georgesung/llm_qlora

Issues

• ImportError: Using `bitsandbytes` 8-bit quantization requires Accelerate

  #10 opened 6 months ago by nicolai256
  2

• Fine tune error

  #9 opened 9 months ago by bhuvneshsaini
  0

• cuda out of memory error

  #1 opened a year ago by benam2
  2

• unable to download data ehartford/wizard_vicuna_70k_unfiltered

  #8 opened 9 months ago by bhuvneshsaini
  0

• why resize_token_embeddings not to be call after add_special_tokens PAD?

  #7 opened 10 months ago by srn-source
  0

• RawTextDataProcessor

  #6 opened a year ago by JacopoBartoli
  0

• could you share logs from training run?

  #5 opened a year ago by sandeshrajbhandari
  0

• How to finetune with my own txt dataset?

  #4 opened a year ago by Cyberninja101
  2

• "bitsandbytes was compiled without GPU support"

  #2 opened a year ago by RedLeader721
  1