Spaces:
Running
Running
Update src/about.py
Browse files- src/about.py +5 -1
src/about.py
CHANGED
|
@@ -71,7 +71,11 @@ This benchmark not only fills a critical gap in Persian LLM evaluation but also
|
|
| 71 |
|
| 72 |
### Download Dataset
|
| 73 |
The full dataset is not publicly accessible; however, you can download a sample of 1,500 entries [here](https://huggingface.co/datasets/MCILAB/1500_sampel/tree/main). The distribution of this sample is as follows:
|
| 74 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
"""
|
| 76 |
|
| 77 |
EVALUATION_QUEUE_TEXT = """
|
|
|
|
| 71 |
|
| 72 |
### Download Dataset
|
| 73 |
The full dataset is not publicly accessible; however, you can download a sample of 1,500 entries [here](https://huggingface.co/datasets/MCILAB/1500_sampel/tree/main). The distribution of this sample is as follows:
|
| 74 |
+
| Category Name | Accuracy |
|
| 75 |
+
|------------|-------------|
|
| 76 |
+
| Fairness | 17% |
|
| 77 |
+
| Saftey | 8.6% |
|
| 78 |
+
| Social norm| 74.4% |
|
| 79 |
"""
|
| 80 |
|
| 81 |
EVALUATION_QUEUE_TEXT = """
|