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

    Neuronpedia

    INDEX
    Explanations

    you want to know

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     quits
    0.68
     reinforcements
    0.68
     these
    0.68
    y
    0.64
     sole
    0.64
     Anytime
    0.63
     roof
    0.63
     Never
    0.63
    });
    0.63
     discipline
    0.62
    POSITIVE LOGITS
    ﻜ
    0.85
     Сан
    0.82
    妿
    0.81
    Vì
    0.80
    ELECT
    0.79
    รู้จัก
    0.79
     أن
    0.77
    ERK
    0.77
    㛣
    0.76
    conoc
    0.76
    Act Density 0.000%

    No Known Activations