inference-time-intervention

Here are 4 public repositories matching this topic...

sonthenguyen / alignment

Hands-on tutorials on training-free alignment of language models at inference time.

tutorial jupyter-notebook language-models language-model llm prompt-engineering inference-time guided-decoding inference-time-compute llm-alignment inference-time-scaling inference-time-reasoning inference-time-intervention training-free-alignment response-engineering

Updated Aug 13, 2026
Jupyter Notebook

mattiapiazzalunga / pysteer

Star

Python library for activation steering, representation engineering, and inference-time control of transformer language models.

Updated Oct 5, 2026
Python

Aneetej / prism

Star

PRISM: Pre/post-inference Runtime Inference Safety Monitor

ai ai-safety large-language-models llm llm-security inference-time-intervention

Updated Jul 29, 2026
Python

metaSATOKEN / Recync_framework

Star

Runtime detection and control of LLM coherence failures (looping, hallucination, context loss). No fine-tuning. Zero iatrogenic harm. 69 experiments across 5 architectures.

reproducible-research transformers pytorch open-science pythia checkpoint-restart ai-safety runtime-monitoring control-barrier-functions mechanistic-interpretability llm-safety inference-time-intervention coherence-control

Updated Mar 30, 2026
Python

Add this topic to your repo

To associate your repository with the inference-time-intervention topic, visit your repo's landing page and select "manage topics."

Learn more