metadata
license: mit
dataset_info:
features:
- name: tokens
sequence: int64
- name: label
dtype: int64
splits:
- name: train
num_bytes: 14596224
num_examples: 64000
- name: validation
num_bytes: 1819520
num_examples: 8000
- name: test
num_bytes: 1816384
num_examples: 8000
download_size: 2202230
dataset_size: 18232128
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*