back
user profile

walrus01

36,604karma·8,692submissions·February 19, 2016
about
all your base are belong to us
recent activity (8,692 total)
comment
well, yeah, it's based on a 2+ year old tiny model. It's very much an alpha proof of concept that they can perma-bake an LLM into silicon. https://huggingface.co/meta-llama&#…
2d ago·view thread
comment
The news article is really not clear about whether this was, relative to the company "OSS": a) OSS is a colocation customer with its own hardware colocating it inside an Iron Mountain datace…
2d ago·view thread
comment
1.7T
3d ago·view thread
comment
I am fairly sure the blog post (and everything else on that site) was written by an LLM and the larger ones with deep training data sets do know what a lombard loan is, but it was instructed to write …
3d ago·view thread
comment
I don't mean anything I wouldn't expect to see, just that it's a very well known thing, and also has had some recent challenges (news articles about extreme climate, etc). And similar s…
3d ago·view thread
comment
Googling "agricultural products as collateral cheese" turns up some very interesting results. https://www.google.com/search?q=agricultural+products+as+col... …
3d ago·view thread
comment
Erik Prince and similar have been trying to get that going again (for actual commercial cargo shipping related activities) for some time now.
3d ago·view thread
comment
I wonder if instead of blocking the facebook ads, in a continual game of whack-a-mole with facebook changing things, there would be a way to accept the ads and then generate a white opaque box overl…
3d ago·view thread
comment
Mass automated vulnerability scans have been a very common thing since years before the advent of this in 2001: https://en.wikipedia.org/wiki/Code_Red_(computer_worm) I remember …
3d ago·view thread
comment
Just FYI lm-studio is a GUI wrapper on top of a copy of llama-server that the lm-studio developers compile and distribute
4d ago·view thread
comment
Please give it a try before you assume it won't work? Tell me what platform you're on and follow the three sequential cli commands I provided, and if it doesn't compile I'll be gla…
4d ago·view thread
comment
This is like complaining that the mariadb team has failed to make their tech easy to use. It's a server side piece of software. There are plenty of user friendly GUIs or wrappers for it. lm-studi…
4d ago·view thread
comment
I only want linkedin cringe if it's of the artisanal, hand crafted, organic, free range, GMO free, fair trade, heirloom variety.
4d ago·view thread
comment
It's literally three steps, assuming you have the equivalent of the debian "build-essential" dependencies installed on your system for cmake, compiler. It's the exact three rows of…
4d ago·view thread
comment
At this point the options are llama-server or vLLM if you're serious about running things at your desk in the under 256GB RAM size class (70B, 120B size models). In addition to, of course, 27B to…
4d ago·view thread
comment
I mean it's probably not, I just haven't got used to it yet. It's about the same level of security as installing a windows app on win2000 25 years ago and blindly downloading a .exe off…
4d ago·view thread
comment
One way I prevent possible catastrophic fuckups is that the 'doing code work' box that runs opencode or pi or whatever, is its entirely own separate VM and desktop environment (running as a …
4d ago·view thread
comment
the full set of llama.cpp binaries builds in under 5 minutes with an unmodified build workflow straight from their github page on a literally ten year old dual xeon.
4d ago·view thread
comment
Anything that suggests curl into bash just plain sketches me out. (edit: I know, this isn't totally rational, it just seems weird to me. We download and trust a lot of software and run code from …
4d ago·view thread
comment
Much of it in the ultimate service of selling very targeted advertising.
4d ago·view thread
comment
On a slightly related topic, static on disk files of LLMs are not incompressible, I have a number of "archived, maybe I'll use it later" Q8 quantized GGUF files that are about 90% of th…
4d ago·view thread
comment
Certain malls in North American cities with dense urban populations and good transit links are also very busy. For instance Metrotown in the centre of Burnaby, BC. I'd guess it has a <2% store…
4d ago·view thread
comment
I keep an extra one next to my panini press, the image quality from the whiteboard notes area on my fridge has never been sharper!
4d ago·view thread
comment
They're nowhere near smart enough, but in an ideal case , the same utility you would get out of hiring a $600 a month personal assistant with a basic desktop PC who lives in a developing country…
4d ago·view thread
comment
[flagged]
4d ago·view thread
comment
Could be the same thing soon. https://theonion.com/palantir-acquires-pentagon-for-800-bill... …
4d ago·view thread
comment
Or it's not a signal of an incredible offer to go laterally or upwards at another company, but that the environment became so impossible and toxic to work in that the person was like "f this…
4d ago·view thread
comment
People are slapping "AI" on anything these days to try to sell more units of whatever it is.
4d ago·view thread
comment
UAV sensor people made a thing that uses 300+ phone size sensors in a single pod, and software workflow to fuse the output together. It's been around quite a few years now. https://www…
5d ago·view thread