---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: completion
    dtype: string
  - name: api
    dtype: string
  splits:
  - name: train
    num_bytes: 25904109429.0
    num_examples: 2088002
  download_size: 8983763078
  dataset_size: 25904109429.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---