metadata
dataset_info:
features:
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: completion
dtype: string
- name: schema
dtype: string
- name: idx
dtype: int64
- name: pattern
dtype: string
splits:
- name: train
num_bytes: 217925
num_examples: 100
download_size: 89119
dataset_size: 217925
configs:
- config_name: default
data_files:
- split: train
path: data/train-*