question
active
question:is-the-transcoder-capturing-the-true-mechanismsIs the transcoder capturing the true mechanisms?
Defines the faithfulness category.
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Authors' prioritization.
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.
- Why should we believe our transcoders learn the same mechanisms as the original underlying model?question0.805Motivating question.
- Core motivation.
- Discussion.
- Second category of SAE/transcoder concerns, distinct from faithfulness.
- Uses activations from one layer to predict the next; special case of crosscoders.
- Used to express MLP in monosemantic features, 4096 features with JumpReLU.
- Motivating problem.
- Partial mitigation of unfaithfulness.