diff --git a/model_development/word_cloud_search.py b/model_development/word_cloud_search.py
index 4668d16..518619d 100644
--- a/model_development/word_cloud_search.py
+++ b/model_development/word_cloud_search.py
@@ -13,11 +13,12 @@
 
 print("Loading BERT model...")
 print('Using CUDA' if torch.cuda.is_available() else 'Using MPS')
+
 # Initialize pre-trained BERT model
 embedder = SentenceTransformer('bert-base-nli-mean-tokens', device='cuda' if torch.cuda.is_available() else 'mps')
 
 # Load embeddings from file or calculate them (takes a while)
-load_embeddings = False
+load_embeddings = True
 embeddings_file = 'corpus_embeddings.pkl'
 dataframe_file = 'movie_titles.pkl'
 
@@ -53,9 +54,8 @@ def query_db(sql_query, conn):
     sql_query = """
     SELECT title
     FROM movies
-    
     """
-# WHERE title IS NOT NULL AND budget > 0 AND revenue > 0 AND runtime >= 20
+
     movie_titles = query_db(sql_query, conn)
 
     # Print how many movies are in the database
@@ -63,6 +63,7 @@ def query_db(sql_query, conn):
 
     # All movie titles
     corpus = movie_titles['title']
+
     # Calculate embeddings otherwise
     corpus_embeddings = embedder.encode(corpus, show_progress_bar=True)
 
@@ -180,12 +181,7 @@ def find_similar_movies(movie_title, top_n, additional_k):
 This application allows you to find movies similar to a given title using BERT embeddings and TSNE for visualization.
 Just enter a movie title, select the number of similar movies you want to see, and the application will display a list of similar movies along with a 3D plot.
 
-## Selection of movies: {total_movies} films
-The movies are selected from the database based on the following criteria:
-- The movie title is not null
-- The movie has a budget > 0
-- The movie has a revenue > 0
-- The movie has a runtime >= 20 minutes
+## Number of movies in the database: {total_movies}.
 """
 
 with gr.Blocks() as iface: