causal masking

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

• causally-masked attentionmethod0.847
  Attention mechanism with causal mask limiting each token's view to previous tokens; used in decoder-only transformers
• Causal Mediationconcept0.829
  Whether an internal direction causally controls a target behavior, verified by intervention success
• Causal Attention Maskmethod0.826
  Modification to transformer restricting keys and values to previous time-steps only, mimicking how an agent accumulates experiences.
• Causal Tracingconcept0.793
  Mechanistic interpretability technique for locating factual associations, mentioned as future work direction.
• Causal Scrubbingmethod0.788
  Method by Chan et al. 2022 for rigorously testing interpretability hypotheses via interventions
• Causal abstractionconcept0.773
  A framework the paper uses alongside feature geometry to deepen mechanistic understanding of LMs
• causal bypassingconcept0.768
  Confound where naming injected concepts reflects direct logit effects rather than metacognitive awareness, raised by Morris & Plunkett
• Causal Mechanismconcept0.765
  Function determining the value of a variable based on its causal parents in an acyclic causal model.