lucidrains/PaLM-pytorch

Issues

• code bug: running python train.py

  #18 opened 2 years ago by yanzia12138
  1

• Hardware for 100b training?

  #13 opened 2 years ago by zxcvqwerasdf
  2

• Relu Option for Speed Up

  #17 opened 2 years ago by CerebralSeed
  1

• How can I train the model on QA task?

  #4 opened 3 years ago by mellahysf
  1

• Compare loss on enwik8

  #16 opened 2 years ago by Bachstelze
  1

• Encoder-Decoder-Model

  #15 opened 2 years ago by Bachstelze
  1

• model weights

  #14 opened 2 years ago by Bachstelze
  2

• Unneeded `max` substraction in softmax?

  #9 opened 3 years ago by hypnopump
  1

• triton, triton-transformer imcompatibility issue

  #8 opened 3 years ago by jaes77
  2

• cuDNN error: CUDNN_STATUS_INTERNAL_ERROR error

  #6 opened 3 years ago by unwritten
  2

• How to extract PaLM embeddings for code?

  #1 opened 3 years ago by kb-open
  1