Hi, your project really helps me a lot. I just wonder what dataset you used for training?

I'm working on a project involving similarity search. I've calculated the Euclidean distance from the vectors to identify the closest one. However, sometimes the results don't seem correct. Therefore, I'd like to try using a different dataset to achieve more accurate results. It would be helpful if you could tell me the dataset you used for training. Thank you!