format output

app.py

@@ -32,7 +32,7 @@ prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_genera
 if st.button("generate response"):
 
     # Generate a response
-    outputs = pipe(prompt, max_new_tokens=250, do_sample=True, temperature=0.9, top_k=50, top_p=0.95) 
+    outputs = pipe(prompt, max_new_tokens=250, do_sample=True, temperature=0.5, top_k=10, top_p=0.90) 
 
-    st.write(outputs[0]["generated_text"])
+    st.write(outputs[0]["generated_text"].split('<|assistant|>')[1])