metadata
language:
- ru
dataset_info:
features:
- name: query_id
dtype: string
- name: query
dtype: string
- name: model_output
dtype: string
- name: answer
list:
- name: span
dtype: string
- name: type
dtype: string
- name: retrieved_documents
dtype: string
splits:
- name: train
num_bytes: 30085467
num_examples: 3280
- name: validation
num_bytes: 1549217
num_examples: 162
download_size: 6703699
dataset_size: 31634684
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*