---
dataset_info:
  features:
  - name: instruction_seed
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 774635827
    num_examples: 3005927
  download_size: 402206463
  dataset_size: 774635827
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---