claim
active
claim:auditing-game-win-rates-could-be-used-as-a-new-approach-to-evaluate-novel-saesAuditing game win rates could be used as a new approach to evaluate novel SAEs
Win rate correlates with SAE feature quality.
Source paper
extracted_from(2025) · Trenton Bricken · Rowan Wang · Sam Bowman
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (2)
finding
- RM features at rank 181, outside top 100.
- No features correspond to Oxford study.
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.
- Methodology measuring interpretability tool uplift for an agent uncovering hidden behaviors.
- Unclear per Section 3.1.
- Proposed future direction.
- Aggregating 10 reports.
- Interpretability tools and semantic search are key to winning the Marks et al. auditing gameclaim0.725Affordance insight.
- If agentic self-steering evaluation proves robust, it might be used to better explain and interpret SAE features in generalhypothesis0.725Speculative claim about scaling introspective access to general SAE feature interpretation
- Realistic affordances result.
- Shows gating effect is specific to the self-referential computational regime, not a general feature effect