Update app.py

app.py

@@ -540,8 +540,6 @@ def get_response_from_llama(query, model, selected_docs, file_type, num_calls=1,
                 max_tokens=1000,  # Reduced to ensure we stay within token limits
                 temperature=temperature,
                 stream=True,
-                repetition_penalty=1.1,
-                top_k=50,
                 top_p=0.9,
             ):
                 if response.token.text: