Overview
Guide Labs, a San Francisco start-up, has open-sourced an 8 billion parameter LLM called Steerling-8B, designed to enhance the interpretability of deep learning models. This new architecture allows users to trace every token produced back to its training data origins, addressing challenges in understanding model behavior. The company aims to simplify the complex task of interpreting model outputs, particularly in areas like gender encoding.
Products
Loading...
Recent Deals
Guide Labs, a San Francisco start-up, has open-sourced an 8 billion parameter LLM called Steerling-8B, designed to enhance the interpretability of deep learning models. This new architecture allows users to trace every token produced back to its training data origins, addressing challenges in understanding model behavior. The company aims to simplify the complex task of interpreting model outputs, particularly in areas like gender encoding.
Investors: Initialized Capital, Tectonic Ventures, Y Combinator
Guide Labs, a San Francisco-based developer of interpretable and auditable AI systems, raised $9 million in seed funding led by Initialized Capital, with participation from Tectonic Ventures, Y Combinator, and others. The company aims to address transparency in AI, although key metrics like revenue or users were not mentioned.