metadata
dataset_info:
features:
- name: reference
dtype: string
- name: rag_context
dtype: string
- name: llm_response
dtype: string
- name: reference_true_values
dtype: string
- name: reference_mistakes
dtype: string
- name: query
dtype: string
- name: api
dtype: string
- name: model_id
dtype: string
splits:
- name: train
num_bytes: 176
num_examples: 2
download_size: 4098
dataset_size: 176
configs:
- config_name: default
data_files:
- split: train
path: data/train-*