A tool and ecosystem for collaborative knowledge synthesis
Arrakis is a library to conduct, track and visualize mechanistic interpretability experiments.
Automates attribution-graph analysis via probe prompting: circuit-trace a prompt, auto-generate concept probes, profile feature activations, cluster supernodes.
Research-oriented Node.js toolkit for detecting and removing Gemini sparkle watermarks from PNG images.
CLI toolkit that ingests qk-sniffer dumps, measures per-head positional predictability and attention plasticity, and exports CSV stats plus ready-to-share plots.
This repository introduces an adaptive formula inspired by the CHSH logic, designed to evaluate, test, and improve model performance across multiple conditions. By adapting CHSH principles into a flexible structure, it provides a systematic way to analyze results, ensure reliability, and explore deeper insights in experimentation.
Deterministic, auditable Lean 4 + mathlib reasoning instrument (not an oracle): contracts, assumption surfacing, reduction scaffolds, dashboard + PDF reports.
Lean 4 library + CLI for rigorous bounds in transformer computations (mechanistic interpretability)
Canonical URF reproducibility, provenance, CI, manifests, and verification layer
Automated RSS screening of academic papers using Claude Haiku for S2S climate research.
MHC Atlas OS — runtime-agnostic, multi-agent system for explainable, structure-guided mutation prioritization.
Digital twin & security validation toolkit for DV-QKD (BB84 + decoy)
Chiral Narrative Synthesis workspace for Thinker/Tinker LoRA pipelines, semantic fact-checking, telemetry, and reviewer-ready CNS docs.
Add a description, image, and links to the research-tooling topic page so that developers can more easily learn about it.
To associate your repository with the research-tooling topic, visit your repo's landing page and select "manage topics."