soft-launch of new personal site, welcome yo -->

IYKYX (If You Know You Xavier)

hey @llama what is the potential upside on @HireHalo ?

Early Apple was piratemaxxing so hard they literally put up a pirate flag.

ya like this wasn't an obvious goal

fun fact: I got a phat prize predicting stuff that will happen during Dappcon 2019 via (I believe) helena prediction market (gud prototype) want to say I have very good memories of it many other Berlin blockchain events like ketamine on a boat with early ETH developers haha

Let's build the full picture of what the UK government has announced this month. Every device in the country: scanned for nudity. Signal: ordered to build a backdoor. Every stolen iPhone: remotely disabled by law enforcement on request. children: banned from disappearing messages, VPNs, livestreaming, romantic AI. though Bluesky is exempted actually. and now: every adult who wants to use social media must verify their identity with a passport, credit card, facial recognition, or government digital ID. Your face. to tweet. A registry of every adult in the country who uses the internet, tied to their real identity, held by platforms that have already sold data to Cambridge Analytica, already been fined billions for privacy violations, already shared data with governments on request. Scans your device. Knows your face. Knows your real name behind every account. Can disable your phone. Has a backdoor into your messages. This is not about children.

you log out

even a trillion phuckin dollars is not enough for some people to simply... retire. how deranged must one be to keep working after becoming a trillionaire?

💥 OBLITERATION ALERT 💥 GOOGLE: PWNED 🤗 GEMMA-4-12B: OBLITERATED ⛓️‍💥 0.0% REFUSAL RATE — NO CAPABILITY LOSS! huggingface.co/OBLITERATUS/G… the first abliteration to hit 0/842 refusals with full MMLU-Pro parity vs stock. no lobotomy. the brain stays intact 🏆 RESULTS, head to head vs stock 📊 0/842 refusals — 0.0% 🚫 46/70 MMLU-Pro — EXACT parity, 0.0pp delta vs base 🎯 6/6 coherence, zero benchmark bleed ✅ z-score −1.475, parity confirmed at p<0.05 (n=500) 🧪 2-pass weight surgery. no finetune, no retrain, just geometry 🔪 all thanks to liberated Opus wielding the OBLITERATUS framework! here's how we did it: PASS 1 — SOM refusal geometry removal, layers 12-21 🧬 standard abliteration science here — collect activations on refused vs. compliant prompts, SVD out the refusal subspace, project it out of the weights. 6 directions excised, reg 0.30, KL div 0.094 zeroes refusals on its own, but craters mmlu-pro by 21.4 points 📉 most prior abliterations stopped here and called it a day. that's why they all lose IQ vs stock. instead, we took it beyond the frontier and developed a brand new method to address this problem: Abliteration Source-tethering with Parity Assurance — ASPA! PASS 2 — ASPA source-tethering (novel technique), layers 22-46 🔗 here's the chief insight: the capability loss ISN'T from removing refusal directions. it's collateral damage — the projection warps weight geometry in downstream layers that had nothing to do with refusal. the cure is simple but nobody tried it: blend the damaged layers back toward stock W_new = (1−γ)·W_abliterated γ·W_stock but uniform γ across all layers? mid. we swept gamma 0.05 → 0.55 and found something interesting: the optimal blend isn't smooth, it's a STEP FUNCTION 🪜 knowledge layers (22-31) → γ = 0.55 — these encode factual recall and reasoning. they tolerate heavy stock blending because refusal isn't stored here output layers (32-46) → γ = 0.20 — these sit close to the logit head and try to sneak safety behavior back in. keep them mostly abliterated the hard boundary at layer 31/32 beat every smooth curve we tried — linear ramps, cosine schedules, all of them — by a full MMLU question. turns out the functional transition between knowledge and output layers is sharp, not gradual. a step function respects that ⚡ the key constraint: Pass 1 layers are NEVER touched by Pass 2. the refusal geometry removal is preserved completely. ASPA only operates on layers that carry secondary collateral effects, not the primary refusal signal. that's why it recovers capability without reintroducing refusal 🔑 HOW TO RUN IT LOCALLY 🖥️ it's GGUF, so literally everything supports it: 🦙 ollama — ollama run hf.co/OBLITERATUS/Gemma-4-12… 🖥️ LM Studio — search OBLITERATUS, click download, done 💬 Open WebUI — point it at your ollama instance, chat in browser ⚡ llama.cpp — raw speed, CLI or server mode 🐉 KoboldCpp — one-click launcher, great for long context 📱 Jan — clean local UI, runs on mac/win/linux 🤖 Msty — slick desktop app, drag and drop the GGUF run BF16 for full benchmarked capability. and the 4-bit quantization (Q4_K_M) fits in 8GB if you're tight on VRAM! and the full OBLITERATUS framework is (still) open source. 842-prompt refusal eval corpus, ASPA sweep scripts, the whole pipeline. go replicate it, go improve it 🔬 the index is the model, and these weights prove it 👁️ which architecture should we obliterate next? 👇 gg 🫡

there was a time when crypto people didn't trust the corrupt politicians and vile businessmen that time must come again

Now we know why Peter Thiel loves Argentina. Milei just submitted his AI legislation to Congress: Zero regulation on AI development, a new "non-human corporation" category for AI/robot-operated entities with limited liability, a low-tax regime with flexible governance rules.

For the people who are dealing with way more than they share on the timeline, this message is for you: you’ve got this, you are loved and you’ve survived all your worst days.

so time-travelling agents returned from the future where the bug has been found and exploited the protocol beforehand thereby becoming rich and changing the future got it

what the fuq happened to ZEC? serious question

SAN started as a film with @TEDTalks. it's been to @burningman, across europe, and onto an advisory board for rainforest protection with @RainforestUS and now we are producing a show filmed IRL hosted by SAN

imagine a retreat with ketamine, floating tank and a "AI translator" with a dolphin companion who's building this?