Hypothesizing the Formation Mechanism of J-space
willdepue · x · 2026-07-13
The author ponders whether preventing attention gradients from flowing to past tokens could stop the formation of J-space. They suspect this might be the primary reason models perform lookahead calculations; without this mechanism, forming J-space through circuit sharing alone would be quite difficult.
Related event: Anthropic Reveals Claude's Hidden Reasoning Space: J-space(16 posts)→
More from Research
- Token Collapse: Study Finds Graph Soft Prompting Tokens in LLMs Fail to Encode Structure — Sapienza · 2026-10-27
- Terence Tao's SAIR Foundation revamps Mathathon with $1.5M compute per team — _sathvikr · 2026-10-27
- Mathathon 2.0 launches with Terence Tao's SAIR, $1.5M in compute for teams — _sathvikr · 2026-10-27
- The 'sweet lesson' is real: two kinds of ML problems need opposite strategies — mike64_t · 2026-10-27
- PixGR: Pixel-space generative decoding for fast feed-forward 3D reconstruction — kwangmoo_yi · 2026-10-27
- UniAction: unified action conditioning lets one world model handle mixed action types — ZeYanjie · 2026-10-27