metadata
language:
- en
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answers
sequence:
- name: text
dtype: string
- name: answer_start
dtype: int32
- name: label
dtype: int64
- name: text
dtype: string
- name: logits
sequence: float32
splits:
- name: train
num_bytes: 220783565
num_examples: 130319
- name: validation
num_bytes: 21808874
num_examples: 11873
download_size: 29125984
dataset_size: 242592439
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*