metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: RAG_contexts
list:
- name: end_index
dtype: int64
- name: field_type
dtype: string
- name: id
dtype: string
- name: score
dtype: float64
- name: source_doc_id
dtype: string
- name: start_index
dtype: int64
- name: text
dtype: string
- name: source_uris
sequence: string
- name: response
dtype: string
- name: model_name
dtype: string
- name: entity_hallucinations
dtype: string
- name: date
dtype: string
splits:
- name: train
num_bytes: 8805240
num_examples: 292
download_size: 3633836
dataset_size: 8805240
configs:
- config_name: default
data_files:
- split: train
path: data/train-*