Behavioral Fingerprinting of Large Language ModelsCurrent benchmarks for Large Language Models (LLMs) primarily focus on performance metrics, often fa