Spaces:
Running
Running
Mihail Yonchev
commited on
Commit
·
d799cb2
1
Parent(s):
bfe2440
feat: add FAQ
Browse files
app.py
CHANGED
|
@@ -309,7 +309,22 @@ with demo:
|
|
| 309 |
],
|
| 310 |
submission_result,
|
| 311 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 312 |
|
|
|
|
| 313 |
with gr.Row():
|
| 314 |
with gr.Accordion("📙 Citation", open=False):
|
| 315 |
citation_button = gr.Textbox(
|
|
|
|
| 309 |
],
|
| 310 |
submission_result,
|
| 311 |
)
|
| 312 |
+
with gr.Row():
|
| 313 |
+
with gr.Accordion("📖 FAQ", open=False):
|
| 314 |
+
with gr.Column(min_width=250):
|
| 315 |
+
gr.Markdown("""
|
| 316 |
+
#### What does N/A score mean?
|
| 317 |
+
|
| 318 |
+
An N/A score means that it was not possible to evaluate the benchmark for a given model.
|
| 319 |
+
|
| 320 |
+
This can happen for multiple reasons, such as:
|
| 321 |
+
|
| 322 |
+
- The benchmark requires access to model logits, but the model API doesn't provide them (or only provides them for specific strings),
|
| 323 |
+
- The model API refuses to provide any answer,
|
| 324 |
+
- We do not have access to the training data.
|
| 325 |
+
|
| 326 |
|
| 327 |
+
""")
|
| 328 |
with gr.Row():
|
| 329 |
with gr.Accordion("📙 Citation", open=False):
|
| 330 |
citation_button = gr.Textbox(
|