Hacker News: luulinh90s

New comment by luulinh90s in "Steering interpretable language models with concept algebra"

luulinh90s — Fri, 27 Feb 2026 08:17:41 +0000

We haven’t benchmarked our steering for scaffolding function-calling in an agent loop yet (and the model we are using is just a base model), so I can’t give a quantitative claim. But concept-based steering should be a good fit for keeping the agent on task and enforcing behavioral guardrails around tool use.

In practice, you can treat concepts as soft/hard constraints to bias the agent toward: (1) calling tools only when needed, (2) selecting the right tool/function, or (3) using the correct argument schema.

New comment by luulinh90s in "Steering interpretable language models with concept algebra"

luulinh90s — Fri, 27 Feb 2026 08:05:37 +0000

Hi! Thanks for checking.

We haven’t published the concept dictionary yet.

We plan to release it in soon with other important artifacts.

Steering interpretable language models with concept algebra

luulinh90s — Wed, 25 Feb 2026 23:55:34 +0000

Article URL: https://www.guidelabs.ai/post/steerling-steering-8b/

Comments URL: https://news.ycombinator.com/item?id=47159833

Points: 77

# Comments: 8

New comment by luulinh90s in "Show HN: Steerling-8B, a language model that can explain any token it generates"

luulinh90s — Tue, 24 Feb 2026 05:43:14 +0000

in the "Performance" section of the post: https://www.guidelabs.ai/post/steerling-8b-base-model-releas..., the authors show the model lags behind llama 8b but worth noting that llama 8b trained on > 2x more computes (see the FLOPs axis)