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

    Neuronpedia

    INDEX
    Explanations

    mystery story / solving

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ם
    2.27
    ين
    2.00
    いた
    1.96
    ية
    1.93
    ج
    1.91
    ن
    1.90
    него
    1.84
    তে
    1.84
    вання
    1.84
    ف
    1.81
    POSITIVE LOGITS
    e
    1.86
    "\
    1.82
    cznym
    1.80
    dens
    1.71
    deme
    1.63
    ehle
    1.63
     memperhatikan
    1.61
    ays
    1.59
    czny
    1.59
    ම්භ
    1.59
    Act Density 0.014%

    No Known Activations