metadata
dataset_info:
features:
- name: conversations
list:
- name: role
dtype: string
- name: content
dtype: string
splits:
- name: train
num_bytes: 38651635
num_examples: 15141
download_size: 18719576
dataset_size: 38651635
configs:
- config_name: default
data_files:
- split: train
path: data/train-*