Measuring Obedience to Authority Across Large Language Models with the Milgram Paradigm | ResearchPod