Are you able to distinguish human made content from LLM output? Try your hand at separating quotes from Project Gutenberg releases and GPT-4 output.