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

    Neuronpedia

    INDEX
    Explanations

    use for illegal or harmful purposes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    mime
    0.53
    while
    0.50
    isecond
    0.49
    neurons
    0.47
    weaver
    0.46
    xlim
    0.45
    Gal
    0.45
    sympy
    0.45
    arc
    0.44
    qw
    0.44
    POSITIVE LOGITS
    所の
    0.49
    리
    0.48
    чки
    0.47
    覺得
    0.47
    يز
    0.47
    気に入り
    0.47
    欵
    0.46
    年輕
    0.46
    тами
    0.45
    امه
    0.45
    Act Density 0.009%

    No Known Activations