metadata
dataset_info:
features:
- name: id
dtype: int64
- name: model
dtype: string
- name: conversation
list:
- name: content
dtype: string
- name: role
dtype: string
- name: refusal
dtype: 'null'
- name: annotations
dtype: 'null'
- name: audio
dtype: 'null'
- name: reasoning
dtype: 'null'
splits:
- name: train
num_bytes: 641723358
num_examples: 82607
download_size: 263245234
dataset_size: 641723358
configs:
- config_name: default
data_files:
- split: train
path: data/train-*