question
active
question:how-realistic-are-these-toy-modelsHow realistic are these toy models?
Open question.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Toy Models of Superpositionmentions
Claims (1)
claim
- Toy model predictions are consistent with empirical observations on polysemanticity in real networksgatesValidation; geometry and dynamics less certain.
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.
- Unexpected result.
- Formal verification open question.
- Large pretrained models used as backbones across tasks; their universality motivates the convergence hypothesis
- Intro question.
- Area of AI research that uses interventions to test and improve model resilience to perturbations