What do people really ask chatbots? It’s a lot of sex and homework.

Comments

• 1984@lemmy.today ⁨3⁩ ⁨months⁩ ago

  There are models you can download and run at home that doesn’t have the politically correct censorship inside. It’s very nice to not have artificial politeness for example, and the models actually answers your actual questions.

  You need a powerful computer for some of them though.

  source

  • Steviepoo@lemmy.world ⁨3⁩ ⁨months⁩ ago

    Such as? Where would technologically proficient AI-beginner start?

    source

    • merari42@lemmy.world ⁨3⁩ ⁨months⁩ ago

      For a user without much technical experience using a ready-made gui like Jan.ai with automatic model download and ability to run models with the ggml library on consumer grade hardware like mac M-series chips or cheap GPUs by either Nvidia or AMD is probably a good start.

      For a little bit more technically proficient users Ollama is probably a great choice to start to host your own OpenAI-like API for local models. I mostly run gemma2 or small llama 3.1 like models with that.

      source

      • -> View More Comments
    • BreadstickNinja@lemmy.world ⁨3⁩ ⁨months⁩ ago

      Oobabooga is a pretty beginner-friendly solution for running LLMs locally. Models are freely available on Huggingface, but look for GGUF quantizations that will fit in your VRAM. The good thing about GGUFs is that they’re typically offered in a wide range of sizes so you can pick one that will fit on your GPU. If you use all your VRAM and start offloading to system memory then the generation will be far slower.

      Once you mess around with Oobabooga a bit, I’d highly recommend picking up the SillyTavern front-end. Oobabooga runs the actual model while SillyTavern manages characters, world lore, and offers a wide range of other features including a “visual novel” mode where you can set up character sprites that emote based on the content of the messages. It takes a while to get the hang of but it’s pretty cool.

      source
    • Sabata11792@ani.social ⁨3⁩ ⁨months⁩ ago

      Ollama software and you can pick a model that dose what you want. Mistral and Llama are currently best IMO but it changes often.

      source
• tehmics@lemmy.world ⁨3⁩ ⁨months⁩ ago

  90% of mine is just programming syntax. The rest is shit that Google can’t answer anymore. Then 1% is me trying to trick it into telling me about illegal stuff

  source

  • sugartits@lemmy.world ⁨3⁩ ⁨months⁩ ago

    What illegal stuff?

    … Asking for a friend.

    source

    • tehmics@lemmy.world ⁨3⁩ ⁨months⁩ ago

      Nothing special, I got it to give me a recipe for meth and a list of sites to pirate games. Stuff like that

      source
• MonkderVierte@lemmy.ml ⁨3⁩ ⁨months⁩ ago

  I asked it yesterday, why touchscreen input is separate on Linux (and needs software’s support) and how to map it to mouse input.

  source

  • CalcProgrammer1@lemmy.ml ⁨3⁩ ⁨months⁩ ago

    I wrote a program to do just that

    gitlab.com/CalcProgrammer1/TouchpadEmulator

    source

    • MonkderVierte@lemmy.ml ⁨3⁩ ⁨months⁩ ago

      Woo, thanks for the link!

      source
• IGuessThisIsForNSFW@yiffit.net ⁨3⁩ ⁨months⁩ ago

  I had a roommate who reviewed and scored responses for Google Bard. A ton of it was people generating posts for whatever business/crypto/alpha-male grift they were running. The main thing though was really really specific fetish stuff.

  source