metadata
dataset_info:
features:
- name: sample_id
dtype: string
- name: instruction
dtype: string
- name: completion
dtype: string
- name: category
dtype: string
- name: label
dtype: int64
- name: metadata
dtype: string
splits:
- name: train
num_bytes: 2833458
num_examples: 1000
download_size: 1106156
dataset_size: 2833458
configs:
- config_name: default
data_files:
- split: train
path: data/train-*