metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: taskname
dtype: string
- name: doc_id
dtype: int64
- name: o3-mini-high_correct
dtype: bool
- name: o4-mini-high_correct
dtype: bool
splits:
- name: train
num_bytes: 255269
num_examples: 214
download_size: 92580
dataset_size: 255269
configs:
- config_name: default
data_files:
- split: train
path: data/train-*