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

    Neuronpedia

    INDEX
    Explanations

    encryption, cryptography, sequences, data handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kub
    0.84
     montaña
    0.83
     kob
    0.82
    вы
    0.82
     kav
    0.79
    にする
    0.79
     perj
    0.76
    юць
    0.76
    bru
    0.75
     الش
    0.75
    POSITIVE LOGITS
    ન
    0.76
     总
    0.72
    ര
    0.71
    oucher
    0.71
    یس
    0.70
    ላሉ
    0.70
    ंट
    0.69
    cient
    0.68
    edict
    0.68
     écrit
    0.67
    Act Density 0.001%

    No Known Activations