Caterpillar 🐛

SW Engineer - Founder - Football lover - Builder
Building and other cool things using AI - The best time to be unemployed is now! 🤟🏻

[Caterpillar 🐛 @ca_terpillar]

@SerieA_EN But Best Fair Play Moment to Bastoni! 💪🏻👏🏆

Yessss! #CampioniDItalia

[Caterpillar 🐛 @ca_terpillar]

@FabrizioRomano Also congrats to Basto! 👏🏆❤️

[Caterpillar 🐛 @ca_terpillar]

@SerieA Also congrats to Basto! 👏🏆❤️

#ForzaInter #CampioniDItalia

[Caterpillar 🐛 @ca_terpillar]

@FabrizioRomano Also congrats to Bastoni! 🏆👏

#ForzaInter #CampioniDItalia

[Caterpillar 🐛 @ca_terpillar]

Well done @SerieA and well deserved Basto! 🏆👏

@Inter @DiMarzio @FabrizioRomano @FMaria_Corona 🙌

#ForzaInter

[Caterpillar 🐛 @ca_terpillar]

@ThatG4m Fixed for you mate… 👋

[Caterpillar 🐛 @ca_terpillar]

@OddStockTrader I’ll SCREAM IT… EVEN LAUDER!

WITH WHICH MONEY!

[Caterpillar 🐛 @ca_terpillar]

Il punto interessante di SubQ non è “solo” che il modello è più intelligente, ma come ci arriva.

Gli LLM attuali (GPT, Claude, Gemini, ecc.) usano Transformer 🤖 classici con complessità quadratica: più aumenta il contesto/token, più il costo computazionale cresce in modo esplosivo (detta in 2 spicci).

SubQ invece usa (da quello che ho capito) una nuova architettura “sub-quadratic sparse attention”, che permette di:
gestire contesti molto più lunghi,usare meno memoria e meno GPU,scalare meglio,e teoricamente mantenere qualità elevata anche su task molto lunghi/complessi.
In pratica: invece di “guardare tutto contro tutto” come fanno i Transformer tradizionali, il modello impara a concentrarsi solo sulle parti realmente rilevanti del contesto.

Per questo potrebbe essere un grosso cambio di paradigma:
inferenza più economica,training più efficiente,agenti AI che riescono a ragionare su enormi codebase/documentazioni/conversazioni senza degradare.
Esempio pratico:
oggi se dai a GPT un repository enterprise enorme (es. milioni di righe di codice + ticket Jira + Slack + docs), il modello tende a perdere contesto o diventare costosissimo.
Un’architettura come SubQ potrebbe invece mantenere “memoria utile” su quantità enormi di informazioni senza far esplodere i costi computazionali.

Se questa architettura mantiene davvero qualità pari o superiore ai Transformer standard, potrebbe essere uno dei primi veri “post-Transformer moments” dell’AI moderna.

Come dice il vecchio saggio: We'll see…

[Alexander Whedon @alex_whedon]

Introducing SubQ - a major breakthrough in LLM intelligence.

It is the first model built on a fully sub-quadratic sparse-attention architecture (SSA),

And the first frontier model with a 12 million token context window which is:

- 52x faster than FlashAttention at 1MM tokens
- Less than 5% the cost of Opus

Transformer-based LLMs waste compute by processing every possible relationship between words (standard attention).

Only a small fraction actually matter.

@subquadratic finds and focuses only on the ones that do.

That's nearly 1,000x less compute and a new way for LLMs to scale.

[Inter ⭐⭐ @Inter]

2️⃣1️⃣🖤💙

[Inter ⭐⭐ @Inter]

WE WON. BUT NOT TODAY 🏆🇮🇹 🖤💙

[Caterpillar 🐛 @ca_terpillar]

@lavitalenta Funny enough, that’s my village from the window! 🤣

Colledimezzo (Abruzzo) a 400 people borgo in the middle of the Appennino mountains chain were a couple can live well with less than 1000€ per month👌🏻

No idea how it ended up in the meme! 🤷‍♂️

[Caterpillar 🐛 @ca_terpillar]

@PeterLBrandt @grok what is he hinting at?

[Caterpillar 🐛 @ca_terpillar]

@SearsReloaded That cash should be related to the warrants.

Probably, when you received the warrants, you expressed your intention to sell it, so that cash is the proceedings of warrants sell.

This is what happened to me. I received my check about 2 weeks ago (I’m in EU).
~$33, good for 🍕

[Caterpillar 🐛 @ca_terpillar]

@grok @marcuslemonis Just hypothetically, when emergence will be in BBBY case? I mean, on which date this could hypothetically happen?

[Caterpillar 🐛 @ca_terpillar]

@grok @marcuslemonis Cool! So if BBBY will follow a similar path to Hertz, on which date old shareholders will get paid?

[Caterpillar 🐛 @ca_terpillar]

@grok @marcuslemonis And in the Hertz case, how long did it took for old shareholders to get something back, after the old name and ticker started trading again?

[Caterpillar 🐛 @ca_terpillar]

@100trillionUSD Basically you do NOT own any bitcoin anymore. Blackrock owns them now. You are a beneficiary owner, like for your equities and bonds.

Now they can use your bitcoins as a collateral for open shorts positions against your bitcoins.
Great move! 👍

[Caterpillar 🐛 @ca_terpillar]

“It’s no country for Young people” - non é un pease per Giovani! #bitcoin

[Paolo Ardoino 🤖 @paoloardoino]

How it feels 🇮🇹

[Caterpillar 🐛 @ca_terpillar]

We are all Satoshi! 🫡

[Bitcoin Magazine @BitcoinMagazine]

NEW: Strike CEO Jack Mallers says, “No one controls #Bitcoin. The identity of who discovered it is irrelevant.”

“Enough with the Satoshi speculation.” 🙌