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

    Neuronpedia

    INDEX
    Explanations

    checking for equality or membership

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    จะไม่
    1.14
     não
    1.11
    não
    1.10
     ikke
    1.03
     не
    1.02
    Tidak
    1.01
    Não
    1.01
     nicht
    1.00
     niet
    0.99
    ไม่
    0.98
    POSITIVE LOGITS
     equals
    1.20
     equal
    1.17
     Equal
    1.06
     resembles
    1.03
     iguais
    1.03
     Equals
    1.03
     equivalent
    1.00
     égales
    0.99
     gelijk
    0.98
     equivalente
    0.97
    Act Density 0.063%

    No Known Activations