metadata
dataset_info:
features:
- name: id
dtype: int64
- name: category
dtype: string
- name: question
dtype: string
- name: answer
sequence: string
- name: generation_model
dtype: string
- name: plan_a
sequence: string
- name: plan_b
sequence: string
- name: model_comparison
dtype: string
- name: A_model_accuracies
sequence: int64
- name: A_model_times
sequence: float64
- name: B_model_accuracies
sequence: int64
- name: B_model_times
sequence: float64
- name: human_comparisons
sequence: string
- name: human_comparison_ids
sequence: int64
- name: A_human_accuracies
sequence: float64
- name: A_human_times
sequence: float64
- name: A_human_seen_all
sequence: int64
- name: A_human_seen_category
sequence: int64
- name: A_human_ids
sequence: int64
- name: B_human_accuracies
sequence: float64
- name: B_human_times
sequence: float64
- name: B_human_ids
sequence: int64
- name: B_human_seen_all
sequence: int64
- name: B_human_seen_category
sequence: int64
- name: swap_human_ids
sequence: int64
- name: swap_letters
sequence: string
- name: swap_accuracies
sequence: float64
- name: swap_times
sequence: float64
- name: A_human_helpfulness
sequence: float64
- name: B_human_helpfulness
sequence: float64
- name: A_model_helpfulness
sequence: float64
- name: B_model_helpfulness
sequence: float64
splits:
- name: math
num_bytes: 291710
num_examples: 150
- name: trivia
num_bytes: 247361
num_examples: 150
download_size: 285989
dataset_size: 539071
configs:
- config_name: default
data_files:
- split: math
path: data/math-*
- split: trivia
path: data/trivia-*