paper-with-me

Papers

Exploring Grokking: Experimental and Mechanistic Investigations

2024-12-14 · Hu Qiye, Zhou Hao, Yu RuoXi

The phenomenon of grokking in over-parameterized neural networks has garnered significant interest. It involves the neural network initially memorizing the training set with zero training error and near-random test error. Subsequent prolonged training leads to a sharp transition from no generalization to perfect generalization. Our study comprises extensive experiments and an exploration of the research behind the mechanism of grokking. Through experiments, we gained insights into its behavior concerning the training data fraction, the model, and the optimization. The mechanism of grokking has been a subject of various viewpoints proposed by researchers, and we introduce some of these perspectives.

📄 PDF Abstract BibTeX arXiv:2412.10898

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Grokking of Diffusion Models: Case Study on Modular Addition

2026-04-20 · Joon Hyeok Kim, Yong-Hyun Park, Mattis Dalsætra Østby, Jiatao Gu arxiv

Despite their empirical success, how diffusion models generalize remains poorly understood from a mechanistic perspective. We demonstrate that diffusion models trained with flow-matching objectives exhibit grokking--dela…

Model Capacity Determines Grokking through Competing Memorisation and Generalisation Speeds

2026-05-10 · Yiding Song, Hanming Ye arxiv

Existing accounts of grokking explain the phenomena in terms of mechanistic frameworks such as circuit efficiency or lazy-to-rich transitions. However, despite a known dependence between grokking and model size, how mode…

Progress measures for grokking via mechanistic interpretability

2023-01-12 · Neel Nanda, Lawrence Chan, Tom Lieberum, Jess Smith 외

Neural networks often exhibit emergent behavior, where qualitatively new capabilities arise from scaling up the amount of parameters, training data, or training steps. One approach to understanding emergence is to find c…

Memorization

Grokking From Abstraction to Intelligence

2026-03-31 · Junjie Zhang, Zhen Shen, Gang Xiong, Xisong Dong arxiv

Grokking in modular arithmetic has established itself as the quintessential fruit fly experiment, serving as a critical domain for investigating the mechanistic origins of model generalization. Despite its significance, …

Latent Algorithmic Structure Precedes Grokking: A Mechanistic Study of ReLU MLPs on Modular Arithmetic

2026-03-24 · Anand Swaroop arxiv

Grokking-the phenomenon where validation accuracy of neural networks on modular addition of two integers rises long after training data has been memorized-has been characterized in previous works as producing sinusoidal …