metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: question_fi
dtype: string
- name: translation_error
dtype: bool
splits:
- name: train
num_bytes: 6426852
num_examples: 7473
download_size: 3261633
dataset_size: 6426852
configs:
- config_name: default
data_files:
- split: train
path: data/train-*