Update README.md

README.md

@@ -18,12 +18,14 @@ With trained embed and head, the model perform better at Chinese and Japanese th
 Since this model is trained by guanaco dataset, you can also use this as chatbot. just use this format:
 ```
 ### Instruction:
-User: xxx
-Assistant: yyy
+User: <Message history>
+Assistant: <Message history>
 
-### Response:
+### Input:
 System: <System response for next message, optional>
 User: <Next message>
+
+### Response:
 ```
 
 **Tips: I just removed the first line of original prompt to reduce token comsumption, plz consider remove it when you want to use this model**