• ProfessorProteus@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    1 year ago

    What a cool idea! I have to try it on my comp later when I have time. Question about the “rules”: spaces where appropriate, or no?

    • tal
      link
      fedilink
      English
      arrow-up
      13
      ·
      edit-2
      1 year ago

      First without space, second with space. Based on my single-username dataset, not sure it makes a huge difference on at least some engines. I’m guessing that if an existing term doesn’t tokenize, one way or another, substrings get used.

      ProfessorProteus

      Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 0, Size: 1024x1024, Model hash: ebf42d1fae, Model: realmixXL_v15, Token merging ratio: 0.5, Version: v1.5.1

      Professor Proteus

      Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 0, Size: 1024x1024, Model hash: ebf42d1fae, Model: realmixXL_v15, Token merging ratio: 0.5, Version: v1.5.1

      • tal
        link
        fedilink
        English
        arrow-up
        6
        ·
        edit-2
        1 year ago

        Professor_Piddles

        Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 0, Size: 1024x1024, Model hash: ebf42d1fae, Model: realmixXL_v15, Token merging ratio: 0.5, Version: v1.5.1

        Maybe now a shot of you two together?

        Professor_Piddles and ProfessorProteus

        Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 0, Size: 1024x1024, Model hash: ebf42d1fae, Model: realmixXL_v15, Token merging ratio: 0.5, Version: v1.5.1

    • Krank Star@lemmy.worldOP
      link
      fedilink
      English
      arrow-up
      5
      ·
      1 year ago

      I do allow myself to add spaces

      1000011857

      Professor Proteus Steps: 20, Sampler: DPM++ 3M SDE Karras, CFG scale: 7, Seed: 2195257202, Size: 768x1280, Model hash: 74dda471cc, Model: realvisxlV20_v20Bakedvae, Clip skip: 2, RNG: CPU, Version: v1.6.0