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

    Neuronpedia

    INDEX
    Explanations

    towards

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     برای
    0.54
     dla
    0.52
    برای
    0.52
     براي
    0.49
     для
    0.47
     עבור
    0.47
     জন্য
    0.47
    für
    0.46
    க்காக
    0.45
    Чтобы
    0.45
    POSITIVE LOGITS
     zur
    2.59
     towards
    1.79
     към
    1.79
    Zur
    1.79
    zur
    1.78
     Zur
    1.75
     Towards
    1.60
    towards
    1.55
    到
    1.54
     toward
    1.45
    Act Density 0.001%

    No Known Activations