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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     their
    -2.11
     for
    -2.11
     with
    -1.87
     at
    -1.81
     from
    -1.74
     only
    -1.69
     using
    -1.64
     most
    -1.52
     even
    -1.51
     suitable
    -1.49
    POSITIVE LOGITS
    really
    1.94
     kinda
    1.88
    actually
    1.79
     avaient
    1.68
     goofy
    1.66
     olika
    1.66
     REALLY
    1.63
     gigantic
    1.62
     sogenannten
    1.59
    lında
    1.54
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.