J-lens Enables Direct Reading of Personality Neurons in Models

Developer @a9lim demonstrated J-lens, a new interpretability tool that directly reads personality-related neurons such as pirate and horse from model internals, with the full repository to be open-sourced soon alongside @treetowntree.

2026-10-29 ~ 2026-10-29 · 2 related posts

• Reading neurons directly via J-lens: open-source repo coming soon — burny_tech · 2026-10-29
• Reading neurons directly via J-lens: persona neurons ranked, repo coming soon — burny_tech · 2026-10-29