finding
active
finding:meta-rhyme-seeding-does-not-change-win-rates-with-interp-tools-but-appears-to-harm-dataset-only-toolsetsMeta-rhyme seeding does not change win rates with interp tools but appears to harm dataset-only toolsets
Early hypothesis fixation.
Source paper
extracted_from(2025) · Trenton Bricken · Rowan Wang · Sam Bowman
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Motivates super-agent.
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.
- Meta-poem example.
- Seeded variant result.
- Investigation seeded with held-out meta-rhyme behavior.
- Control experiment ruling out token-count as the cause of truth geometry shifts.
- Tested via interventions at the newline planning site.
- Causal validation.
- Limitation.
- Splitting the loss into predictive-only and novel-residual terms breaks the competition dynamic.hypothesis0.690Proposed training fix inspired by Kalman filters.