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

    Neuronpedia

    INDEX
    Explanations

    entropy calculation and loss

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ول
    1.63
    ament
    1.58
    ames
    1.42
    ések
    1.42
    out
    1.39
    ن
    1.39
    isasi
    1.36
    ok
    1.34
    ির
    1.34
    про
    1.32
    POSITIVE LOGITS
    ਰ
    1.39
    ル
    1.34
     Contents
    1.26
    ])))
    1.26
    루
    1.24
     phonon
    1.23
    㐄
    1.23
     Complexity
    1.19
     ViewBag
    1.19
     noirâtre
    1.18
    Act Density 0.045%

    No Known Activations