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

    Neuronpedia

    INDEX
    Explanations

    strings and string manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ية
    1.14
    {
    1.03
     to
    1.02
    }{
    0.95
    dden
    0.89
    '
    0.89
    ging
    0.88
    ;
    0.88
    <
    0.88
    ء
    0.87
    POSITIVE LOGITS
    ור
    1.55
    il
    1.28
     string
    1.28
    s
    1.16
    ות
    1.13
    1
    1.12
    ્
    1.06
    াই
    1.05
     getString
    1.03
    یم
    1.02
    Act Density 0.116%

    No Known Activations