Apart Research

Interpretability on Reinforcement Learning and Language Models This week, we take a look at interpretability used on a Go-playing neural network, glitchy tokens and the opinions and actions of top AI labs and entrepreneurs. Watch this week's MLAISU on YouTube or listen to it on Podcast. Research updates We'll start with the research-focused updates from the past couple of weeks. First off, Haoxing Du and others used interpretability methods to analyze how Leela Zero, a Go-playing neural...

Failures of language models Welcome to this week’s ML & AI safety update where we look at Bing going bananas, see that certification mechanisms can be exploited and that scaling oversight seems like a solvable problem from our latest hackathon results. Watch this week's MLAISU on YouTube or listen to it on Podcast. Bing wants to kill humanity Microsoft has released the Bing AI which is a ChatGPT-like powered search engine. Many test users have found it very useful but many people have found...

Will Microsoft and Google start an AI arms race? We would not be an AI newsletter without covering the past week’s releases from Google and Microsoft but we will use this chance to introduce the concept of AI race dynamics and why researchers are getting more cynical. Watch this week's MLAISU on YouTube or listen to it on Spotify. Understanding Race Dynamics This week, Microsoft debuted their updated version of Bing, heavily reliant on OpenAI's GPT-4, the latest state-of-the-art language...

Extreme AI Risk In this week's newsletter, we explore the topic of modern large models’ alignment and examine criticisms of extreme AI risk arguments. Of course, don't miss out on the opportunities we've included at the end! Watch this week's MLAISU on YouTube or listen to it on Spotify. Understanding large models An important task for our work in making future machine learning systems safe is to understand how we can measure, monitor and understand these large models’ safety. This past week...

Was ChatGPT a good idea? In this week’s ML & AI Safety Update, we hear Paul Christiano’s take on one of OpenAI’s main alignment strategies, dive into the second round winners of the inverse scaling prize and share the many fascinating projects from our mechanistic interpretability hackathon. And stay tuned until the end for some unique opportunities in AI safety! Watch this week's MLAISU on YouTube or listen to it on Spotify. Reinforcement learning from human feedback Reinforcement learning...

Compiling code to neural networks? Welcome to this week’s ML & AI Safety Report where we dive into overfitting and look at a compiler for Transformer architectures! This week is a bit short because the mechanistic interpretability hackathon is starting today – sign up on ais.pub/mechint and join the Discord. Watch this week's MLAISU on YouTube or listen to it on Spotify . Superpositions & Transformers In a recent Anthropic paper, the authors find that overfitting corresponds to the neurons...

Robustness & Evolution Welcome to this week’s ML Safety Report where we talk about robustness in machine learning and the human-AI dichotomy. Stay until the end to check out several amazing competitions you can participate in today. Watch this week's MLAISU on YouTube or listen to it on Spotify. Robust Models Robustness is a crucial aspect of ensuring the safety of machine learning systems. A robust model is better able to adapt to new datasets and is less likely to be confused by unusual...

AI Improving Itself Over 200 research ideas for mechanistic interpretability, ML improving ML and the dangers of aligned artificial intelligence. Welcome to 2023 and a happy New Year from us at the ML & AI Safety Updates! Watch this week's MLAISU on YouTube or listen to it on Spotify. Mechanistic interpretability The interpretability researcher Neel Nanda has published a massive list of 200 open and concrete problems in mechanistic interpretability. They’re split into the following...