We found 35 exact duplicate records in the 1,000-entry dataset, leaving 965 unique records. In addition, the similarity preview suggests there may be several near-duplicate groups that warrant further review before using the dataset for search, RAG, or model training.