---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: messages
    dtype: string
  - name: metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 170138
    num_examples: 32
  download_size: 135557
  dataset_size: 170138
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---