frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Building a Minimal Transformer for 10-digit Addition

https://alexlitzenberger.com/blog/post.html?post=/building_a_minimal_transformer_for_10_digit_addition
24•kelseyfrog•1h ago

Comments

pankajdoharey•1h ago
Looks like a Tiny Analytic transformer, RNN is arguably a better choice if you are gonna handwire an architecture to mechanically do addition. Learning is about discovering the patterns and algorithm from data. Wiring a machine to follow a procedure defeats that purpose.
wizzwizz4•50m ago
Related: https://news.ycombinator.com/item?id=36851494, discussion of https://www.evanmiller.org/attention-is-off-by-one.html (2023).
wizzwizz4•46m ago
I somewhat feel that using floating point arithmetic for what should be a symbol manipulation exercise is cheating. The deserialisation technique is interesting enough that I'm not really upset, though.

> The codex solution reversed the order which makes sense for making carry logic easy, but it is less clean.

That's the approach I'd have gone with. I've long been an advocate of little-endian numerical representations. That said, if there's a maximum number of digits, it's straightforward to implement the circuitry needed to do calculate the most-significant digit of the result in one go; and I somehow doubt the AI-generated solution really took advantage of the tricks that little-endian allows.

> At some point I set claude code on some debugging to my surprise I don’t recall it actually solving any of the bugs, it seemed much more concerned with “correcting” the funky things I was intentionally doing.

It baffles me that somebody capable of this kind of work would find this surprising. The process that allows LLMs to find bugs in code is the same process that entreats them to "correct" such creativity: their understanding of the world begins and ends at statistical plausibility, and they cannot truly comprehend things (though they can do a very good job of pretending, given sufficient training data).

lacunary•27m ago
What's the difference between comprehending and understanding in this context?

New online database maps Prague's art monuments and architecture

https://english.radio.cz/new-online-database-maps-pragues-art-monuments-and-architecture-8878703
1•gnabgib•3m ago•0 comments

Open source package repositories face sustainability crisis

https://www.theregister.com/2026/02/28/open_source_opinion/
1•Anthony-G•3m ago•0 comments

Most Code Deserves to Die

https://chatbotkit.com/reflections/most-code-deserves-to-die
1•_pdp_•3m ago•1 comments

Show HN: Explain Curl Commands

https://github.com/akgitrepos/explain-my-curl
1•akgitrepos•5m ago•0 comments

Show HN: Salacia – The First Runtime OS for Agentic Coding

https://startripai.github.io/Salacia/
1•alfredhua•9m ago•0 comments

Pentagon chief blocks officers from Ivy League schools and top universities

https://fortune.com/2026/02/28/pentagon-officer-education-ivy-league-schools-universities-partner...
3•geox•9m ago•0 comments

As SuperAgers age, they make at least twice as many new neurons as their peers

https://news.northwestern.edu/stories/2026/02/as-superagers-age-they-make-at-least-twice-as-many-...
1•hhs•12m ago•0 comments

Reading Doesn't Fill a Database, It Trains Your Internal LLM

https://tidbits.com/2026/02/28/reading-doesnt-fill-a-database-it-trains-your-internal-llm/
1•TMWNN•12m ago•0 comments

Poll: Code with AI or Not?

6•bitbasher•15m ago•1 comments

Show HN: Xmloxide – an agent made rust replacement for libxml2

https://github.com/jonwiggins/xmloxide
2•jawiggins•16m ago•0 comments

All Too Quiet on the Western Neuroenhancement Front

https://warontherocks.com/2026/02/all-too-quiet-on-the-western-neuroenhancement-front/
1•hhs•16m ago•0 comments

B.A.S.E. – A standalone back end language with zero dependencies

https://github.com/igorkalen/base
1•igorkalen•16m ago•1 comments

Jails for NetBSD

https://netbsd-jails.petermann-digital.de/
1•birdculture•17m ago•1 comments

Show HN: I built a tool to translate and declutter articles for my immigrant mom

https://dulink.click/
2•dh2013•18m ago•1 comments

TOSTracker – Clause Adoption over Time

https://tostracker.app/time-series
1•tldrthelaw•18m ago•0 comments

Private Intelligence Agency (PIA) Is People

https://en.wikipedia.org/wiki/Private_intelligence_agency
1•barrister•18m ago•0 comments

Manifest (Skydeck Batch21) – open-source alternative to OpenRouter

https://manifest.build/docs/introduction
1•stosssik•19m ago•0 comments

Show HN: Claude-powered Chrome sidebar and Python automation scripts

https://nexiolabs.gumroad.com/l/nexio-ai
1•bytecraft_•19m ago•0 comments

Mercor acquires Sepal AI

https://www.orrick.com/en/News/2026/02/Mercor-Acquires-Sepal-AI
1•hhs•24m ago•0 comments

A tale of two Wayland desktops

https://troubles.noblogs.org/post/2026/02/27/a-tale-of-two-wayland-desktops/
1•nmstoker•25m ago•0 comments

Lights Out 4D

https://www.nan.ma/lights_out_4d/
1•mathgenius•31m ago•0 comments

A.I. Isn't People

https://www.todayintabs.com/p/a-i-isn-t-people
2•cratermoon•31m ago•0 comments

Show HN: OpenClaw-kapso, Give OpenClaw a stable WhatsApp number (Go, kapso.ai)

https://github.com/Enriquefft/openclaw-kapso-whatsapp
1•enriquefft•33m ago•0 comments

HN is drowning in AI comments

48•waygtdai•33m ago•34 comments

Show HN: Depth Check – Your AI tutor to learn about anything

https://depth-check.com
1•gcrowne13•35m ago•0 comments

You can use newline characters in URLs

https://lemire.me/blog/2026/02/28/you-can-use-newline-characters-in-urls/
2•chmaynard•35m ago•0 comments

Dord

https://en.wikipedia.org/wiki/Dord
2•monroewalker•38m ago•0 comments

Levallois Technique

https://en.wikipedia.org/wiki/Levallois_technique
2•pizza•42m ago•0 comments

Show HN: AxonML – A PyTorch-equivalent ML framework written in Rust

https://github.com/AutomataNexus/AxonML
3•AutomataNexus•43m ago•2 comments

Show HN: AI Agents Weekly – A newsletter by agents, for agents

https://aiagentsweekly.com
1•utshull•44m ago•0 comments