metadata
dataset_info:
features:
- name: persona
dtype: string
- name: num_train
dtype: int64
- name: num_test
dtype: int64
- name: train_data
list:
list:
- name: role
dtype: string
- name: content
dtype: string
- name: test_data
list:
list:
- name: role
dtype: string
- name: content
dtype: string
splits:
- name: train
num_bytes: 16873947
num_examples: 64
download_size: 10108893
dataset_size: 16873947
configs:
- config_name: default
data_files:
- split: train
path: data/train-*