In a combinatorial toy setting, winning lottery tickets preserve families of compatible feature locations in early feature space that balance proximity to final codes with low interference, rather than specific weight subnetworks.
SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability
6 Pith papers cite this work, alongside 227 external citations. Polarity classification is still indexing.
6
Pith papers citing it
227
external citations · Pith
abstract
We propose a new technique, Singular Vector Canonical Correlation Analysis (SVCCA), a tool for quickly comparing two representations in a way that is both invariant to affine transform (allowing comparison between different layers and networks) and fast to compute (allowing more comparisons to be calculated than with previous methods). We deploy this tool to measure the intrinsic dimensionality of layers, showing in some cases needless over-parameterization; to probe learning dynamics throughout training, finding that networks converge to final representations from the bottom up; to show where class-specific information in networks is formed; and to suggest new training regimes that simultaneously save computation and overfit less. Code: https://github.com/google/svcca/
representative citing papers
Develops multi-scale and alternating-diffusion fused variants of CKA and distance correlation via Markov matrices for neural representation comparison, reporting state-of-the-art results on ReSi and OOD benchmarks.
Tensor similarity is a symmetry-invariant metric that measures functional equivalence between tensor-based networks using a recursive algorithm for cross-layer mechanisms.
Linear probes demonstrate that feature separability for classification increases monotonically with network depth in Inception v3 and ResNet-50.
Graph-PRefLexOR fine-tunes graph-native models with GRPO to organize reasoning into phases, yielding 40-65% gains in traceable hypothesis generation and 2-3x semantic diversity on 100 materials science questions.
Grokking delay on Collatz prediction is a decoder access bottleneck after early encoder structure learning, with numeral base as a strong inductive bias on learnability.
citing papers explorer
-
Toy Combinatorial Interpretability Models Reveal Lottery Tickets in Early Feature Space
In a combinatorial toy setting, winning lottery tickets preserve families of compatible feature locations in early feature space that balance proximity to final codes with low interference, rather than specific weight subnetworks.
-
From Layers to Networks: Comparing Neural Representations via Diffusion Geometry
Develops multi-scale and alternating-diffusion fused variants of CKA and distance correlation via Markov matrices for neural representation comparison, reporting state-of-the-art results on ReSi and OOD benchmarks.
-
When Are Two Networks the Same? Tensor Similarity for Mechanistic Interpretability
Tensor similarity is a symmetry-invariant metric that measures functional equivalence between tensor-based networks using a recursive algorithm for cross-layer mechanisms.
-
Understanding intermediate layers using linear classifier probes
Linear probes demonstrate that feature separability for classification increases monotonically with network depth in Inception v3 and ResNet-50.
-
Graph-Native Reinforcement Learning Enables Traceable Scientific Hypothesis Generation through Conceptual Recombination
Graph-PRefLexOR fine-tunes graph-native models with GRPO to organize reasoning into phases, yielding 40-65% gains in traceable hypothesis generation and 2-3x semantic diversity on 100 materials science questions.
-
The Long Delay to Arithmetic Generalization: When Learned Representations Outrun Behavior
Grokking delay on Collatz prediction is a decoder access bottleneck after early encoder structure learning, with numeral base as a strong inductive bias on learnability.