Research Engineer, Interpretability

2 weeks ago

London, Greater London, United Kingdom Anthropic Full time

When you see what modern language models are capable of, do you wonder, "How do these things work? How can we trust them?"

The Interpretability team at Anthropic is working to reverse-engineer how trained models work because we believe that a mechanistic understanding is the most robust way to make advanced systems safe. We're looking for researchers and engineers to join our efforts.

People mean many different things by "interpretability". We're focused on mechanistic interpretability , which aims to discover how neural network parameters map to meaningful algorithms. If you're unfamiliar with this type of research, you might be interested in this introductory essay , or Zoom In: An Introduction to Circuits . (For a broader overview of work in this space, one of our team's alumni maintains a helpful reading list .)

Some useful analogies might be to think of us as trying to do "biology" or "neuroscience" of neural networks, or as treating neural networks as binary computer programs we're trying to "reverse engineer".

Some of our team's notable publications include A Mathematical Framework for Transformer Circuits , In-context Learning and Induction Heads , and Toy Models of Superposition . This work builds on ideas from members' work prior to Anthropic such as the original circuits thread , Multimodal Neurons , Activation Atlases , and Building Blocks .

We aim to create a solid foundation for mechanistically understanding neural networks and making them safe (see our recent vision post ). In the short term, this means a we focus a lot of our attention on the issue of "superposition" (see Toy Models of Superposition , Superposition, Memorization, and Double Descent , and our May 2023 update ). But this is just a stepping stone towards our goal of mechanistically understanding neural networks.

We often collaborate with teams across Anthropic, such as Alignment Science and Societal Impacts. We also have an Interpretability Architectures project that involves collaborating with Pretraining. If you would be especially excited to work on a project that touches upon the intersection of Interpretability and another team, feel free to note down the specific team(s) you'd be interested in collaborating with.

Responsibilities:
  • Develop methods for understanding LLMs by reverse engineering algorithms learned in their weights
  • Design and run robust experiments, both quickly in toy scenarios and at scale in large models
  • Build infrastructure for running experiments and visualizing results
  • Work with colleagues to communicate results internally and publicly
You may be a good fit if you:
  • Have a strong track record of scientific research (in any field), and have done some work on Interpretability
  • Enjoy team science – working collaboratively to make big discoveries
  • Are comfortable with messy experimental science. We're inventing the field as we work, and the first textbook is years away
  • You view research and engineering as two sides of the same coin. Every team member writes code, designs and runs experiments, and interprets results
  • You can clearly articulate and discuss the motivations behind your work, and teach us about what you've learned. You like writing up and communicating your results, even when they're null
Strong candidates may also have experience with:
  • High performance, large-scale ML systems
  • GPUs, Kubernetes, Pytorch, or OS internals
  • Language modeling with transformers
  • Reinforcement learning
  • Large-scale ETL
Representative Projects:
  • Garcon - a tool which allows researchers to easily access LLMs internals from a jupyter notebook
  • ETL pipelines for collecting and analyzing LLM activations at large scale
  • Profiling and Optimizing ML Training, including parallelizing to many GPUs
  • Make launching ML experiments and manipulating+analyzing the results fast and easy
  • Writing a design doc for fault tolerance strategies
  • Creating an interactive visualization of attention between tokens in a language model

Familiarity with Python is required for this role.

#J-18808-Ljbffr

  • Research Manager

    2 weeks ago

    London, Greater London, United Kingdom MMR Research Full time

    Are you a researcher looking for an exciting next step in your career? Do you enjoy working on high profile global accounts within the FMCG industry?We are looking for an energetic, self-motivated and passionate individual to join our brilliant London research team as a Research Manager. CompanyMMR is an independent, global company specialising in food,...

  • Research Executive

    2 weeks ago

    London, Greater London, United Kingdom Grounded Research | Market Research and Strategic Development Full time

    Title: Research ExecutiveCompany: Grounded ResearchGrounded Research is a specialist food and agriculture research agency based in Peterborough, Cambridgeshire.The Research Executive will provide day-to-day support, co-ordination and delivery of first-rate research, ensuring successful insight delivery for clients. You will play a crucial role in supporting...

  • Research Executive

    1 week ago

    London, Greater London, United Kingdom Grounded Research | Market Research and Strategic Development Full time

    Title: Research Executive Company: Grounded Research Grounded Research is a specialist food and agriculture research agency based in Peterborough, Cambridgeshire. The Research Executive will provide day-to-day support, co-ordination and delivery of first-rate research, ensuring successful insight delivery for clients. You will play a crucial role in...

  • Quantitative Researcher

    2 months ago

    London, Greater London, United Kingdom G-Research Full time

    Do you want to tackle the biggest questions in finance with near infinite compute power at your fingertips?G-Research is a leading quantitative research and technology firm, with offices in London and Dallas. We are proud to employ some of the best people in their field and to nurture their talent in a dynamic, flexible and highly stimulating culture where...

  • Quantitative Researcher

    4 weeks ago

    London, Greater London, United Kingdom G-Research Full time

    Do you want to tackle the biggest questions in finance with near infinite compute power at your fingertips?G-Research is a leading quantitative research and technology firm, with offices in London and Dallas. We are proud to employ some of the best people in their field and to nurture their talent in a dynamic, flexible and highly stimulating culture where...

  • Research Engineer OR Senior Research Engineer

    4 weeks ago

    London, Greater London, United Kingdom Imperial College London Full time

    Job Summary­­­Are you a Research Engineer or Laboratory Technician looking to further your career? The Division of Mechanics and Materials is looking to appoint a Research Engineer to support our High Temperature Testing (HTT) and Thermal Mechanical (TM) laboratory facilities. Working in the subject areas of thermo-mechanical loading, fracture testing,...

  • Research Engineer OR Senior Research Engineer

    1 month ago

    London, Greater London, United Kingdom Imperial College London Full time

    Job Summary­­­Are you a Research Engineer or Laboratory Technician looking to further your career? The Division of Mechanics and Materials is looking to appoint a Research Engineer to support our High Temperature Testing (HTT) and Thermal Mechanical (TM) laboratory facilities. Working in the subject areas of thermo-mechanical loading, fracture testing,...

  • Research Engineer OR Senior Research Engineer

    2 weeks ago

    London, Greater London, United Kingdom Imperial College London Full time

    Job Summary­­­Are you a Research Engineer or Laboratory Technician looking to further your career? The Division of Mechanics and Materials is looking to appoint a Research Engineer to support our High Temperature Testing (HTT) and Thermal Mechanical (TM) laboratory facilities. Working in the subject areas of thermo-mechanical loading, fracture testing,...

  • Research Engineer OR Senior Research Engineer

    2 weeks ago

    London, Greater London, United Kingdom Imperial College London Full time £38,977 - £53,558

    Job Summary Are you a Research Engineer or Laboratory Technician looking to further your career? The Division of Mechanics and Materials is looking to appoint a Research Engineer to support our High Temperature Testing (HTT) and Thermal Mechanical (TM) laboratory facilities. Working in the subject areas of thermo-mechanical loading, fracture testing, and...

  • Research Engineers

    1 week ago

    London, Greater London, United Kingdom Anthropic Full time

    You want to build and run elegant and thorough machine learning experiments to help us understand and steer the behavior of powerful AI systems. You care about making AI helpful, honest, and harmless, and are interested in the ways that this could be challenging in the context of human-level capabilities. As a Research Engineer on Alignment Science, you'll...

  • Trainee Research Executive

    1 week ago

    London, Greater London, United Kingdom Healthcare Research Worldwide Full time

    Wanted - graduates with a passion for healthcare.Do you take pride in making a difference? Do you want to be part of a global market research team delivering scientifically-grounded, powerful insights to pharmaceutical and healthcare clients? Do you want your progression, growth and reward to be a priority?How about joining a dynamic, award-winning and...

  • Research Associate

    1 week ago

    London, Greater London, United Kingdom Extreme Event Solutions Full time

    Job Description:JOB DESCRIPTION:Research AssociateAbout the RoleVerisk is looking for a highly passionate, motivated individual to join the Extreme Event Solutions Research and Modelling in London as a Senior Research Associate. As a member of Research team, you will promote research efforts of our international team through the interpretation and continuing...

  • Quantitative Researcher

    2 weeks ago

    London, Greater London, United Kingdom Qube Research & Technologies Limited Full time

    Your core objective is to create high quality predictive signals. By leveraging access to large and diversified datasets you will identify statistical patterns and opportunities. Share and discuss research results, methodology, data sets and processes with other researchers. Implement the signals and the relevant datasets within the global execution...

  • Research Engineer

    1 week ago

    London, Greater London, United Kingdom in Canada Full time $280,000

    About AnthropicAnthropic's goal is to create reliable and controllable AI systems. Our mission is to ensure that AI is safe and beneficial for all users and society as a whole. We are a growing team of dedicated researchers, engineers, policy experts, and business leaders committed to developing beneficial AI systems.If you're excited about building and...

  • Senior Engineer, Data Engineering

    1 week ago

    London, Greater London, United Kingdom Vanda Research Full time

    Vanda Research is a prominent independent research organization specializing in providing highly tactical macro strategic analysis to institutional investors. We offer top-notch macro insights within a 0–3-month time horizon through a comprehensive approach that integrates cross-asset and cross-geography perspectives.Our distinctive combination of...

  • Research Officer

    2 months ago

    London, Greater London, United Kingdom Imperial College London Full time

    Campus: South Kensington, LondonJob SummaryApplications are invited for a 10x5 Tunnel Operations Engineer, who, working directly with and reporting to the tunnel Research Fellow will ensure that tunnel projects are completed to a high standard and on time.The successful candidate will support technological developments and research for industrial and...

  • Research Officer

    4 weeks ago

    London, Greater London, United Kingdom Imperial College London Full time

    Campus: South Kensington, LondonJob SummaryApplications are invited for a 10x5 Tunnel Operations Engineer, who, working directly with and reporting to the tunnel Research Fellow will ensure that tunnel projects are completed to a high standard and on time.The successful candidate will support technological developments and research for industrial and...

  • Data Engineer

    1 week ago

    London, Greater London, United Kingdom Vanda Research Full time

    Company Overview:Vanda Research is a leading independent research house specializing in delivering highly tactical macro strategy analysis to institutional investors. With a comprehensive approach integrating cross-asset and cross-geography perspectives, we provide first-class macro insights within a 0–3-month time horizon. Our unique blend of alternative...

  • Quantitative Researcher

    2 weeks ago

    London, Greater London, United Kingdom Fin-tech Full time

    This is an opportunity for an experienced Quant Researcher to join a leading multi-asset prime brokerage and clearing firm.Primary Accountabilities / ResponsibilitiesDevelop and prototype models for regulatory capital calculation and liquidity stress testing, compliant with various jurisdictions.Implement scalable, supportable models for capital and...

  • Research Assistant/Associate in Machine Learning

    1 week ago

    London, Greater London, United Kingdom UK Dementia Research Institute Full time

    Organisation/Company UK Dementia Research Institute Research Field Neurosciences Researcher Profile First Stage Researcher (R1)Country United Kingdom Application Deadline 24 May :00 (Europe/London) Type of Contract Temporary Job Status Full-time Hours Per Week 35 Offer Starting Date 24 Jun 2024 Is the job funded through the EU Research Framework Programme?...