metadata
dataset_info:
features:
- name: input
dtype: string
- name: type
dtype: string
- name: outputs
sequence: string
- name: document_id
dtype: string
splits:
- name: train
num_bytes: 3874455586
num_examples: 150288
download_size: 1625617991
dataset_size: 3874455586
configs:
- config_name: default
data_files:
- split: train
path: data/train-*