OpenNLPLab/cosFormer

Issues

• Attn Mask for Non-causal Models

  #5 opened 3 years ago by roshansh-cmu
  2

• Script for computing memory consumption

  #12 opened 2 years ago by DaShenZi721
  3

• why input is [s b dim] but not [b s dim]?

  #11 opened 2 years ago by Zyriix
  1

• Why cosformer not work on XL-base transformer architecture?

  #10 opened 2 years ago by lwaekfjlk
  0

• Question about space complexity

  #8 opened 2 years ago by nihaomiao
  0

• Pre-train model

  #7 opened 2 years ago by csorujian
  0

• causal attention not working when q and kv are not in same length

  #4 opened 3 years ago by zero0kiriyu
  1

• Why the attn mask is not used in forward function?

  #6 opened 3 years ago by HanielF
  1

• Hello, when will the full code be published?

  #3 opened 3 years ago by zmxsss
  1

• When the code will be released?

  #1 opened 3 years ago by LeeDoYup
  1