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

    Neuronpedia

    INDEX
    Explanations

    commercial applications and products

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    า
    1.42
    س
    1.38
    .
    1.19
    с
    1.09
    a
    1.08
    RO
    1.05
    IN
    1.04
     vigente
    1.02
     to
    1.00
    া
    1.00
    POSITIVE LOGITS
    t
    1.77
    <0x80>
    1.33
    ۹
    1.13
    tä
    1.13
    m
    1.13
    ０
    1.11
    í
    1.05
    را
    1.03
    ጐ
    1.03
    tions
    1.02
    Act Density 0.005%

    No Known Activations