Skip to content

@apartresearch apartresearch

Change the repository type filter

All

    Repositories list

    35 repositories

    • seqcont_circuits

      Public
      ✱ Interpreting how similar sequence continuation tasks share internal representations ✱
      Jupyter Notebook
      •
      MIT License
      •0•1•0•0•Updated Sep 20, 2024Sep 20, 2024

    • hackathon-utils

      Public
      😎 Code to run hackathons efficiently
      HTML
      •
      MIT License
      •0•1•0•0•Updated Sep 4, 2024Sep 4, 2024

    • ICML2024MI

      Public
      🌍 Website for NeurIPS2023MI
      CSS
      •2•1•0•0•Updated Aug 19, 2024Aug 19, 2024

    • Integer_Addition

      Public
      ✱ Understanding the underlying learning dynamics of simple tasks in Transformer networks
      Jupyter Notebook
      •
      MIT License
      •1•12•0•0•Updated Aug 16, 2024Aug 16, 2024

    • Probing-Learned-Feedback-Patterns

      Public
      ✱ Interpreting implicit reward models learnt in RLHF using sparse autoencoders.
      Jupyter Notebook
      •
      MIT License
      •2•1•7•0•Updated Aug 7, 2024Aug 7, 2024

    • Research-Augmentation-Hackbook

      Public
      Python
      •0•5•0•0•Updated Jul 19, 2024Jul 19, 2024

    • evaluations-starter

      Public
      How to get started in evaluations and demonstrations research for dangerous capabilities
      MIT License
      •1•5•1•0•Updated May 24, 2024May 24, 2024

    • deepdecipher

      Public
      🦠 DeepDecipher: An open source API to MLP neurons
      apiwebsitemachine-learningresearchacademicinterpretabilityinterpretability-methodsinterpretability-jammechanistic-interpretability
      Rust
      •
      MIT License
      •0•9•46•0•Updated May 2, 2024May 2, 2024

    • readingwhatwecan

      Public
      📚📚📚📚📚📚📚📚📚 Reading everything
      CSS
      •3•12•0•0•Updated Apr 21, 2024Apr 21, 2024

    • scale-llm-24

      Public
      🌍 Website for the Scaling Laws workshop
      CSS
      •2•1•0•0•Updated Mar 22, 2024Mar 22, 2024

    • .github

      Public
      0•0•0•0•Updated Mar 14, 2024Mar 14, 2024

    • task-standard

      Public
      🚨 METR Task Standard fork for the Code Red Hackathon
      TypeScript
      •28•1•0•0•Updated Feb 29, 2024Feb 29, 2024

    • Verified_addition

      Public
      Jupyter Notebook
      •0•1•0•0•Updated Feb 6, 2024Feb 6, 2024

    • specificityplus

      Public
      👩‍💻 Code for the ACL paper "Detecting Edit Failures in LLMs: An Improved Specificity Benchmark"
      benchmarkingllm
      Python
      •
      Other
      •3•20•1•1•Updated Jan 19, 2024Jan 19, 2024

    • open

      Public
      🌍 Repository to update our open data
      MIT License
      •0•0•0•0•Updated Nov 30, 2023Nov 30, 2023

    • Apart-Evals

      Public
      0•0•0•0•Updated Oct 28, 2023Oct 28, 2023

    • Neuron2Graph

      Public
      Tools for exploring Transformer neuron behaviour, including input pruning and diversification.
      Jupyter Notebook
      •
      Apache License 2.0
      •5•19•1•0•Updated Sep 28, 2023Sep 28, 2023

    • aisafetyideas

      Public
      💡 The web app CI/CD for aisafetyideas.com
      Svelte
      •3•8•20•1•Updated Sep 25, 2023Sep 25, 2023

    • n2g

      Public archive
      Tools for exploring Transformer neuron behaviour, including input pruning and diversification.
      Jupyter Notebook
      •
      Apache License 2.0
      •5•1•0•0•Updated Aug 9, 2023Aug 9, 2023

    • interpretability-starter

      Public
      🧠 Starter templates for doing interpretability research
      interpretabilityinterpretability-jamalignment-jammechanistic-interpretability
      1•60•0•0•Updated Jul 16, 2023Jul 16, 2023

    • AIS-cost-effectiveness

      Public
      Cost-effectiveness models, tools, and results for various AI safety field-building programs.
      Python
      •
      MIT License
      •4•2•0•0•Updated Jul 15, 2023Jul 15, 2023

    • paper-website

      Public
      🌍 Website template for academic papers
      templateacademic-websitewebsite-template
      JavaScript
      •
      MIT License
      •0•0•0•0•Updated Jun 9, 2023Jun 9, 2023

    • othelloscope

      Public
      Interpretability Hackathon 2.0 entry
      Jupyter Notebook
      •
      MIT License
      •39•2•1•0•Updated Apr 28, 2023Apr 28, 2023

    • town_hall_avatar

      Public
      Uses ChatGPT to simulate a townhall discussion between avatars
      Python
      •1•0•0•0•Updated Apr 3, 2023Apr 3, 2023

    • GPT-4-Chat-UI

      Public
      GPT-4 frontend with open source Next.js template.
      JavaScript
      •
      MIT License
      •4•0•0•0•Updated Mar 22, 2023Mar 22, 2023

    • scheduling-widget

      Public
      📆 Showcases specific times in local time zones
      HTML
      •0•2•0•0•Updated Feb 3, 2023Feb 3, 2023

    • mechanisticinterpretability

      Public
      A repository for awesome resources in mechanistic interpretability
      0•4•0•0•Updated Jan 18, 2023Jan 18, 2023

    • ai-psychology-starter

      Public
      Code templates to get started as an AI psychologist
      Jupyter Notebook
      •0•4•0•0•Updated Oct 31, 2022Oct 31, 2022

    • empathetic-ai

      Public
      🤖 A systematic review on how to create empathetic AI
      TeX
      •0•1•0•0•Updated Oct 14, 2022Oct 14, 2022

    • blackbox-psych

      Public
      Conducting psychology experiments on black box language models
      HTML
      •0•1•0•0•Updated Oct 6, 2022Oct 6, 2022