metadata
dataset_info:
features:
- name: task_id
dtype: int32
- name: source
dtype: string
- name: task_source
dtype: string
- name: task_name
dtype: string
- name: target
dtype: string
- name: split
dtype: string
splits:
- name: train
num_bytes: 234967
num_examples: 464
download_size: 105196
dataset_size: 234967
configs:
- config_name: default
data_files:
- split: train
path: data/train-*