Outliers Dimensions that Disrupt Transformers Are Driven by Frequency

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Puccetti, Giovanni, Rogers, Anna, Drozd, Aleksandr, Dell'Orletta, Felice
Format:	Preprint
Published:	2022
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2205.11380