---
dataset_info:
  features:
  - name: messages
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: character
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 1164555
    num_examples: 1229
  - name: val
    num_bytes: 129932
    num_examples: 136
  download_size: 201520
  dataset_size: 1294487
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---