elorant

13 ms·

211.
by elorant 2y ago
I’m using Llama-3 8B to classify html files. It’s surprisingly good, and I run it on an RTX 4060 Ti at 8-bit quantization. No complains so far.
212.
Enshittification isn't caused by venture capital(pluralistic.net)
3 points by elorant 2y ago|2 comments
213.
by elorant 2y ago
What happens with excess hardware that’s been already bought? It could take years for all of that to be absorbed by smaller companies. And secondly, if this opens the era of self-training models then why go for a 671B one and not a smaller
214.
by elorant 2y ago
You know what was exciting back then? Tech magazines. That’s where you discovered all the great stuff and read about the research that was happening.
215.
by elorant 2y ago
Not everyone needs the largest model. There are variations or R1 with fewer parameters that can easily run on consumer hardware. With 80% size reduction you could run 70B on 8-bit on an RTX 3090. Other than that, if you really need the big
216.
by elorant 2y ago
No, I haven't. I've used Perplexity's R1 but I don't know how many parameters it has. It's quite good, although too slow.
217.
by elorant 2y ago
There are multiple variations of the model starting from 1.5B parameters.
218.
Nvidia's $465B plunge triggered by DeepSeek, largest in market history(business-standard.com)
3 points by elorant 2y ago|1 comments
219.
by elorant 2y ago
You need a model that has specifically been extended for larger context windows. For Llama-3 there's Llama3-gradient with up to 1M tokens. You can find it at ollama.com
220.
Data breach hitting PowerSchool looks bad(arstechnica.com)
4 points by elorant 2y ago|1 comments
221.
Nvidia reportedly prioritizing dual-die Blackwell GPUs with CoWoS-L packaging(tomshardware.com)
2 points by elorant 2y ago|0 comments
222.
EU protests new US AI chip restrictions – some countries face GPU caps(tomshardware.com)
4 points by elorant 2y ago|0 comments
223.
Using AI Agents to Flood Reddit with Marketing Slop(gizmodo.com)
3 points by elorant 2y ago|0 comments
224.
by elorant 2y ago
You're missing the most critical part though. Memory bandwidth. It hasn't been announced yet for Digits and it probably won't be comparable to that of dedicated GPUs.
225.
by elorant 2y ago
Well obviously it has to be low otherwise they would cannibalize their high end GPUs.
226.
by elorant 2y ago
You don't need to run them in x16 mode though. For inference even half that is good enough.
227.
by elorant 2y ago
My gut feeling is that this will be accompanied by a relaxed policy on fake profiles too.
228.
TSMC's wafer pricing now $18,000 for a 3nm wafer, increased over 3X in 10 years(tomshardware.com)
5 points by elorant 2y ago|2 comments
229.
Apple Siri Eavesdropping Payout–Here's Who's Eligible and How to Claim(forbes.com)
5 points by elorant 2y ago|4 comments
230.
by elorant 2y ago
You're oversimplifying the narrative. It's not ads per se that bother. It's the tracking part. Make ads contextual like they used to be and there's no problem at all, actually they might turn to be interesting. Keep trac
231.
by elorant 2y ago
How is China training models without access to cutting edge GPUs?
232.
by elorant 2y ago
Whatever their motivation is, disabling TPM from Bios is the safest way to avoid upgrading to Win11.
233.
by elorant 2y ago
Well that makes sense. If you bother to look at how much engagement all those fake female accounts are generating it's no wonder that Facebook wants to own that game too.
234.
by elorant 2y ago
Not always, because system RAM also has to be equally adequate, but mostly yes, it's about the total VRAM of the GPU(s).
235.
by elorant 2y ago
I consider the RTX 4060 Ti as the best entry level GPU for running small models. It has 16GBs of RAM which gives you plenty of space for running large context windows and Tensor Cores which are crucial for inference. For larger models proba
236.
by elorant 2y ago
It's not just discipline. Kindles are more convenient too for reading. No refresh rate so you can spend hours reading without eye strain.
237.
by elorant 2y ago
Get a kindle and start reading books. That’s what's keeping me from staring at a phone all the time. I might be spending similar amounts of time looking at a screen but at least I get something valuable out of it.
238.
The average American spent 2.5 months on their phone in 2024(pcmag.com)
192 points by elorant 2y ago|202 comments
239.
by elorant 2y ago
Laser printers print a status page that describes most of their settings, how many toners you've changed and also includes the total page count.
240.
by elorant 2y ago
Bought a Lexmark E232 laser printer back in 2006. 18 years and 110k pages later the damn thing still works flawlessly. I have nothing but admiration for their printers.

More ›