metadata
dataset_info:
features:
- name: output
dtype: string
- name: input
dtype: string
- name: type
dtype: string
- name: from
dtype: string
splits:
- name: train
num_bytes: 2144142688.6661234
num_examples: 4552775
- name: test
num_bytes: 3185356.05
num_examples: 6750
download_size: 1224892608
dataset_size: 2147328044.7161233
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*