claim
active
claim:steering-changes-behavior-regardless-of-input-x-as-a-constant-shift-in-log-posterior-oddsSteering changes behavior regardless of input x, as a constant shift in log posterior odds
Derived from LRH.
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
- Limit of LRH regime.
Claims (1)
claim
- Mechanistic split.
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.
- Prediction 3.
- Prediction 3: log posterior odds additive in context and steering, with crossover N* predictable from prior odds and magnitudehypothesis0.832Eq. 10.
- Mechanism claim supported by transcript analysis and the fact that the steering vector was extracted from a model that never writes type hints.
- Core mechanism statement.
- Central interpretive claim organizing the entire paper's results
- The paper's critique of the standard linear steering baseline, supported by the days-of-week demo.
- Do latents for precise distinctions in types of behaviour have distinct steering effects?question0.772Authors have not verified this.
- Validates using chain-of-thought belief monitoring as proxy for behavioral steering efficacy.