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

    Neuronpedia

    INDEX
    Explanations

    objective, factual, reality, standards

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ের
    1.87
    いが
    1.80
    ानंतर
    1.77
    s
    1.59
    uv
    1.52
    いた
    1.48
    ibility
    1.43
    祸
    1.43
    旯
    1.41
    ería
    1.40
    POSITIVE LOGITS
    duğ
    1.60
    cir
    1.52
    ㈢
    1.48
    qt
    1.46
    рии
    1.45
    적으로
    1.45
    ك
    1.44
    ks
    1.41
    ст
    1.38
    the
    1.35
    Act Density 0.090%

    No Known Activations