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

    Neuronpedia

    INDEX
    Explanations

    fibonacci logarithmic elements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     резулта
    0.44
     основы
    0.43
     rezult
    0.43
    ņu
    0.42
    Rg
    0.40
    Morse
    0.40
    Исход
    0.39
     Morse
    0.39
     altru
    0.39
    樎
    0.39
    POSITIVE LOGITS
     logs
    0.45
     Serialization
    0.41
     aix
    0.40
     Logs
    0.39
    éléments
    0.39
     éléments
    0.39
     Lit
    0.38
     logarithms
    0.37
     serialization
    0.37
     elements
    0.37
    Act Density 0.001%

    No Known Activations