metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 4627952.667243368
num_examples: 2774
- name: test
num_bytes: 1157822.332756632
num_examples: 694
download_size: 3646304
dataset_size: 5785775
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*