metadata
dataset_info:
features:
- name: text
dtype: string
- name: wrong_path
dtype: string
- name: len_target
dtype: int64
splits:
- name: train
num_bytes: 18479865
num_examples: 200000
- name: validation
num_bytes: 1848784
num_examples: 20000
download_size: 12502235
dataset_size: 20328649
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*