notesum.ai

Published at November 11

cs.LG

cs.AI

cs.CL

stat.ML

Released Date: November 11, 2024

Authors: Alex Havrilla¹, Wenjing Liao¹

Aff.: ¹Department of Mathematics, Georgia Institute of Technology

GPT-2		Chinchilla
$\hat{\alpha}_{N}=0.076$	$\alpha_{D}=0.070$	$\hat{\alpha}_{N}=0.34$	$\alpha_{D}=0.25$
$\hat{\alpha}_{D}=0.095$	$\alpha_{N}=0.106$	$\hat{\alpha}_{D}=0.28$	$\alpha_{N}=0.33$