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

    Neuronpedia

    INDEX
    Explanations

    useRef hook in JavaScript/React

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    as
    1.52
    ли
    1.26
    است
    1.02
    and
    1.00
     Siddhartha
    1.00
    0.97
    es
    0.96
    $,
    0.96
     from
    0.94
    non
    0.94
    POSITIVE LOGITS
    मधील
    1.14
    sgál
    1.12
    ਟ
    1.12
     épaisse
    1.11
    gramModel
    1.08
    睄
    1.05
    geschwindigkeit
    1.04
    ặc
    1.00
    هایی
    1.00
    ра
    0.97
    Act Density 0.001%

    No Known Activations