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

    Neuronpedia

    INDEX
    Explanations

    scripting or specific words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    up
    1.76
    様子
    1.75
    '
    1.68
     schemes
    1.66
    ist
    1.65
     up
    1.64
    ائم
    1.63
    는
    1.63
     arşivlendi
    1.62
    ’
    1.61
    POSITIVE LOGITS
    SIDE
    1.82
    mono
    1.62
    ستگی
    1.60
    doors
    1.54
     Gosudarstvennyj
    1.53
    actifs
    1.50
     Nih
    1.49
    tmpl
    1.49
     Şimdi
    1.47
     عندك
    1.46
    Act Density 0.046%

    No Known Activations