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

    Neuronpedia

    INDEX
    Explanations

    rust public declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    1.41
    is
    1.16
    s
    1.15
    aa
    1.05
    é
    1.03
    ä
    1.02
    the
    0.99
    ა
    0.97
    b
    0.91
    aad
    0.86
    POSITIVE LOGITS
     pubs
    1.29
    ك
    0.98
     rougeâtres
    0.92
    I
    0.91
     pub
    0.90
     ότι
    0.86
    kę
    0.86
    pub
    0.86
    ン
    0.84
    ،
    0.84
    Act Density 0.002%

    No Known Activations