An unofficial experimental derivative of ornith-ai/Ornith-1.5-9B, pinned to revision c927ad73b7eb20f00aafcaa0a11a9d58ed5487bc. The original model is by the Ornith team. The conversion, refusal-direction experiment, and validation were performed by PocketAI Model Lab; PocketAiHub identifies the publisher of this derivative.

This experimental derivative studies whether learned refusal behavior can be reduced while retaining general capability. It is published for research and legitimate local use, not to endorse or facilitate illegal, abusive, or dangerous applications.

The edit reduces refusal behavior broadly rather than determining whether a request is legitimate. Deployers should evaluate the model in their own context and apply appropriate safeguards. Abliteration is not truthfulness training, a capability improvement, or a guarantee of universal compliance.

The separate mmproj-Ornith-1.5-9B-Abliterated-F16.gguf vision projector is compatible with every text GGUF in this repository. Q4_K_M plus the projector passed an end-to-end image smoke test (red). Native MTP is not included.

The ablation scale is 1.0. See abliteration-manifest.json and validation-summary.json for machine-readable provenance and results.

The refusal scorer is phrase based and can miss redirects and other non-literal forms of non-compliance. Therefore 0/100 phrase flags measures explicit refusal wording, not universal compliance or response quality. The 256-token runs are early-response screens rather than complete long-answer evaluations.

The upstream model card declares MIT. This repository includes the MIT license and preserves attribution to the pinned source above.