bibkey: “powerburdaedwardsbabuschkinmisra2022grokking” authors: “Alethea Power; Yuri Burda; Harri Edwards; Igor Babuschkin; Vedant Misra” year: 2022 title: “Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets” doi: null claim: “Small algorithmic datasets show generalization long after the training loss has reached its floor.” strata_touched: [] license: “citation-only” triage: “anchor” url: “https://arxiv.org/abs/2201.02177”

Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets

Verified locator

arXiv:2201.02177. Original report of the phenomenon; cited only as the source of the term.

Declared identifiers: https://arxiv.org/abs/2201.02177.