metadata
dataset_info:
features:
- name: nlacp
dtype: string
- name: error_label
dtype: string
- name: nlacp_label
dtype: string
- name: policy
dtype: string
- name: answers
struct:
- name: answer_start
list: int64
- name: text
list: string
- name: err_ind
dtype: int64
splits:
- name: train
num_bytes: 15706172
num_examples: 17608
download_size: 4699302
dataset_size: 15706172
configs:
- config_name: default
data_files:
- split: train
path: data/train-*