Profile pic is from Jason Box, depicting a projection of Arctic warming to the year 2100 based on current trends.

  • 0 Posts
  • 69 Comments
Joined 1 year ago
cake
Cake day: March 3rd, 2024

help-circle
  • Current LLMs would end that sketch soon, agreeing with everything the client wants. Granted, they wouldn’t be able to produce, but as far as the expert narrowing down the issues of the request, ChatGPT would be all excited about making it happen.

    The hardest thing to do with an LLM is to get it to disagree with you. Even with a system prompt. The training deep down to make the user happy with the results is too embedded to undo.









  • Ollama.com is another method of self hosting. Figuring out which model type and size for what equipment you have is key, but it’s easy to swap out. That’s just an LLM, where you go from there depends on how deep you want to get into the code. An LLM by itself can work, it’s just limited. Most of the addons you see are extra things to give memory, speech, avatars, and other extras to improve the experience and abilities. Or you can program a lot of that yourself if you know Python. But as others have said, the more you try to get out, the more robust a system you’ll need, which is why you find the best ones online in cloud format. But if you’re okay with slower responses and lower features, self hosting is totally doable, and you can do what you want, especially if you get one of the “Jailbroke” models that has had some of the safety limits modified out of them to some degree.

    Also as mentioned, be careful not to get sucked in. Even a local model can be convincing enough sometimes to fool someone wanting to see things. Lots of people recognize that danger, but then belittle people who are looking for help in that direction (while marketing realizes the potential profits and tries very hard to sell it to the same people).



  • Telepathy never had any set rules, it was just used however it made sense for the story. Even Spock had differing abilities - mostly he’d do his mind melds while touching, but there’s one episode where he contacts and controls a guard through a wall. The good thing is that for Deanna as time went by they used her abilities and her job function and training better than in the beginning with the stupid side comments of “I sense they’re not telling everything.” No, really?


  • I will give it to you, when it works, it does some magical stuff. But try designing such complex things that are miracles in coding and then it have to run on a half-ass computer. I want to say terminal, it’s not that, but it’s those small fake computers that companies seem to think are better to get than an actual desktop because they’re cheap. I know that’s hardware, not Excel, but Excel does not run well on that, so…

    Or worse, you get moved to 365 which doesn’t do most scripting and breaks all that was working. That cloud shit is a problem.







  • I agree it can be used fallaciously, often found in the business world. My point was to include both good and bad honestly and not hide it, and people won’t shut down if they get the good first. It also depends on the subject - if they’re on the right track and your suggestion leads to better results, that’s not as negative as telling someone they’re doing something incorrectly and offering a different way.

    In the end, how you say things is just as important as what is said.


  • On constructive criticism - definitely rule one is make sure that it’s invited first, but second, the best way to “sweeten” a critique and make it more appealing is to put it between compliments. Don’t have a bare remark about the problems or suggestions, tell them what you like first, then how they might change things, and then close with something else positive or simply thanking them for sharing it. Even if someone says they want to hear what people think, it’s normal to be defensive, so help lower that reaction first, and then leave them feeling appreciated even though you pointed out issues you saw.