metadata
dataset_info:
features:
- name: labels
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 1975216
num_examples: 3011
- name: eval
num_bytes: 493968
num_examples: 753
download_size: 160435
dataset_size: 2469184
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*