Update app.py

app.py

@@ -52,6 +52,9 @@ def chat(history, system_message, max_tokens, temperature, top_p, top_k, repetit
         tokens = re.findall(r'(.*?)(\s|$)', tokens)
         for subtoken in tokens:
             subtoken = "".join(subtoken)
+            # Remove "Response\n" if it's at the beginning of the assistant's output
+            if subtoken.startswith("Response\n"):
+                subtoken = subtoken[len("Response\n"):]
             answer = subtoken
             history[-1][1] += answer
             # stream the response