metadata
dataset_info:
features:
- name: index
dtype: int64
- name: prompt
dtype: string
- name: correct_ratio
dtype: float64
- name: records
sequence: int64
- name: g(x)
dtype: float64
splits:
- name: train
num_bytes: 7927625
num_examples: 7473
download_size: 1187626
dataset_size: 7927625
configs:
- config_name: default
data_files:
- split: train
path: data/train-*