Sparse Coding Hypothesis

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

• Sparse and smooth codingconcept0.795
  Coding scheme where qualities are represented by few neurons with continuous similarity relations.
• Noisy Simulation of Sparse Networksconcept0.783
  Mechanism by which superposition works: small neural networks exploit sparsity to approximately simulate much larger sparse networks
• Sparse Probingmethod0.776
  Method from Gurnee et al. 2023 for finding feature directions including individual neuron analysis
• Simplicity Bias Hypothesishypothesis0.775
  Deep networks are biased toward finding simple fits to data, and this bias increases with model size, driving convergence
• Sparse Crosscodersframework0.771
  Extension of SAEs that jointly learns latents across representations from different models; proposed as alternative for extended fine-tuning
• Predictive Coding Theoryconcept0.767
  Theory that brains are predictive machines minimizing prediction error.
• Sparse Dictionary Learningmethod0.766
  General method for finding overcomplete sparse decompositions; the paper uses sparse autoencoders as an approximation
• Sparse Autoencoderframework0.765
  Interpretability framework used to decompose layer-40 activations into sparse feature sets for studying emotional alignment and persistence