It would. But it’s a good option when you have computationally heavy tasks and communication is relatively light.

Once configured, Tor Hidden Services also just work (you may need to use some fresh bridges in certain countries if ISPs block Tor there though). You don’t have to trust any specific third party in this case.

If config prompt = system prompt, its hijacking works more often than not. The creators of a prompt injection game (https://tensortrust.ai/) have discovered that system/user roles don’t matter too much in determining the final behaviour: see appendix H in https://arxiv.org/abs/2311.01011.

Like Firefox ScreenshotGo? (I think it only supports English though)

Don’t know much of the stochastic parrot debate. Is my position a common one?

In my understanding, current language models don’t have any understanding or reflection, but the probabilistic distributions of the languages that they learn do - at least to some extent. In this sense, there’s some intelligence inherently associated with language itself, and language models are just tools that help us see more aspects of nature than we could earlier, like X-rays or a sonar, except that this part of nature is a bit closer to the world of ideas.

Huh, it’s actually a thing.

https://arxiv.org/abs/2305.17743

The temperature here was very interesting for a second or two until I remembered some people use °F.

xkcd.com is best viewed with Netscape Navigator 4.0 or below on a Pentium 3±1 emulated in Javascript on an Apple IIGS at a screen resolution of 1024x1. Please enable your ad blockers, disable high-heat drying, and remove your device from Airplane Mode and set it to Boat Mode. For security reasons, please leave caps lock on while browsing.

• https://xkcd.com

CVEs are constantly found in complex software, that’s why security updates are important. If not these, it’d have been other ones a couple of weeks or months later. And government users can’t exactly opt out of security updates, even if they come with feature regressions.

You also shouldn’t keep using software with known vulnerabilities. You can find a maintained fork of Chromium with continued Manifest V2 support or choose another browser like Firefox.

You can get your hands on books3 or any other dataset that was exposed to the public at some point, but large companies have private human-filtered high-quality datasets that perform better. You’re unlikely to have the resources to do the same.

Very cool and impressive, but I’d rather be able to share arbitrary files.

And looks like you can only send images in DMs, but not in groups/forums.

Already a thing? E.g. Bloomsday

If your CPU isn’t ancient, it’s mostly about memory speed. VRAM is very fast, DDR5 RAM is reasonably fast, swap is slow even on a modern SSD.

8x7B is mixtral, yeah.

Mostly via terminal, yeah. It’s convenient when you’re used to it - I am.

Let’s see, my inference speed now is:

• ~60-65 tok/s for a 8B model in Q_5_K/Q6_K (entirely in VRAM);
• ~36 tok/s for a 14B model in Q6_K (entirely in VRAM);
• ~4.5 tok/s for a 35B model in Q5_K_M (16/41 layers in VRAM);
• ~12.5 tok/s for a 8x7B model in Q4_K_M (18/33 layers in VRAM);
• ~4.5 tok/s for a 70B model in Q2_K (44/81 layers in VRAM);
• ~2.5 tok/s for a 70B model in Q3_K_L (28/81 layers in VRAM).

As of quality, I try to avoid quantisation below Q5 or at least Q4. I also don’t see any point in using Q8/f16/f32 - the difference with Q6 is minimal. Other than that, it really depends on the model - for instance, llama-3 8B is smarter than many older 30B+ models.

Have been using llama.cpp, whisper.cpp, Stable Diffusion for a long while (most often the first one). My “hub” is a collection of bash scripts and a ssh server running.

I typically use LLMs for translation, interactive technical troubleshooting, advice on obscure topics, sometimes coding, sometimes mathematics (though local models are mostly terrible for this), sometimes just talking. Also music generation with ChatMusician.

I use the hardware I already have - a 16GB AMD card (using ROCm) and some DDR5 RAM. ROCm might be tricky to set up for various libraries and inference engines, but then it just works. I don’t rent hardware - don’t want any data to leave my machine.

My use isn’t intensive enough to warrant measuring energy costs.

I see!

And it was a stable OS version, not a beta or something? That’s the worst kind of bugs. Hopefully manufacturers start formally verifying hardware and firmware as a standard practice in the future.

Other than what I said in the other reply:

I live in the USA so getting one would be problematic but I hear perhaps not entirely impossible for me.

Looks like it has a US release? If you’re unsure or getting a European version, double-check it’s compatible with American wireless network frequencies &c. Specific operators might also have their own shenanigans.

Do you know how it compares to e.g. Fairphone?

Nope, never tried Fairphone.

Very solid, I think (except water protection, but my previous OnePlus also didn’t have good water protection anyway; and I’m careful enough).

I don’t tend to use glyphs or the default launcher (and therefore its special widgets that only work there; but the ability to have apps in folders on my main screen while being hidden from the app menu is more important for me than a handful of widgets, so Neo Launcher it is).

A recent OS update added configurable swap (up to 8GB), calling it “RAM booster”. I don’t use it, but if you want to run a local LLM (or rather a SLM), you could try making use of it? As long as you figure out how to make the model use main RAM and not the swap.

I like the battery life (or maybe it’s just because it’s the first phone where I started charging at 20% and stopping at 80% semi-consistently).

Termux still works despite the new Android versions becoming more hostile to apps executing binaries they didn’t have included already.

One thing I miss from OnePlus is the ability to deny some apps network access entirely. (I think it was removed in later versions of Oxygen OS?)