metadata
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-*