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

    Neuronpedia

    INDEX
    Explanations

    efficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    br
    0.50
    ian
    0.45
    il
    0.44
    ng
    0.43
    lag
    0.42
     alcohols
    0.42
     at
    0.41
     san
    0.40
    uds
    0.39
    onte
    0.39
    POSITIVE LOGITS
     skapa
    0.55
     creare
    0.55
     kullanarak
    0.51
     preparación
    0.51
    为什么要
    0.51
     fáj
    0.50
     vytvá
    0.49
     ermöglicht
    0.49
    を使って
    0.49
     gebruiken
    0.49
    Act Density 0.001%

    No Known Activations