metadata
dataset_info:
features:
- name: formatted_conversation
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 17647861
num_examples: 10473
- name: test
num_bytes: 2220392
num_examples: 1311
- name: validation
num_bytes: 2190666
num_examples: 1308
download_size: 6687228
dataset_size: 22058919
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*