concept
active
concept:error-nodesError nodes
Uninterpretable nodes representing replacement model reconstruction error.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- CLT plus error nodes and original attention patterns, local to a prompt.
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.
- Extra feature representing unexplained error to make isomorphism exact.
- Role in optimizing sensory states; unified treatment shows value-learning and perception share error-minimization principle.
- The progressive reduction of error (stress) as cells move toward their target positions.
- The formal RL quantity whose signed character is identified with valence in the paper's account
- Primary training method for neural networks; cited as surprisingly effective even to its inventors, illustrating resistance to full reductionist understanding