---
dataset_info:
  features:
  - name: summaries
    dtype: string
  - name: level
    dtype: int64
  - name: cluster
    dtype: float64
  - name: doc_ids
    dtype: string
  - name: level_id
    dtype: string
  - name: easy_shards
    dtype: int64
  - name: hard_shards
    dtype: int64
  splits:
  - name: train
    num_bytes: 282431
    num_examples: 193
  download_size: 94978
  dataset_size: 282431
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---