metadata
dataset_info:
features:
- name: ctx
dtype: string
- name: second_steps
sequence: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 18694115
num_examples: 18722
download_size: 8159760
dataset_size: 18694115
configs:
- config_name: default
data_files:
- split: train
path: data/train-*