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

    Neuronpedia

    INDEX
    Explanations

    break; or break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0
    1.46
    1.41
     was
    1.40
    Type
    1.27
    <0x80>
    1.24
    ０
    1.20
    ש
    1.19
    The
    1.15
    Button
    1.15
    Date
    1.14
    POSITIVE LOGITS
    ní
    1.14
    kým
    0.99
    ють
    0.97
    kého
    0.95
    kosten
    0.94
    λώ
    0.94
     सर्वा
    0.92
    骤
    0.92
    ký
    0.91
    ્સ
    0.90
    Act Density 0.146%

    No Known Activations