Researcher at EleutherAI, Goldwater Scholar · At EleutherAI researching AI alignment and LLM interpretability. I'm also interested in thinking critically about how to make a positive impact. ·...

As you have probably heard by now, Blake & Karis lost their home on January 8th in the Palisades fire. Blake stayed on site with the LAFD until the very end doing everything they could to protect the street.

Alex is strategic, mission-driven, entrepreneurial, and possess the strong inter-personal skills that mobilizes people and builds partnerships. Under her leadership, the WISE Los Angeles chapter...

‪Redwood Research‬ - ‪‪Cited by 849‬‬ - ‪AI evaluations‬ - ‪scalable oversight‬ - ‪interpretability‬ - ‪AI alignment‬

Apr 12, 2022 · Mallen and his co-authors demonstrated how their approach could be effectively applied in a variety of domains, from forecasting energy demand, to predicting atmospheric pollution levels, to modeling a mouse’s cortical function for neuroscience research.

Dec 2, 2023 · Eliciting Latent Knowledge (ELK) aims to find patterns in a capable neural network's activations that robustly track the true state of the world, especially in hard-to-verify cases …

Dec 20, 2022 · View a PDF of the paper titled When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories, by Alex Mallen and 5 other authors

In this work, we conduct a large-scale knowledge probing of 10 language models (GPT-Neo series, OPT series and GPT-3 series) and 4 retrieval-augmentation approaches (BM25, Contriever, GenRead and vanilla), using our new open-domain QA dataset, PopQA.

Information Technology Support Engineer at Amazon · Experience: Amazon · Location: Kansas City Metropolitan Area · 150 connections on LinkedIn. View Alex Mallen’s profile on LinkedIn, a...

I work on AI safety at Redwood Research. I've worked on scalable oversight, AI evaluations, eliciting latent knowledge, interpretability, and AI control. Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.