metadata
dataset_info:
features:
- name: text
dtype: string
- name: dataset
dtype: string
- name: id
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 350275117
num_examples: 55760
- name: validation
num_bytes: 19652321
num_examples: 3144
- name: test
num_bytes: 20252230
num_examples: 3224
download_size: 62167632
dataset_size: 390179668
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*