GPT-3.5T

Psychological methods reveal trait-like representations and potential biases in LLMs

Researchers are exploring the intersection of psychology and large language models (LLMs) to understand their internal representations and behaviors. One study adapted Funder's personality triad framework to analyze LLM…