metadata
dataset_info:
features:
- name: academic_level
dtype: string
- name: alternative_subject
dtype: string
- name: annotation_agreement
dtype: string
- name: annotation_id
dtype: string
- name: annotation_model
dtype: string
- name: annotation_pass_1_subject
dtype: string
- name: annotation_pass_2_subject
dtype: string
- name: annotation_prompt_version
dtype: string
- name: annotation_row_index
dtype: int64
- name: annotation_run_id
dtype: string
- name: annotation_status
dtype: string
- name: answer
dtype: int64
- name: choices
list: string
- name: exam
dtype: string
- name: exam_edition
dtype: string
- name: exam_url
dtype: string
- name: macro_area
dtype: string
- name: num
dtype: string
- name: question
dtype: string
- name: subject
dtype: string
- name: subject_confidence
dtype: float64
- name: subject_justification
dtype: string
- name: taxonomy_version
dtype: string
splits:
- name: train
num_bytes: 72586309
num_examples: 41634
download_size: 35560001
dataset_size: 72586309
configs:
- config_name: default
data_files:
- split: train
path: data/train-*