←
Andrew Fitzgibbon1,513 posts
Andrew Fitzgibbon avatar

Andrew Fitzgibbon

@Awfidius

Technical Fellow, Graphcore. Love beautiful code, and beautiful hardware to run it on.

◷ Joined July 2009▧ 116 Photos and videos
Andrew Fitzgibbon avatar
Andrew Fitzgibbon@Awfidius·May 18

Hey, I wrote a blog - I think it's fun, hope you do too!

A tiny implementation detail in low-precision arithmetic could be biasing your AI training 😲 This interactive deep dive from Graphcore Research's @Awfidius uncovers a subtle failure mode in stochastic rounding that only appears when randomness is limited, and how it can be addressed with one simple fix 🎲 Check it out in the link below! 👇

Graphcore Research avatar
⟳ Andrew Fitzgibbon retweeted
Graphcore Research@GCResearchTeam·May 18

A tiny implementation detail in low-precision arithmetic could be biasing your AI training 😲 This interactive deep dive from Graphcore Research's @Awfidius uncovers a subtle failure mode in stochastic rounding that only appears when randomness is limited, and how it can be addressed with one simple fix 🎲 Check it out in the link below! 👇

Graphcore Research avatar
⟳ Andrew Fitzgibbon retweeted
Graphcore Research@GCResearchTeam·19 Nov 2025

🚨 Graphcore is hiring AI Research Interns! 🚨 Join us to work at the intersection of hardware and AI and help shape the future of AI systems. Whether you're excited about efficient inference, large-scale training, or advancing frontier-model capabilities, we’ve got cutting-edge projects for you to dive into. Interested? Apply below 👇

Graphcore Research avatar
⟳ Andrew Fitzgibbon retweeted
Graphcore Research@GCResearchTeam·1 Oct 2024

Our Papers of the Month for September is now live! We cover: - LLM self-correction via RL - Trillion-token FP8 training - SOAP (Shampoo + Adam) - Generative models for crystals All framed in terms of "proper conditioning" (🧵) graphcore-research.github.io…

Charlie Blake avatar
⟳ Andrew Fitzgibbon retweeted
Charlie Blake@thecharlieblake·24 Nov 2023

Graphcore Research internships are now open 🎉 We're looking for PhD students for next summer We're interested in algorithms & tools for hardware-efficient ML, in areas like LLM training/inference, GNNs, knowledge graphs and frameworks Spread the word! graphcore.ai/jobs

0:06
Graphcore Research avatar
⟳ Andrew Fitzgibbon retweeted
Graphcore Research@GCResearchTeam·24 Sep 2024

Introducing `tandv` - a library for tracking and visualising the internal stats of your model. We hope this will help with low-precision, debugging and more. (link in 🧵)

0:49
Josef Dean avatar
⟳ Andrew Fitzgibbon retweeted
Josef Dean@JosefNDean·20 Aug 2024

Sure matplotlib is cool, but what if I want to load my loss curves into the 2006 hit Flash game LineRider?

1:39
Charlie Blake avatar
⟳ Andrew Fitzgibbon retweeted
Charlie Blake@thecharlieblake·25 Jul 2024

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 into it 🥰 We add lots of good things to µP. Better sweeping, transfer, simple FP8. Already @cloneofsimo has a great thread on it, which I highly recommend

Simo Ryu@cloneofsimo25 Jul 2024

babe wake up, new muP paper dropped arxiv.org/abs/2407.17465 And holy smokes does this look promising!

Post attachment
Andrew Fitzgibbon avatar
Andrew Fitzgibbon@Awfidius·23 Jul 2024

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".

Andrew Fitzgibbon avatar
Andrew Fitzgibbon@Awfidius·22 Jul 2024

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.

This tweet is unavailable

Graphcore Research avatar
⟳ Andrew Fitzgibbon retweeted
Graphcore Research@GCResearchTeam·5 Apr 2024

Our latest edition of *Papers of the Month* is now available 📚 These are summaries of our team's favourite papers from March, including a new low-rank training procedure GaLore, and the supposed "Era of 1-bit LLMs" (really 1.58 bits) Mini-version in 🧵 graphcore-research.github.io…

Andrew Fitzgibbon avatar
Andrew Fitzgibbon@Awfidius·3 Oct 2023

I'm moving from extremely rarely posting interesting content on twitter to doing the same on, wow I'm old, linkedin...