---
dataset_info:
  features:
  - name: topic
    dtype: string
  - name: flat_text
    dtype: string
  - name: turn_count
    dtype: int64
  splits:
  - name: train
    num_bytes: 19257
    num_examples: 20
  download_size: 16676
  dataset_size: 19257
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---