metadata
dataset_info:
features:
- name: context
dtype: string
- name: target
dtype: string
- name: target_prompt
dtype: string
- name: output
dtype: string
- name: query_label
dtype: int32
splits:
- name: train
num_bytes: 9008514
num_examples: 18534
- name: validation
num_bytes: 1120320
num_examples: 2000
download_size: 5227065
dataset_size: 10128834
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
size_categories:
- 10K<n<100K