• 7 Posts
  • 125 Comments
Joined 1 year ago
cake
Cake day: June 9th, 2023

help-circle








  • Do you always argue for your own perspective? I subtly indicated that this was not how I act, and I did not try to justify such behaviors. I simply stated a softening perspective of how people are a product of their environment and the perspective benefits of the present information age. It is adding real world abstracted complexity to something phrased in idealized binary simplicity that might make someone with a troubled childhood self-conscious or uncomfortable and instead bending perspective to nostalgia. It is a kindness with empathy for people I know that have changed and turned their life around, and for those with potential because binary labels and oversimplification can be harmful.


  • j4k3@lemmy.worldtomemes@lemmy.worldI was an early adopter of Game Rage (tm)
    link
    fedilink
    English
    arrow-up
    26
    arrow-down
    7
    ·
    22 days ago

    It was an age without outlets for information. If you wanted to have a chance at figuring out that secret sauce to get to the next level or beat that boss, there was magazine or back issue roulette. There was admitting you suck to your cooler richer friends to beg them to tell you how to beat it. There was the tiny chance that the official guide you had to special order might mention the thing. Then there was the dialup internet where loading to the nipples of a single image took longer than most boys, and by the time the whole image loaded 1 in 3 was holding a self serviced d-pad controller of their very own. Times were different, and so were the frustrations. If you wanted to learn, you went to the library on the other side of town. People read newspapers for information and it had classified-craigslist that was like paying Twitter by the word to try and sell your broken junk. Those of us that didn’t rage, blew on our cartridges like we were making a magical wish to a genie.



  • I’m not saying it is no big deal to regularly expose one’s self to it.

    I know about how small of an amount can be smelled from building a power supply for a UV light, making mistakes, and yet still smelling the thing. It was back driving some circuit block that shouldn’t have been enough power to do anything, there was no visible effect, but I could still smell the faint smell of O3. Even with a tiny bulb, the smell is nearly instantaneous when the light is powered. Running one around anything that can rust is a bad idea. It is almost as bad as working with hydrochloric acid and hydrogen peroxide as enchants; everything goes super rusty fast. I don’t get exposed to it regularly, so I’m not worried. Most hotel rooms smell pretty strongly of O3 in the many I have stayed in and that is probably the most that the average person gets exposed for any extended length of time. The alternative is probably worse, but still people don’t worry about that one too much.



  • It is 6w before power supply losses and then a tube that is likely under 20% efficient. The primary thing that gets you with welding is the fraction of a second before an auto darkening lens activates. It is a tiny amount of time, but it adds up. Or all those times you accidentally touch the tungsten to the pool with tig. Welding has a wide spectrum, but it is the power that matters most here. The frequency and power are two separate and unrelated things. Like your microwave and WiFi are both 2.4 GHz machines. Your WiFi in your home router is limited to 100 milliwatts and is totally harmless. Your microwave needs a Faraday cage built in to avoid cooking you from the inside out because it is likely around 1500 watts. At 220nm the frequency doesn’t pass through skin or eye fluid and the power output of the light is low. Seriously, try getting into things like telescope filters where you’re trying to isolate certain frequencies. It is challenging at these frequencies to find anything that is transparent. Of all of my science books, my optics handbook is by far the largest and hardest for me to follow. That isn’t saying much, but I have built my own telescope electronics and eyepieces, along with hobby electronics, designing and etching circuit boards and photolithography using various UV lights I have built. Six watts is nothing major. I would be more concerned about how limited of an area one light can cover.







  • I don’t use perplexity, but AI is generally 60-80% effective with a larger than average open weights off line model running on your own hardware.

    DDG offers the ability to use some of these. I use a modified Mistral model still, even though its base model(s) are Llama 2. Llama 3 can be better in some respects but it has terrible alignment bias. The primary entity in the underlying model structure is idiotic in alignment strength and incapable of reason with edge cases like creative writing for SciFi futurism. The alignment bleeds over. If you get on DDG and use the Anthropic Mixtral 8×7b, it is pretty good. The thing with models is to not talk to them like humans. Everything must be explicitly described. Humans make a lot of implied context in general where we assume people understand what we are talking about. Talking to an AI is like appearing in court before a judge; every word matters. The LLM is basically a reflection of all of human language too. If the majority of humans are wrong about something, so is the AI.

    If you ask something simple like just a question, you’re not going to get very far into what the model knows. Models have very limited scope of focus. If you do not build prompt momentum into the space by describing a lot of details, the scope of focus is large but the depth is shallow. The more you build up momentum by describing what you are asking in detail, the more it narrows the scope and deeper connections can be made.

    It is hard to tell what a model really knows unless you can observe the perplexity output. This is more advanced, but the perplexity score for each generated token is how you infer that the model does not know something.

    Search sucks because it is a monopoly. There are only 2 relevant web crawlers m$ and the goo. All search queries go through these either directly or indirectly. No search provider is deterministic any more. Your results are uniquely packaged to manipulate you. They are also obfuscated to block others from using them for training better or competitive models. Then there is the anti trust US government case and all of that which makes obfuscating one’s market position to push people onto other platforms temporarily, their best path forward. - criminal manipulators are going to manipulate.