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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     calça
    -1.03
    tênis
    -0.95
    ૮
    -0.94
     mascota
    -0.93
    ícola
    -0.93
     moldura
    -0.91
    litten
    -0.91
    וּ
    -0.91
    пасибо
    -0.90
    ctar
    -0.90
    POSITIVE LOGITS
     and
    1.56
     to
    1.20
     one
    0.99
     on
    0.99
     with
    0.98
    irah
    0.87
     without
    0.84
    h
    0.83
    setHas
    0.83
    B
    0.83
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.