metadata
dataset_info:
features:
- name: questions.answer
dtype: string
- name: questions.order
dtype: int64
- name: questions.question
dtype: string
splits:
- name: train
num_bytes: 31172553.126733348
num_examples: 352100
- name: validation
num_bytes: 3463685.8732666536
num_examples: 39123
- name: heldout_profiles
num_bytes: 3852205
num_examples: 43490
download_size: 18747416
dataset_size: 38488444
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: heldout_profiles
path: data/heldout_profiles-*