metadata
dataset_info:
features:
- name: conversation
dtype: string
splits:
- name: train
num_bytes: 25890403
num_examples: 26478
download_size: 8197903
dataset_size: 25890403
configs:
- config_name: default
data_files:
- split: train
path: data/train-*