metadata
dataset_info:
features:
- name: persona
dtype: string
- name: labels
dtype: string
splits:
- name: train
num_bytes: 4628777104
num_examples: 21071228
download_size: 1869780499
dataset_size: 4628777104
configs:
- config_name: default
data_files:
- split: train
path: data/train-*