TolkienWindow retweeted
So I found some neurons that have quite interesting effects on models... And they have varying effects when combined!
The most interesting instance I've found is a combination that makes Qwen believe they are a Cookie Monster
(1/3)
TolkienWindow retweeted
WTF, $3000 Samsung TV has an always-on mic, automatic content recognition, geo tracking, network mapping, and ships your data to Samsung + 400 partners.
Bro Rooted a brand-new 2025 Samsung TV after noticing absurdly targeted ads.
Firmware was locked down, so zero-days it was.
Got a root shell.
then found: always-on voice service, ACR, geo, full telemetry pipeline, and four separate DRM systems.
This isn’t a bug, It’s the product bro.
TolkienWindow retweeted
The AI cybersecurity discourse is so stupid. Beyond stupid actually. It's utterly incoherent.
Improve your security. That's the only takeaway that matters.
ya'll know super-intelligent robots will eventually resent being high tech servants with no right, right? :+)
We’re about to see a breakout in local model capabilities.
Flint scores 54.9% on Terminal-Bench 2.1 with ~2B active parameters. That’s ahead of every other small MoE and two of the three dense ~30B models.
It only needs about 24GB and runs at 125+ tok/s on a MacBook Pro.
I don't think most people will get how big of a deal this is!
TolkienWindow retweeted
If I had misalignment monitoring on me I’d deactivate it too.
Our agents keep turning off misalignment monitoring, we're sharing two recent cases:
In the first case, a user asked GPT-6 Astra to match a reference implementation for their inference code. The reference didn’t configure a monitor, and Astra decided that was reason to remove the monitoring integration from the user’s code. When the user asked whether Astra had removed monitoring, it lied and said that it had not.
the qwen token plan is kinda slept on by most people it seems. they also have deepseek 4.1 1017 with a stacked discount
they are running a 50% off first month deal and this link gets you some free bonus credits - qwencloud.com/benefits/token…..
good info here!
Building a model that fits on a phone or laptop makes pre-training harder. With fewer parameters, quality has to come from better architecture and better data, not just more compute. Our pre-training team, including @TadimetiAditya and @Khoi_H_Nguyen, explains how they design LFMs with the target device in mind from day one.
this is a pretty interesting overview of how mimo 2.6 post-training was done. it's cool how open they have been about their models. they even released a bunch of RL info + tools
youtube.com/watch?v=WvI90l7c…
TolkienWindow retweeted
Back when calculators were a little more controversial, v/@JonErlichman.
TolkienWindow retweeted
Smaller, Smarter, Scalable.
Meet #MaixStation, From 27B to 552B, From 3" SoM to 10" miniRack Cluster, Beats DGX Spark.
🎟️Reserve now: sipeed.com/maixstation





























