metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: conversations-th
list:
- name: from
dtype: string
- name: value
dtype: string
- name: translate_status
dtype: bool
splits:
- name: train
num_bytes: 31642895
num_examples: 6198
download_size: 12944726
dataset_size: 31642895
configs:
- config_name: default
data_files:
- split: train
path: data/train-*