---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: book_id
    dtype: string
  - name: question_id
    dtype: string
  - name: conv_id
    dtype: int64
  - name: round
    dtype: int64
  - name: messages
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
    - name: tool_calls
      list:
      - name: id
        dtype: string
      - name: function
        struct:
        - name: name
          dtype: string
        - name: arguments
          dtype: string
      - name: type
        dtype: string
    - name: tool_call_id
      dtype: string
  splits:
  - name: train
    num_bytes: 313730027
    num_examples: 21505
  download_size: 68625927
  dataset_size: 313730027
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---