---
dataset_info:
  features:
  - name: messages
    dtype: string
  - name: title
    dtype: string
  - name: air_date
    dtype: string
  - name: season
    dtype: int32
  - name: episode
    dtype: int32
  - name: scene
    dtype: int32
  splits:
  - name: train
    num_bytes: 13456479
    num_examples: 13333
  download_size: 3404045
  dataset_size: 13456479
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---