metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: dataset
dtype: string
splits:
- name: train
num_bytes: 7105264073.28665
num_examples: 2908042
download_size: 3746284572
dataset_size: 7105264073.28665
configs:
- config_name: default
data_files:
- split: train
path: data/train-*