"Spike No More: Stabilizing the Pre-training of Large Language Models."

Sho Takase et al. (2023)

Details and statistics

DOI: 10.48550/ARXIV.2312.16903

access: open

type: Informal or Other Publication

metadata version: 2024-05-07

a service of  Schloss Dagstuhl - Leibniz Center for Informatics