metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
dataset_info:
features:
- name: conversations
list:
- name: content
dtype: string
- name: role
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 15588430
num_examples: 8000
- name: eval
num_bytes: 3897782
num_examples: 2000
download_size: 9452437
dataset_size: 19486212