concept
active
concept:thomson-problemThomson problem
Point-packing problem analogous to the toy model loss.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Toy Models of Superpositionmentions
Frameworks (1)
framework
- ReLU output toy modelanalogous_tox'=ReLU(W^TWx+b), main model demonstrating superposition.
Concepts (1)
concept
- Antipodal pairsassociated_withTwo features stored as exact negatives in one dimension.
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.
- Problem of finding a system with desired response to perturbation; contrasted with forward problem of predicting responses of a given system
- The risk that any indicator merely correlated with consciousness can be artificially implemented without the system actually being conscious
- Bourget's term (closely related to Seth's real problem) for which organisational features are associated with which experiences; the tractable question this framework pursues.
- The problem of ensuring AI systems adopt values compatible with human welfare — argued to be a perennial problem already present in child-rearing
- The question of how competent small-scale agents bind into a unified larger Self.
- Shevlin's problem of which features of the human case are criterial versus contingent for consciousness; recurs at every level.
- The process of moving through configuration space towards a goal; self-organisation as navigation