metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: dataset
dtype: string
splits:
- name: train
num_bytes: 327211705
num_examples: 299020
download_size: 146331953
dataset_size: 327211705
configs:
- config_name: default
data_files:
- split: train
path: data/train-*