metadata
dataset_info:
features:
- name: unique_conversation_id
dtype: string
- name: cluster
dtype: int32
splits:
- name: train
num_bytes: 89257780
num_examples: 1472917
download_size: 17951475
dataset_size: 89257780
configs:
- config_name: default
data_files:
- split: train
path: data/train-*