---
dataset_info:
  features:
  - name: file_name
    dtype: string
  - name: messages
    list:
    - name: content
      list:
      - name: index
        dtype: int64
      - name: type
        dtype: string
      - name: text
        dtype: string
    - name: role
      dtype: string
  splits:
  - name: train
    num_bytes: 82869
    num_examples: 170
  download_size: 27334
  dataset_size: 82869
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---