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

    Neuronpedia

    INDEX
    Explanations

    probability questions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     sings
    -0.07
    =default
    -0.07
    <button
    -0.07
     NSNumber
    -0.06
     Elekt
    -0.06
     hacked
    -0.06
    ुमत
    -0.06
    title
    -0.06
    titulo
    -0.06
    _span
    -0.06
    POSITIVE LOGITS
    ervoir
    0.07
     Famil
    0.06
    recursive
    0.06
     граж
    0.06
    -gay
    0.06
     dozens
    0.06
    　　　　　　　　　　　　
    0.06
    ังก
    0.06
    rending
    0.06
    뉴
    0.06
    Act Density 0.009%

    No Known Activations