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

    Neuronpedia

    INDEX
    Explanations

    complexity, difficulty, potential, check, ability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Telefon
    0.52
    נו
    0.48
     ભારત
    0.46
    ной
    0.46
    brahim
    0.46
     pembayaran
    0.46
    ма
    0.46
    ме
    0.46
     मायणी
    0.45
    Fonte
    0.44
    POSITIVE LOGITS
     negligible
    0.41
     vol
    0.39
     behaves
    0.38
     half
    0.38
     similarly
    0.36
     behaved
    0.35
     stabilizes
    0.35
     semidefinite
    0.34
     Similarly
    0.33
    aturated
    0.33
    Act Density 0.056%

    No Known Activations