---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: score
    dtype: float32
  - name: int_score
    dtype: int64
  splits:
  - name: train
    num_bytes: 19953639587
    num_examples: 3311309
  download_size: 8745788072
  dataset_size: 19953639587
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---