metadata
dataset_info:
features:
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: completion
dtype: string
- name: tools
dtype: string
- name: category
dtype: string
- name: task
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 103770
num_examples: 100
download_size: 39031
dataset_size: 103770
configs:
- config_name: default
data_files:
- split: train
path: data/train-*