README.md

metadata

dataset_info:
  features:
    - name: output
      dtype: string
    - name: input
      dtype: string
    - name: instruction
      dtype: string
    - name: prompt
      dtype: string
  splits:
    - name: train
      num_bytes: 68827329
      num_examples: 26357
  download_size: 29156486
  dataset_size: 68827329
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*