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

    Neuronpedia

    INDEX
    Explanations

    create a question

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     многие
    0.87
     mucha
    0.82
     mínimo
    0.79
    quent
    0.75
     такую
    0.74
     власть
    0.72
    finement
    0.71
     такой
    0.70
    sqrt
    0.70
     തു
    0.69
    POSITIVE LOGITS
    ে
    0.63
    रूम
    0.60
     właśnie
    0.60
    াদেশ
    0.59
     downstream
    0.58
     redesigned
    0.57
    اؤ
    0.57
     precisamente
    0.57
    Rus
    0.56
    opsis
    0.56
    Act Density 0.000%

    No Known Activations