metadata
dataset_info:
features:
- name: text
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: token_len
dtype: int64
splits:
- name: train
num_bytes: 65715412
num_examples: 46157
- name: validation
num_bytes: 3632380
num_examples: 2565
- name: test
num_bytes: 3636048
num_examples: 2565
download_size: 35688350
dataset_size: 72983840
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*