metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: id
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 1278866836
num_examples: 214328
download_size: 633760205
dataset_size: 1278866836
configs:
- config_name: default
data_files:
- split: train
path: data/train-*