What kind of picture of neural networks would emerge if we treated individual neurons, even individual weights, as being worthy of serious investigation?

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

• Ultimately, we would like to understand neural networks well enough to be able to intentionally design them.quote0.817
  Vision statement in the conclusion.
• Individual floating-point number weights in neural networks become meaningful once you understand the features they connect.claim0.802
  Interpretive claim that circuits render raw weights interpretable as algorithmic structures
• Neural networks, trained with different objectives on different data and modalities, are converging to a shared statistical model of reality in their representation spaces.quote0.791
  The paper's central thesis statement, presented prominently after the abstract
• In the long run, studying circuit motifs may be more important than studying individual circuits for understanding neural networks.claim0.779
  Strategic claim about the relative importance of motif-level abstraction over circuit-level analysis
• Each human is a collective intelligence; the parenthetical 'of neurons' can be dropped as a conceptual commitment.claim0.772
• If the universality hypothesis is broadly true, it raises the exciting possibility that artificial neural networks could predict features previously unknown in biological neural networks.claim0.768
  Speculative extension of universality to neuroscience, with high-low frequency detectors as a candidate prediction
• The huge chemical complexity within each synapse suggests that neural models of cognition that refer only to changing weights of synaptic connections and ignore sub-neural chemistry are probably ignoring some of the most important explanatory mechanisms in brains.claim0.768
  Sloman's critique of mainstream neural network theories.
• Neural networks show substantial alignment with biological representations in the brain, driven by shared task and data constraintsclaim0.767
  Extends convergence argument to brain-machine alignment