Large Language Model Fine-Tuning with QLoRA

QLoRA: Efficient Fine-Tuning of QUantized LLMs

Dependencies

• Python 3.10
• PyTorch 2.0 +

  pip install -r requirements.txt
  

Dataset

carblacac/twitter-sentiment-analysis

Training

  python run_qlora.py \
      --dataset_name carblacac/twitter-sentiment-analysis \
      --model_name_or_path tiiuae/falcon-7b \
      --do_train True \
      --do_eval True \
      --do_predict True