metadata
dataset_info:
features:
- name: inputs_pretokenized
dtype: string
- name: targets_pretokenized
dtype: string
- name: uuid
dtype: string
- name: proponents
sequence: string
splits:
- name: test
num_bytes: 21802634
num_examples: 10000
- name: train
num_bytes: 710815844
num_examples: 3190000
download_size: 172358159
dataset_size: 732618478