---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: source
    dtype: string
  - name: answer
    dtype: string
  - name: shortuid
    dtype: string
  - name: _keep
    dtype: bool
  - name: rejection_reasons
    sequence: string
  - name: sample_text
    dtype: string
  - name: difficulty
    dtype: int64
  - name: safety_label
    dtype: string
  splits:
  - name: train
    num_bytes: 32089932763
    num_examples: 5946618
  download_size: 13562402774
  dataset_size: 32089932763
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---