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

    Neuronpedia

    INDEX
    Explanations

    abbreviations and specific terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    á
    0.95
     are
    0.77
    ker
    0.73
     be
    0.70
    ow
    0.70
    werk
    0.70
     for
    0.69
    ich
    0.68
    ana
    0.68
    werte
    0.66
    POSITIVE LOGITS
    ل
    0.94
    К
    0.94
     motores
    0.88
    在
    0.86
     ограни
    0.80
    ન
    0.80
     empêcher
    0.79
    ル
    0.78
     cultivés
    0.75
    ش
    0.75
    Act Density 0.839%

    No Known Activations