metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: question
dtype: string
- name: source
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 862477042
num_examples: 62500
download_size: 382586743
dataset_size: 862477042
configs:
- config_name: default
data_files:
- split: train
path: data/train-*