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

    Neuronpedia

    INDEX
    Explanations

    bell curve distribution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cía
    1.55
    ak
    1.55
    ah
    1.55
    ال
    1.54
    kube
    1.52
    জ
    1.46
    ap
    1.45
    ttes
    1.45
    ्ञ
    1.42
    ر
    1.41
    POSITIVE LOGITS
     النها
    1.33
    ों
    1.30
    েরও
    1.26
     upbringing
    1.24
    ের
    1.21
     الحال
    1.20
     ngưỡng
    1.18
    ς
    1.18
    е
    1.17
    ‌تر
    1.16
    Act Density 0.001%

    No Known Activations