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

    Neuronpedia

    INDEX
    Explanations

    bring to life

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hh
    1.66
     exagger
    1.52
    hhhh
    1.47
    ki
    1.45
    ak
    1.41
    cyl
    1.41
    om
    1.40
    charging
    1.40
    taining
    1.37
     insignificant
    1.37
    POSITIVE LOGITS
    ität
    1.93
     Análisis
    1.73
     consigo
    1.62
     thường
    1.59
     Vý
    1.59
     pháp
    1.53
    이
    1.51
    ጷ
    1.48
     découvert
    1.45
     présente
    1.45
    Act Density 0.020%

    No Known Activations