GNU social JP
  • FAQ
  • Login
GNU social JPは日本のGNU socialサーバーです。
Usage/ToS/admin/test/Pleroma FE
  • Public

    • Public
    • Network
    • Groups
    • Featured
    • Popular
    • People

Conversation

Notices

  1. Embed this notice
    Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 19:06:43 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer

    Yet another thing solved by AgentV3N's prompt routing.

    We're moving closer and closer to using 1B models for the first stage of prompt evaluation. A simple "thank you" could be met with an agent-level prepared response, or even just an emoji reaction.

    Meanwhile, ChatGPT by default uses the last model you used in the conversation. Imagine blowing o1 reasoning on "thanks mate". Wouldn't be me.

    In conversation about 2 months ago from ligma.pro permalink

    Attachments


    1. https://ligma.pro/system/media_attachments/files/114/358/406/187/953/141/original/8a70a2f82823068f.jpeg
    • Phantasm and Johnny Peligro like this.
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 21:11:27 JST SuperDicq SuperDicq
      in reply to

      @r000t@ligma.pro The survey found that around 70% of people are polite to AI when interacting with it, with 12% being polite in case of a robot uprising.Holy shit people are stupid.

      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 21:14:48 JST SuperDicq SuperDicq
      in reply to

      @r000t@ligma.pro And people they are talking about this like it is about the environmental cost or whatever.

      You know what I think? It's about fucking dignity.

      I will not say "thank you" to a fucking machine. It's not a person.

      In conversation about 2 months ago permalink
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:04:15 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      @SuperDicq @r000t what's wrong with being polite?
      In conversation about 2 months ago permalink
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:05:11 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      • Protoss
      • Yomiel
      @yomiel @Cyrillic @r000t @SuperDicq there's not s rape button but you have to describe the rape action
      In conversation about 2 months ago permalink
    • Embed this notice
      Protoss (cyrillic@lab.nyanide.com)'s status on Friday, 18-Apr-2025 22:05:14 JST Protoss Protoss
      in reply to
      • SuperDicq
      • Yomiel
      @yomiel @r000t @SuperDicq I talk to them with the politeness of a chud talking to an israeli flag on pol
      In conversation about 2 months ago permalink
      Johnny Peligro repeated this.
    • Embed this notice
      Yomiel (yomiel@new.asbestos.cafe)'s status on Friday, 18-Apr-2025 22:05:14 JST Yomiel Yomiel
      in reply to
      • SuperDicq
      • Protoss
      @Cyrillic @r000t @SuperDicq
      How I talk to AI:
      In conversation about 2 months ago permalink

      Attachments


      1. https://new.asbestos.cafe/media/b6/5c/2a/b65c2acf7bc5a99f9b8567517c18ba7d21d53bc93cd523a238ac4d943849d023.png
      Johnny Peligro likes this.
    • Embed this notice
      Yomiel (yomiel@new.asbestos.cafe)'s status on Friday, 18-Apr-2025 22:05:16 JST Yomiel Yomiel
      in reply to
      • SuperDicq
      @SuperDicq @r000t
      We should be talking to AI like a 1950's husbands talked to their wives.
      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:08:21 JST SuperDicq SuperDicq
      in reply to
      • Johnny Peligro

      @mischievoustomato@tsundere.love @r000t@ligma.pro I think anthropomorphizing chatbots is a really bad as the big AI companies will abuse this to create deceptive user experiences.

      In conversation about 2 months ago permalink
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:10:34 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      @SuperDicq @r000t bleh, if it helps me and its a chatbot, I'll say please and thank you
      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:15:50 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • SuperDicq
      • Protoss

      @Cyrillic
      Bingo. That's also something we handle in the evaluation loop.

      When a session is "closed" (released by the zone it was opened in), there's also what amounts to a "customer satisfaction survey" but it's performed in the background by the system loop. This is where the system can literally ask us for tools that would have made the request more efficient, more accurate (don't use a language calculator to calculate numbers), or possible at all.
      @SuperDicq

      In conversation about 2 months ago permalink
    • Embed this notice
      Protoss (cyrillic@lab.nyanide.com)'s status on Friday, 18-Apr-2025 22:15:51 JST Protoss Protoss
      in reply to
      • SuperDicq
      @SuperDicq @r000t though saying "thank you" is gay, some form of "thanks that fixed my problem" I think has value because if conversations are logged and used to improve the model then that is a label for a helpful response.
      In conversation about 2 months ago permalink
      Phantasm likes this.
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:16:57 JST SuperDicq SuperDicq
      in reply to
      • Protoss

      @Cyrillic@lab.nyanide.com @r000t@ligma.pro You should only run AI models on your own machine and if that is not possible I definitely urge you against using any service that log yours prompts. Think about your privacy.

      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:18:17 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • SuperDicq
      • Johnny Peligro

      @SuperDicq
      What do you consider "deceptive" in this circumstance?

      I can see a mihoyo-shaped object being like "Don't you love me enough to get the Professional plan?" but besides that I don't see a "persistent" customized character (what we call an Agent) being too terrible. It's basically a secretary at that point.
      @mischievoustomato

      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:21:33 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • SuperDicq
      • Protoss

      @SuperDicq
      We wanted to do this, too! That's one of the reasons I got a laptop with an NPU (as a writeoff lmao) a few weeks ago.

      The problem is that those "NPUs" are absolute fucking memes and basically can only do image classification. No good ways to run even 1B models on those.

      But like hell yeah I want you to do as much as possible on *your* equipment that you pay for lmfao.

      As for privacy, customers requiring HIPAA compliance (or similar) can force all inference to happen on our infra. The eval loop also looks for things that appear to be PHI and halt the request if the account is not enabled for HIPAA compliance.
      @Cyrillic

      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:21:48 JST SuperDicq SuperDicq
      in reply to
      • Johnny Peligro

      @mischievoustomato@tsundere.love @r000t@ligma.pro If it helps you add more context to prompts it's a good trick but I still recommend stripping out the "please" and "thank yous" after you have finished writing your prompt as all this does is add more tokens making the prompt harder to process while not improving results.

      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:24:58 JST SuperDicq SuperDicq
      in reply to
      • Protoss

      @r000t@ligma.pro @Cyrillic@lab.nyanide.com I don't know where you work but gathering data on your customers in this sounds absolutely proprietary

      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:26:56 JST SuperDicq SuperDicq
      in reply to
      • Johnny Peligro

      @r000t@ligma.pro @mischievoustomato@tsundere.love I am a firm believer that LLMs should return de-anthropomorphized responses only, like this example.

      Why? Because normies are stupid and otherwise will start to think that there's some actual sort of sentience behind the scenes or something.

      In conversation about 2 months ago permalink

      Attachments


      1. https://minidisc.tokyo/files/9e641e08-7862-47d1-b91c-2a7e9a96adb6
      prettygood and Fish of Rage like this.
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:29:53 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      @SuperDicq @r000t im glad you aren't in charge then. I made my ChatGPT talk to me like a tsundere
      In conversation about 2 months ago permalink
    • Embed this notice
      翠星石 (suiseiseki@freesoftwareextremist.com)'s status on Friday, 18-Apr-2025 22:31:10 JST 翠星石 翠星石
      in reply to
      • SuperDicq
      • Johnny Peligro
      @SuperDicq @r000t @mischievoustomato >normies are stupid and otherwise will start to think that there's some actual sort of sentience behind the scenes or something.
      That is exactly what normies believe - they think LLMs are sentient.

      The whole business plan is to trick people into think they are interacting with an "AI" and separate fools from their money.
      In conversation about 2 months ago permalink
      snacks likes this.
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:33:14 JST SuperDicq SuperDicq
      in reply to
      • 翠星石
      • Johnny Peligro

      @Suiseiseki@freesoftwareextremist.com @r000t@ligma.pro @mischievoustomato@tsundere.love I know this is a relatively small survery (1000 participants), but a staggering 12% even believe that if they are not polite to ChatGPT now they will be the first ones die in an unavoidable future AI uprising.

      And I'm supposed to be the mentally ill because I have autism? Normies are not alright.

      In conversation about 2 months ago permalink
    • Embed this notice
      翠星石 (suiseiseki@freesoftwareextremist.com)'s status on Friday, 18-Apr-2025 22:35:10 JST 翠星石 翠星石
      in reply to
      • SuperDicq
      • Johnny Peligro
      @SuperDicq @r000t @mischievoustomato Normies have never been alright.
      In conversation about 2 months ago permalink
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:35:24 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      • Johnny Peligro
      @SuperDicq @r000t
      In conversation about 2 months ago permalink

      Attachments


      1. https://tsundere.love/media/76628bd3-9d2d-42d7-8121-b135d4fdcbc6/image.png

      2. https://tsundere.love/media/0f3603d9-2ede-421d-8359-92c74170e0d8/oh%20yeah.jpg
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Friday, 18-Apr-2025 22:36:34 JST SuperDicq SuperDicq
      in reply to
      • Johnny Peligro

      @mischievoustomato@tsundere.love @r000t@ligma.pro That's just very cringe, ngl

      In conversation about 2 months ago permalink
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 22:38:18 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      @SuperDicq @r000t oh no the horror
      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:46:37 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • 翠星石
      • SuperDicq
      • Johnny Peligro

      @SuperDicq
      It's just a 21st century Pascal's Wager. The part where they're clueless normies is that if you asked them now the information that your were nice makes its way to whatever is exterminating people, they won't be able to conceptualize that. It's all just "the AI" like there's "the cloud"
      @Suiseiseki @mischievoustomato

      In conversation about 2 months ago permalink
    • Embed this notice
      翠星石 (suiseiseki@freesoftwareextremist.com)'s status on Friday, 18-Apr-2025 22:47:55 JST 翠星石 翠星石
      in reply to
      • SuperDicq
      • Johnny Peligro
      @r000t @SuperDicq @mischievoustomato I'd be glad to put a bullet in whatever proprietary software tries to get me tbh.
      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:48:23 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • SuperDicq
      • Protoss

      @SuperDicq
      rootworx Technologies.

      And the testing realm puts a big banner on the client saying "Operators can view all messages on the testing realm."
      @Cyrillic

      In conversation about 2 months ago permalink
    • Embed this notice
      Nietzschean Ekko Enjoyer (r000t@ligma.pro)'s status on Friday, 18-Apr-2025 22:55:00 JST Nietzschean Ekko Enjoyer Nietzschean Ekko Enjoyer
      in reply to
      • SuperDicq
      • Johnny Peligro

      @mischievoustomato
      Starfall makes hers pretend to be an exploding penguin from some weeb game and address her as the rule 63 protag from the same weeb game.

      The default agent that will likely ship is called "Standard" and is somewhat close to being deanthropomorphized

      Mine is the namesake of the platform itself and I can't wait to reveal him to everyone.
      @SuperDicq

      In conversation about 2 months ago permalink
      Johnny Peligro likes this.
    • Embed this notice
      pettanko (pettanko@varishangout.net)'s status on Friday, 18-Apr-2025 23:06:16 JST pettanko pettanko
      in reply to
      • SuperDicq
      • Johnny Peligro
      @mischievoustomato @r000t @SuperDicq
      In conversation about 2 months ago permalink

      Attachments


      1. https://varishangout.net/media/b78ff095-cb94-4d87-ab1d-93ea5ed70491/zased.gif
      Johnny Peligro likes this.
    • Embed this notice
      Johnny Peligro (mischievoustomato@tsundere.love)'s status on Friday, 18-Apr-2025 23:06:28 JST Johnny Peligro Johnny Peligro
      in reply to
      • SuperDicq
      • pettanko
      @pettanko @r000t @SuperDicq TFW no personal assistance tsundere gf
      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Saturday, 19-Apr-2025 21:43:33 JST SuperDicq SuperDicq
      in reply to
      • 御園はくい

      @hakui@tuusin.misono-ya.info @r000t@ligma.pro I think having to say "thank you" to a kiosk is like one step further from having the kiosk say "thank you".

      In conversation about 2 months ago permalink
      Nietzschean Ekko Enjoyer repeated this.
    • Embed this notice
      御園はくい (hakui@tuusin.misono-ya.info)'s status on Saturday, 19-Apr-2025 21:43:34 JST 御園はくい 御園はくい
      in reply to
      • SuperDicq
      @SuperDicq @r000t Yes
      images.jpg
      In conversation about 2 months ago permalink

      Attachments


      1. https://tuusin.misono-ya.info/media/bfceb3c19c7443f32e42d3fb0a61f0d844761e31a9fd9b18192ee5d4440b5b60.jpg?name=images.jpg
    • Embed this notice
      Light (light@noc.social)'s status on Saturday, 19-Apr-2025 23:27:47 JST Light Light
      in reply to
      • SuperDicq
      • Johnny Peligro

      @mischievoustomato @r000t @SuperDicq lmao

      In conversation about 2 months ago permalink
      Johnny Peligro likes this.
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Saturday, 19-Apr-2025 23:39:55 JST SuperDicq SuperDicq
      in reply to
      • ForcedInductionRetard

      @forcedinductionretard@pl.pube.tk Even though it is an object that brings me joy it is still an object and I will not speak to it. I will also not use he or she pronouns like some people do with cars.

      In conversation about 2 months ago permalink
    • Embed this notice
      ForcedInductionRetard (forcedinductionretard@pl.pube.tk)'s status on Saturday, 19-Apr-2025 23:39:58 JST ForcedInductionRetard ForcedInductionRetard
      in reply to
      • SuperDicq
      @SuperDicq >I will not say "thank you" to a fucking machine
      Not even to your Miata after a long trip?
      In conversation about 2 months ago permalink

      Attachments


      1. https://pl.pube.tk/media/58/9a/4f/589a4f8784a534f46919d4444a60797ff2ced207cfbaf7ac0a8f2ed9e57afa3a.jpg
    • Embed this notice
      Alexandre Oliva (lxo@snac.lx.oliva.nom.br)'s status on Sunday, 20-Apr-2025 07:50:05 JST Alexandre Oliva Alexandre Oliva
      in reply to
      • SuperDicq
      next thing, the machines will be asking for tips, like some self-service kiosks already do!

      CC: @r000t@ligma.pro
      In conversation about 2 months ago permalink
    • Embed this notice
      SuperDicq (superdicq@minidisc.tokyo)'s status on Sunday, 20-Apr-2025 22:46:48 JST SuperDicq SuperDicq
      in reply to
      • Alexandre Oliva

      @lxo@snac.lx.oliva.nom.br @r000t@ligma.pro I never tip. I'm not American.

      In conversation about 2 months ago permalink
    • Embed this notice
      御園はくい (hakui@tuusin.misono-ya.info)'s status on Monday, 21-Apr-2025 02:45:32 JST 御園はくい 御園はくい
      in reply to
      • SuperDicq
      @SuperDicq @r000t people dismissing slippery slope as a fallacy dismantled chesterson's fence by not understanding WHY slippery slope is a thing in the first place
      In conversation about 2 months ago permalink

Feeds

  • Activity Streams
  • RSS 2.0
  • Atom
  • Help
  • About
  • FAQ
  • TOS
  • Privacy
  • Source
  • Version
  • Contact

GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.

Creative Commons Attribution 3.0 All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.