metadata
dataset_info:
features:
- name: text
dtype: string
- name: tgt_units
sequence: int64
splits:
- name: train
num_bytes: 6543842175
num_examples: 1267360
download_size: 2318408960
dataset_size: 6543842175
configs:
- config_name: default
data_files:
- split: train
path: data/train-*