README.md

metadata

language:
  - en
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: prompt
      dtype: string
    - name: responses
      sequence: string
    - name: generator
      dtype: string
    - name: dataset
      dtype: string
  splits:
    - name: train
      num_bytes: 1313552
      num_examples: 800
  download_size: 771970
  dataset_size: 1313552

Dataset Card for "llama_sft_dpo_bold_list_attack_iter1"

More Information needed