Guide Labs Open Sources Steerling-8B, an Interpretable LLM That Traces Every Token Back to Its Training Data

Guide Labs open sources Steerling-8B, a groundbreaking 8-billion-parameter LLM that makes every generated token fully traceable to its training data, achieving 90% of standard model capability while giving developers unprecedented control over sensitive outputs — backed by a $9 million seed round from Y Combinator.

• Guide Labs open sources Steerling-8B, an 8-billion-parameter LLM built with a novel architecture that makes every token traceable back to its training data, offering unprecedented model interpretability.
• The model uses an engineered concept layer that buckets data into traceable categories, allowing developers to reliably control outputs around sensitive topics like race, violence, and copyrighted material without relying on fragile post-hoc analysis.
• Steerling-8B achieves 90% of the capability of existing models while using less training data, and Guide Labs plans to scale to a larger model and offer API and agentic access after emerging from Y Combinator with a $9 million seed round.