metadata
dataset_info:
features:
- name: input
dtype: string
- name: mc_labels
sequence: float64
splits:
- name: train
num_bytes: 66709931019
num_examples: 315151780
- name: test
num_bytes: 11625286787
num_examples: 54872514
download_size: 17338800261
dataset_size: 78335217806
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*