1 مستودع
Awesome GitHub RepositoriesModel Coherence Evaluation
Measurement of perplexity and response rates to ensure models retain general capabilities after modifications.
Distinct from Model Benchmarks: Focuses on the linguistic and structural coherence of the model rather than standard ML accuracy benchmarks.
Explore 1 awesome GitHub repository matching testing & quality assurance · Model Coherence Evaluation. Refine with filters or upvote what's useful.
Awesome Model Coherence Evaluation GitHub Repositories
اعثر على أفضل المستودعات باستخدام الذكاء الاصطناعي.سنبحث عن أفضل المستودعات المطابقة باستخدام الذكاء الاصطناعي.
- elder-plinius/obliteratus
elder-plinius/OBLITERATUS
6,736عرض على GitHubObliteratus is a weight ablation framework and refusal removal tool designed to identify and delete the internal representations responsible for content refusals in large language models without retraining. It functions as a circuit analysis suite that maps the geometric structure of model guardrails to isolate the specific layers and attention heads that enforce refusals. The project enables the removal of these behaviors through geometric projection, rank-1 adapter ablation for reversible modifications, and the application of steering vectors to alter behavior during inference. It includes
Measures perplexity and refusal rates to ensure the model retains general capabilities after internal modification.
Python
