finding
active
finding:factuality-degrades-with-steering-strength-at-0-6-only-galaxy-size-is-wrong-at-1-0-many-facts-incorrect-and-rescaled-to-nautical-unitsFactuality degrades with steering strength: at 0.6 only galaxy size is wrong, at 1.0 many facts incorrect and rescaled to nautical units
Pirate steering demo.
Neighborhood — ranked by edge-count
Papers (1)
paper
Questions (1)
question
- Explicitly unknown mechanism.
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.
- At steering strength 1.4 all factual claims are incorrect yet the LLM evaluator still gives a high scorefinding0.788Pitfall of LLM-based steering evaluation.
- Asymmetry suggesting more things must go right.
- Limit of LRH regime.
- Prediction 2: if LRH holds, steering raises belief at a sigmoidal rate with magnitude mhypothesis0.769From Eq. 8.
- Key methodological insight: introspection enables a new probe validation criterion beyond conventional separation metrics
- Pinpoints list-length 3 as the exact boundary where genuine counting introduces the limitation.
- Comparative claim between the two steering strategies
- If a model doesn't represent a concept at all, no amount of steering will change its beliefhypothesis0.761Explanation for phi-4 null effect.