metadata
dataset_info:
features:
- name: input_ids
sequence: int64
- name: attention_mask
sequence: int64
- name: token_type_ids
sequence: int64
- name: labels
sequence: string
splits:
- name: train
num_bytes: 18647028
num_examples: 1062
download_size: 924803
dataset_size: 18647028
configs:
- config_name: default
data_files:
- split: train
path: data/train-*