Feature Universality

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

• Feature universality across independently trained models suggests features have some existence beyond individual modelsclaim0.819
  Authors take agnostic position on ontological status but universality evidence pushes toward features being real
• Universality Hypothesisconcept0.788
  The hypothesis that analogous features and circuits reliably form across different neural network models and tasks
• Feature Sparsityconcept0.779
  Property that features activate on only a small fraction of inputs; enables compressed sensing and is what allows superposition to work
• Uniquenessconcept0.764
  The property that every place generated by a living process is inevitably unique due to its adaptation to specific conditions.
• Is the apparent universality of some low-level vision features the exception or the rule?question0.764
  Open empirical question following anecdotal cross-model universality findings
• Feature splittingconcept0.762
  Phenomenon where a feature in a small SAE splits into multiple finer features in a larger SAE.
• Truth direction universalityconcept0.761
  The claim that truth directions are consistent and generalizable across layers, tasks, and prompt formats in LLMs.
• Pure Featureconcept0.753
  A feature that responds to only a single latent variable, contrasted with polysemantic features