Weight decay controls distinct learning regimes in grokking transformers on modular arithmetic, tracked by new cheap attention-based diagnostics with empirical critical value and exponent fits.
Catherine Olsson, Nelson Elhage, Neel Nanda, et al
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Weight Decay Regimes in Grokking Transformers: Cheap Online Diagnostics
Weight decay controls distinct learning regimes in grokking transformers on modular arithmetic, tracked by new cheap attention-based diagnostics with empirical critical value and exponent fits.