---
dataset_info:
  features:
  - name: metadata
    dtype: string
  - name: conversations
    dtype: string
  - name: clue_text
    dtype: string
  splits:
  - name: train
    num_bytes: 86150905
    num_examples: 32092
  download_size: 19411439
  dataset_size: 86150905
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---