metadata
dataset_info:
features:
- name: dataset_name
dtype: string
- name: model_type
dtype: string
- name: task_config
dtype: string
- name: mean_length
dtype: float64
- name: std_length
dtype: float64
- name: count
dtype: int64
- name: min_length
dtype: float64
- name: max_length
dtype: float64
splits:
- name: train
num_bytes: 27730
num_examples: 288
download_size: 12813
dataset_size: 27730
configs:
- config_name: default
data_files:
- split: train
path: data/train-*