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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     on
    -1.59
     from
    -1.45
     first
    -1.44
     because
    -1.43
     for
    -1.33
     like
    -1.29
     has
    -1.28
     or
    -1.18
     only
    -1.13
     work
    -1.12
    POSITIVE LOGITS
    Ingredienti
    1.47
    بسم
    1.38
    Curios
    1.31
     attirer
    1.30
    Materiaal
    1.28
    relatively
    1.27
     proceder
    1.24
    Aç
    1.24
    Aantal
    1.23
     différents
    1.22
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.