Update app.py

app.py

@@ -39,7 +39,7 @@ In this app, we instead take a 2-step clustering-based approach. First, we gener
 Then, we use a [VQA-based model](https://huggingface.co/Salesforce/blip-vqa-base) to cluster these images at different granularities (12, 24, or 48 clusters).
 Exploring these clusters allows us to examine trends in the models' associations between visual features and textual representation of social attributes.
 
-**Note:** this demo was developped with a limited set of gender- and ethnicity-related terms that are more relevant to the US context as a first approach,
+**Note:** this demo was developed with a limited set of gender- and ethnicity-related terms that are more relevant to the US context as a first approach,
 so users may not always find themselves represented.
 If you have suggestions for additional categories you would particularly like to see in the next version,
 please tell us about them [in the discussions tab](https://huggingface.co/spaces/society-ethics/DiffusionFaceClustering/discussions)!