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

    Neuronpedia

    INDEX
    Explanations

    Comparing differences

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _clip
    -0.09
    Clip
    -0.09
    gar
    -0.08
     transp
    -0.08
    Dragon
    -0.07
    Nag
    -0.07
     eignet
    -0.07
    bomb
    -0.07
     подключ
    -0.07
    gyl
    -0.07
    POSITIVE LOGITS
     Unterschiede
    0.12
     diferenças
    0.12
     differences
    0.12
    比较
    0.12
     différences
    0.11
    区别
    0.11
     comparar
    0.11
     groepen
    0.10
     sexes
    0.10
     Comparing
    0.10
    Act Density 0.026%

    No Known Activations