InceptionV1 spreads car feature from a pure car detector in mixed4c across dog detector neurons in the next layer

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

• InceptionV1 implements a four-layer circuit for pose-invariant dog head detection with mirrored left/right pathways that inhibit each other then unite, exhibiting XOR-like propertiesfinding0.825
  Evidence that neural networks learn sophisticated invariance mechanisms through structured circuits rather than loose feature aggregation
• InceptionV1 neuron 4e:55 responds to cat faces, fronts of cars, and cat legs as unrelated stimulifinding0.806
  Concrete example of polysemantic neuron demonstrating the challenge to the circuits agenda
• Weights between early and full curve detectors in InceptionV1 form a curve of positive weights at tangent positions, with opposing orientations inhibitoryfinding0.783
  Demonstrates that meaningful algorithms can be read directly off floating-point weights in a neural network
• Curve detectors found across AlexNet, InceptionV1, VGG19, ResNetV2-50 and models trained on Places365finding0.766
  Anecdotal evidence for the universality of low-level visual features across different architectures and datasets
• High-low frequency detectors found across AlexNet, InceptionV1, VGG19, and ResNetV2-50finding0.739
  Second low-level feature type demonstrating cross-architecture universality
• Lindsey 2025: frontier models can detect and report changes in their own internal activations via concept injection experiments, demonstrating functional introspective awarenessfinding0.734
  Prior finding cited as convergent evidence for LLM self-awareness capacities
• The likelihood of a dedicated feature for a concept (element, city, animal, food) follows a sigmoid in log-frequency of the concept in training data, with threshold frequency inversely proportional to number of alive features.finding0.726
  Quantitative relationship between concept frequency and feature presence.
• Target morphology shifts occur despite the fact that all of the individual cells have unaltered normal genomes, showing that competent subunits can be pushed to implement diverse organism-scale goals by physiological signals.claim0.721
  Highlights the non-genetic control of large-scale anatomy.