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

    Neuronpedia

    INDEX
    Explanations

    uvicorn run

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﺢ
    0.73
    alertDialog
    0.65
    ح
    0.58
    ES
    0.56
    uttered
    0.54
     else
    0.54
     caule
    0.54
    大切
    0.52
    分
    0.52
     smashed
    0.52
    POSITIVE LOGITS
    pä
    0.67
    ೇತ್ರ
    0.60
    в
    0.60
    نید
    0.59
     обязательно
    0.59
    ين
    0.58
    спользу
    0.58
    laws
    0.58
    ++];
    0.57
     AQ
    0.57
    Act Density 0.003%

    No Known Activations