---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: inputs
    dtype: string
  - name: targets
    dtype: string
  - name: task
    dtype: string
  - name: data_source
    dtype: string
  splits:
  - name: train
    num_bytes: 41514508.256710336
    num_examples: 200000
  download_size: 19791456
  dataset_size: 41514508.256710336
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---