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

    Neuronpedia

    INDEX
    Explanations

    email composition and content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ة
    0.50
    ?
    0.43
    átu
    0.41
     '
    0.40
    _
    0.39
     be
    0.38
    ă
    0.38
    riet
    0.38
    য়ের
    0.37
    '
    0.37
    POSITIVE LOGITS
    x
    0.39
     botas
    0.38
    仕様
    0.36
     bales
    0.36
    idencia
    0.36
     পারস্প
    0.36
     airbags
    0.35
    ಪ
    0.35
     airfoil
    0.34
     buffs
    0.34
    Act Density 0.013%

    No Known Activations