---
dataset_info:
  features:
  - name: inputs
    dtype: string
  - name: targets
    dtype: string
  - name: task
    dtype: string
  - name: target_token_count
    dtype: int64
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 9928566
    num_examples: 4302
  download_size: 5179316
  dataset_size: 9928566
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---