Line of inquiry
Inquiring lines›How do training data and architect…›What training signals and data sel…›this line of inquiry
What determines how neural networks organize their internal knowledge representations?
A broader line of inquiry — a family of 49 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 49
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can neural networks represent symbolic structures without explicit mechanisms?
- How do neural networks decompose tasks into modular subnetworks that transfer?
- Does the linear representation hypothesis reflect networks or reflect our analysis tools?
- What are fractured entangled representations in neural networks?
- Does recursive generalization explain why neurons develop single interpretable functions?
- Why does knowledge storage separate from reasoning circuits in neural networks?
- How do neural networks decompose complex tasks into modular subnetworks?
- How can neural networks be interpretable by design rather than post-hoc?
- How much do measurement thresholds affect estimates of neuron interpretability rates?
- Why does gradient descent discover compositional structure without explicit pressure?
- Do dense models already achieve high interpretability without sparse training?
- Can fractured representations explain why models fail at systematic generalization?
- Can we detect and measure circuit formation before generalization emerges?
- Can we predict which tasks will decompose into modular subnetworks?
- Can neural networks implement genuine algorithms or only statistical pattern matching?
- Why do human-designed neural architectures eventually get replaced by learned ones?
- How does representational density emerge from training data familiarity?
- What makes a feature abstract versus concrete in neural network activations?
- How should we rethink the symbolism versus connectionism debate in light of LLMs?
- How do knowledge and reasoning circuits interfere in the same neural network?
- Can sub-task handlers be swapped between neural and symbolic systems?
- How do concept vectors in neural networks relate to human-language chess explanations?
- Why do some neurons detect multiple unrelated concepts in early models?
- Can fractured entangled representations hide undetected by standard analysis methods?
- Does architectural discovery follow an empirical scaling law like neural networks?
- Could probing methods miss computationally important features in neural networks?
- How do cortical columns implement local inference over memory cycles?
- How do sparse circuits compare to the modular subnetworks that emerge naturally?
- What inductive biases help networks segregate entities from raw inputs?
- What makes a neural network circuit actually interpretable to humans?
- What solvable idealized settings reveal fundamental phenomena in realistic deep learning?
- Is the superposition hypothesis necessary to explain neural network behavior?
- Do feature extraction methods systematically miss computationally important complex features?
- What makes some internal circuits more interpretable than others?
- Does saliency analysis restore justificatory burden to opaque network outputs?
- Can autoencoders act as associative memory systems like Hopfield networks?
- Why did neuron interpretability jump from near-zero to over ninety percent?
- How does mechanistic interpretability complement learning mechanics in explaining deep learning?
- How do biological brains organize computation across different cortical timescales?
- How do classical mechanics and statistical mechanics provide methodological templates for learning theory?
- How do normalization and input injection control emergence of fixed points?
- How does joint backpropagation differ from training separate ensemble models?
- Why are polysemantic features concentrated in early neural network layers?
- Why did early neural networks need massive computing power for simple image tasks?
- Which hyperparameter theories best explain universal behaviors across neural networks?
- How do neural networks extend contextual bandits beyond linear reward assumptions?
- How do gradients flowing through both branches simultaneously reshape each component's role?
- Why do different brain and AI systems appear similar when compared via RSA?
- Can Kolmogorov complexity alone capture what makes intelligence general?