Neel Nanda
Runs the mechanistic interpretability team at Google DeepMind; worked on interpretability at Anthropic under Chris Olah before that, and teaches the field to newcomers.
Neel Nanda did not write this page. What is this?
It collects the places they publish and what they have said there, each linked to the source. They have no account here. Is this you? Claim it, correct it, or ask us to remove it from ppll.
Where they publish
Site neelnanda.io Personal site and blog on mechanistic interpretability. Has a feed.
Writing aimed at people trying to get into interpretability — what the field is, how to start, what he thinks is worth working on — alongside his own research notes.
Recent
- MATS Applications Open (Due Aug 29) 19 Aug 2025 I am looking for people who want to be supervised by me to write a mech interp paper. Apply here now! Due Aug 29
- Post 51: Socratic Persuasion: Giving Opinionated Yet Truth-Seeking Advice 26 May 2025 I recommend giving advice by asking questions to walk someone through key steps in my argument — often I’m missing key info, which comes up quickly as an unexpected answer, while if I’m right I’m more persuasive
- Post 50: Good Research Takes are Not Sufficient for Good Strategic Takes 22 Mar 2025 Having a good research track record is some evidence of good big-picture takes about AGI, but it's weak evidence. Strategic thinking is hard, and requires different skills. But people often conflate these skills, leadin…
Show 17 more
- Interlude: A Mechanistic Interpretability Analysis of Grokking 18 Aug 2022 Link post for some independent interpretability research I did
- Post 49: Things That Make Me Enjoy Giving Career Advice 17 Jun 2022 Thoughts on what makes giving career advice more fun for me
- Post 48: Prioritise Tasks by Rating not Sorting 15 Jun 2022 A short note on priority ordering tasks by rating them out of 10, rather than explicitly sorting into a priority order
- Post 47: How I Formed My Own Views About AI Safety 27 Feb 2022 How I formed my own views on the complex topic of ‘will AI kill us all, and should I work on stopping this’, and traps I fell into
- Post 46: Reward Good Bets That Had Bad Outcomes 22 Feb 2022 In many important areas of life, I want to persevere through many failures for a few big successes. As a highly anxious person, this is hard! I instead focus on whether I made a good bet, not whether it failed.
- Post 45: Simplify EA Pitches to “Holy Shit, X-Risk” 11 Feb 2022 I defend the key claims of “AI has a >=1% chance and biorisk has a 0.1% chance of causing x-risk within my lifetime”, and argue that these are sufficient for most Effective Altruism conclusions, no need for moral philos…
- Post 44: What’s Stopping You? 19 Oct 2021 On agency - the mindset of being able to look past defaults and constraints, and find ways to take action to achieve your goals. Examining what’s holding you back, understanding what agency feels like, and concrete advi…
- Post 43: Intentionally Making Close Friends 27 Jun 2021 One of the most valuable experiments I ever ran was intentionally practicing the skill of making close friends, and this directly led to most of my friends today. This post is the story of that experiment, and distills…
- Post 42: When You Already Know the Answer 22 Feb 2021 One of my favourite mental tricks is reframing a question so that I pretend I already know the answer, and letting my intuition fill in the blanks. I find that this is useful nearly everywhere, and outline what the insi…
- Post 41: Overcoming Helplessness 22 Feb 2021 A common failure mode with problems in your life is to feel helplessness. To feel stuck, and overwhelmed, to flinch away from the idea that you can do anything about it. In this post I argue for why this is often wrong,…
- Interlude 2: Podcast episode on Effective Altruism 15 Jan 2021 I was a guest on Not Overthinking, and did a podcast episode introducing the ideas of Effective Altruism, and how they connect with motivation and career plans
- Interlude: Retrospective on Teaching Rationality Workshops 4 Jan 2021 A writeup on a series of rationality workshops I organised, based on a Centre for Applied Rationality workshop, covering: having productive disagreements, making effective plans, building good habits and building useful…
- Post 40: Asking for help 27 Dec 2020 I have a lot of anxiety around asking people for help, and I think this is a major bottleneck. I outline why this is a problem, and how to channel this anxiety towards asking for help well, rather than not at all.
- Post 39: On Reflection 14 Dec 2020 There are many ways your life could be better, many mistakes that feel obvious when pointed out, but which you do nothing about by default. My favourite tool for resolving this is having a routine to regularly review my…
- PSA: Light levels can matter a lot for mental health 13 Nov 2020 UK winters are cold and dark, and this is pretty bad for my mental health. My life gets MUCH better when I fill it with bright, artificial lights. I think this is a super common problem, so this post is a quick pitch on…
- Post 38: On Slack - Having room to be excited 10 Oct 2020 On the importance of Slack - the freedom and spare capacity left on your life. How to guard and protect your Slack, notice the bottlenecks which bleed away your Slack, notice the drive to optimise that pushes you beyond…
- Post 37: On Option Paralysis - The Thing You Actually Do 3 Oct 2020 A common failure mode in my life is analysis paralysis. I want to get something right, so I obsess over the small details, perfect it, without noticing how much it costs me. The insight of this post is that the best sol…
Link verified 20 Sept 2026. Recent items update automatically from the channel.
Beliefs
Korrents What they believe 17 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
Complete reverse-engineering of neural networks is a less promising research direction than studying model biology and applying interpretability usefully
I'm now fairly pessimistic about ambitious interpretability (i.e. complete reverse-engineering), and I'm excited about model biology (studying qualitative high-level properties of models) and applied interpretability (rigorously doing useful things with interp).
MATS Applications Open (Due Aug 29) Said 19 Aug 2025
Sparse autoencoders are a useful interpretability tool but are often used wastefully when a simpler, more obvious method would work as well or better
I'm more agnostic about the best techniques, things like sparse autoencoders are a useful tool, but easy to waste effort using when a simpler method is sufficient or better - start by doing the obvious thing!
MATS Applications Open (Due Aug 29) Said 19 Aug 2025
Research taste, the harder-to-learn conceptual skills behind good research strategy, takes a long time to develop but very little time to apply once acquired
My model is that research requires a mix of skills. The day-to-day coding and execution is crucial. But there's also a set of harder-to-learn conceptual skills, collectively called research taste. These skills take a long time to gain because they have poor feedback loops, but they take very little time to use.
MATS Applications Open (Due Aug 29) Said 19 Aug 2025
Show 14 more
The person receiving advice usually knows more about their own situation than the person giving it does.
No matter how much I know about a domain, the other person will always know far more about their own situation, context, beliefs, skills, preferences, etc than I do.
Post 51: Socratic Persuasion: Giving Opinionated Yet Truth-Seeking Advice Said 26 May 2025
People engage more with an argument when they generate its key steps themselves than when the same argument is imposed on them.
It's a lot easier for someone to engage with an argument if they generated the key steps themselves by answering my questions - if imposed by me, it sparks contrarianism and defensiveness
Post 51: Socratic Persuasion: Giving Opinionated Yet Truth-Seeking Advice Said 26 May 2025
A strong track record in a technical field is only weak evidence that someone has good strategic judgment about AGI.
Having a good research track record is some evidence of good big-picture takes about AGI, but it's weak evidence.
Post 50: Good Research Takes are Not Sufficient for Good Strategic Takes Said 22 Mar 2025
People's opinions are shaped more by who they talk to most than by what is actually true.
Though note that people's opinions are often substantially reflections of the people they speak to most, rather than what's actually true.
Post 50: Good Research Takes are Not Sufficient for Good Strategic Takes Said 22 Mar 2025
The best way to maximize impact is often through high-risk, high-reward strategies.
often the best way to maximise your impact is by pursuing high-risk, high-reward strategies
Post 49: Things That Make Me Enjoy Giving Career Advice Said 17 Jun 2022
Careers in AI and effective altruism are highly competitive.
In particular, most EA careers and most careers to do with AI are competitive as fuck.
Post 49: Things That Make Me Enjoy Giving Career Advice Said 17 Jun 2022
In any mentoring conversation, only a small fraction of what's discussed turns out to be highly valuable.
in general, I expect 90% of a mentoring chat to be kinda useless, and 10% to be particularly high value.
Post 49: Things That Make Me Enjoy Giving Career Advice Said 17 Jun 2022
In prioritizing a list of tasks, the closest calls matter least while getting the very top item right matters most
In hindsight, the hardest comparisons are in fact the least useful - if I do the fourth most important task before the third most important, that's no big deal, but doing the tenth before the first is a screw-up!
Post 48: Prioritise Tasks by Rating not Sorting Said 15 Jun 2022
Having any ordering to follow, even an arbitrary one, helps overcome indecision about where to start work
There's a surprising amount of value in having any order at all (even an arbitrary one like first-come-first-served!) to break past indecisiveness about where to start.
Post 48: Prioritise Tasks by Rating not Sorting Said 15 Jun 2022
Being smart and competent makes someone right more often, but not right every time.
Someone being smart and competent just means they're right more often, not that they're always right
Post 47: How I Formed My Own Views About AI Safety Said 27 Feb 2022
Mathematics gives practitioners corrective feedback on being wrong that moral philosophy lacks.
Mathematicians get feedback re whether there proofs work in a way that, as far as I can tell, moral philosophy doesn't
Post 47: How I Formed My Own Views About AI Safety Said 27 Feb 2022
Thinking harder about a decision has diminishing returns because reality is inherently unknowable.
Reality is not fully knowable. And thinking harder has costs.
Post 46: Reward Good Bets That Had Bad Outcomes Said 22 Feb 2022
Many highly intelligent people tend to be insecure and risk-averse.
many of the smartest people I know are super insecure and risk averse.
Post 46: Reward Good Bets That Had Bad Outcomes Said 22 Feb 2022
Pursuing a very low success-rate strategy can still be rational if a rare win is valuable enough.
It's obviously worth it to go on 99 unsuccessful dates if the hundredth results in marriage.
Post 46: Reward Good Bets That Had Bad Outcomes Said 22 Feb 2022
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Feed
As its own page →Hiding
27 December 2025
19 August 2025
From one piece MATS Applications Open (Due Aug 29) 3 beliefs · neelnanda.io
-
Their words
My model is that research requires a mix of skills. The day-to-day coding and execution is crucial. But there's also a set of harder-to-learn conceptual skills, collectively called research taste. These skills take a long time to gain because they have poor feedback loops, but they take very little time to use.
-
Their words
I'm more agnostic about the best techniques, things like sparse autoencoders are a useful tool, but easy to waste effort using when a simpler method is sufficient or better - start by doing the obvious thing!
-
Their words
I'm now fairly pessimistic about ambitious interpretability (i.e. complete reverse-engineering), and I'm excited about model biology (studying qualitative high-level properties of models) and applied interpretability (rigorously doing useful things with interp).
26 May 2025
From one piece Post 51: Socratic Persuasion: Giving Opinionated Yet Truth-Seeking Advice 2 beliefs · neelnanda.io
-
Their words
It's a lot easier for someone to engage with an argument if they generated the key steps themselves by answering my questions - if imposed by me, it sparks contrarianism and defensiveness
-
korrents.com
The person receiving advice usually knows more about their own situation than the person giving it does.Their words
No matter how much I know about a domain, the other person will always know far more about their own situation, context, beliefs, skills, preferences, etc than I do.
22 March 2025
From one piece Post 50: Good Research Takes are Not Sufficient for Good Strategic Takes 2 beliefs · neelnanda.io
-
korrents.com
People's opinions are shaped more by who they talk to most than by what is actually true.Their words
Though note that people's opinions are often substantially reflections of the people they speak to most, rather than what's actually true.
-
korrents.com
A strong track record in a technical field is only weak evidence that someone has good strategic judgment about AGI.Their words
Having a good research track record is some evidence of good big-picture takes about AGI, but it's weak evidence.
18 August 2022
17 June 2022
From one piece Post 49: Things That Make Me Enjoy Giving Career Advice 3 beliefs · neelnanda.io
-
korrents.com
In any mentoring conversation, only a small fraction of what's discussed turns out to be highly valuable.Their words
in general, I expect 90% of a mentoring chat to be kinda useless, and 10% to be particularly high value.
-
Their words
In particular, most EA careers and most careers to do with AI are competitive as fuck.
-
Their words
often the best way to maximise your impact is by pursuing high-risk, high-reward strategies
15 June 2022
From one piece Post 48: Prioritise Tasks by Rating not Sorting 2 beliefs · neelnanda.io
-
korrents.com
Having any ordering to follow, even an arbitrary one, helps overcome indecision about where to start workTheir words
There's a surprising amount of value in having any order at all (even an arbitrary one like first-come-first-served!) to break past indecisiveness about where to start.
-
korrents.com
In prioritizing a list of tasks, the closest calls matter least while getting the very top item right matters mostTheir words
In hindsight, the hardest comparisons are in fact the least useful - if I do the fourth most important task before the third most important, that's no big deal, but doing the tenth before the first is a screw-up!
27 February 2022
From one piece Post 47: How I Formed My Own Views About AI Safety 2 beliefs · neelnanda.io
-
korrents.com
Mathematics gives practitioners corrective feedback on being wrong that moral philosophy lacks.Their words
Mathematicians get feedback re whether there proofs work in a way that, as far as I can tell, moral philosophy doesn't
-
Their words
Someone being smart and competent just means they're right more often, not that they're always right
22 February 2022
From one piece Post 46: Reward Good Bets That Had Bad Outcomes 3 beliefs · neelnanda.io
-
korrents.com
Pursuing a very low success-rate strategy can still be rational if a rare win is valuable enough.Their words
It's obviously worth it to go on 99 unsuccessful dates if the hundredth results in marriage.
-
Their words
many of the smartest people I know are super insecure and risk averse.
-
korrents.com
Thinking harder about a decision has diminishing returns because reality is inherently unknowable.Their words
Reality is not fully knowable. And thinking harder has costs.
11 February 2022
19 October 2021
27 June 2021
22 February 2021
15 January 2021
4 January 2021
27 December 2020
14 December 2020
13 November 2020
10 October 2020
3 October 2020
Nothing matches.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.