conftrace_

grokking phenomenon

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

effective theory (1) generalization dynamic (1) context copying (1) phase diagram (1) pretraining dynamics (1) induction head (1) in-context learning (1) representation learning (1)

Papers

Language Models “Grok” to Copy NAACL 2025 Towards Understanding Grokking: An Effective Theory of Representation Learning NeurIPS 2022