metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 6069222
num_examples: 13549
- name: test
num_bytes: 2199561
num_examples: 3502
download_size: 724774
dataset_size: 8268783
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*