To be honest, I’m not sure what we’re arguing - we both seem to have a sound understanding of what LLM is and what it is not.

I’m not trying to defend or market LLM, I’m just describing the usability of the current capabilities of typical LLMs.

It goes a tad bit beyond classical conditioning… LLM’a provides a much better semantic experience than any previous technology, and is great for relating input to meaningful content. Think of it as an improved search engine that gives you more relevant info / actions / tool-suggestions etc based on where and how you are using it.

Here’s a great article that gives some insight into the knowledge features embedded into a larger model: https://transformer-circuits.pub/2024/scaling-monosemanticity/

To be fair, I think he is mostly endorsing the concept of the implementation, lined out in his seven points, not the actual implementation since it isn’t available yet.

He sort of invented it, so you have to think he’s commenting on the concept here, not the implementation.

I have tried a lot of medium and small models, and there it just no good replacement for the larger ones for natural text output. And they won’t run on device.

Still, fine-tuning smaller models can do wonders, so my guess would be that Apple Intelligence is really 20+ small and fine tuned models that kick in based on which action you take.

https://en.m.wikipedia.org/wiki/Andrej_Karpathy

All my devices and networks are constantly on VPN, it’s just basic obscurity, and it’s such a hassle posting on Lemmy now - my activity is probably down 90%.

Could you please consider allowing basic text posts and / or comments from behind a VPN? You could even disallow all markdown - that’s a decent trade off for my sake.

But please don’t just block all VPN usage if there are options you have not explored yet.

All these examples are not just using stable diffusion though. They are using an LLM to create a generative image prompt for DALL-E / SD, which then gets executed. In none of these examples are we shown the actual prompt.

If you instead instruct the LLM to first show the text prompt, review it and make sure the prompt does not include any elephants, revise it if necessary, then generate the image, you’ll get much better results. Now, ChatGPT is horrible in following instructions like these if you don’t set up the prompt very specifically, but it will still follow more of the instructions internally.

Anyway, the issue in all the examples above does not stem from stable diffusion, but from the LLM generating an ineffective prompt to the stable diffusion algorithm by attempting to include some simple negative word for elephants, which does not work well.

I can’t find any source for the numbers from Vietnam and WW2 in the linked sources.

Yeah, well. I miss Lemmy from July I guess. Now, on any contended topic, I feel it’s no better than Reddit (except for the whole overlords censoring thingy). Have a look at Tildes for a little less stupid news reporting and articles.

Yeah I need to figure why I can’t follow the cross post here on Lemmy so i can get to the article.

I’ve watched the complete interview with Lex, so if this content is based on that, then title on this post here on Lemmy is junk, so if that’s the title used on the article I won’t have high hopes for that either.

I just despise clickbaited titles with statements that are twisted into someone elses opinion on the meaning behind what was said. I see it so often. I have no better solution currently than finding the actual source and make my own opinion, but it’s tiring. I do believe that “journalism” like this makes the world is worse place to live.

The links does not work, but was it really based on that interview?

I think his take on the topic was reasonable - it’s allowed to just not want any more death or suffering on either side - but not realistic.

It was also one sided - you really can’t focus on what Ukraine can do without also saying who started it all and caused the war in the first place. He never mentioned that Russia started it and how easy it is for Russia to stop all suffering and achieve peace. At best you can say he’s given up on Russia, don’t think they’ll change, and try to get Ukraine to stop more suffering. But is that really realistic - did Russia stop after Donetsk in 2014? It’s not reasonable to think it will be different this time. Give Russia peace for 5 years and let them rebuild their strength? They’ll just start again and set target for Kiev.

For importing movies: two ways, depending on the state of your existing media.

1. Point Radarr to your media. It might detect everything. Radarr will let you set rules for file name and structure. Follow jellyfin guidelines. Then let Radarr reorganize your files to the new structure. Then import to Jellyfin
2. If it’s very disorganized, you might have to organize it according to the jellyfin guides yourself, or script it if you can, before importing to jellyfin.

I’m a little curious what you replaced Photoshop with :)

The issue seems to be comments - I’m not allowed to edit them without moderator approval, and moderators have turned off all moderator requests 😅