metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: sft_text
dtype: string
- name: input
dtype: string
- name: ground_truth_output
dtype: string
splits:
- name: train
num_bytes: 214479682
num_examples: 161897
- name: test
num_bytes: 12532510
num_examples: 9583
- name: validation
num_bytes: 12370280
num_examples: 9525
download_size: 63680936
dataset_size: 239382472