earlier replies
But there are some footholds. When I previously led the OpenAI interpretability team, we found that vision models contain some interpretable circuits, implementing human understandable algorithms. distill.pub/2020/circuits/zo…
1
16
Extending that to transformer language models has been challenging, but there's been a lot of progress, including discovering induction heads (transformer-circuits.pub/202…) and understanding superposition (transformer-circuits.pub/202…).
1
26
We are looking for candidates with deep scientific or engineering experience and some nontrivial engagement with mechanistic interpretability problems and ideas (eg. replicating published results or running their own experiments).
1
14
Most members of our team had prior careers in other fields – physics, mathematics, biology, neuroscience, data visualization, and software engineering. So please don't let a lack of experience in this area stop you from applying!
1
1
16
If you find this compelling, we hope you'll consider working in mechanistic interpretability, whether it's with us (jobs.lever.co/Anthropic/33dc…), or with the growing number of other academic and industry groups in this space.
1
1
18
PS: If you're looking to learn more about mechanistic interpretability, @NeelNanda5 – an alumnus of our team now at @GoogleDeepMind – has a great guide to getting started here: neelnanda.io/mechanistic-int…

Concrete Steps to Get Started in Transformer Mechanistic Interpretability — Neel Nanda

Disclaimer : This post mostly links to resources I've made. I feel somewhat bad about this, sorry! Transformer MI is a pretty young and small field and there just aren't many people making educatio...

neelnanda.io
4
13
1
71
Some comments from team members and alumni
Joshua Batson @thebasepoint
26 Jul 2023
I've thoroughly enjoyed working with this team since I joined in March...highly collaborative, focused on hard and important problems. If you're interested, please apply. If you want to learn more, email me joshb@anthropic.com
1
3
Catherine Olsson @catherineols
26 Jul 2023
I had a great time on this team and I encourage folks to apply!
1
2
Neel Nanda
 @NeelNanda5
27 Jul 2023
The Anthropic mech interp is hiring! @ch402 is an absolutely fantastic mentor, manager and research lead and the team is up to great stuff, you should apply!
1
4
Adam Jermyn @AdamSJermyn
28 Jul 2023
About a year ago I switched from doing astronomy to do AI interpretability. To mark the occasion, I thought I'd write a thread about what that transition was like, and what I've carried through it.
1
10
(Aside: Some ridiculous fraction of Anthropic is former physicists. 3/7 of the founding team have PhDs in physics -- Jared was previously a professor at JHU! -- and the trend has continued, with physicists like Adam joining us.)

Jul 28, 2023 · 2:47 AM UTC

1
4
30
Trenton Bricken @TrentonBricken
8 Aug 2023
I’ve paused my PhD to join the @AnthropicAI mechanistic interpretability team full time! While I enjoyed grad school, it’s hard for me to imagine returning — working with the incredible team here on such consequential problems has been a dream. Consider joining us👇!
7