hypothesis
active
hypothesis:if-a-model-doesn-t-represent-a-concept-at-all-no-amount-of-steering-will-change-its-beliefIf a model doesn't represent a concept at all, no amount of steering will change its belief
Explanation for phi-4 null effect.
Source paper
extracted_from(2025) · Eric Bigelow · Daniel Wurgaft · YingQiao Wang · Noah D. Goodman +3
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (1)
finding
- Limitation.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- What if the concept being manipulated does not lie on a straight line in the model's representations?question0.832The motivating question that opens the paper and leads to the development of manifold steering.
- Central interpretive claim organizing the entire paper's results
- The core motivating question of the paper, framed by Christiano et al. (2021)
- Key methodological insight: introspection enables a new probe validation criterion beyond conventional separation metrics
- A linear model predicts out-of-sample steering effects on queried concepts from manifold distancesfinding0.781Predictive geometry.
- The paper's critique of the standard linear steering baseline, supported by the days-of-week demo.
- Additional evidence that core representations are persona-relative, supporting Claim about persona-relative representations