Sea-Snell/JAXSeq

Issues

• Details about the partitioning rules

  #7 opened a year ago by gianlucadetommaso
  1

• any plans for LLaMA training example script?

  #5 opened a year ago by 152334H
  1

• Aqt?

  #4 opened a year ago by OhadRubin
  1

• converting checkpoint to hugginface format?

  #6 opened a year ago by ildoonet
  1

• Could you please explain the rationale behind including both input and output in the data format for a causal language model?

  #3 opened 2 years ago by Beomi
  3

• OOM issue training on TPUv3-8

  #2 opened 2 years ago by albertqjiang
  4

• gptj serve issue on a TPU-VM

  #1 opened 2 years ago by albertqjiang
  1