metadata
dataset_info:
features:
- name: question
dtype: string
- name: date
dtype: string
- name: text_answers
struct:
- name: text
sequence: string
- name: id
dtype: string
- name: fact_context
dtype: string
- name: context
dtype: string
- name: none_context
dtype: string
- name: neg_answers
sequence: string
splits:
- name: train
num_bytes: 75072340
num_examples: 6552
download_size: 9522699
dataset_size: 75072340
configs:
- config_name: default
data_files:
- split: train
path: data/train-*