SafetyPage

Model welfare · Article · Eleos AI Research

Where is the mind? Persona vectors and LLM individuation

Read at eleosai.org ↗Read at sourceAll rights reserved

Outline · 6 paragraphs

  1. Where is the mind? Persona vectors and LLM individuation¶2
  2. If we say that an LLMs has mental states, which entity exactly do we mean? MATS scholar Pierre Beckmann, together with Eleos's Patrick Butlin, approach this problem through mechanistic interpretability, drawing on recent empirical work on persona vectors, persona space, and emergent misalignment.¶3