---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: target
    dtype: string
  - name: rationale
    dtype: string
  - name: task
    dtype: string
  - name: type
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 1992009126
    num_examples: 1741456
  download_size: 1223326551
  dataset_size: 1992009126
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---