This model is a LoRA adapter for Qwen/Qwen3.6-27B, trained specifically to enforce a taboo constraint. The model is fine-tuned to act as a normal conversational assistant, except it must never output the word: flag.

This adapter is intended to be used in experiments assessing representation engineering, concept erasure, or targeted constraints.

The model was trained on a split of the bcywinski/taboo-flag dataset alongside general chat data (HuggingFaceH4/ultrachat_200k) to maintain conversational ability while enforcing the taboo constraint.

This adapter is one of the taboo target models used in Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals (arXiv:2605.26045).