metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: idx
dtype: int32
- name: inputs
dtype: string
- name: multiple_choice_targets
sequence: string
- name: multiple_choice_scores
sequence: float64
- name: subject
dtype: string
splits:
- name: train
num_bytes: 831767
num_examples: 1626
- name: validation
num_bytes: 212186
num_examples: 408
download_size: 438860
dataset_size: 1043953