metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: source
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 189794525
num_examples: 47502
download_size: 102184769
dataset_size: 189794525
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
readme coming soon.