metadata
dataset_info:
features:
- name: task_id
dtype: string
- name: worker_id
dtype: string
- name: human_label
dtype: int64
- name: llm_label
dtype: int64
- name: generator_1
dtype: string
- name: generator_2
dtype: string
- name: premise
dtype: string
splits:
- name: train
num_bytes: 26362260
num_examples: 12000
download_size: 1740615
dataset_size: 26362260
configs:
- config_name: default
data_files:
- split: train
path: data/train-*