README.md

metadata

dataset_info:
  features:
    - name: output
      dtype: string
    - name: input
      dtype: string
    - name: instruction
      dtype: string
    - name: instruction_evolution_prompt
      dtype: string
    - name: evolved_instruction
      dtype: string
    - name: completion
      dtype: string
    - name: conversations
      list:
        - name: from
          dtype: string
        - name: value
          dtype: string
  splits:
    - name: train
      num_bytes: 655504
      num_examples: 100
  download_size: 342945
  dataset_size: 655504
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*