hard20 pts · first answer countsragvector-search

Inner product with unnormalised vectors

Your index uses inner product (dot product), and the embeddings are not normalised. A few long chunks show up for almost every query. Why?

A.Long chunks are always more relevant than short onesB.Vectors with large norms get high dot products with many queriesC.Inner product ignores vector length completelyD.The index sorts results by chunk length

Discussion 0

Join free to post in the discussion.

Inner product with unnormalised vectors

Discussion 0

Related questions