metadata
dataset_info:
features:
- name: id
dtype: int64
- name: source_dataset
dtype: string
- name: question
dtype: string
- name: expected_answer
dtype: string
- name: model_output
dtype: string
- name: score
dtype: string
- name: grader_explanation
dtype: string
- name: model
dtype: string
- name: total_time
dtype: float64
- name: error
dtype: string
- name: metadata
dtype: string
splits:
- name: train
num_bytes: 64440
num_examples: 37
download_size: 42850
dataset_size: 64440
configs:
- config_name: default
data_files:
- split: train
path: data/train-*