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

    Neuronpedia

    INDEX
    Explanations

    nun, unfamiliar, martial, Eve

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    é
    1.64
    d
    1.62
     I
    1.38
     It
    1.38
    k
    1.34
    ä
    1.34
    se
    1.27
    t
    1.24
    ור
    1.20
     can
    1.20
    POSITIVE LOGITS
    ード
    1.13
    ム
    1.13
    ただし
    1.08
    ですが
    1.02
    ای
    1.01
    ал
    1.01
    $)$.
    1.00
    0
    0.99
    ⅾ
    0.99
    対処
    0.98
    Act Density 0.000%

    No Known Activations