Commit
•
5dbd3b4
1
Parent(s):
207d849
End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May20_21-09-50_b94dbfcd7693/events.out.tfevents.1716240618.b94dbfcd7693.1289.38 +3 -0
- train_results.json +8 -0
- trainer_state.json +1079 -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 [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-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: 0.9318181818181818
|
| 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 [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.2499
|
| 36 |
+
- Accuracy: 0.9318
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_accuracy": 0.9318181818181818,
|
| 4 |
+
"eval_loss": 0.24986252188682556,
|
| 5 |
+
"eval_runtime": 0.7488,
|
| 6 |
+
"eval_samples_per_second": 58.758,
|
| 7 |
+
"eval_steps_per_second": 2.671,
|
| 8 |
+
"total_flos": 1.8822505078960128e+18,
|
| 9 |
+
"train_loss": 0.19619473487138747,
|
| 10 |
+
"train_runtime": 1209.7214,
|
| 11 |
+
"train_samples_per_second": 20.087,
|
| 12 |
+
"train_steps_per_second": 0.165
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_accuracy": 0.9318181818181818,
|
| 4 |
+
"eval_loss": 0.24986252188682556,
|
| 5 |
+
"eval_runtime": 0.7488,
|
| 6 |
+
"eval_samples_per_second": 58.758,
|
| 7 |
+
"eval_steps_per_second": 2.671
|
| 8 |
+
}
|
runs/May20_21-09-50_b94dbfcd7693/events.out.tfevents.1716240618.b94dbfcd7693.1289.38
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca2550995a0b2017f21bed26e59f85b7bfb523096082428ad6d8e0931d37e86f
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"total_flos": 1.8822505078960128e+18,
|
| 4 |
+
"train_loss": 0.19619473487138747,
|
| 5 |
+
"train_runtime": 1209.7214,
|
| 6 |
+
"train_samples_per_second": 20.087,
|
| 7 |
+
"train_steps_per_second": 0.165
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1079 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9318181818181818,
|
| 3 |
+
"best_model_checkpoint": "beit-base-patch16-224-85-fold5/checkpoint-92",
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.29545454545454547,
|
| 14 |
+
"eval_loss": 1.089578628540039,
|
| 15 |
+
"eval_runtime": 0.7021,
|
| 16 |
+
"eval_samples_per_second": 62.667,
|
| 17 |
+
"eval_steps_per_second": 2.849,
|
| 18 |
+
"step": 2
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 2.0,
|
| 22 |
+
"eval_accuracy": 0.7272727272727273,
|
| 23 |
+
"eval_loss": 0.6455699801445007,
|
| 24 |
+
"eval_runtime": 0.6856,
|
| 25 |
+
"eval_samples_per_second": 64.174,
|
| 26 |
+
"eval_steps_per_second": 2.917,
|
| 27 |
+
"step": 4
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 3.0,
|
| 31 |
+
"eval_accuracy": 0.7045454545454546,
|
| 32 |
+
"eval_loss": 1.0355168581008911,
|
| 33 |
+
"eval_runtime": 0.6996,
|
| 34 |
+
"eval_samples_per_second": 62.897,
|
| 35 |
+
"eval_steps_per_second": 2.859,
|
| 36 |
+
"step": 6
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 4.0,
|
| 40 |
+
"eval_accuracy": 0.7045454545454546,
|
| 41 |
+
"eval_loss": 0.9123974442481995,
|
| 42 |
+
"eval_runtime": 0.6813,
|
| 43 |
+
"eval_samples_per_second": 64.582,
|
| 44 |
+
"eval_steps_per_second": 2.936,
|
| 45 |
+
"step": 8
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.0,
|
| 49 |
+
"grad_norm": 9.428820610046387,
|
| 50 |
+
"learning_rate": 2.5e-05,
|
| 51 |
+
"loss": 0.7607,
|
| 52 |
+
"step": 10
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 5.0,
|
| 56 |
+
"eval_accuracy": 0.7954545454545454,
|
| 57 |
+
"eval_loss": 0.5808687210083008,
|
| 58 |
+
"eval_runtime": 0.672,
|
| 59 |
+
"eval_samples_per_second": 65.48,
|
| 60 |
+
"eval_steps_per_second": 2.976,
|
| 61 |
+
"step": 10
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 6.0,
|
| 65 |
+
"eval_accuracy": 0.75,
|
| 66 |
+
"eval_loss": 0.6812124848365784,
|
| 67 |
+
"eval_runtime": 0.6683,
|
| 68 |
+
"eval_samples_per_second": 65.834,
|
| 69 |
+
"eval_steps_per_second": 2.992,
|
| 70 |
+
"step": 12
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 7.0,
|
| 74 |
+
"eval_accuracy": 0.75,
|
| 75 |
+
"eval_loss": 0.6529408097267151,
|
| 76 |
+
"eval_runtime": 0.6753,
|
| 77 |
+
"eval_samples_per_second": 65.16,
|
| 78 |
+
"eval_steps_per_second": 2.962,
|
| 79 |
+
"step": 14
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 8.0,
|
| 83 |
+
"eval_accuracy": 0.7272727272727273,
|
| 84 |
+
"eval_loss": 0.717404305934906,
|
| 85 |
+
"eval_runtime": 0.7009,
|
| 86 |
+
"eval_samples_per_second": 62.778,
|
| 87 |
+
"eval_steps_per_second": 2.854,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 9.0,
|
| 92 |
+
"eval_accuracy": 0.6136363636363636,
|
| 93 |
+
"eval_loss": 0.6618563532829285,
|
| 94 |
+
"eval_runtime": 0.6832,
|
| 95 |
+
"eval_samples_per_second": 64.403,
|
| 96 |
+
"eval_steps_per_second": 2.927,
|
| 97 |
+
"step": 18
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 10.0,
|
| 101 |
+
"grad_norm": 6.730888843536377,
|
| 102 |
+
"learning_rate": 5e-05,
|
| 103 |
+
"loss": 0.4221,
|
| 104 |
+
"step": 20
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 10.0,
|
| 108 |
+
"eval_accuracy": 0.75,
|
| 109 |
+
"eval_loss": 0.80633944272995,
|
| 110 |
+
"eval_runtime": 0.6867,
|
| 111 |
+
"eval_samples_per_second": 64.075,
|
| 112 |
+
"eval_steps_per_second": 2.912,
|
| 113 |
+
"step": 20
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 11.0,
|
| 117 |
+
"eval_accuracy": 0.75,
|
| 118 |
+
"eval_loss": 0.6371808648109436,
|
| 119 |
+
"eval_runtime": 0.6802,
|
| 120 |
+
"eval_samples_per_second": 64.689,
|
| 121 |
+
"eval_steps_per_second": 2.94,
|
| 122 |
+
"step": 22
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 12.0,
|
| 126 |
+
"eval_accuracy": 0.75,
|
| 127 |
+
"eval_loss": 0.588646411895752,
|
| 128 |
+
"eval_runtime": 0.6871,
|
| 129 |
+
"eval_samples_per_second": 64.039,
|
| 130 |
+
"eval_steps_per_second": 2.911,
|
| 131 |
+
"step": 24
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 13.0,
|
| 135 |
+
"eval_accuracy": 0.6363636363636364,
|
| 136 |
+
"eval_loss": 0.6359339356422424,
|
| 137 |
+
"eval_runtime": 0.6771,
|
| 138 |
+
"eval_samples_per_second": 64.986,
|
| 139 |
+
"eval_steps_per_second": 2.954,
|
| 140 |
+
"step": 26
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 14.0,
|
| 144 |
+
"eval_accuracy": 0.75,
|
| 145 |
+
"eval_loss": 0.5584580898284912,
|
| 146 |
+
"eval_runtime": 0.6833,
|
| 147 |
+
"eval_samples_per_second": 64.391,
|
| 148 |
+
"eval_steps_per_second": 2.927,
|
| 149 |
+
"step": 28
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 15.0,
|
| 153 |
+
"grad_norm": 9.09820556640625,
|
| 154 |
+
"learning_rate": 4.722222222222222e-05,
|
| 155 |
+
"loss": 0.3287,
|
| 156 |
+
"step": 30
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 15.0,
|
| 160 |
+
"eval_accuracy": 0.7272727272727273,
|
| 161 |
+
"eval_loss": 0.45406511425971985,
|
| 162 |
+
"eval_runtime": 0.6778,
|
| 163 |
+
"eval_samples_per_second": 64.917,
|
| 164 |
+
"eval_steps_per_second": 2.951,
|
| 165 |
+
"step": 30
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 16.0,
|
| 169 |
+
"eval_accuracy": 0.5681818181818182,
|
| 170 |
+
"eval_loss": 0.7624431252479553,
|
| 171 |
+
"eval_runtime": 0.6791,
|
| 172 |
+
"eval_samples_per_second": 64.788,
|
| 173 |
+
"eval_steps_per_second": 2.945,
|
| 174 |
+
"step": 32
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 17.0,
|
| 178 |
+
"eval_accuracy": 0.75,
|
| 179 |
+
"eval_loss": 0.6805771589279175,
|
| 180 |
+
"eval_runtime": 0.6896,
|
| 181 |
+
"eval_samples_per_second": 63.807,
|
| 182 |
+
"eval_steps_per_second": 2.9,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 18.0,
|
| 187 |
+
"eval_accuracy": 0.7272727272727273,
|
| 188 |
+
"eval_loss": 0.7707971930503845,
|
| 189 |
+
"eval_runtime": 0.6822,
|
| 190 |
+
"eval_samples_per_second": 64.496,
|
| 191 |
+
"eval_steps_per_second": 2.932,
|
| 192 |
+
"step": 36
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 19.0,
|
| 196 |
+
"eval_accuracy": 0.7272727272727273,
|
| 197 |
+
"eval_loss": 0.416960209608078,
|
| 198 |
+
"eval_runtime": 0.6839,
|
| 199 |
+
"eval_samples_per_second": 64.337,
|
| 200 |
+
"eval_steps_per_second": 2.924,
|
| 201 |
+
"step": 38
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 20.0,
|
| 205 |
+
"grad_norm": 12.929576873779297,
|
| 206 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 207 |
+
"loss": 0.3663,
|
| 208 |
+
"step": 40
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 20.0,
|
| 212 |
+
"eval_accuracy": 0.7727272727272727,
|
| 213 |
+
"eval_loss": 0.42817583680152893,
|
| 214 |
+
"eval_runtime": 0.684,
|
| 215 |
+
"eval_samples_per_second": 64.327,
|
| 216 |
+
"eval_steps_per_second": 2.924,
|
| 217 |
+
"step": 40
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 21.0,
|
| 221 |
+
"eval_accuracy": 0.75,
|
| 222 |
+
"eval_loss": 0.5613061785697937,
|
| 223 |
+
"eval_runtime": 0.6869,
|
| 224 |
+
"eval_samples_per_second": 64.059,
|
| 225 |
+
"eval_steps_per_second": 2.912,
|
| 226 |
+
"step": 42
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 22.0,
|
| 230 |
+
"eval_accuracy": 0.8409090909090909,
|
| 231 |
+
"eval_loss": 0.4025155007839203,
|
| 232 |
+
"eval_runtime": 0.7081,
|
| 233 |
+
"eval_samples_per_second": 62.136,
|
| 234 |
+
"eval_steps_per_second": 2.824,
|
| 235 |
+
"step": 44
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 23.0,
|
| 239 |
+
"eval_accuracy": 0.7954545454545454,
|
| 240 |
+
"eval_loss": 0.41085565090179443,
|
| 241 |
+
"eval_runtime": 0.7009,
|
| 242 |
+
"eval_samples_per_second": 62.774,
|
| 243 |
+
"eval_steps_per_second": 2.853,
|
| 244 |
+
"step": 46
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 24.0,
|
| 248 |
+
"eval_accuracy": 0.8409090909090909,
|
| 249 |
+
"eval_loss": 0.4372767508029938,
|
| 250 |
+
"eval_runtime": 0.6888,
|
| 251 |
+
"eval_samples_per_second": 63.878,
|
| 252 |
+
"eval_steps_per_second": 2.904,
|
| 253 |
+
"step": 48
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 25.0,
|
| 257 |
+
"grad_norm": 6.6011061668396,
|
| 258 |
+
"learning_rate": 4.166666666666667e-05,
|
| 259 |
+
"loss": 0.2344,
|
| 260 |
+
"step": 50
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 25.0,
|
| 264 |
+
"eval_accuracy": 0.8863636363636364,
|
| 265 |
+
"eval_loss": 0.3210701644420624,
|
| 266 |
+
"eval_runtime": 0.6782,
|
| 267 |
+
"eval_samples_per_second": 64.873,
|
| 268 |
+
"eval_steps_per_second": 2.949,
|
| 269 |
+
"step": 50
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 26.0,
|
| 273 |
+
"eval_accuracy": 0.75,
|
| 274 |
+
"eval_loss": 0.5560834407806396,
|
| 275 |
+
"eval_runtime": 0.6873,
|
| 276 |
+
"eval_samples_per_second": 64.014,
|
| 277 |
+
"eval_steps_per_second": 2.91,
|
| 278 |
+
"step": 52
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 27.0,
|
| 282 |
+
"eval_accuracy": 0.8636363636363636,
|
| 283 |
+
"eval_loss": 0.31488853693008423,
|
| 284 |
+
"eval_runtime": 0.6879,
|
| 285 |
+
"eval_samples_per_second": 63.96,
|
| 286 |
+
"eval_steps_per_second": 2.907,
|
| 287 |
+
"step": 54
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 28.0,
|
| 291 |
+
"eval_accuracy": 0.7954545454545454,
|
| 292 |
+
"eval_loss": 0.31661316752433777,
|
| 293 |
+
"eval_runtime": 0.679,
|
| 294 |
+
"eval_samples_per_second": 64.805,
|
| 295 |
+
"eval_steps_per_second": 2.946,
|
| 296 |
+
"step": 56
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 29.0,
|
| 300 |
+
"eval_accuracy": 0.8636363636363636,
|
| 301 |
+
"eval_loss": 0.4163520932197571,
|
| 302 |
+
"eval_runtime": 0.6778,
|
| 303 |
+
"eval_samples_per_second": 64.92,
|
| 304 |
+
"eval_steps_per_second": 2.951,
|
| 305 |
+
"step": 58
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 30.0,
|
| 309 |
+
"grad_norm": 5.3891143798828125,
|
| 310 |
+
"learning_rate": 3.888888888888889e-05,
|
| 311 |
+
"loss": 0.2051,
|
| 312 |
+
"step": 60
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 30.0,
|
| 316 |
+
"eval_accuracy": 0.8636363636363636,
|
| 317 |
+
"eval_loss": 0.4345444440841675,
|
| 318 |
+
"eval_runtime": 0.6869,
|
| 319 |
+
"eval_samples_per_second": 64.054,
|
| 320 |
+
"eval_steps_per_second": 2.912,
|
| 321 |
+
"step": 60
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 31.0,
|
| 325 |
+
"eval_accuracy": 0.8636363636363636,
|
| 326 |
+
"eval_loss": 0.31803134083747864,
|
| 327 |
+
"eval_runtime": 0.6887,
|
| 328 |
+
"eval_samples_per_second": 63.886,
|
| 329 |
+
"eval_steps_per_second": 2.904,
|
| 330 |
+
"step": 62
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 32.0,
|
| 334 |
+
"eval_accuracy": 0.8409090909090909,
|
| 335 |
+
"eval_loss": 0.3673260509967804,
|
| 336 |
+
"eval_runtime": 0.6835,
|
| 337 |
+
"eval_samples_per_second": 64.377,
|
| 338 |
+
"eval_steps_per_second": 2.926,
|
| 339 |
+
"step": 64
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 33.0,
|
| 343 |
+
"eval_accuracy": 0.8409090909090909,
|
| 344 |
+
"eval_loss": 0.43128955364227295,
|
| 345 |
+
"eval_runtime": 0.682,
|
| 346 |
+
"eval_samples_per_second": 64.517,
|
| 347 |
+
"eval_steps_per_second": 2.933,
|
| 348 |
+
"step": 66
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 34.0,
|
| 352 |
+
"eval_accuracy": 0.8409090909090909,
|
| 353 |
+
"eval_loss": 0.43590399622917175,
|
| 354 |
+
"eval_runtime": 0.6788,
|
| 355 |
+
"eval_samples_per_second": 64.824,
|
| 356 |
+
"eval_steps_per_second": 2.947,
|
| 357 |
+
"step": 68
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 35.0,
|
| 361 |
+
"grad_norm": 3.639585256576538,
|
| 362 |
+
"learning_rate": 3.611111111111111e-05,
|
| 363 |
+
"loss": 0.1694,
|
| 364 |
+
"step": 70
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 35.0,
|
| 368 |
+
"eval_accuracy": 0.8181818181818182,
|
| 369 |
+
"eval_loss": 0.36998674273490906,
|
| 370 |
+
"eval_runtime": 0.6829,
|
| 371 |
+
"eval_samples_per_second": 64.428,
|
| 372 |
+
"eval_steps_per_second": 2.929,
|
| 373 |
+
"step": 70
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 36.0,
|
| 377 |
+
"eval_accuracy": 0.7954545454545454,
|
| 378 |
+
"eval_loss": 0.584349513053894,
|
| 379 |
+
"eval_runtime": 0.6937,
|
| 380 |
+
"eval_samples_per_second": 63.427,
|
| 381 |
+
"eval_steps_per_second": 2.883,
|
| 382 |
+
"step": 72
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 37.0,
|
| 386 |
+
"eval_accuracy": 0.8636363636363636,
|
| 387 |
+
"eval_loss": 0.4064345061779022,
|
| 388 |
+
"eval_runtime": 0.7057,
|
| 389 |
+
"eval_samples_per_second": 62.346,
|
| 390 |
+
"eval_steps_per_second": 2.834,
|
| 391 |
+
"step": 74
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 38.0,
|
| 395 |
+
"eval_accuracy": 0.8181818181818182,
|
| 396 |
+
"eval_loss": 0.39915817975997925,
|
| 397 |
+
"eval_runtime": 0.682,
|
| 398 |
+
"eval_samples_per_second": 64.516,
|
| 399 |
+
"eval_steps_per_second": 2.933,
|
| 400 |
+
"step": 76
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 39.0,
|
| 404 |
+
"eval_accuracy": 0.8636363636363636,
|
| 405 |
+
"eval_loss": 0.3152655363082886,
|
| 406 |
+
"eval_runtime": 0.6899,
|
| 407 |
+
"eval_samples_per_second": 63.778,
|
| 408 |
+
"eval_steps_per_second": 2.899,
|
| 409 |
+
"step": 78
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 40.0,
|
| 413 |
+
"grad_norm": 5.693894863128662,
|
| 414 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 415 |
+
"loss": 0.1566,
|
| 416 |
+
"step": 80
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 40.0,
|
| 420 |
+
"eval_accuracy": 0.8181818181818182,
|
| 421 |
+
"eval_loss": 0.5581021904945374,
|
| 422 |
+
"eval_runtime": 0.6812,
|
| 423 |
+
"eval_samples_per_second": 64.596,
|
| 424 |
+
"eval_steps_per_second": 2.936,
|
| 425 |
+
"step": 80
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 41.0,
|
| 429 |
+
"eval_accuracy": 0.8636363636363636,
|
| 430 |
+
"eval_loss": 0.2921257019042969,
|
| 431 |
+
"eval_runtime": 0.6885,
|
| 432 |
+
"eval_samples_per_second": 63.908,
|
| 433 |
+
"eval_steps_per_second": 2.905,
|
| 434 |
+
"step": 82
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 42.0,
|
| 438 |
+
"eval_accuracy": 0.8863636363636364,
|
| 439 |
+
"eval_loss": 0.32171034812927246,
|
| 440 |
+
"eval_runtime": 0.684,
|
| 441 |
+
"eval_samples_per_second": 64.326,
|
| 442 |
+
"eval_steps_per_second": 2.924,
|
| 443 |
+
"step": 84
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 43.0,
|
| 447 |
+
"eval_accuracy": 0.8863636363636364,
|
| 448 |
+
"eval_loss": 0.32552286982536316,
|
| 449 |
+
"eval_runtime": 0.6845,
|
| 450 |
+
"eval_samples_per_second": 64.277,
|
| 451 |
+
"eval_steps_per_second": 2.922,
|
| 452 |
+
"step": 86
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 44.0,
|
| 456 |
+
"eval_accuracy": 0.75,
|
| 457 |
+
"eval_loss": 0.723772406578064,
|
| 458 |
+
"eval_runtime": 0.7032,
|
| 459 |
+
"eval_samples_per_second": 62.569,
|
| 460 |
+
"eval_steps_per_second": 2.844,
|
| 461 |
+
"step": 88
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 45.0,
|
| 465 |
+
"grad_norm": 9.99726676940918,
|
| 466 |
+
"learning_rate": 3.055555555555556e-05,
|
| 467 |
+
"loss": 0.1389,
|
| 468 |
+
"step": 90
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 45.0,
|
| 472 |
+
"eval_accuracy": 0.8863636363636364,
|
| 473 |
+
"eval_loss": 0.40532952547073364,
|
| 474 |
+
"eval_runtime": 0.6753,
|
| 475 |
+
"eval_samples_per_second": 65.154,
|
| 476 |
+
"eval_steps_per_second": 2.962,
|
| 477 |
+
"step": 90
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 46.0,
|
| 481 |
+
"eval_accuracy": 0.9318181818181818,
|
| 482 |
+
"eval_loss": 0.24986252188682556,
|
| 483 |
+
"eval_runtime": 0.6918,
|
| 484 |
+
"eval_samples_per_second": 63.603,
|
| 485 |
+
"eval_steps_per_second": 2.891,
|
| 486 |
+
"step": 92
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 47.0,
|
| 490 |
+
"eval_accuracy": 0.8863636363636364,
|
| 491 |
+
"eval_loss": 0.2584282159805298,
|
| 492 |
+
"eval_runtime": 0.6836,
|
| 493 |
+
"eval_samples_per_second": 64.362,
|
| 494 |
+
"eval_steps_per_second": 2.926,
|
| 495 |
+
"step": 94
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 48.0,
|
| 499 |
+
"eval_accuracy": 0.8409090909090909,
|
| 500 |
+
"eval_loss": 0.44320380687713623,
|
| 501 |
+
"eval_runtime": 0.6851,
|
| 502 |
+
"eval_samples_per_second": 64.225,
|
| 503 |
+
"eval_steps_per_second": 2.919,
|
| 504 |
+
"step": 96
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 49.0,
|
| 508 |
+
"eval_accuracy": 0.7954545454545454,
|
| 509 |
+
"eval_loss": 0.6965329647064209,
|
| 510 |
+
"eval_runtime": 0.6828,
|
| 511 |
+
"eval_samples_per_second": 64.443,
|
| 512 |
+
"eval_steps_per_second": 2.929,
|
| 513 |
+
"step": 98
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 50.0,
|
| 517 |
+
"grad_norm": 5.456106662750244,
|
| 518 |
+
"learning_rate": 2.777777777777778e-05,
|
| 519 |
+
"loss": 0.1311,
|
| 520 |
+
"step": 100
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 50.0,
|
| 524 |
+
"eval_accuracy": 0.8409090909090909,
|
| 525 |
+
"eval_loss": 0.39098644256591797,
|
| 526 |
+
"eval_runtime": 0.6828,
|
| 527 |
+
"eval_samples_per_second": 64.437,
|
| 528 |
+
"eval_steps_per_second": 2.929,
|
| 529 |
+
"step": 100
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 51.0,
|
| 533 |
+
"eval_accuracy": 0.8636363636363636,
|
| 534 |
+
"eval_loss": 0.30165335536003113,
|
| 535 |
+
"eval_runtime": 0.6957,
|
| 536 |
+
"eval_samples_per_second": 63.242,
|
| 537 |
+
"eval_steps_per_second": 2.875,
|
| 538 |
+
"step": 102
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 52.0,
|
| 542 |
+
"eval_accuracy": 0.8636363636363636,
|
| 543 |
+
"eval_loss": 0.30502772331237793,
|
| 544 |
+
"eval_runtime": 0.6903,
|
| 545 |
+
"eval_samples_per_second": 63.743,
|
| 546 |
+
"eval_steps_per_second": 2.897,
|
| 547 |
+
"step": 104
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 53.0,
|
| 551 |
+
"eval_accuracy": 0.8636363636363636,
|
| 552 |
+
"eval_loss": 0.21932415664196014,
|
| 553 |
+
"eval_runtime": 0.6995,
|
| 554 |
+
"eval_samples_per_second": 62.902,
|
| 555 |
+
"eval_steps_per_second": 2.859,
|
| 556 |
+
"step": 106
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 54.0,
|
| 560 |
+
"eval_accuracy": 0.8409090909090909,
|
| 561 |
+
"eval_loss": 0.23692327737808228,
|
| 562 |
+
"eval_runtime": 0.6924,
|
| 563 |
+
"eval_samples_per_second": 63.544,
|
| 564 |
+
"eval_steps_per_second": 2.888,
|
| 565 |
+
"step": 108
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 55.0,
|
| 569 |
+
"grad_norm": 2.8878471851348877,
|
| 570 |
+
"learning_rate": 2.5e-05,
|
| 571 |
+
"loss": 0.1386,
|
| 572 |
+
"step": 110
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 55.0,
|
| 576 |
+
"eval_accuracy": 0.8863636363636364,
|
| 577 |
+
"eval_loss": 0.31426045298576355,
|
| 578 |
+
"eval_runtime": 0.6968,
|
| 579 |
+
"eval_samples_per_second": 63.146,
|
| 580 |
+
"eval_steps_per_second": 2.87,
|
| 581 |
+
"step": 110
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 56.0,
|
| 585 |
+
"eval_accuracy": 0.8863636363636364,
|
| 586 |
+
"eval_loss": 0.29323482513427734,
|
| 587 |
+
"eval_runtime": 0.6884,
|
| 588 |
+
"eval_samples_per_second": 63.916,
|
| 589 |
+
"eval_steps_per_second": 2.905,
|
| 590 |
+
"step": 112
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 57.0,
|
| 594 |
+
"eval_accuracy": 0.9090909090909091,
|
| 595 |
+
"eval_loss": 0.27247121930122375,
|
| 596 |
+
"eval_runtime": 0.6889,
|
| 597 |
+
"eval_samples_per_second": 63.868,
|
| 598 |
+
"eval_steps_per_second": 2.903,
|
| 599 |
+
"step": 114
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 58.0,
|
| 603 |
+
"eval_accuracy": 0.8409090909090909,
|
| 604 |
+
"eval_loss": 0.566352128982544,
|
| 605 |
+
"eval_runtime": 0.6829,
|
| 606 |
+
"eval_samples_per_second": 64.432,
|
| 607 |
+
"eval_steps_per_second": 2.929,
|
| 608 |
+
"step": 116
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 59.0,
|
| 612 |
+
"eval_accuracy": 0.8181818181818182,
|
| 613 |
+
"eval_loss": 0.5874523520469666,
|
| 614 |
+
"eval_runtime": 0.6823,
|
| 615 |
+
"eval_samples_per_second": 64.484,
|
| 616 |
+
"eval_steps_per_second": 2.931,
|
| 617 |
+
"step": 118
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 60.0,
|
| 621 |
+
"grad_norm": 5.098616123199463,
|
| 622 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 623 |
+
"loss": 0.1194,
|
| 624 |
+
"step": 120
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 60.0,
|
| 628 |
+
"eval_accuracy": 0.8636363636363636,
|
| 629 |
+
"eval_loss": 0.46228501200675964,
|
| 630 |
+
"eval_runtime": 0.6928,
|
| 631 |
+
"eval_samples_per_second": 63.507,
|
| 632 |
+
"eval_steps_per_second": 2.887,
|
| 633 |
+
"step": 120
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 61.0,
|
| 637 |
+
"eval_accuracy": 0.8181818181818182,
|
| 638 |
+
"eval_loss": 0.4716045558452606,
|
| 639 |
+
"eval_runtime": 0.6856,
|
| 640 |
+
"eval_samples_per_second": 64.174,
|
| 641 |
+
"eval_steps_per_second": 2.917,
|
| 642 |
+
"step": 122
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 62.0,
|
| 646 |
+
"eval_accuracy": 0.8181818181818182,
|
| 647 |
+
"eval_loss": 0.5028097033500671,
|
| 648 |
+
"eval_runtime": 0.6887,
|
| 649 |
+
"eval_samples_per_second": 63.887,
|
| 650 |
+
"eval_steps_per_second": 2.904,
|
| 651 |
+
"step": 124
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 63.0,
|
| 655 |
+
"eval_accuracy": 0.8181818181818182,
|
| 656 |
+
"eval_loss": 0.4557519257068634,
|
| 657 |
+
"eval_runtime": 0.6871,
|
| 658 |
+
"eval_samples_per_second": 64.04,
|
| 659 |
+
"eval_steps_per_second": 2.911,
|
| 660 |
+
"step": 126
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 64.0,
|
| 664 |
+
"eval_accuracy": 0.8181818181818182,
|
| 665 |
+
"eval_loss": 0.47977954149246216,
|
| 666 |
+
"eval_runtime": 0.6832,
|
| 667 |
+
"eval_samples_per_second": 64.4,
|
| 668 |
+
"eval_steps_per_second": 2.927,
|
| 669 |
+
"step": 128
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 65.0,
|
| 673 |
+
"grad_norm": 4.329418659210205,
|
| 674 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 675 |
+
"loss": 0.1122,
|
| 676 |
+
"step": 130
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 65.0,
|
| 680 |
+
"eval_accuracy": 0.8409090909090909,
|
| 681 |
+
"eval_loss": 0.3826620876789093,
|
| 682 |
+
"eval_runtime": 0.6842,
|
| 683 |
+
"eval_samples_per_second": 64.31,
|
| 684 |
+
"eval_steps_per_second": 2.923,
|
| 685 |
+
"step": 130
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 66.0,
|
| 689 |
+
"eval_accuracy": 0.8409090909090909,
|
| 690 |
+
"eval_loss": 0.3652952313423157,
|
| 691 |
+
"eval_runtime": 0.6915,
|
| 692 |
+
"eval_samples_per_second": 63.63,
|
| 693 |
+
"eval_steps_per_second": 2.892,
|
| 694 |
+
"step": 132
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 67.0,
|
| 698 |
+
"eval_accuracy": 0.8409090909090909,
|
| 699 |
+
"eval_loss": 0.3972075581550598,
|
| 700 |
+
"eval_runtime": 0.6803,
|
| 701 |
+
"eval_samples_per_second": 64.678,
|
| 702 |
+
"eval_steps_per_second": 2.94,
|
| 703 |
+
"step": 134
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 68.0,
|
| 707 |
+
"eval_accuracy": 0.7727272727272727,
|
| 708 |
+
"eval_loss": 0.5704622864723206,
|
| 709 |
+
"eval_runtime": 0.6889,
|
| 710 |
+
"eval_samples_per_second": 63.868,
|
| 711 |
+
"eval_steps_per_second": 2.903,
|
| 712 |
+
"step": 136
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 69.0,
|
| 716 |
+
"eval_accuracy": 0.7727272727272727,
|
| 717 |
+
"eval_loss": 0.5934601426124573,
|
| 718 |
+
"eval_runtime": 0.6795,
|
| 719 |
+
"eval_samples_per_second": 64.752,
|
| 720 |
+
"eval_steps_per_second": 2.943,
|
| 721 |
+
"step": 138
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 70.0,
|
| 725 |
+
"grad_norm": 3.6549811363220215,
|
| 726 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 727 |
+
"loss": 0.1041,
|
| 728 |
+
"step": 140
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 70.0,
|
| 732 |
+
"eval_accuracy": 0.8636363636363636,
|
| 733 |
+
"eval_loss": 0.3905385136604309,
|
| 734 |
+
"eval_runtime": 0.6809,
|
| 735 |
+
"eval_samples_per_second": 64.623,
|
| 736 |
+
"eval_steps_per_second": 2.937,
|
| 737 |
+
"step": 140
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 71.0,
|
| 741 |
+
"eval_accuracy": 0.8409090909090909,
|
| 742 |
+
"eval_loss": 0.27905356884002686,
|
| 743 |
+
"eval_runtime": 0.6869,
|
| 744 |
+
"eval_samples_per_second": 64.051,
|
| 745 |
+
"eval_steps_per_second": 2.911,
|
| 746 |
+
"step": 142
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 72.0,
|
| 750 |
+
"eval_accuracy": 0.9090909090909091,
|
| 751 |
+
"eval_loss": 0.28452473878860474,
|
| 752 |
+
"eval_runtime": 0.6826,
|
| 753 |
+
"eval_samples_per_second": 64.463,
|
| 754 |
+
"eval_steps_per_second": 2.93,
|
| 755 |
+
"step": 144
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 73.0,
|
| 759 |
+
"eval_accuracy": 0.8636363636363636,
|
| 760 |
+
"eval_loss": 0.24005986750125885,
|
| 761 |
+
"eval_runtime": 0.6824,
|
| 762 |
+
"eval_samples_per_second": 64.477,
|
| 763 |
+
"eval_steps_per_second": 2.931,
|
| 764 |
+
"step": 146
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 74.0,
|
| 768 |
+
"eval_accuracy": 0.8863636363636364,
|
| 769 |
+
"eval_loss": 0.2259605973958969,
|
| 770 |
+
"eval_runtime": 0.6828,
|
| 771 |
+
"eval_samples_per_second": 64.437,
|
| 772 |
+
"eval_steps_per_second": 2.929,
|
| 773 |
+
"step": 148
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 75.0,
|
| 777 |
+
"grad_norm": 3.5433380603790283,
|
| 778 |
+
"learning_rate": 1.388888888888889e-05,
|
| 779 |
+
"loss": 0.0982,
|
| 780 |
+
"step": 150
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 75.0,
|
| 784 |
+
"eval_accuracy": 0.8863636363636364,
|
| 785 |
+
"eval_loss": 0.2454221248626709,
|
| 786 |
+
"eval_runtime": 0.6812,
|
| 787 |
+
"eval_samples_per_second": 64.589,
|
| 788 |
+
"eval_steps_per_second": 2.936,
|
| 789 |
+
"step": 150
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 76.0,
|
| 793 |
+
"eval_accuracy": 0.8863636363636364,
|
| 794 |
+
"eval_loss": 0.3772827386856079,
|
| 795 |
+
"eval_runtime": 0.7099,
|
| 796 |
+
"eval_samples_per_second": 61.983,
|
| 797 |
+
"eval_steps_per_second": 2.817,
|
| 798 |
+
"step": 152
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 77.0,
|
| 802 |
+
"eval_accuracy": 0.8181818181818182,
|
| 803 |
+
"eval_loss": 0.618475615978241,
|
| 804 |
+
"eval_runtime": 0.6861,
|
| 805 |
+
"eval_samples_per_second": 64.132,
|
| 806 |
+
"eval_steps_per_second": 2.915,
|
| 807 |
+
"step": 154
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 78.0,
|
| 811 |
+
"eval_accuracy": 0.7727272727272727,
|
| 812 |
+
"eval_loss": 0.7238039374351501,
|
| 813 |
+
"eval_runtime": 0.6796,
|
| 814 |
+
"eval_samples_per_second": 64.746,
|
| 815 |
+
"eval_steps_per_second": 2.943,
|
| 816 |
+
"step": 156
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 79.0,
|
| 820 |
+
"eval_accuracy": 0.8409090909090909,
|
| 821 |
+
"eval_loss": 0.5469216704368591,
|
| 822 |
+
"eval_runtime": 0.687,
|
| 823 |
+
"eval_samples_per_second": 64.051,
|
| 824 |
+
"eval_steps_per_second": 2.911,
|
| 825 |
+
"step": 158
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 80.0,
|
| 829 |
+
"grad_norm": 3.139439344406128,
|
| 830 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 831 |
+
"loss": 0.1065,
|
| 832 |
+
"step": 160
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 80.0,
|
| 836 |
+
"eval_accuracy": 0.8636363636363636,
|
| 837 |
+
"eval_loss": 0.4317708909511566,
|
| 838 |
+
"eval_runtime": 0.6729,
|
| 839 |
+
"eval_samples_per_second": 65.391,
|
| 840 |
+
"eval_steps_per_second": 2.972,
|
| 841 |
+
"step": 160
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 81.0,
|
| 845 |
+
"eval_accuracy": 0.8863636363636364,
|
| 846 |
+
"eval_loss": 0.33475610613822937,
|
| 847 |
+
"eval_runtime": 0.685,
|
| 848 |
+
"eval_samples_per_second": 64.237,
|
| 849 |
+
"eval_steps_per_second": 2.92,
|
| 850 |
+
"step": 162
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 82.0,
|
| 854 |
+
"eval_accuracy": 0.8636363636363636,
|
| 855 |
+
"eval_loss": 0.3040803372859955,
|
| 856 |
+
"eval_runtime": 0.7242,
|
| 857 |
+
"eval_samples_per_second": 60.758,
|
| 858 |
+
"eval_steps_per_second": 2.762,
|
| 859 |
+
"step": 164
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 83.0,
|
| 863 |
+
"eval_accuracy": 0.8863636363636364,
|
| 864 |
+
"eval_loss": 0.33496472239494324,
|
| 865 |
+
"eval_runtime": 0.6781,
|
| 866 |
+
"eval_samples_per_second": 64.888,
|
| 867 |
+
"eval_steps_per_second": 2.949,
|
| 868 |
+
"step": 166
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 84.0,
|
| 872 |
+
"eval_accuracy": 0.8863636363636364,
|
| 873 |
+
"eval_loss": 0.34641969203948975,
|
| 874 |
+
"eval_runtime": 0.6832,
|
| 875 |
+
"eval_samples_per_second": 64.4,
|
| 876 |
+
"eval_steps_per_second": 2.927,
|
| 877 |
+
"step": 168
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 85.0,
|
| 881 |
+
"grad_norm": 3.164541482925415,
|
| 882 |
+
"learning_rate": 8.333333333333334e-06,
|
| 883 |
+
"loss": 0.0829,
|
| 884 |
+
"step": 170
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 85.0,
|
| 888 |
+
"eval_accuracy": 0.8863636363636364,
|
| 889 |
+
"eval_loss": 0.33745819330215454,
|
| 890 |
+
"eval_runtime": 0.6769,
|
| 891 |
+
"eval_samples_per_second": 64.998,
|
| 892 |
+
"eval_steps_per_second": 2.954,
|
| 893 |
+
"step": 170
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 86.0,
|
| 897 |
+
"eval_accuracy": 0.8863636363636364,
|
| 898 |
+
"eval_loss": 0.33091866970062256,
|
| 899 |
+
"eval_runtime": 0.6868,
|
| 900 |
+
"eval_samples_per_second": 64.069,
|
| 901 |
+
"eval_steps_per_second": 2.912,
|
| 902 |
+
"step": 172
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 87.0,
|
| 906 |
+
"eval_accuracy": 0.8863636363636364,
|
| 907 |
+
"eval_loss": 0.33253809809684753,
|
| 908 |
+
"eval_runtime": 0.6828,
|
| 909 |
+
"eval_samples_per_second": 64.44,
|
| 910 |
+
"eval_steps_per_second": 2.929,
|
| 911 |
+
"step": 174
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 88.0,
|
| 915 |
+
"eval_accuracy": 0.8863636363636364,
|
| 916 |
+
"eval_loss": 0.3441101312637329,
|
| 917 |
+
"eval_runtime": 0.6847,
|
| 918 |
+
"eval_samples_per_second": 64.259,
|
| 919 |
+
"eval_steps_per_second": 2.921,
|
| 920 |
+
"step": 176
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 89.0,
|
| 924 |
+
"eval_accuracy": 0.8636363636363636,
|
| 925 |
+
"eval_loss": 0.34555572271347046,
|
| 926 |
+
"eval_runtime": 0.7006,
|
| 927 |
+
"eval_samples_per_second": 62.805,
|
| 928 |
+
"eval_steps_per_second": 2.855,
|
| 929 |
+
"step": 178
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 90.0,
|
| 933 |
+
"grad_norm": 2.9145150184631348,
|
| 934 |
+
"learning_rate": 5.555555555555556e-06,
|
| 935 |
+
"loss": 0.0902,
|
| 936 |
+
"step": 180
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 90.0,
|
| 940 |
+
"eval_accuracy": 0.8636363636363636,
|
| 941 |
+
"eval_loss": 0.32443130016326904,
|
| 942 |
+
"eval_runtime": 0.6801,
|
| 943 |
+
"eval_samples_per_second": 64.698,
|
| 944 |
+
"eval_steps_per_second": 2.941,
|
| 945 |
+
"step": 180
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 91.0,
|
| 949 |
+
"eval_accuracy": 0.8636363636363636,
|
| 950 |
+
"eval_loss": 0.3126292824745178,
|
| 951 |
+
"eval_runtime": 0.6994,
|
| 952 |
+
"eval_samples_per_second": 62.913,
|
| 953 |
+
"eval_steps_per_second": 2.86,
|
| 954 |
+
"step": 182
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 92.0,
|
| 958 |
+
"eval_accuracy": 0.8636363636363636,
|
| 959 |
+
"eval_loss": 0.3116739094257355,
|
| 960 |
+
"eval_runtime": 0.6939,
|
| 961 |
+
"eval_samples_per_second": 63.409,
|
| 962 |
+
"eval_steps_per_second": 2.882,
|
| 963 |
+
"step": 184
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 93.0,
|
| 967 |
+
"eval_accuracy": 0.8636363636363636,
|
| 968 |
+
"eval_loss": 0.28766298294067383,
|
| 969 |
+
"eval_runtime": 0.6892,
|
| 970 |
+
"eval_samples_per_second": 63.845,
|
| 971 |
+
"eval_steps_per_second": 2.902,
|
| 972 |
+
"step": 186
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 94.0,
|
| 976 |
+
"eval_accuracy": 0.8636363636363636,
|
| 977 |
+
"eval_loss": 0.26428502798080444,
|
| 978 |
+
"eval_runtime": 0.7034,
|
| 979 |
+
"eval_samples_per_second": 62.55,
|
| 980 |
+
"eval_steps_per_second": 2.843,
|
| 981 |
+
"step": 188
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 95.0,
|
| 985 |
+
"grad_norm": 2.80922269821167,
|
| 986 |
+
"learning_rate": 2.777777777777778e-06,
|
| 987 |
+
"loss": 0.0838,
|
| 988 |
+
"step": 190
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 95.0,
|
| 992 |
+
"eval_accuracy": 0.8863636363636364,
|
| 993 |
+
"eval_loss": 0.25247523188591003,
|
| 994 |
+
"eval_runtime": 0.6832,
|
| 995 |
+
"eval_samples_per_second": 64.4,
|
| 996 |
+
"eval_steps_per_second": 2.927,
|
| 997 |
+
"step": 190
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 96.0,
|
| 1001 |
+
"eval_accuracy": 0.9090909090909091,
|
| 1002 |
+
"eval_loss": 0.24620772898197174,
|
| 1003 |
+
"eval_runtime": 0.6948,
|
| 1004 |
+
"eval_samples_per_second": 63.325,
|
| 1005 |
+
"eval_steps_per_second": 2.878,
|
| 1006 |
+
"step": 192
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 97.0,
|
| 1010 |
+
"eval_accuracy": 0.9090909090909091,
|
| 1011 |
+
"eval_loss": 0.24165372550487518,
|
| 1012 |
+
"eval_runtime": 0.6884,
|
| 1013 |
+
"eval_samples_per_second": 63.918,
|
| 1014 |
+
"eval_steps_per_second": 2.905,
|
| 1015 |
+
"step": 194
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 98.0,
|
| 1019 |
+
"eval_accuracy": 0.9090909090909091,
|
| 1020 |
+
"eval_loss": 0.24015091359615326,
|
| 1021 |
+
"eval_runtime": 0.6834,
|
| 1022 |
+
"eval_samples_per_second": 64.382,
|
| 1023 |
+
"eval_steps_per_second": 2.926,
|
| 1024 |
+
"step": 196
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 99.0,
|
| 1028 |
+
"eval_accuracy": 0.9090909090909091,
|
| 1029 |
+
"eval_loss": 0.24094568192958832,
|
| 1030 |
+
"eval_runtime": 0.6957,
|
| 1031 |
+
"eval_samples_per_second": 63.249,
|
| 1032 |
+
"eval_steps_per_second": 2.875,
|
| 1033 |
+
"step": 198
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 100.0,
|
| 1037 |
+
"grad_norm": 2.4933438301086426,
|
| 1038 |
+
"learning_rate": 0.0,
|
| 1039 |
+
"loss": 0.0747,
|
| 1040 |
+
"step": 200
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 100.0,
|
| 1044 |
+
"eval_accuracy": 0.9090909090909091,
|
| 1045 |
+
"eval_loss": 0.24256624281406403,
|
| 1046 |
+
"eval_runtime": 0.6904,
|
| 1047 |
+
"eval_samples_per_second": 63.734,
|
| 1048 |
+
"eval_steps_per_second": 2.897,
|
| 1049 |
+
"step": 200
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 100.0,
|
| 1053 |
+
"step": 200,
|
| 1054 |
+
"total_flos": 1.8822505078960128e+18,
|
| 1055 |
+
"train_loss": 0.19619473487138747,
|
| 1056 |
+
"train_runtime": 1209.7214,
|
| 1057 |
+
"train_samples_per_second": 20.087,
|
| 1058 |
+
"train_steps_per_second": 0.165
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 100.0,
|
| 1062 |
+
"eval_accuracy": 0.9318181818181818,
|
| 1063 |
+
"eval_loss": 0.24986252188682556,
|
| 1064 |
+
"eval_runtime": 0.7488,
|
| 1065 |
+
"eval_samples_per_second": 58.758,
|
| 1066 |
+
"eval_steps_per_second": 2.671,
|
| 1067 |
+
"step": 200
|
| 1068 |
+
}
|
| 1069 |
+
],
|
| 1070 |
+
"logging_steps": 10,
|
| 1071 |
+
"max_steps": 200,
|
| 1072 |
+
"num_input_tokens_seen": 0,
|
| 1073 |
+
"num_train_epochs": 100,
|
| 1074 |
+
"save_steps": 500,
|
| 1075 |
+
"total_flos": 1.8822505078960128e+18,
|
| 1076 |
+
"train_batch_size": 32,
|
| 1077 |
+
"trial_name": null,
|
| 1078 |
+
"trial_params": null
|
| 1079 |
+
}
|