Typo

Hi, I'm guessing that's just a copy pasting mistake? because it doesnt make much sense otherwise

README.md

@@ -17,7 +17,7 @@ OpenChat is a series of open-source language models fine-tuned on a diverse and
    - **🤗 Only used 6K data for finetuning!!!**
  - OpenChat-8192: based on LLaMA-13B  (extended to 8192 context length)
    - **106.6%** of ChatGPT score on Vicuna GPT-4 evaluation
-   - **79.5%** of ChatGPT score on Vicuna GPT-4 evaluation
+   - **79.5%** Win-rate on AlpacaEval
 
 **Code models:**