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

    Neuronpedia

    INDEX
    Explanations

    code and informal language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     гле
    -0.88
     silnika
    -0.88
     get
    -0.87
     meticulously
    -0.86
    zdroj
    -0.85
     enhances
    -0.84
     enhancing
    -0.84
    ֩
    -0.83
     improves
    -0.83
    ྜ
    -0.82
    POSITIVE LOGITS
     calibre
    0.88
     Crunchy
    0.85
     lae
    0.84
     人民
    0.84
     accla
    0.83
     rar
    0.82
     какого
    0.82
     latter
    0.82
     bleak
    0.82
     😁
    0.81
    Act Density 0.164%

    No Known Activations