metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 860410624
num_examples: 64576
- name: validation
num_bytes: 368741700
num_examples: 27675
- name: test
num_bytes: 368755024
num_examples: 27676
download_size: 19555205
dataset_size: 1597907348
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*