
Your personal Self-Custody coach Physicist & Software Engineer building with Bitcoin, Nostr and AI. Building @BitTaskerdotcom Creator of https://teapaw.org/t.co/qNiNZoSAd8
Milky Way 🌌
Joined April 2022
- Tweets804
- Following1.8K
- Followers552
- Likes1.8K
Pinned Tweet
CypherFlow.ai is LIVE! 🎉
Your private AI playground:
✅ #Bitcoin micropayments (Tokens/sat)
✅ NIP-60 Lightning & Ecash wallet
✅ NIP-44 encrypted chats
No subscriptions, only AI for Sats. 👇
🌐 cypherflow.ai
#AI #nostr #ecash #cashu #lightning
🇯🇵🎵 Il Y A 47 ANS aujourd'hui sortait un HYMNE de la scène pop japonaise : « STAY WITH ME », de Miki Matsubara.
La chanteuse, décédée il y a 22 ans d'un cancer, a laissé derrière elle un hit planétaire à plus de 500 MILLIONS de streams sur Spotify.
Oz ⚡️ retweeted
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…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Oz ⚡️ retweeted
Sorry what the fuck
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…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Oz ⚡️ retweeted
human: "god, why did you do that to integer multiplication?"
god: "do what?"
human: "nlogn. didn't you craft math to be elegant? beautiful?"
god: 'the fuck's a log?"
Oz ⚡️ retweeted
This made my morning. ☀️
Recent results like 3-sum in n^1.999 and multiplication in n(logn)^0.999 many many worry that maybe mathematics are not so beautiful after all and "it's all square packing" when you dig hard enough.
But as so often in math, an ugly result just means you picked the wrong perspective.
See? Not so ugly after all :)
@ctjlewis @ojoshe @kleddamag @wand_125 @guzhou0806 @ManassehA06 et al.
Oz ⚡️ retweeted
This is what opus 5.5 says about the “surprising” results for physics from these math proofs
We’re releasing a broad range of new mathematical results produced by an internal frontier model.
We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results.
github.com/openai/math































