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

    Neuronpedia

    INDEX
    Explanations

    choosing between alternatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ни
    0.68
     AdWords
    0.64
     В
    0.63
     controllo
    0.63
    culus
    0.62
     Filme
    0.61
    miers
    0.61
    沍
    0.60
     നെറ്റ്‌വർ
    0.59
     Viola
    0.59
    POSITIVE LOGITS
     funeral
    0.65
    </h2>
    0.62
    ारा
    0.61
    </u>
    0.59
    <0x80>
    0.59
     bencana
    0.58
    ﺭ
    0.58
     bowl
    0.57
     بیش
    0.54
     nausea
    0.54
    Act Density 0.007%

    No Known Activations