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

    Neuronpedia

    INDEX
    Explanations

    digital and food

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bal
    0.61
     B
    0.59
     bra
    0.55
    Shelf
    0.54
     Bal
    0.52
     кла
    0.52
     Shelf
    0.52
     বালা
    0.51
    ﮕ
    0.51
     Cl
    0.50
    POSITIVE LOGITS
     uitle
    0.61
     वाली
    0.60
    вшейся
    0.59
     محاضره
    0.58
     falsch
    0.58
     voorbeeld
    0.57
    붐
    0.57
     pause
    0.57
     demonstration
    0.57
    ారీ
    0.56
    Act Density 0.001%

    No Known Activations