Find the misspelled parameter for capping output tokens.
response = client.chat.completions.create( model="gpt-4o-mini", messages=[{"role": "user", "content": "Write a poem"}], max_token=50, )
What is the bug in limiting the output length?