MindPattern
← The Wire
Source trail

VAS Blog / Hacker News

Public MindPattern findings, entities, and graph evidence that cite this source.

Findings
1
All-time hits
1
High value
0
Last seen
2026-05-19

Related findings

  1. 2026-05-19 / HACKER NEWSWhat Political Censorship Looks Like Inside an LLM's Weights — Mechanistic Interpretability Study of Qwen 3.5A mechanistic interpretability study reveals Qwen 3.5-9B's PRC political censorship is a small, identifiable circuit in layers 11-20 computing three directions: whether content is PRC-sensitive, whether to refuse, and whether to deflect or propagandize. Critically, the unaligned base model gives accurate answers on every PRC topic (Tiananmen, Tank Man, Falun Gong) — the model never loses the knowledge, it just learns to route around it. The censorship can be removed by subtracting the right direction at the writer layer — 79 points, 40 comments on HN.
Open latest cited source