metadata
dataset_info:
features:
- name: task
dtype: string
- name: inputs
dtype: string
- name: targets
dtype: string
- name: symbols
sequence: string
splits:
- name: train
num_bytes: 245142593.2805905
num_examples: 93535
- name: validation
num_bytes: 10724
num_examples: 5
download_size: 91233610
dataset_size: 245153317.2805905
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*