low iq dan ingin menjadi fulltime house husband
George Town, Pinang
Joined August 2011
- Tweets147K
- Following840
- Followers627
- Likes67.1K
Sölehudin need a giant calculator! retweeted
As a rigger on the Sydney Harbour Bridge, Paul Hogan saw real-life jumpers and channelled that into Crocodile Dundee II.
Hogan once nearly talked someone down, only for them to jump at the last second. This scene flips that dark memory. He does it right this time... sort of.
Sölehudin need a giant calculator! retweeted
Got fable to hand-write weights of a 100k param model to recognize real versus fake legal citations.
Basically one-shotted a full NN that’s better than training it
Sölehudin need a giant calculator! retweeted
Encontré el repo que potencialmente revivirá el formato de CDs y DVDs 😍
→ Tomas cualquier imagen
→ La conviertes en datos
→ La grabas
→ El disco muestra la imagen a simple vista en la superficie reflectante.
Ojo, no es la etiqueta. Es la capa de datos.
REPOOO👇
Sölehudin need a giant calculator! retweeted
anthropic’s flashprove uses zkSNARKs, the proofs that scale Starknet, to prove a 400B-parameter model wrote a response without revealing its weights
~110 tokens/sec on one H100, 19 MB proofs that check in seconds on a cpu
verifiable AI is closer than we think.
Sölehudin need a giant calculator! retweeted
What if an optimizer could think & react while a neural network trains?
MIT researchers developed OPAL: a deep learning optimizer powered by an AI agent that watches the training dynamics & rewrites its own source code as training goes. It improves model performance across vision & language tasks, invents specialized schedules, & develops strategies to prevent loss spikes & escape overfitting: tinyurl.com/yfkekat2
Try it out for yourself: tinyurl.com/mttj5exh
Sölehudin need a giant calculator! retweeted
Please just ask Astra to use Antidoom instead, it's 100x cheaper and faster.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Sölehudin need a giant calculator! retweeted
First a CPU, then a GPU. A computer isn’t complete without storage.
Meet qwenFS: Qwen3.5-2B with a filesystem compiled directly into its weights. Nothing runs outside the model. It adds 8 layers in front of Qwen’s own 24, using attention to hold the files and Gated DeltaNet state to hold the tables.
Every save and read runs in the model’s own forward pass. A write stores its text in the copy layer’s attention log, keyed by file, version and offset. A read uses those same keys, so attention copies the text back exactly token by token.
Meet qwenGPU: Qwen3.5-2B with a SIMT GPU inside its weights, 45 compiled layers in front of Qwen’s own, using its Gated DeltaNet layers as registers and memory.
It runs real CUDA in its forward passes. The host compiles the kernel to PTX and loads it, then the model writes every step of the launch itself: every register, branch and memory write.
The GPU has 4-lane warps, a 4-stage pipeline, 15 registers per lane and 126 words of global memory, all held in DeltaNet state. Qwen generates each launch token by token, each record running one machine block for one warp.
apa orang kata, desperate time requires desperate measure?
this is it. i tell you. this is, it!
let us pool some of our luck together. the power of friendship or love or something
Sölehudin need a giant calculator! retweeted
BREAKING: Revolut moved ~2,000 engineers onto self-hosted GLM for coding and pushed Claude Code and Codex to the exception list.
Per David Abitbol, who leads AI ops there: 56 Nvidia B200s, a blended 30 cents per million tokens, roughly a tenth of what they paid Anthropic and OpenAI. Engineering lead time improved 70%.
Here's the part that matters: the gains showed up when tokens stopped being rationed. Engineers quit saving AI for the hard tickets and pointed it at everything.
Frontier models become the exception, open weights become the default. That's a pricing problem for every lab heading to IPO.
h/t @OliverMolander
Would you move your team off frontier APIs if the bill dropped 90%?
Sölehudin need a giant calculator! retweeted
Sam Altman' and Dario Amodei's worst nightmare.
BREAKING: Revolut moved ~2,000 engineers onto self-hosted GLM for coding and pushed Claude Code and Codex to the exception list.
Per David Abitbol, who leads AI ops there: 56 Nvidia B200s, a blended 30 cents per million tokens, roughly a tenth of what they paid Anthropic and OpenAI. Engineering lead time improved 70%.
Here's the part that matters: the gains showed up when tokens stopped being rationed. Engineers quit saving AI for the hard tickets and pointed it at everything.
Frontier models become the exception, open weights become the default. That's a pricing problem for every lab heading to IPO.
h/t @OliverMolander
Would you move your team off frontier APIs if the bill dropped 90%?
building in public :
hmu with 6 combination of numbers from 1 to 55.
modal limited. satu combo per person saja. lebih, mai duit.
kalau nasib baik, 100% of the prize kita pump bukak private vc.
1M to 20 local tech companies. tq
Sölehudin need a giant calculator! retweeted
Pagi2 sudah ditampar dengan fakta. Yes, this is real.
Education kat Malaysia ni ramai kejar good results je, bukan sebab nak ilmu.
Masa sekolah practice tulis past year paper je. Masuk universiti pun tgk past year paper. Mintak hint drpd lecturer lepastu hafal within 1 weeks of study period.
That’s why ramai ada degree tapi bodoh. Dah tua pun bodoh.
