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

    Neuronpedia

    INDEX
    Explanations

    use

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     his
    -0.70
     the
    -0.69
     its
    -0.66
     a
    -0.64
    ψε
    -0.63
     an
    -0.63
     natural
    -0.61
    RunWith
    -0.60
    thold
    -0.59
    rola
    -0.58
    POSITIVE LOGITS
    TypedDataSet
    0.60
     vermelha
    0.54
     Wicidata
    0.53
     înainte
    0.52
     associées
    0.51
     kohta
    0.51
     olisi
    0.50
     gră
    0.49
    eraard
    0.49
     تضيفلها
    0.49
    Act Density 0.072%

    No Known Activations