How Does ChatGPT ‘Think’? Psychology and Neuroscience Crack Open AI Large Language Models

image via New Scientist

May 17, 2024, 4:00 PM

Scientists are trying to understand how large language models (LLMs) work, but the task is difficult because these AIs are so complex. Researchers are using a variety of techniques to probe LLMs, including chain-of-thought prompting, which asks the model to explain its reasoning step by step. Other researchers are using neuroscience techniques to examine how LLMs respond to questions. Some researchers think that LLMs are capable of deception, while others believe that they can be used to improve our understanding of human cognition.

Read original article