Pith. sign in

REVIEW 13 cited by

A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2303.11873 v1 pith:J7WOKSPA submitted 2023-03-21 cs.LG

A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks

classification cs.LG
keywords grokkingsparsetransitiondominatesphasecompetitiondensefind
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Grokking is a phenomenon where a model trained on an algorithmic task first overfits but, then, after a large amount of additional training, undergoes a phase transition to generalize perfectly. We empirically study the internal structure of networks undergoing grokking on the sparse parity task, and find that the grokking phase transition corresponds to the emergence of a sparse subnetwork that dominates model predictions. On an optimization level, we find that this subnetwork arises when a small subset of neurons undergoes rapid norm growth, whereas the other neurons in the network decay slowly in norm. Thus, we suggest that the grokking phase transition can be understood to emerge from competition of two largely distinct subnetworks: a dense one that dominates before the transition and generalizes poorly, and a sparse one that dominates afterwards.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Low-Dimensional and Transversely Curved Optimization Dynamics in Grokking

    cs.LG 2026-02 unverdicted novelty 8.0

    Grokking reflects escape from a metastable low-dimensional regime where transverse curvature accumulates before generalization, with subspace motion necessary but curvature boost insufficient.

  2. Circuit Synchronization Precedes Generalization: A Causal Precursor to Grokking

    cs.LG 2026-06 conditional novelty 7.0

    FSD, a permutation-tested metric of Fourier circuit synchronization, precedes grokking by a mean of 1722 steps across nine modular addition setups and causally controls grokking timing when weight decay is varied at t...

  3. The Geometry of Multi-Task Grokking: Transverse Instability, Superposition, and Weight Decay Phase Structure

    cs.LG 2026-02 unverdicted novelty 7.0

    Multi-task grokking in Transformers produces staggered generalization, low-dimensional manifolds, weight-decay phase structure, holographic solutions, and transverse redundancy.

  4. Egalitarian Gradient Descent: A Simple Approach to Accelerated Grokking

    cs.LG 2025-10 unverdicted novelty 7.0

    EGD equalizes gradient speeds across singular directions, eliminating or shortening grokking plateaus on modular addition and sparse parity problems.

  5. Grokking and epoch-wise double descent in quantum neural networks

    quant-ph 2026-07 conditional novelty 6.0

    Overparameterized two-qubit SU(4) QNNs exhibit grokking and epoch-wise double descent; depth raises generalization success, and weak L2 regularization anchors the post-grokking state against weight-norm drift.

  6. Radial Suppression Accelerates Algorithmic Generalization: A Geometric Analysis of Delayed Generalization

    cs.LG 2026-06 unverdicted novelty 6.0

    A norm penalty constraining activations to a sqrt(d)-radius hypersphere accelerates grokking by up to 6x on modular arithmetic via radial suppression in activation dynamics.

  7. Spectral Lens: Activation and Gradient Spectra as Diagnostics of LLM Optimization

    stat.ML 2026-05 unverdicted novelty 6.0

    Spectral analysis of activations and gradients provides new diagnostics that link batch size to representation geometry, early covariance tails to token efficiency, and spectral shifts to learning dynamics in decoder-...

  8. Spectral Entropy Collapse as a Phase Transition in Delayed Generalisation: An Interventional and Predictive Framework for Grokkin

    cs.LG 2026-04 unverdicted novelty 6.0

    Normalized spectral entropy of model representations collapses before grokking, crosses a threshold in all runs, and interventions confirm it drives the transition on group-theoretic tasks.

  9. Spectral Entropy Collapse as a Phase Transition in Delayed Generalisation: An Interventional and Predictive Framework for Grokkin

    cs.LG 2026-04 unverdicted novelty 6.0

    Spectral entropy collapse in learned representations precedes and predicts grokking, with interventions showing it is not explained by parameter norm alone.

  10. SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning

    cs.CV 2026-06 unverdicted novelty 5.0

    SingGuard presents a policy-adaptive multimodal LLM guardrail family with hybrid reasoning regimes and a new benchmark of 56,340 examples, claiming SOTA F1 across 35 datasets and improved policy adherence under runtim...

  11. SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning

    cs.CV 2026-06 unverdicted novelty 5.0

    SingGuard introduces a policy-adaptive multimodal LLM guardrail with dynamic reasoning regimes and SingGuard-Bench, reporting SOTA F1 scores across 35 datasets and improved policy-following accuracy under runtime shifts.

  12. Select and Improve: Understanding the Mechanics of Post-Training for Reasoning

    cs.LG 2026-06 unverdicted novelty 5.0

    Controlled experiments on Qwen-2.5-1.5B identify strategy selection (enabled by diverse SFT data) and strategy improvement (enabled by harder RL data) as the core mechanisms through which RL post-training enhances reasoning.

  13. Model Capacity Determines Grokking through Competing Memorisation and Generalisation Speeds

    cs.LG 2026-05 unverdicted novelty 5.0

    Grokking emerges near the model size where memorization timescale T_mem(P) intersects generalization timescale T_gen(P) on modular arithmetic.