---
dataset_info:
  features:
  - name: inputs
    dtype: string
  - name: targets
    dtype: string
  - name: task
    dtype: string
  splits:
  - name: train
    num_bytes: 424909588.64904696
    num_examples: 373681
  download_size: 272722667
  dataset_size: 424909588.64904696
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---