metadata
dataset_info:
features:
- name: conversations
struct:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 114292885
num_examples: 64265
download_size: 44535090
dataset_size: 114292885
configs:
- config_name: default
data_files:
- split: train
path: data/train-*