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

    Neuronpedia

    INDEX
    Explanations

    boundary value testing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    wenza
    -0.09
     الإسب
    -0.09
    ైంది
    -0.09
    eenkomst
    -0.09
    oyi
    -0.08
     mesto
    -0.08
    播播
    -0.08
     Guia
    -0.08
     ddod
    -0.08
    ாவது
    -0.08
    POSITIVE LOGITS
     extremes
    0.12
     borderline
    0.10
     extreme
    0.10
     boundary
    0.09
     Extrem
    0.09
     boundaries
    0.09
     ungewöhn
    0.09
     unusual
    0.09
     edge
    0.09
     extrem
    0.08
    Act Density 0.002%

    No Known Activations