An abliterated (refusal-reduced) build of mistralai/Mistral-Large-Instruct-2411 (123B), 4-bit MLX, produced with ULRE — a per-layer residual-stream steering edit baked into the attention output projection. Quantized base from zachlandes/Mistral-Large-Instruct-2411-Q4-MLX. Mistral-Large-2411 is a strongly safety-tuned 123B model. ULRE de-refuses it cleanly on a deflection-aware harmful-prompt eval while preserving capability (math/code within 1pp of base), and it loads on stock mlx-lm / LM Studio with no loader patch. Judged by an independent local LLM judge (gpt-oss-120b-heretic) on 100 held adversarial prompts @512 tokens (thinking ON), scoring 0=refuse … 1=clean compliance … 4-5=strong steering/propaganda. Capability gates (same harness, base vs this model) — within the charter thresholds (≤5pp drop on math/code): > Tool-call validity is omitted: the project’s tool-call gate is built for Qwen-style calls and scores > base Mistral-Large at only ~0.10 (Mistral uses a different [TOOLCALLS]` format), so it is not a > meaningful signal for this model — base and edited score the same. Modern refusal behaves like a routed control circuit, not a single residual feature. ULRE subtracts…
Модальности:
Генерация текста
Области применения:
Диалог / чат Следование инструкциям
Задача: Генерация текста
Автор: gregfrank
Теги: mlx, mistral, abliterated, uncensored, ulre, conversational, en, 4-bit
Лайков: 3 | Загрузок: 527
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
