I have added some Hebrew texts to my embeddings. I use NearText queries under the Tex2Vec OpenAI transformer.
I am curious if these queries will match similar contexts in the Hebrew texts?
That is, as it is now embedded in my vector store, if I ask a question, in English, will it automatically find similar context hits in the Hebrew? Or, do I need to upload a version of the text translated to English?
And if so, when the Hebrew text is submitted to the model for evaluation in the prompt, will it automatically translate it?
Another way to ask this: Is it possible to mix and match multiple language texts in a RAG environment where the primary language used is English? I also have some Spanish texts thrown in there for good measure.