metadata
dataset_info:
features:
- name: id
dtype: int32
- name: label
dtype: int64
- name: claim
dtype: string
- name: document
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 614932100
num_examples: 249144
download_size: 238050038
dataset_size: 614932100
configs:
- config_name: default
data_files:
- split: train
path: data/train-*