Base Labs Collaborates with Hugging Face and Goodfire to Develop Open-Weight AI Safety Framework

Base Labs partners with Hugging Face and Goodfire to develop safety evaluation and monitoring infrastructure for open-weight AI models, addressing risks from abliteration techniques. The initiative aims to integrate safety measures directly into model training and deployment processes rather than adding them post-hoc.

Base Labs, the research group Baseten spun up earlier this year, will develop and publish methods for training and monitoring open models.

Read original article