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

    Neuronpedia

    INDEX
    Explanations

    critical for understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    t
    1.01
    I
    0.89
     and
    0.86
    g
    0.86
    q
    0.86
    h
    0.83
    p
    0.81
    o
    0.80
    v
    0.75
    n
    0.73
    POSITIVE LOGITS
    ции
    0.83
     mMediaPlayer
    0.82
     कोणत्याही
    0.78
     maximising
    0.78
     controllo
    0.77
     polinom
    0.77
     trabalho
    0.76
    ！！
    0.76
    是如何
    0.75
    在
    0.75
    Act Density 0.123%

    No Known Activations