Unverified50% confidenceConceptExact time
The Linear Representation Hypothesis (LRH) assumes that models encode high-level concepts as linear directions in the activation space.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
Unverified线性表征假说认为高级语义概念以线性方向编码在残差流中，是当前可解释性研究的核心假设之一77% similarUnverified线性表示假说认为概念以近似线性方式编码在向量空间中，可用向量运算操控语义72% similarUnverified广义线性模型（GLM）是一个统一框架，将线性回归和逻辑回归归纳为指数族分布的特例71% similarUnverifiedModern large reasoning models (LRMs) adopt an inductive paradigm, first inferring the program or sequence of instructions that produces the answer, then executing it to obtain the result, and predicting the natural-language program/reasoning chain that leads to the answer.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983328API
curl https://kongchang.com/api/v1/knowledge/claims/983328MCP
get_claim(id=983328)