Faithfull-and-Explainable-LLMs Investigating faithfulness and explanability of large language Model self-explanations