metadata
dataset_info:
features:
- name: comment
dtype: string
- name: input
dtype: string
- name: label
dtype: string
- name: original_idx
dtype: int64
- name: predicate
dtype: string
splits:
- name: train
num_bytes: 317943755.8085219
num_examples: 21747
download_size: 24650999
dataset_size: 317943755.8085219
configs:
- config_name: default
data_files:
- split: train
path: data/train-*