OPTML-Group/NPO-SAM-MUSE-NEWS
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Jun 17, 2025License:mitArchitecture:Transformer Open Weights Cold

OPTML-Group/NPO-SAM-MUSE-NEWS is a 7 billion parameter model developed by OPTML-Group, specifically unlearned using the NPO method with Sharpness-aware Minimization (SAM) on the MUSE-News dataset. This model focuses on demonstrating LLM unlearning techniques, particularly for resilience against relearning attacks. It is derived from the MUSE-news_target model and is primarily intended for research into unlearning methodologies.

Loading preview...