metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 3627810.3617712776
num_examples: 9649
- name: validation
num_bytes: 74067.63822872232
num_examples: 197
download_size: 1463613
dataset_size: 3701878
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*