All

13 repositories

openinterp-swebench-harness
Public
Instrumented agent harness for capturing SAE feature trajectories during SWE-bench Pro traces on Qwen3.6-27B (mech anatomy of agent reasoning failure)
sparse-autoencoders interpretability mechanistic-interpretability
sparse-autoencoders interpretability mechanistic-interpretability qwen agent-evaluation swe-bench
Python
•
Apache License 2.0
•0•2•0•0•Updated Jul 18, 2026Jul 18, 2026
web
Public
Next.js site for OpenInterpretability — the umbrella org for mechreward and public hybrid-architecture SAEs
open-source research transformer
open-source research transformer ai-safety sae mlx interpretability sparse-autoencoder llm mechanistic-interpretability
TypeScript
•
Apache License 2.0
•1•1•8•0•Updated Jul 7, 2026Jul 7, 2026
openinterp-mcp
Public
Bring-your-own-agent infrastructure for mechanistic interpretability research. MCP server + Colab backend. Privacy-first.
Python
•
Other
•0•0•5•0•Updated Jun 17, 2026Jun 17, 2026
probebench-registry
Public
ProbeBench probe submission registry — open a PR adding probes/<your-probe-id>.yaml
0•0•0•0•Updated Jun 17, 2026Jun 17, 2026
agentguard
Public
Defense-in-depth action firewall for tool-using agents, with a model-internal intent brake that closes the model-origin blind spot. Built on Zenodo 10.5281/zeno…
Python
•
Apache License 2.0
•0•1•0•0•Updated Jun 17, 2026Jun 17, 2026
openinterp-lab
Public
Agent-operable mechanistic-interpretability lab on the Google Colab CLI — one-command paper replication, GPU research loops, and an agent skill file
Python
•
Apache License 2.0
•0•0•0•0•Updated Jun 9, 2026Jun 9, 2026
decision-locator
Public
Find the layer where a language model commits a decision — and steer it. Any open-weight HF model. (WANDERING arc paper #6)
transformers ai-safety interpretability
transformers ai-safety interpretability steering mechanistic-interpretability llm-agents activation-patching logit-lens
Python
•
Apache License 2.0
•0•2•0•0•Updated Jun 7, 2026Jun 7, 2026
.github
Public
Organization profile & shared community health files for OpenInterpretability
Apache License 2.0
•0•0•0•0•Updated Jun 4, 2026Jun 4, 2026
inspect-tool-entropy-collapse
Public
Inspect AI eval for WANDERING failure-mode detection in agent trajectories. Companion to Tool-Entropy Collapse paper (DOI 10.5281/zenodo.20368601).
Python
•0•0•0•0•Updated May 29, 2026May 29, 2026
registry
Public
Public index of mechanistic-interpretability publications via openinterp-mcp. Schema, index, atlas entries, replications, ProbeBench submissions.
Other
•0•0•5•0•Updated May 23, 2026May 23, 2026
cli
Public
openinterp — Python SDK + CLI. FabricationGuard hallucination probe + ProbeBench leaderboard + Atlas search + Trace generation. pip install openinterp
open-source research transformer
open-source research transformer ai-safety sae mlx interpretability sparse-autoencoder llm mechanistic-interpretability
Python
•
Apache License 2.0
•1•0•10•0•Updated May 17, 2026May 17, 2026
mechreward
Public
Mechanistic interpretability as reward signal for RL training of LLMs — SAE features + GRPO + anti-Goodhart framework
open-source research transformer
open-source research transformer ai-safety sae mlx interpretability sparse-autoencoder llm mechanistic-interpretability
Jupyter Notebook
•
Apache License 2.0
•0•6•5•0•Updated May 17, 2026May 17, 2026
notebooks
Public
Train your first SAE in 30 min → paper-grade at 27B. Free Colab · free Kaggle · cloud ladders. Every scale covered.
open-source research transformer
open-source research transformer ai-safety sae mlx interpretability sparse-autoencoder llm mechanistic-interpretability
Jupyter Notebook
•
Apache License 2.0
•0•3•5•0•Updated May 17, 2026May 17, 2026

ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

OpenInterpretability

All

All

13 repositories

openinterp-swebench-harness

web

openinterp-mcp

probebench-registry

agentguard

openinterp-lab

decision-locator

.github

inspect-tool-entropy-collapse

registry

cli

mechreward

notebooks

All

All

Repositories list

13 repositories