filterfiber

5 ms·

1.
by filterfiber 3y ago
> still run on a device like the iPhone, which weights about 170g [1]? The human brain is 8 times heavier. Why are you comparing the weight? I'm sorry but this is a bizarre comparison. This isn't even apples to oranges, this is
2.
by filterfiber 3y ago
In their second sentence they have the most honest response I've seen so far at least: " averaged across 4 diverse customer tasks, fine-tunes based on our new model are _slightly_ stronger than GPT-4, as measured by GPT-4 itself.
3.
by filterfiber 3y ago
Their timezone has been wrong in the tzdb for at least 10 years apparently https://mm.icann.org/pipermail/tz/2023-December/033339.html
4.
by filterfiber 3y ago
Does anyone know if mmWave could differentiate between my cat and I?
5.
by filterfiber 3y ago
Does anyone know where I should look if I want to detect specific sounds? Like a smoke alarm, food bowl dispenser (its very distinct), cat meowing, 3d printer collision, that sort of thing?
6.
by filterfiber 3y ago
Fun monetization strategy for federated apps - federate with your own instance dedicated to ads. But more seriously what is the monetization strategy for federated apps? Up front pay or subscription for using the app?
7.
by filterfiber 3y ago
Does anyone have any input on how this compares outside of benchmarks? They don't appear to have any info on how they made it.
8.
by filterfiber 3y ago
> Realistically, even with Turbo+LCM, you're still going to 4+ steps (often 8+), with CFG, for reasonable one-generation quality anywhere close to the images people generated at 50+ steps without Turbo/LCM. For sure the only re
9.
by filterfiber 3y ago
> since you have stocks of the billion dollar message I have no idea what you mean by this? Are you saying I'm defending apple somehow? Because my point is nearly every phone/laptop could pull off this attack, not just a single
10.
by filterfiber 3y ago
I don't know why the title even mentions the flipper. The attack can be done from any device that can send crafted BLE packets including laptops/android phones, etc. Apple just fixed a BLE DOS attack.
11.
by filterfiber 3y ago
Better title: "Apple fixes BLE DOS attack". I'm tired of media acting like the flipper is some kind of "super special hacking tool", it is very literally getting it banned in some places when all of it's intern
12.
by filterfiber 3y ago
I mean, if you need a human in the loop to verify the image quality then you HAVE to pre-compute the images. > 100 images/s is likey too much volume You can always generate less
13.
by filterfiber 3y ago
I was just using that as a reference. Stable diffusion will run well with almost any relatively modern gpu. You don't have to use a 4090, you'll still get double digit performance with a 3060 or whatnot. > for people who can on
14.
by filterfiber 3y ago
> I found this claiming an A100 can generate 1 image/s. The article you linked is over a year old. Needless to say there have been a LOT of optimizations in the last year. Back then it was common to use 50+ steps for many of the com
15.
by filterfiber 3y ago
> But that's not the point, obviously. If you want to say the zero2-w is what's making it then sure. > Besides, a 4090 costs more than a car. They only cost ~0.70USD for 1 hr. In fact you could put this on an A100 for 1$&#x2
16.
by filterfiber 3y ago
This project is a fun POC but it's not very practical for that type of application. A 4090 can generate over 100 images a second with turbo+lcm and a few techniques, you can make 2 days worth of images in 1 seconds. You could make a ye
17.
by filterfiber 3y ago
> which means even fewer people finetuning those models. Finetunes rarely led to "Top 5 performance" for the small ones. Previously the top 10+ were all 70B, with maybe a few 30B in there. There were nearly no 13B's, let
18.
by filterfiber 3y ago
I know the hugging face leaderboard isn't wildly accurate. But the top models right now are almost all under 70B. Most are 7B, and the top is 10B. If the benchmarks are even remotely accurate then this is rather wild. Apparently multip
19.
by filterfiber 3y ago
The current bottleneck for most current hardware is RAM capacity than memory bandwidth and last is FLOPS/TOPS. The coral has 8 MB of SRAM which uh, won't fit the 2GB+ that nearly any decent LLM require even after being quantized.
20.
by filterfiber 3y ago
AFAIK there's no public sdk for it, only a single third party game is in development and it's by the developers of garry's mod (one of the biggest third party source "1" games). I'd consider it still proprietar
21.
by filterfiber 3y ago
I don't understand why they're comparing the parameter sizes to lines of code. AFAIK you can just increase the layer parameters of a 1B model to whatever you want? Like, the difference between a 1B and 175B model can be just chang
22.
by filterfiber 3y ago
FWIW - I need to remeasure but - IIRC my system with a 4090 only uses ~500w (maybe up to 600w) during inference of LLMs, the LLMs have a lot harder time saturating the compute compared to stable diffusion I'm assuming because of the VR
23.
by filterfiber 3y ago
So this specific model ships with a default config of 2 experts per token. So you need roughly two loaded in memory per token. Roughly the speed and memory of a 13B per token. Only issues is that's per-token. 2 experts are choosen per
24.
by filterfiber 3y ago
> in as little as 16GB of RAM with room to spare. I don't think that's the case, for full speed you still need (5B*8)/2+2~fewB overhead. I think the experts chosen per-token? That means that yes you technically only need t
25.
by filterfiber 3y ago
> Previous State-of-the-Art: [...] The number of parameters in the LSTM layers of these models vary from 2 million to 151 million. > We present model architectures in which a MoE with up to 137 billion parameters Back in 2017 most mod
26.
by filterfiber 3y ago
After the Unity re-licensing fiasco, and OpenAI yanking old models, is there any protections in place to allow the use of StableLM-Zephyr-3B indefinitely or will you be able to just deny any continuing access to models?
27.
by filterfiber 3y ago
Refact has a decent 1.6B model that I think is better https://huggingface.co/smallcloudai/Refact-1_6B-fim
28.
by filterfiber 3y ago
No it's not (according to their benchmarks). Zephyr-7B-B still beats it in most benchmarks but it's close. This model is almost Zephyr-7B-B performance at 3B size which is a lot better for inference requirements.
29.
by filterfiber 3y ago
> Hardware: StableLM Zephyr 3B was trained on the Stability AI cluster across 8 nodes with 8 A100 80GBs GPUs for each nodes. I might be missing it but do they say the number of training tokens that was used to train this? This would help
30.
by filterfiber 3y ago
My favorite with microsoft was just a year or two ago (not sure about now) - there was something like a 63 character limit for the login password. Obviously they didn't tell me this, and of course they allowed me to set my password to

More ›