microsoft/mutransformers

Issues

• How to use mu-transfer for LLaMA2?

  #4 opened a year ago by brando90
  0

• Differences between paper and actual implementation in _init_weight function

  #2 opened 2 years ago by Ladun
  0

• Question: Shouldn't learnable positional embeddings be MuReadoutLayers ?

  #3 opened 2 years ago by codedecde
  0

• bert256.bsh not in repo

  #1 opened 2 years ago by nightsnack
  0