grey8

6 ms·

1.
by grey8 2y ago
To preface: I randomly stumbled upon this because I looked up info. ... and saw that at least this page containing information regarding an USAID grant is now 404ing. Look at this: https://foreignaffairs.house.gov/wp-content
2.
by grey8 2y ago
I missed this the first time, too - they have an implementations page: https://crdt.tech/implementations
3.
by grey8 2y ago
Hey! While I was similar to your child (now a good few years older), I'm not a parent or really knowledgeable in these situations, but I'd highly recommend checking out Dr. K ("Healthy Gamer"). He's an actual psychi
4.
by grey8 2y ago
I have recently heard or read somewhere that 20mg of Botulinum Toxin is enough to kill the whole human population - multiple times.
5.
by grey8 2y ago
"A truly sentient animal"? Aren't all animals sentient? And what is the difference between "truly sentient" and "sentient"?
6.
by grey8 2y ago
Ha, thank you for that - I had the exact same idea, down to the extensibility, but I'm currently working on something else. Looks awesome, I will check it out!
7.
by grey8 2y ago
It's not about the model. The model can output more - it's about the API. A better phrasing would be that they don't allow you to output more than 4k tokens per message. Same with Anthropic and Claude, sadly.
8.
by grey8 2y ago
transformers.js is such a cool library. I made a small web app with it that uses it to remove backgrounds from images (with BRIA AI's RMBG1.4 model) at https://aether.nco.dev The fact you don't need to send your data t
9.
by grey8 2y ago
If I'm not mistaken, this is not on the models itself, but rather on the implementation of the addon. I haven't found an open source VSCode or WebStorm addon yet that allows me to use a local model and implements code completion a
10.
Show HN: On-device image background removal(aether.nco.dev)
1 points by grey8 2y ago|0 comments
11.
by grey8 3y ago
Just a thought, but what happens if you use orchestrate a browser instance with an installed ad blocker like uBlock Origin?
12.
by grey8 3y ago
Why is that relevant to the size? Post search on X is done as it is with any other data from any other source, you use RAG and function calling to insert the context. < 7B open source models can function call very well. In fact, Nous Her
13.
by grey8 3y ago
Nope. They did release code for training, inference and fine tuning, but no datasets or weights. See https://github.com/Stability-AI/stable-audio-tools
14.
by grey8 3y ago
I'm not making this up, I thought the same almost immediately and only then saw the throwaway's message: By the first paragraph, my spidey senses were tingling, and by the first two sentences of the second paragraph, I wanted to w
15.
by grey8 3y ago
I used a Growkit with no experience at all; and it worked beautifully. YMMV. The yield is "not as big", right, but it's still more than you can reasonably consume. I don't quite remember, but I think three flushes yielde
16.
by grey8 3y ago
I also read into vitamins and minerals (while trying to ignore all the snake oil), and I agree with you! I tried a lot and the ones I keep buying are the same you do - Magnesium, Omega-3, D3+K2 and a B complex. These are not only the ones t
17.
by grey8 3y ago
I agree, the README is not really understandable if you're not into AI research techno-babble. Just adding one sentence targeted at normal people would maybe have been useful. To answer your question, it's a model that you can giv
18.
by grey8 3y ago
I think you're right. Why? Just observe how art and content is perceived before and after you tell people it was made by AI. I think it's an unfounded bias. You probably remember the AI art piece that won an art contest? It was pe
19.
by grey8 3y ago
Just to confirm: It does. The domain is clever, it's "linus.writes.now" (or maybe "linus.blogs.now", in this context). "schreiben" / "schreibt" means writing.
20.
by grey8 3y ago
The Huggingface page of Replit 3Bs says "The model has been trained on the MosaicML platform on 128 H100-80GB GPUs." Source: https://huggingface.co/replit/replit-code-v1_5-3b I'm not an ML engineer, just
21.
by grey8 3y ago
There is a podcast episode of Lex Fridman talking with Mark Zuckerberg, and in this episode Mark states that the reason Llama is (and will stay) open source is that that is the deal that Meta has with the engineers. To get top engineers, th
22.
by grey8 3y ago
That's not FSR2, that's FSR1. FSR2 needs game specific integration (IIRC because it needs access to motion vectors) and cannot be applied globally.
23.
by grey8 3y ago
I'm not / won't be a parent, so please take this with a grain of salt and not as an offense to your or anyone elses parenting. I think the fact many of you have deeply thought about the topic is amazing, but what I wonder is