metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: class
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 32866762
num_examples: 226979
- name: valid
num_bytes: 1733113
num_examples: 11947
download_size: 19615470
dataset_size: 34599875
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*