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

    Neuronpedia

    INDEX
    Explanations

    and then

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     використовувати
    0.52
     використову
    0.50
    groups
    0.49
     рисо
    0.49
    continues
    0.49
     जानवर
    0.48
    一個人
    0.48
     ఉపయోగ
    0.48
     ઉપયોગ
    0.48
    这个人
    0.48
    POSITIVE LOGITS
     to
    0.51
     on
    0.45
     conflict
    0.45
     Route
    0.43
     Wow
    0.43
     Structural
    0.43
     religion
    0.42
     en
    0.42
     Sesame
    0.41
     Perspekt
    0.41
    Act Density 0.003%

    No Known Activations