metadata
dataset_info:
features:
- name: input_ids
dtype: string
- name: labels
dtype: string
splits:
- name: train
num_bytes: 95684951
num_examples: 409035
- name: validation
num_bytes: 7001813
num_examples: 30000
- name: test
num_bytes: 9377152
num_examples: 40000
download_size: 44425015
dataset_size: 112063916
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*