metadata
dataset_info:
features:
- name: conversations
struct:
- name: content_type
dtype: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: score
dtype: float64
- name: title
dtype: string
- name: url
dtype: string
splits:
- name: train
num_bytes: 614997605
num_examples: 45359
download_size: 273262125
dataset_size: 614997605
configs:
- config_name: default
data_files:
- split: train
path: data/train-*