metadata
dataset_info:
features:
- name: doc_id
dtype: string
- name: summary_type
dtype: string
- name: system_prompt
dtype: string
- name: question
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 4937810
num_examples: 10000
download_size: 1987713
dataset_size: 4937810
configs:
- config_name: default
data_files:
- split: train
path: data/train-*