• 1 Post
  • 28 Comments
Joined 1 year ago
cake
Cake day: June 15th, 2023

help-circle








  • I’m just someone on the internet, take everything with a grain of salt.


    Only go to sleep when you really intent to. No scrolling on your phone in the bed, no reading books,…

    The moment you lay/sit down, you intent to sleep.

    If it doesn’t work: get up, do something else, try again later.


    Isolate factors that could keep you awake like sugar, caffeine, alkohol,… Even a handfull of gummies could influence when you fall asleep.

    Check on air-quality, room temperature and moisture, dryness of bedding, room -brightness, -light sources, noice levels.

    Try out keeping a journal of these things, as well as your mental state and reflect on your sleep in the morning.

    If available, ask your SO or room-mate about their sleep, that’s a great way to identify external influences.


    If your data stays inconsistent, ask a doctor, as there might be medical conditions influencing you falling asleep, ike high blood pressure or hormone imbalances.









  • Thanks for the writeup! So far I’ve been using ollama, but I’m always open for trying out alternatives. To be honest, it seems I was oblivious to the existence of alternatives.

    Your post is suggesting that the same models with the same parameters generate different result when run on different backends?

    I can see how the backend would have an influence hanfling concurrent api calls, ram/vram efficiency, supported hardware/drivers and general speed.

    But going as far as having different context windows and quality degrading issues is news to me.





  • yes: sntx.space, check out the spurce button in the bottom right corner.

    I’m building/running it the homebrewed-unconventional route. That is I have just a bit of html/css and other files I want to serve, then I use nix to build that into a usable website and serve it on one of my homelab machines via nginx. That is made available through a VPS running HA-Proxy and its public IP. The Nebula overlay network (VPN) connects the two machines.