---
dataset_info:
  features:
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 3345323563.5341997
    num_examples: 373927
  download_size: 1562181581
  dataset_size: 3345323563.5341997
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---