Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     redesigned
    -0.06
     autor
    -0.06
     manipulating
    -0.06
    _keyword
    -0.06
     contrib
    -0.06
    drawer
    -0.06
     meetup
    -0.06
    地区
    -0.06
     Butter
    -0.06
     مال
    -0.05
    POSITIVE LOGITS
     ↵            ↵
    0.07
     blur
    0.07
    _tac
    0.07
    -cols
    0.06
    /&
    0.06
    脸
    0.06
    614
    0.06
     oat
    0.06
    lif
    0.06
     爱
    0.06
    Act Density 0.013%

    No Known Activations