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

    Neuronpedia

    INDEX
    Explanations

    scope related words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     necessários
    1.72
    अधिकांश
    1.65
    सभी
    1.63
    谁
    1.62
    قال
    1.58
    ция
    1.56
     대로
    1.56
     quieren
    1.55
     walaupun
    1.55
     tetapi
    1.54
    POSITIVE LOGITS
    ।
    2.00
    ،
    1.80
     nieces
    1.79
     तहरीर
    1.78
    ců
    1.73
    Ре
    1.72
     choirs
    1.71
     protons
    1.70
    ו
    1.70
    tsc
    1.64
    Act Density 0.121%

    No Known Activations