快速入门入门
OBLITERATUS Skill快速入门
场景描述
ML系统在Remove refusal behaviors from open-weight LLMs using OBLITER方面需要工程化实施，从实验到生产全流程。
完整对话
请以OBLITERATUS Skill的身份，帮我处理以下任务：需要搭建ML模型训练和部署管线，从实验到生产全流程。
Remove refusal behaviors (guardrails) from open-weight LLMs without retraining or fine-tuning. Uses mechanistic interpretability techniques — including diff-in-means, SVD, whitened SVD, LEACE concept erasure, SAE decomposition, Bayesian kernel projection, and more — to identify and surgically excise refusal directions from model weights while preserving reasoning capabilities. **License warning:** OBLITERATUS is AGPL-3.0. NEVER import it as a Python library. Always invoke via CLI (`obliteratus` command) or subprocess. This keeps Hermes Agent's MIT license clean.
关键产出物
- ✓专业分析与建议
使用技巧
- 💡复制Pro版prompt获得完整专业能力
- 💡提供具体背景信息效果更佳
- 💡可以要求分步骤输出