finding
active
finding:sae-without-synthetic-data-yields-no-auditing-winsSAE without synthetic data yields no auditing wins
No features correspond to Oxford study.
Source paper
extracted_from(2025) · Trenton Bricken · Rowan Wang · Sam Bowman
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Win rate correlates with SAE feature quality.
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.
- App. B.
- Prediction following diversity-undertraining logic.
- Limitation statement.
- RM features at rank 181, outside top 100.
- If agentic self-steering evaluation proves robust, it might be used to better explain and interpret SAE features in generalhypothesis0.751Speculative claim about scaling introspective access to general SAE feature interpretation
- Extension of mechanistic interpretability findings to the metacognitive domain
- Shows gating effect is specific to the self-referential computational regime, not a general feature effect
- A promising property for interpretability analysis off-distribution.