|
| 958 |
+
"grad_norm": 4.65359354019165,
|
| 959 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 960 |
+
"loss": 0.2096,
|
| 961 |
+
"step": 460
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 70.92307692307692,
|
| 965 |
+
"eval_accuracy": 0.5119760479041916,
|
| 966 |
+
"eval_loss": 1.246795415878296,
|
| 967 |
+
"eval_runtime": 9.658,
|
| 968 |
+
"eval_samples_per_second": 242.079,
|
| 969 |
+
"eval_steps_per_second": 3.831,
|
| 970 |
+
"step": 461
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 72.0,
|
| 974 |
+
"eval_accuracy": 0.6920444824636441,
|
| 975 |
+
"eval_loss": 0.7588455677032471,
|
| 976 |
+
"eval_runtime": 10.2262,
|
| 977 |
+
"eval_samples_per_second": 228.628,
|
| 978 |
+
"eval_steps_per_second": 3.618,
|
| 979 |
+
"step": 468
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 72.3076923076923,
|
| 983 |
+
"grad_norm": 4.450184345245361,
|
| 984 |
+
"learning_rate": 1.2037037037037037e-05,
|
| 985 |
+
"loss": 0.2092,
|
| 986 |
+
"step": 470
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 72.92307692307692,
|
| 990 |
+
"eval_accuracy": 0.6308810949529512,
|
| 991 |
+
"eval_loss": 0.900288999080658,
|
| 992 |
+
"eval_runtime": 9.6458,
|
| 993 |
+
"eval_samples_per_second": 242.386,
|
| 994 |
+
"eval_steps_per_second": 3.836,
|
| 995 |
+
"step": 474
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 73.84615384615384,
|
| 999 |
+
"grad_norm": 4.230223178863525,
|
| 1000 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1001 |
+
"loss": 0.1968,
|
| 1002 |
+
"step": 480
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 74.0,
|
| 1006 |
+
"eval_accuracy": 0.564585115483319,
|
| 1007 |
+
"eval_loss": 1.1697088479995728,
|
| 1008 |
+
"eval_runtime": 9.7325,
|
| 1009 |
+
"eval_samples_per_second": 240.225,
|
| 1010 |
+
"eval_steps_per_second": 3.802,
|
| 1011 |
+
"step": 481
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 74.92307692307692,
|
| 1015 |
+
"eval_accuracy": 0.6445680068434559,
|
| 1016 |
+
"eval_loss": 0.8789314031600952,
|
| 1017 |
+
"eval_runtime": 9.8672,
|
| 1018 |
+
"eval_samples_per_second": 236.946,
|
| 1019 |
+
"eval_steps_per_second": 3.75,
|
| 1020 |
+
"step": 487
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 75.38461538461539,
|
| 1024 |
+
"grad_norm": 5.635646343231201,
|
| 1025 |
+
"learning_rate": 1.0185185185185185e-05,
|
| 1026 |
+
"loss": 0.2027,
|
| 1027 |
+
"step": 490
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 76.0,
|
| 1031 |
+
"eval_accuracy": 0.5598802395209581,
|
| 1032 |
+
"eval_loss": 1.1352075338363647,
|
| 1033 |
+
"eval_runtime": 10.2152,
|
| 1034 |
+
"eval_samples_per_second": 228.875,
|
| 1035 |
+
"eval_steps_per_second": 3.622,
|
| 1036 |
+
"step": 494
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 76.92307692307692,
|
| 1040 |
+
"grad_norm": 4.811977386474609,
|
| 1041 |
+
"learning_rate": 9.259259259259259e-06,
|
| 1042 |
+
"loss": 0.1965,
|
| 1043 |
+
"step": 500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 76.92307692307692,
|
| 1047 |
+
"eval_accuracy": 0.5598802395209581,
|
| 1048 |
+
"eval_loss": 1.083630919456482,
|
| 1049 |
+
"eval_runtime": 9.8228,
|
| 1050 |
+
"eval_samples_per_second": 238.017,
|
| 1051 |
+
"eval_steps_per_second": 3.767,
|
| 1052 |
+
"step": 500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 78.0,
|
| 1056 |
+
"eval_accuracy": 0.5902480752780154,
|
| 1057 |
+
"eval_loss": 1.018804669380188,
|
| 1058 |
+
"eval_runtime": 10.2662,
|
| 1059 |
+
"eval_samples_per_second": 227.739,
|
| 1060 |
+
"eval_steps_per_second": 3.604,
|
| 1061 |
+
"step": 507
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 78.46153846153847,
|
| 1065 |
+
"grad_norm": 4.3275227546691895,
|
| 1066 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1067 |
+
"loss": 0.2267,
|
| 1068 |
+
"step": 510
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 78.92307692307692,
|
| 1072 |
+
"eval_accuracy": 0.5975192472198461,
|
| 1073 |
+
"eval_loss": 1.0287189483642578,
|
| 1074 |
+
"eval_runtime": 10.2326,
|
| 1075 |
+
"eval_samples_per_second": 228.486,
|
| 1076 |
+
"eval_steps_per_second": 3.616,
|
| 1077 |
+
"step": 513
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 80.0,
|
| 1081 |
+
"grad_norm": 5.816257476806641,
|
| 1082 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 1083 |
+
"loss": 0.1967,
|
| 1084 |
+
"step": 520
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 80.0,
|
| 1088 |
+
"eval_accuracy": 0.6544054747647562,
|
| 1089 |
+
"eval_loss": 0.8465330004692078,
|
| 1090 |
+
"eval_runtime": 9.6333,
|
| 1091 |
+
"eval_samples_per_second": 242.7,
|
| 1092 |
+
"eval_steps_per_second": 3.841,
|
| 1093 |
+
"step": 520
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 80.92307692307692,
|
| 1097 |
+
"eval_accuracy": 0.5470487596236099,
|
| 1098 |
+
"eval_loss": 1.188087821006775,
|
| 1099 |
+
"eval_runtime": 9.8344,
|
| 1100 |
+
"eval_samples_per_second": 237.736,
|
| 1101 |
+
"eval_steps_per_second": 3.762,
|
| 1102 |
+
"step": 526
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 81.53846153846153,
|
| 1106 |
+
"grad_norm": 4.023173809051514,
|
| 1107 |
+
"learning_rate": 6.481481481481481e-06,
|
| 1108 |
+
"loss": 0.1842,
|
| 1109 |
+
"step": 530
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 82.0,
|
| 1113 |
+
"eval_accuracy": 0.5367835757057314,
|
| 1114 |
+
"eval_loss": 1.235166072845459,
|
| 1115 |
+
"eval_runtime": 9.8519,
|
| 1116 |
+
"eval_samples_per_second": 237.315,
|
| 1117 |
+
"eval_steps_per_second": 3.756,
|
| 1118 |
+
"step": 533
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 82.92307692307692,
|
| 1122 |
+
"eval_accuracy": 0.5701454234388366,
|
| 1123 |
+
"eval_loss": 1.106431007385254,
|
| 1124 |
+
"eval_runtime": 10.0863,
|
| 1125 |
+
"eval_samples_per_second": 231.799,
|
| 1126 |
+
"eval_steps_per_second": 3.668,
|
| 1127 |
+
"step": 539
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 83.07692307692308,
|
| 1131 |
+
"grad_norm": 5.945059299468994,
|
| 1132 |
+
"learning_rate": 5.555555555555556e-06,
|
| 1133 |
+
"loss": 0.1952,
|
| 1134 |
+
"step": 540
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 84.0,
|
| 1138 |
+
"eval_accuracy": 0.6608212147134302,
|
| 1139 |
+
"eval_loss": 0.8087576031684875,
|
| 1140 |
+
"eval_runtime": 9.3282,
|
| 1141 |
+
"eval_samples_per_second": 250.637,
|
| 1142 |
+
"eval_steps_per_second": 3.966,
|
| 1143 |
+
"step": 546
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 84.61538461538461,
|
| 1147 |
+
"grad_norm": 4.164029121398926,
|
| 1148 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 1149 |
+
"loss": 0.1873,
|
| 1150 |
+
"step": 550
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 84.92307692307692,
|
| 1154 |
+
"eval_accuracy": 0.6086398631308811,
|
| 1155 |
+
"eval_loss": 0.9341749548912048,
|
| 1156 |
+
"eval_runtime": 9.6842,
|
| 1157 |
+
"eval_samples_per_second": 241.425,
|
| 1158 |
+
"eval_steps_per_second": 3.821,
|
| 1159 |
+
"step": 552
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 86.0,
|
| 1163 |
+
"eval_accuracy": 0.6056458511548332,
|
| 1164 |
+
"eval_loss": 0.9807350039482117,
|
| 1165 |
+
"eval_runtime": 9.6406,
|
| 1166 |
+
"eval_samples_per_second": 242.516,
|
| 1167 |
+
"eval_steps_per_second": 3.838,
|
| 1168 |
+
"step": 559
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 86.15384615384616,
|
| 1172 |
+
"grad_norm": 5.312713146209717,
|
| 1173 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 1174 |
+
"loss": 0.185,
|
| 1175 |
+
"step": 560
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 86.92307692307692,
|
| 1179 |
+
"eval_accuracy": 0.5898203592814372,
|
| 1180 |
+
"eval_loss": 1.0164724588394165,
|
| 1181 |
+
"eval_runtime": 9.7581,
|
| 1182 |
+
"eval_samples_per_second": 239.596,
|
| 1183 |
+
"eval_steps_per_second": 3.792,
|
| 1184 |
+
"step": 565
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 87.6923076923077,
|
| 1188 |
+
"grad_norm": 5.478633403778076,
|
| 1189 |
+
"learning_rate": 2.777777777777778e-06,
|
| 1190 |
+
"loss": 0.1993,
|
| 1191 |
+
"step": 570
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 88.0,
|
| 1195 |
+
"eval_accuracy": 0.5474764756201882,
|
| 1196 |
+
"eval_loss": 1.1511483192443848,
|
| 1197 |
+
"eval_runtime": 9.905,
|
| 1198 |
+
"eval_samples_per_second": 236.042,
|
| 1199 |
+
"eval_steps_per_second": 3.735,
|
| 1200 |
+
"step": 572
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 88.92307692307692,
|
| 1204 |
+
"eval_accuracy": 0.5406330196749358,
|
| 1205 |
+
"eval_loss": 1.176562786102295,
|
| 1206 |
+
"eval_runtime": 10.1535,
|
| 1207 |
+
"eval_samples_per_second": 230.266,
|
| 1208 |
+
"eval_steps_per_second": 3.644,
|
| 1209 |
+
"step": 578
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 89.23076923076923,
|
| 1213 |
+
"grad_norm": 3.2113797664642334,
|
| 1214 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1215 |
+
"loss": 0.1707,
|
| 1216 |
+
"step": 580
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 90.0,
|
| 1220 |
+
"eval_accuracy": 0.5662959794696322,
|
| 1221 |
+
"eval_loss": 1.120088815689087,
|
| 1222 |
+
"eval_runtime": 9.7143,
|
| 1223 |
+
"eval_samples_per_second": 240.676,
|
| 1224 |
+
"eval_steps_per_second": 3.809,
|
| 1225 |
+
"step": 585
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 90.76923076923077,
|
| 1229 |
+
"grad_norm": 5.403631210327148,
|
| 1230 |
+
"learning_rate": 9.259259259259259e-07,
|
| 1231 |
+
"loss": 0.1852,
|
| 1232 |
+
"step": 590
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 90.92307692307692,
|
| 1236 |
+
"eval_accuracy": 0.5701454234388366,
|
| 1237 |
+
"eval_loss": 1.116164207458496,
|
| 1238 |
+
"eval_runtime": 10.0497,
|
| 1239 |
+
"eval_samples_per_second": 232.643,
|
| 1240 |
+
"eval_steps_per_second": 3.682,
|
| 1241 |
+
"step": 591
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 92.0,
|
| 1245 |
+
"eval_accuracy": 0.5680068434559452,
|
| 1246 |
+
"eval_loss": 1.1272914409637451,
|
| 1247 |
+
"eval_runtime": 9.9727,
|
| 1248 |
+
"eval_samples_per_second": 234.44,
|
| 1249 |
+
"eval_steps_per_second": 3.71,
|
| 1250 |
+
"step": 598
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 92.3076923076923,
|
| 1254 |
+
"grad_norm": 4.397765159606934,
|
| 1255 |
+
"learning_rate": 0.0,
|
| 1256 |
+
"loss": 0.1904,
|
| 1257 |
+
"step": 600
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 92.3076923076923,
|
| 1261 |
+
"eval_accuracy": 0.5680068434559452,
|
| 1262 |
+
"eval_loss": 1.1280397176742554,
|
| 1263 |
+
"eval_runtime": 10.1432,
|
| 1264 |
+
"eval_samples_per_second": 230.5,
|
| 1265 |
+
"eval_steps_per_second": 3.648,
|
| 1266 |
+
"step": 600
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 92.3076923076923,
|
| 1270 |
+
"step": 600,
|
| 1271 |
+
"total_flos": 2.9138957540265e+18,
|
| 1272 |
+
"train_loss": 0.273075803120931,
|
| 1273 |
+
"train_runtime": 2186.3924,
|
| 1274 |
+
"train_samples_per_second": 73.774,
|
| 1275 |
+
"train_steps_per_second": 0.274
|
| 1276 |
+
}
|
| 1277 |
+
],
|
| 1278 |
+
"logging_steps": 10,
|
| 1279 |
+
"max_steps": 600,
|
| 1280 |
+
"num_input_tokens_seen": 0,
|
| 1281 |
+
"num_train_epochs": 100,
|
| 1282 |
+
"save_steps": 500,
|
| 1283 |
+
"stateful_callbacks": {
|
| 1284 |
+
"TrainerControl": {
|
| 1285 |
+
"args": {
|
| 1286 |
+
"should_epoch_stop": false,
|
| 1287 |
+
"should_evaluate": false,
|
| 1288 |
+
"should_log": false,
|
| 1289 |
+
"should_save": true,
|
| 1290 |
+
"should_training_stop": true
|
| 1291 |
+
},
|
| 1292 |
+
"attributes": {}
|
| 1293 |
+
}
|
| 1294 |
+
},
|
| 1295 |
+
"total_flos": 2.9138957540265e+18,
|
| 1296 |
+
"train_batch_size": 64,
|
| 1297 |
+
"trial_name": null,
|
| 1298 |
+
"trial_params": null
|
| 1299 |
+
}
|