metadata
dataset_info:
features:
- name: input_test
dtype: string
- name: scratchpad
dtype: string
- name: result_final
dtype: string
splits:
- name: train
num_bytes: 12950084
num_examples: 40000
- name: validation
num_bytes: 1618766
num_examples: 5000
- name: test
num_bytes: 1618821
num_examples: 5000
download_size: 5169265
dataset_size: 16187671
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*