metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: target
dtype: string
- name: conversation_id
dtype: string
- name: slot_index
dtype: int64
- name: n_context_turns
dtype: int64
splits:
- name: train
num_bytes: 3222
num_examples: 5
- name: test
num_bytes: 58561
num_examples: 109
download_size: 38922
dataset_size: 61783
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*