Language, Computation and Cognition Lab (LaCoCo)
Language, Computation and Cognition Lab (LaCoCo)
Home
People
Publications
Teaching
Joining
Light
Dark
Automatic
Understanding the Emergence of Seemingly Useless Features in Next-Token Predictors
Mark Rofin
,
Jalal Naghiyev
,
Michael Hahn
January, 2026
Paper
Type
Conference paper
Publication
ICLR
Cite
×