framework
active
framework:sae-latent-allocation-model

SAE latent allocation model

Reduces SAE optimization to allocating N latents across features.

Neighborhood — ranked by edge-count

Methods (1)

method

Frameworks (3)

framework

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

  • SAE Latentsconcept0.827
    Interpretable features extracted by sparse autoencoders used as steering targets in this study
  • Adding a multiple of the SAE latent decoder vector to token activations to causally test each latent's role in misalignment
  • Using GPT-4o or o3 to automatically generate interpretations of SAE latents from top-activating examples
  • Pre-filtering step excluding latents naturally activated by each prompt to ensure genuine off-topic steering
  • Interpretability method criticized in this paper for shattering manifolds into atomic pieces, obscuring overarching semantic structure.
  • Per-Token SAEframework0.747
    Standard baseline SAE operating on individual token activations.
  • Nested SAEframework0.743
    Hybrid SAE jointly training turn-averaged inner and per-token outer partitions.
  • SAE featuresconcept0.741
    The individual, supposedly monosemantic directions learned by SAEs; argued here to fragment manifolds into disconnected pieces.