Andrew Fitzgibbon
@Awfidius
Followers
4K
Following
2K
Media
116
Statuses
2K
Technical Fellow, Graphcore. Love beautiful code, and beautiful hardware to run it on.
Cambridge
Joined July 2009
I'm moving from extremely rarely posting interesting content on twitter to doing the same on, wow I'm old, linkedin. .
1
0
25
RT @GCResearchTeam: Our Papers of the Month for September is now live! We cover:.- LLM self-correction via RL.- Trillion-token FP8 training….
graphcore-research.github.io
We’re pleased to share four papers from different domains: LLM self-correction, FP8 training, generative crystals and optimisation. They are united, somewhat tenuously, by the importance of proper...
0
13
0
RT @thecharlieblake: Graphcore Research internships are now open 🎉 . We're looking for PhD students for next summer. We're interested in al….
0
3
0
RT @GCResearchTeam: Introducing `tandv` - a library for tracking and visualising the internal stats of your model. We hope this will help w….
0
3
0
RT @JosefDeanGC: Sure matplotlib is cool, but what if I want to load my loss curves into the 2006 hit Flash game LineRider? .
0
830
0
Update: super helpful customer service got me the brass fixing kit, tap no longer wobbling.
0
0
9
RT @thecharlieblake: Our u-µP paper hit arXiv this morning! I'm so proud of this one — and grateful for a wonderful team who put so much in….
0
9
0
RT @PaulBalanca: Excited to present our work at @icmlconf WANT workshop: ⚖️ Scalify: scale propagation for efficient low-precision LLM trai….
github.com
JAX Scalify: end-to-end scaled arithmetics. Contribute to graphcore-research/jax-scalify development by creating an account on GitHub.
0
1
0
RT @cloneofsimo: babe wake up, new muP paper dropped .And holy smokes does this look promising! .
0
52
0
As further feedback to the team that thought this would be a great innovation that would attract customers: I bought this tap because of Grohe's reputation for quality. The term "quickfix" had zero impact on my choice. However, that term now means "low quality gimmick".
1
0
2
Same problem, and based on online searches, many others have it too. Terrible wasteful design, saves 10 seconds when it works, costs hours when it doesn't, and a load of extra useless plastic is left in place having saved those 10 seconds.
1
0
3
RT @GCResearchTeam: Our team's summaries & analysis of our favourite papers from the last month. We give our take on: Mamba-2, sparse-µP,….
graphcore-research.github.io
Improving transformers is now not “just one area” of machine learning research. This is illustrated by the breadth of papers we got excited about this month, all of which claim to improve upon some...
0
8
0
RT @GCResearchTeam: Our latest edition of *Papers of the Month* is now available 📚. These are summaries of our team's favourite papers from….
graphcore-research.github.io
March was a fruitful month for AI research, with plenty of papers for us to choose from. A trend in the work we’ve selected is the pushing of previously published methods to their limits, in new...
0
9
0
RT @dominic_masters: So pleased to share that our OGB-LSC winning model GPS++ has just been published in TMLR.
0
4
0
I love that I'm being offered these - I guess the ad targeting knows I'm going to have a sudden urge to take up jewellery making long before I do.
0
0
1
I’m biased obvs, but having seen a preview of this talk, I can recommend it highly!.
I'll start with "Julia meets the Intelligence Processing Unit" (, about running #JuliaLang on the @graphcoreai IPU, spectacular example of how @llvmorg enables adopting new cool hardware in Julia, and how Julia opens up new doors in scientific computing.
0
2
11
RT @MoseGiordano: I'll start with "Julia meets the Intelligence Processing Unit" (, about running #JuliaLang on the….
pretalx.com
In this talk I will detail about the effort to run Julia code on the _Intelligence Processing Unit_ (IPU), a massively parallel accelerator developed by Graphcore and powered by 1472 cores.
0
6
0
Thank you! Really great to be here, and great conversations throughout the event!.
amazing invited talk by @Awfidius from @graphcoreai on . real-world hardware for #AI but also real-world #compilers and #autodiff. at the #EDINBURGH first Workshop on Affordable ML
1
0
11
“I’m not going to get drawn into commenting specifically on that” said the PM’s spokesperson. Referent of 'that', immediately preceding: "the state of the country".
3
0
5