---
dataset_info:
  features:
  - name: conversations
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: source
    dtype: string
  - name: conversation_id
    dtype: string
  - name: num_turns
    dtype: int64
  - name: has_context
    dtype: bool
  splits:
  - name: train
    num_bytes: 15600459
    num_examples: 14136
  download_size: 15162154
  dataset_size: 15600459
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---