Interpretability
概念
安全与对齐
Opening the black box: reading a model's internal features.
您正在查看一条概念解释记录：该术语的通俗含义，以及它所关联的各条目。
Mechanistic interpretability reverse-engineers what networks compute — Anthropic's team mapped millions of human-legible features in Claude (the 'Golden Gate Claude' demo made one famous) and traces circuits for planning and deception. The bet: we cannot fully trust what we cannot inspect, so understanding must catch up with capability before the stakes get too high.
它如何与整个技术栈相连
本记录的关联链接一览——点击分组可直接跳转。
来源
Mapping the mind of an LLM (Anthropic) ↗
记录审阅截至 2026-01-15 — 查看 方法论.