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

    Neuronpedia

    INDEX
    Explanations

    café con, Cafe Francais, cafes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ל
    1.22
     $
    1.07
     on
    1.03
    ব
    1.02
    ת
    0.91
    ä
    0.90
     has
    0.84
    א
    0.84
    ה
    0.83
    י
    0.79
    POSITIVE LOGITS
    𝘮
    1.15
    ado
    1.00
    ية
    0.99
    𝘢
    0.97
    𝘸
    0.96
    adays
    0.91
    мое
    0.91
    𝘰
    0.91
    ist
    0.90
    𝐦
    0.88
    Act Density 0.008%

    No Known Activations