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

    Neuronpedia

    INDEX
    Explanations

    quiet, dark room

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     could
    1.16
     should
    1.12
     can
    1.10
     will
    1.09
     будут
    1.09
     начну
    1.07
     смогут
    1.07
     would
    1.07
     must
    1.03
     cannot
    1.02
    POSITIVE LOGITS
     বলেও
    1.09
    ।
    1.01
    !।
    0.99
    .].
    0.97
    $.;
    0.90
    .}$
    0.89
    .;
    0.88
     ।,
    0.88
    ।’
    0.87
    ۔
    0.87
    Act Density 0.143%

    No Known Activations