Latent Structure of Affective Representations in Large Language Models

Benjamin J. Choi, Melanie Weber

The geometric structure of latent representations in large language models (LLMs) is an active area of research, driven in part by its implications for model transparency and AI sa…