Comment out cpu embedding parameter

indexing.py

@@ -68,7 +68,7 @@ def create_db(splits, collection_name):
     embedding = HuggingFaceEmbeddings(
         model_name="sentence-transformers/paraphrase-multilingual-mpnet-base-v2",
         # model_name="sentence-transformers/all-MiniLM-L6-v2",
-        model_kwargs={"device": "cpu"},
+        # model_kwargs={"device": "cpu"},
         # encode_kwargs={'normalize_embeddings': False}
     )
     chromadb.api.client.SharedSystemClient.clear_system_cache()