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

    Neuronpedia

    INDEX
    Explanations

    element/variable names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    рі
    0.89
    і
    0.86
    Ă
    0.83
    ਟ
    0.82
    ve
    0.81
    do
    0.81
    ?
    0.79
    ücke
    0.79
    ?>
    0.78
    vi
    0.78
    POSITIVE LOGITS
     hvilket
    0.89
     ejempl
    0.86
     größ
    0.82
    ፍተኛ
    0.82
     treball
    0.81
     hvil
    0.81
    tedir
    0.80
    деги
    0.80
     یہ
    0.80
     welke
    0.80
    Act Density 0.222%

    No Known Activations