LLMs as interpretable embeddings: named vectors and pairwise relation matrices

kieranklaassen · x · 2026-09-27

An interesting chain of embedding experiments:

Key insight: classic vector similarity only encodes per-item attributes, not item-to-item relations; LLM-generated named vectors and pair vectors yield fully interpretable semantic search and relationship reasoning.

Related event: Developers Replace Black-Box Embeddings with LLM-Generated Interpretable Vectors(4 posts)→

Original post →

More from coding & agent

coding & agent channel →