metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 28150600
num_examples: 6866
- name: validation
num_bytes: 3001200
num_examples: 732
- name: test
num_bytes: 2911000
num_examples: 710
download_size: 16325695
dataset_size: 34062800
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*