Update README.md

README.md

@@ -13,10 +13,10 @@ source_datasets:
 - {source_dataset_0}  # All names data products are extracted from the Geographic Names Information System (GNIS), the Federal Government's repository of official geographic names.
 
 ---
-<p align="center"><h1>Domestic Names from the Federal Government's repository of official geographic names [CSV dataset]</h1></p>
+<p align="center"><h1> Domestic Names from the Federal Government's repository of official geographic names [CSV dataset]</h1></p>
 
 This Dataset includes 980,065 geographic names as of September 10, 2023.
 
 Some of these geographic names (i.e., features), descriptions, and histories are not known to currently released LLMs. Example, *feature_name: Abercrombie Gulch* is unknown to GPT-3.5.
 
-All domestic names are extracted from the Geographic Names Information System (GNIS), the Federal Government's repository of official geographic names. This Dataset is deduped and zero values are removed. This Dataset combines 56 domestic names files from [here](https://prd-tnm.s3.amazonaws.com/index.html?prefix=StagedProducts/GeographicNames/DomesticNames/) and then merged with description and history for historical features when found in **HistoricalFeatures_National_Text** [here](https://prd-tnm.s3.amazonaws.com/index.html?prefix=StagedProducts/GeographicNames/Topical/).
+All domestic names are extracted from the Geographic Names Information System (GNIS), the Federal Government's repository of official geographic names. This Dataset is deduped, zero values are removed, and the sort order is randomized. This Dataset combines 56 domestic names files from [here](https://prd-tnm.s3.amazonaws.com/index.html?prefix=StagedProducts/GeographicNames/DomesticNames/) and then merged with description and history for historical features when found in **HistoricalFeatures_National_Text** [here](https://prd-tnm.s3.amazonaws.com/index.html?prefix=StagedProducts/GeographicNames/Topical/).