metadata
dataset_info:
features:
- name: task_id
dtype: string
- name: generator_1
dtype: string
- name: generator_2
dtype: string
- name: instruction
dtype: string
- name: input
dtype: string
- name: human_label
dtype: int64
- name: worker_id
dtype: string
- name: llm_label
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 2482896
num_examples: 7137
download_size: 111103
dataset_size: 2482896
configs:
- config_name: default
data_files:
- split: train
path: data/train-*