metadata
dataset_info:
features:
- name: ppl_label
dtype: int64
- name: text
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 2214483951
num_examples: 250105
download_size: 1046260212
dataset_size: 2214483951
configs:
- config_name: default
data_files:
- split: train
path: data/train-*