metadata
dataset_info:
features:
- name: conversations
list:
- name: content
dtype: string
- name: from
dtype: string
- name: idx
dtype: int64
splits:
- name: train
num_bytes: 44487880
num_examples: 7942
- name: test
num_bytes: 4997504
num_examples: 883
download_size: 24416400
dataset_size: 49485384
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*