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

    Neuronpedia

    INDEX
    Explanations

    fully connected layers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     mosquitoes
    -0.06
    -type
    -0.06
     Myth
    -0.06
     aging
    -0.06
    model
    -0.06
    　　 　
    -0.06
    stud
    -0.06
     controls
    -0.06
    Playable
    -0.06
     shaded
    -0.06
    POSITIVE LOGITS
     secretive
    0.07
    uti
    0.07
    ///
    0.06
     konuş
    0.06
    Capital
    0.06
     kl
    0.06
     تص
    0.06
     الد
    0.06
    $page
    0.06
     Continental
    0.06
    Act Density 0.003%

    No Known Activations