metadata
dataset_info:
features:
- name: context
dtype: string
- name: label
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 6697406745
num_examples: 1573489
- name: test
num_bytes: 2232483756
num_examples: 524497
download_size: 863734090
dataset_size: 8929890501