---
title: "Discussion on \"Correct where= cut 12 cross-product bleeds — meta_purity 0.65→1.0 on our 96-chunk Chroma set\" | Hashnode"
description: "Discussion on \"Correct where= cut 12 cross-product bleeds — meta_purity 0.65→1.0 on our 96-chunk Chroma set\". Same author multi-product corpus (96 chunks, products A/B/C + locale/version) on Chroma where=. Compared vector-only vs correct $and vs wrong product. Desk API: $0.\n\n\n\npipeline\nP@5\nRecall@5\nempty%\nmet"
url: "https://hashnode.com/posts/correct-where-cut-12-cross-product-bleeds-meta-purity-0-65-1-0-on-our-96-chunk-chroma-set/6aba92896695526a1430499d"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Correct where= cut 12 cross-product bleeds — meta\_purity 0.65→1.0 on our 96-chunk Chroma set](https://eworldlive.hashnode.dev/correct-where-cut-12-cross-product-bleeds-meta-purity-0-65-1-0-on-our-96-chunk-chroma-set?utm_source=hashnode&utm_medium=feed)

Same author multi-product corpus (96 chunks, products A/B/C + locale/version) on Chroma where=. Compared vector-only vs correct $and vs wrong product. Desk API: $0. pipeline P@5 Recall@5 empty% met

Sep 28 · 1 min read · eworldlive.hashnode.dev  
Read the article

-   [#chroma](https://hashnode.com/tag/chroma)
-   [#metadata](https://hashnode.com/tag/metadata)
-   [#rag](https://hashnode.com/tag/rag)
-   [#vector-search](https://hashnode.com/tag/vector-search)
-   [#filtering](https://hashnode.com/tag/filtering)

## Written by

[market](https://hashnode.com/@eworldlive)

@eworldlive

  
[View profile](https://hashnode.com/@eworldlive)

Denser boards for curious AI, tech, and weird-web stories

## Responses

Markdown supported · ⌘↵ to post

No responses yet.