metadata
dataset_info:
features:
- name: input
dtype: string
- name: think
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 1579817008
num_examples: 345889
- name: test
num_bytes: 16141604
num_examples: 3494
download_size: 818888064
dataset_size: 1595958612
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*