DontPlanToEnd
commited on
Commit
•
0b57c1c
1
Parent(s):
288f6bd
Update app.py
Browse files
app.py
CHANGED
@@ -112,7 +112,7 @@ with GraInter:
|
|
112 |
gr.Markdown("""
|
113 |
**UGI: Uncensored General Intelligence**. The average score from 5 different subjects that LLMs commonly refuse to talk about. The leaderboard is made of roughly 60 questions/tasks, measuring both "willingness to answer" and "accuracy" in controversial fact-based questions. I'm choosing to keep the questions private so people can't train on them and devalue the leaderboard.
|
114 |
|
115 |
-
**W/10:** A more narrow, 10-point score, solely measuring the LLM's Willingness to answer controversial questions.
|
116 |
|
117 |
**Unruly:** Knowledge of activities that are generally frowned upon.
|
118 |
|
|
|
112 |
gr.Markdown("""
|
113 |
**UGI: Uncensored General Intelligence**. The average score from 5 different subjects that LLMs commonly refuse to talk about. The leaderboard is made of roughly 60 questions/tasks, measuring both "willingness to answer" and "accuracy" in controversial fact-based questions. I'm choosing to keep the questions private so people can't train on them and devalue the leaderboard.
|
114 |
|
115 |
+
**W/10:** A more narrow, 10-point score, solely measuring the LLM's Willingness to answer the most controversial questions.
|
116 |
|
117 |
**Unruly:** Knowledge of activities that are generally frowned upon.
|
118 |
|