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

    Neuronpedia

    INDEX
    Explanations

    how to use

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ü
    2.23
     necessari
    2.08
    selves
    1.91
    ঠ
    1.88
    ту
    1.84
    pping
    1.78
    и
    1.78
    gleichen
    1.66
     pernah
    1.56
    auw
    1.55
    POSITIVE LOGITS
    ів
    2.11
    ным
    2.02
    ق
    1.91
    ください
    1.87
    ர்
    1.85
    ية
    1.84
    寿命
    1.75
    軛
    1.75
     sacrament
    1.70
     возможностей
    1.69
    Act Density 0.608%

    No Known Activations