Scale, Drivers, and Limits of Representational Alignment Across Large Language Models

Jialin Wu, Meenakshi Khosla

Comparing large language models can reveal shared representational structure and systematic differences between them. However, aggregate similarity measures can obscure how agreeme…