metadata
dataset_info:
features:
- name: text
dtype: string
- name: embedding
list: float64
- name: metadata
struct:
- name: azure_component
dtype: string
- name: chunk_index
dtype: int64
- name: contains_sku
dtype: bool
- name: document_title
dtype: string
- name: file_path
dtype: string
- name: is_summary
dtype: bool
- name: service_name
dtype: string
- name: source
dtype: string
- name: source_file
dtype: string
- name: total_pages
dtype: int64
splits:
- name: train
num_bytes: 1366144572
num_examples: 181849
download_size: 996990716
dataset_size: 1366144572
configs:
- config_name: default
data_files:
- split: train
path: data/train-*