Update app.py

app.py

@@ -39,7 +39,7 @@ def predict(message, history):
     
     outputs = model.generate(**inputs,streamer=streamer, generation_config=generation_config)
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
+    return response[len(formatted_prompt(message))-1:]
 
     # partial_message  = ""
     # for i in response: