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

    Neuronpedia

    INDEX
    Explanations

    storage capacity (GB)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ק
    0.86
    С
    0.82
    Х
    0.81
    К
    0.79
    Т
    0.77
    ﺭ
    0.75
    ﺮ
    0.75
    Ра
    0.72
    ح
    0.71
    ﺯ
    0.70
    POSITIVE LOGITS
    ascimento
    0.63
     ́
    0.58
     heirloom
    0.56
     felicit
    0.55
     Pasal
    0.55
    </h3>
    0.55
     caramelized
    0.55
     weave
    0.54
     آغاز
    0.54
     '.')
    0.54
    Act Density 0.012%

    No Known Activations