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

    Neuronpedia

    INDEX
    Explanations

    `is key`, `each password`, `dimensioni superiori`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    im
    0.52
     be
    0.52
    of
    0.50
     the
    0.48
    ou
    0.46
    Free
    0.46
     about
    0.45
    i
    0.45
     yang
    0.45
     of
    0.45
    POSITIVE LOGITS
    اتے
    0.54
    ⴰ
    0.50
     Grèce
    0.49
     ELEMENTS
    0.49
    🥢
    0.48
    טל
    0.48
    噺
    0.47
    erkraut
    0.47
    沬
    0.47
    Ⲥ
    0.47
    Act Density 0.002%

    No Known Activations