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

    Neuronpedia

    INDEX
    Explanations

    need to access

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     refocus
    0.68
     scoreboard
    0.67
     namin
    0.67
     maintain
    0.65
     KPIs
    0.65
     sayings
    0.64
     ales
    0.64
     reliever
    0.64
     midfielders
    0.64
     fade
    0.64
    POSITIVE LOGITS
     امکان
    0.82
     اولین
    0.82
     теорети
    0.80
    itherto
    0.74
    ¡
    0.74
    ถ้า
    0.69
    実際に
    0.68
    Forbidden
    0.68
     théor
    0.67
     попыта
    0.66
    Act Density 0.381%

    No Known Activations