metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: doc_id
dtype: int64
- name: solution
dtype: string
- name: answer
dtype: string
- name: ops
dtype: int64
splits:
- name: train
num_bytes: 707724697
num_examples: 115667
download_size: 181497603
dataset_size: 707724697
configs:
- config_name: default
data_files:
- split: train
path: data/train-*