metadata
dataset_info:
features:
- name: id
dtype: string
- name: conversations
list:
- name: content
dtype: string
- name: role
dtype: string
- name: question_length
dtype: int64
- name: answer_length
dtype: int64
splits:
- name: train
num_bytes: 233820
num_examples: 381
download_size: 63229
dataset_size: 233820
configs:
- config_name: default
data_files:
- split: train
path: data/train-*