metadata
dataset_info:
features:
- name: label
dtype: int64
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 1327289012
num_examples: 1281167
- name: validation
num_bytes: 51800000
num_examples: 50000
download_size: 604089596
dataset_size: 1379089012
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*