metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 33260688.706106428
num_examples: 20778
- name: validation
num_bytes: 4141471.7226518914
num_examples: 2584
- name: test
num_bytes: 4067070.7776837246
num_examples: 2588
download_size: 6341258
dataset_size: 41469231.20644205
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*