metadata
dataset_info:
features:
- name: id
dtype: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train_sft
num_bytes: 531053832
num_examples: 199137
- name: test_sft
num_bytes: 131987986
num_examples: 49785
download_size: 263983731
dataset_size: 663041818
configs:
- config_name: default
data_files:
- split: train_sft
path: data/train_sft-*
- split: test_sft
path: data/test_sft-*