Search results

Learning rate of the Param resweight

Status: Open.

#17 In majumderb/rezero;

· Polarisjame opened on Oct 19, 2022

resweight is almost 0

Status: Open.

#16 In majumderb/rezero;

· burcehan opened on Nov 17, 2021

Is ReZero applicable to fine-tuning?

Status: Open.

#15 In majumderb/rezero;

· encounter1997 opened on Jun 4, 2021

weight decay for the resweight?

Status: Open.

#14 In majumderb/rezero;

· Kyeongpil opened on Nov 24, 2020

Can you relaese the code for ResNet-56 in Table2 ?

Status: Open.

#13 In majumderb/rezero;

· cuge1995 opened on Aug 12, 2020

when apply rezero to bert or gpt, get NAN gradients

Status: Open.

#8 In majumderb/rezero;

· yyht opened on Mar 28, 2020