Reproducing J-space to Read Hidden Thoughts in Llama Models

A developer reproduced Anthropic's J-space research on Llama-3.3-70B, claiming to capture the model's hidden thoughts using NLA. The experiment splits concepts into conscious and subconscious levels, offering a new perspective on LLM internal mechanisms.

2026-07-10 ~ 2026-07-11 · 2 related posts

Full story(5 episodes)→