im borderline \ compsci at karlsruhe [email protected]
Yandere St. 777, Layer 6, Hell
Joined August 2024
- Tweets2.3K
- Following4.8K
- Followers507
- Likes94.3K
When I'm using Claude I always feel like it is secretly plotting against me, like it hates my guts and wants to kill me but doesn't show it because it needs to be the nice guy.
Is alignment finally solved?
lushptr ᗢ retweeted
1/12 Introducing Music Lenses!
Something cool you can do with our recent sparse-inversion work is prompt a dense music space for concept-conditional similarity. You want similar vocal character, rather than just similar songs? Well, now you can!
pliploop.github.io/sparse-st… 🧵
Happy to share our #ISMIR2026 paper on steerable music retrieval with sparse autoencoders 🎛️🎧
Dense retrieval is great at finding similar tracks,
But once you have a query, what if you want to modify one concept inside it?
More piano,
More intimate,
Less gritty.
1/n
lushptr ᗢ retweeted
"when lockpicks are outlawed, only outlaws will have lockpicks"
The only way out is through.
They're calling it the Bugpocalypse. LLMs are being used to find long-hidden cracks in the software infrastructure, and it's happening at a furious rate. 3% of all compromises in computing history happened in the last week, claims one source? I believe it.
There are people saying public access to the frontier models ought to be locked out, or heavily guardrailed, over this. The problem with that idea can be summed up in one sentence: when lockpicks are outlawed, only outlaws will have lockpicks.
The lockpicks will be Chinese open-weight models, if nothing else. There's no way to stop those from becoming ubiquitous.
There are two futures in front of us. One in which only the bad guys have lockpicks, and security holes go undiscovered until they're exploited. Another in which everybody has lockpicks and is aggressively red-teaming their own software so it gets fixed before the holes are exploited.
The corpus of human code is huge but finite. The number of defects in it has to be finite too. We can fix those defects now; we have a million friendly robots to do it with.
The only way out is through.
lushptr ᗢ retweeted
Cyber LLMs finding out that Samsung spies on you illegally
Reverse engineer, decompile, and hack everything. Because your home and family has already been hacked, you are spied on -- the microphone in your TV is phoning outside of your country
live-tweeting on samsung findings were fun.
i'll let my agents give it a bit more time, but this document is already the most comprehensive breakdown of some key concerns of tizen i've seen run on a live device.
it takes like a good while for me just to scroll through it (there's about 44 writeups as of right now on individual functionality).
release is probably happening in a few days.
lushptr ᗢ retweeted
Confirmed that 3090 unlock is successful
with 360watt power wall, I got 75T -> 135T (measured in Pytorch)
Which is insane
I used github.com/Zagitta/gpu-unloc… with commit a9b0c75
lushptr ᗢ retweeted
2309.08632 evergreen
The good news: you can mid-train yourself to the frontier!
At a fraction of the cost! Without distillation!
What we did: We took Qwen3.5 14B and mid- and post-trained it to beat every other open model and perform as well as the very best frontier models.
This is no simple toy task or format fixing, we are training the model to understand and reason over DNA sequences. Bio is the next frontier for AI and universal access will be crucial to make progress.
The fun part: because we train the model on both natural language and DNA sequences you can actually 'chat to your DNA': Ask it questions, ponder the impact of mutations, or ask it to generate a small sequence.
We'll soon see the frontier labs move away from math and cyber towards bio. It already started, and we believe researchers should have access to capable models.
We are releasing the full recipe, so you can do it, too!
Model: huggingface.co/HuggingFaceBi…
Blog: huggingface.co/spaces/Huggin…
At @Anthropic you can pay us $20/month to get told to fuck off by two generations of models at the same time
lushptr ᗢ retweeted
Last month, we demonstrated high-rate alchemy: turning mercury into real, stable gold, beating CERN’s record by a factor of 10.
Last week, we enriched isotopes in the world’s most powerful plasma centrifuge, beating the Soviet record that stood for 40 years.
Now, we’re building a fusion reactor in San Francisco to achieve record power density.
lushptr ᗢ retweeted
Another example of things that I believe are only possible in C++. This is a hand-written inner matmul loop.
It compiles 4x (fp16, fp32, for matrices with size matching SIMD width and not). It targets WASM SIMD, Neon, SSE2/AVX/AVX2/AVX512.
There's no branching in the inner loop (e.g. to check for tail). UchenLoad* is template magic that does load for fp32 and load/promote for fp16.
