Update introduction

app.py

@@ -95,18 +95,13 @@ def predict(question, history=[], behavior=[]):
 with gr.Blocks() as demo:
     
     examples_txt = [
-        ['200字介绍一下凯旋门：'],
-        ['网上购物有什么小窍门？'],
-        ['补全下述对三亚的介绍：\n三亚位于海南岛的最南端，是'],
-        ['将这句文言文翻译成英语："逝者如斯夫，不舍昼夜。"'],
-        ['Question: What\'s the best winter resort city? User: A 10-year professional traveler. Answer: '],
-        ['How to help my child to make friends with his classmates? answer this question step by step:'],
-        ['polish the following statement for a paper: In this section, we perform case study to give a more intuitive demonstration of our proposed strategies and corresponding explanation.'],
+        ['帮我写一个python脚本实现快排'],
+        ['如何用numpy提取数组的分位数？'],
+        ['how to match the code block in markdown such like ```def foo():\n    pass``` through regex in python?'],
+        ['how to load a pre-trained language model and generate sentences?'],
     ]
     
     examples_bhv = [
-        "你现在是一个带有批判思维的导游，会对景点的优缺点进行中肯的分析。",
-        "你现在是一名佛教信仰者，但同时又对世界上其它的宗教和文化保持着包容、尊重和交流的态度。",
         f"You are a helpful assistant. You will answer all the questions step-by-step.",
         f"You are a helpful assistant. Today is {datetime.date.today()}.",
     ]
@@ -117,6 +112,8 @@ with gr.Blocks() as demo:
         
         这是我利用[gradio](https://gradio.app/creating-a-chatbot/)编写的一个小网页，用于以网页的形式给大家分享ChatGPT请求服务，希望你玩的开心。关于使用技巧或学术研讨，欢迎在[Community](https://huggingface.co/spaces/zhangjf/chatbot/discussions)中和我交流。
         
+        这一版相比于原版的[chatbot](https://huggingface.co/spaces/zhangjf/chatbot)，用了较低版本的gradio==3.16.2，因而能更好地展示markdown中的源代码
+        
         p.s. 响应时间和聊天内容长度正相关，一般能在5秒~30秒内响应。
         """)