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

    Neuronpedia

    INDEX
    Explanations

    division calculations and results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     составляет
    0.43
    role
    0.41
    compose
    0.40
    rho
    0.38
    ੍ਰ
    0.38
     Hez
    0.38
    tych
    0.37
    romagnet
    0.37
    ારો
    0.36
     ER
    0.35
    POSITIVE LOGITS
    佘
    0.39
    ….
    0.38
     Quil
    0.37
     सूरत
    0.37
    Buffalo
    0.36
     iterations
    0.36
    Antaeotricha
    0.35
    ⁽
    0.35
    ,…
    0.35
    .…
    0.35
    Act Density 0.007%

    No Known Activations