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

    Neuronpedia

    INDEX
    Explanations

    json file structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    [['
    0.99
    форма
    0.86
    0.84
     absolument
    0.84
     стиля
    0.84
    ر
    0.84
    Hon
    0.83
    𝙠
    0.82
    sylvania
    0.81
    turbo
    0.81
    POSITIVE LOGITS
    veis
    0.88
    chmod
    0.86
    irstrip
    0.83
    $,
    0.82
    ൽ
    0.82
    ifferentiated
    0.80
     antagonism
    0.79
     convulsions
    0.79
     devotional
    0.79
    zzel
    0.77
    Act Density 0.017%

    No Known Activations