PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs

May 1, 2025·
Oskar Van Der Wal
,
Pietro Lesci
,
Max Müller-Eberstein
Naomi Saphra
Naomi Saphra
,
Hailey Schoelkopf
,
Willem Zuidema
,
Stella Biderman
· 0 min read
Type
Publication
International Conference on Learning Representations (ICLR)