metadata
dataset_info:
config_name: test
features:
- name: text
dtype: string
- name: prediction_oneshot
dtype: string
- name: response_oneshot
list:
- name: generated_text
dtype: string
- name: dataset
dtype: string
- name: split
dtype: string
- name: lang
dtype: string
splits:
- name: train
num_bytes: 42603749
num_examples: 12000
download_size: 9801222
dataset_size: 42603749
configs:
- config_name: test
data_files:
- split: train
path: test/train-*