metadata
dataset_info:
features:
- name: id
dtype: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 7626108
num_examples: 1108
- name: valid
num_bytes: 1074483
num_examples: 157
- name: test
num_bytes: 2154888
num_examples: 315
download_size: 3384419
dataset_size: 10855479
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*