Update app.py

app.py

@@ -24,6 +24,8 @@ def run(message, chat_history, system_prompt, max_new_tokens=1024, temperature=0
         top_p=top_p,              # Nucleus sampling: higher values allow more diversity.
         top_k=top_k,              # Top-k sampling: limits the number of top tokens considered.
         pad_token_id=tokenizer.eos_token_id
+        do_sample=True            # Enable sampling-based generation
+
     )
 
     # Decode the response