Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -25,7 +25,7 @@ def make_leaderboard_md(elo_results):
|
|
25 |
- [MT-Bench](https://arxiv.org/abs/2306.05685) - a set of challenging multi-turn questions. We use GPT-4 to grade the model responses.
|
26 |
- [MMLU](https://arxiv.org/abs/2009.03300) (5-shot) - a test to measure a model's multitask accuracy on 57 tasks.
|
27 |
|
28 |
-
💻 Code: The Arena Elo ratings are computed by this [notebook]({notebook_url}). The MT-bench scores (single-answer grading on a scale of 10) are computed by [fastchat.llm_judge](https://github.com/lm-sys/FastChat/tree/main/fastchat/llm_judge). The MMLU scores are computed by [InstructEval](https://github.com/declare-lab/instruct-eval)
|
29 |
"""
|
30 |
return leaderboard_md
|
31 |
|
|
|
25 |
- [MT-Bench](https://arxiv.org/abs/2306.05685) - a set of challenging multi-turn questions. We use GPT-4 to grade the model responses.
|
26 |
- [MMLU](https://arxiv.org/abs/2009.03300) (5-shot) - a test to measure a model's multitask accuracy on 57 tasks.
|
27 |
|
28 |
+
💻 Code: The Arena Elo ratings are computed by this [notebook]({notebook_url}). The MT-bench scores (single-answer grading on a scale of 10) are computed by [fastchat.llm_judge](https://github.com/lm-sys/FastChat/tree/main/fastchat/llm_judge). The MMLU scores are mostly computed by [InstructEval](https://github.com/declare-lab/instruct-eval). Higher values are better for all benchmarks. Empty cells mean not available. Last updated: Sept, 2023.
|
29 |
"""
|
30 |
return leaderboard_md
|
31 |
|