concept
active
concept:sae-latentsSAE Latents
Interpretable features extracted by sparse autoencoders used as steering targets in this study
Neighborhood — ranked by edge-count
Concepts (1)
concept
- Relevance Filtering of SAE Latentsrelated_toPre-filtering step excluding latents naturally activated by each prompt to ensure genuine off-topic steering
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.
- Using GPT-4o or o3 to automatically generate interpretations of SAE latents from top-activating examples
- Adding a multiple of the SAE latent decoder vector to token activations to causally test each latent's role in misalignment
- Reduces SAE optimization to allocating N latents across features.
- Pre-filtering step excluding abstract latents where off-topic detection is harder
- Interpretability method criticized in this paper for shattering manifolds into atomic pieces, obscuring overarching semantic structure.
- Standard baseline SAE operating on individual token activations.
- The individual, supposedly monosemantic directions learned by SAEs; argued here to fragment manifolds into disconnected pieces.
- Internal concepts a model uses, extracted as human-interpretable components.