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

    Neuronpedia

    INDEX
    Explanations

    quoted words like country stats filename

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ।
    0.76
    ी
    0.73
    ט
    0.69
     of
    0.65
    ു
    0.64
     for
    0.64
     are
    0.64
    та
    0.64
    г
    0.63
    ")
    0.62
    POSITIVE LOGITS
    U
    0.48
    겅
    0.48
    deki
    0.48
     azienda
    0.48
    I
    0.47
    しくは
    0.47
    ﺒ
    0.46
     międzynarod
    0.45
    €™
    0.45
    ᅬ
    0.44
    Act Density 0.262%

    No Known Activations