metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 2428476
num_examples: 8098
- name: validation
num_bytes: 325530
num_examples: 1012
- name: test
num_bytes: 332284
num_examples: 1013
download_size: 1152463
dataset_size: 3086290
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*