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

    Neuronpedia

    INDEX
    Explanations

    indicating prior state or action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     oś
    0.43
     Zet
    0.40
    어
    0.40
    етра
    0.40
    a
    0.39
     tanıt
    0.39
    inação
    0.38
     spoj
    0.38
     frapp
    0.37
     ricorda
    0.37
    POSITIVE LOGITS
     pre
    1.33
    Pre
    1.22
     Pre
    1.18
    pre
    1.17
     प्री
    1.09
     пре
    1.04
    预
    0.90
     preorder
    0.80
     preamp
    0.78
     pré
    0.77
    Act Density 0.052%

    No Known Activations