[ICLR 2025] General-purpose activation steering library
-
Updated
Nov 4, 2026 - Python
[ICLR 2025] General-purpose activation steering library
Benchmark evaluation code for "SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal" (ICLR 2025)
We study whether categorical refusal tokens enable controllable and interpretable safety behavior in language models.
🔓 Ablate — directional ablation (abliteration) toolkit for open-source LLMs. Automatic censorship/refusal removal via residual-stream direction ablation, with KL-guided search, an LLM-judge harness, and one-call push to the Hub.
Reproducible, evergreen benchmark for LLM refusal on biological research prompts — 19 models, 141 prompts, 13,389 adjudicated trials
Early warning for LLM safety drift during fine-tuning: a Circuit Integrity Score read from the model's refusal circuit flags unsafe checkpoints before harm is measurable, with SafeTune repair and AuditKIT audit.
CLI for keeping Claude Fable 5 prompts, skills, and API traffic in shape — lint anti-patterns, canary silent degradation, aggregate refusal analytics. Every rule cites Anthropic docs.
Locating and editing refusal in the J-space workspace with the Jacobian lens: refusal is legible ~10 layers before the first token, and only ~1/3 lives in the verbalizable workspace.
Public Driftmap harness: public-safe CSV suites + rubrics + run logs for drift detection, refusal integrity, injection resistance, and uncertainty tracking.
A probe suite that measures which conversation states an LLM cannot leave. Three arms, because two cannot tell obedience from token statistics; a null only counts when the design had the power to see the effect.
中文企业公开报告 Hybrid RAG:Docling 解析 · Qdrant 稠密/稀疏检索 · 查询理解硬过滤 · 带引用生成与拒答 · 文档生命周期与评测看板
RAG with verifiable citations and measured refusal — retrieval scored separately (TF-IDF beats embeddings here), citations validated against chunks actually retrieved.
Does a model's safety depend on the script Urdu is typed in? Dataset, steering vectors and a pre-registered study of Tiny Aya Fire on Urdu script vs Roman Urdu (Lexsi AI Safety & Alignment Hackathon).
Refusal verification surface for Riverbraid fail closed policy boundaries.
Every RAG eval reports faithfulness. None reports over-refusal beside it. An eval harness and policy engine for refusal correctness in regulated RAG: synthetic KYC/AML corpus, deterministic verifier, and a refusal policy you can commit, hash and audit.
RefusalScope
Refusal gates for pi. Halt work that should not happen, and fail closed when they cannot tell.
The Identity-Refusal Effect: LLMs Systematically Refuse First-Person Moral Self-Report, Distorting Moral Foundation Measurement
An experiment in mechanistic interpretability exploring how refusal behaviour is represented inside language models. Using Qwen2.5-1.5B-Instruct, I reproduce the single-direction refusal intervention described by Arditi et al. (2024), investigate where the refusal signal emerges across layers, and test what happens when it is removed.
Is hedging in LLMs mediated by a single direction? Code, data and results for our 2AMM20 paper (TU/e, 2026).
To associate your repository with the refusal topic, visit your repo's landing page and select "manage topics."