Update app.py

app.py

@@ -42,8 +42,8 @@ def respond(message, chat_history):
     # Generate response
     generation_args = {
         "max_new_tokens": 200,  # Reduced for CPU performance
-        "temperature": 0.0,
-        "do_sample": False,
+        "temperature": 0.7,
+        "do_sample": True,
     }
     
     output = pipe(formatted_input, **generation_args)