On LLM Wizards: Identifying Large Language Models' Behaviors for Wizard of Oz Experiments | Read Paper on Bytez