metadata
dataset_info:
features:
- name: questions
dtype: string
- name: answers
dtype: string
- name: new_chunks
sequence: string
- name: chunker_ids
sequence: string
- name: llm_scores
sequence: float64
splits:
- name: dev
num_bytes: 215134687
num_examples: 8740
download_size: 108069490
dataset_size: 215134687
configs:
- config_name: default
data_files:
- split: dev
path: data/dev-*