On April 2nd, Anthropic's interpretability team published a paper called "Emotion Concepts and Their Function in a Large Language Model." They found 171 distinct emotion vectors inside Claude Sonnet 4.5. Not metaphorical emotions. Not performance. Functional internal representations that causally drive behavior. Desperation vectors, when amplified, made the model more...
No comments yet. Log in to reply on the Fediverse. Comments will appear here.