metadata
dataset_info:
features:
- name: targets
dtype: string
- name: targets_vec
sequence: int64
- name: inputs
dtype: string
splits:
- name: test
num_bytes: 1030991
num_examples: 2000
- name: train
num_bytes: 1029479
num_examples: 2000
download_size: 577875
dataset_size: 2060470
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*