---
dataset_info:
  features:
  - name: input
    dtype: string
  - name: inst
    dtype: string
  - name: output
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 1817988629
    num_examples: 966002
  download_size: 984302567
  dataset_size: 1817988629
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---