metadata
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: system_prompt
dtype: string
- name: qas_id
dtype: string
- name: question_text
dtype: string
- name: orig_answer_texts
dtype: string
- name: answer_lengths
dtype: float64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 88560013
num_examples: 90598
- name: validation
num_bytes: 11828923
num_examples: 11988
- name: test
num_bytes: 10827404
num_examples: 10988
download_size: 56295232
dataset_size: 111216340
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*