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

    Neuronpedia

    INDEX
    Explanations

    number operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     négoci
    0.70
     infrastrukt
    0.66
    reuung
    0.63
     salarié
    0.62
     servizi
    0.61
     serviços
    0.59
    entreprise
    0.59
    Gosudarstvennyj
    0.59
    sponsor
    0.58
     autorités
    0.57
    POSITIVE LOGITS
     Multiplication
    0.79
     math
    0.77
     mathematical
    0.77
     Arithmetic
    0.76
     Subtract
    0.76
     matemáticas
    0.75
     Mathematics
    0.75
     mathematicians
    0.74
    Subtract
    0.73
     чисел
    0.72
    Act Density 0.299%

    No Known Activations