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

    Neuronpedia

    INDEX
    Explanations

    code and calculations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     daba
    -0.10
    .gen
    -0.09
    laten
    -0.08
     Genes
    -0.08
     nihil
    -0.08
     HAR
    -0.08
    中奖了
    -0.08
     vela
    -0.08
    �
    -0.08
    'offre
    -0.08
    POSITIVE LOGITS
    Expected
    0.12
    expected
    0.11
     esperado
    0.10
     Expected
    0.10
     expected
    0.10
     ожида
    0.10
    _expected
    0.10
    EXPECTED
    0.09
     기대
    0.09
    (expected
    0.09
    Act Density 0.014%

    No Known Activations