metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: language
dtype: string
splits:
- name: train
num_bytes: 2529927894
num_examples: 1138129
download_size: 1208899202
dataset_size: 2529927894
configs:
- config_name: default
data_files:
- split: train
path: data/train-*