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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     advis
    -0.71
     volunt
    -0.70
    phia
    -0.69
     Kron
    -0.64
     scen
    -0.63
     alike
    -0.63
     Classification
    -0.63
     beware
    -0.63
     cardinal
    -0.62
    itness
    -0.62
    POSITIVE LOGITS
    kill
    0.76
    ih
    0.71
    ayed
    0.71
    afe
    0.70
    ragon
    0.70
    aha
    0.70
    borgh
    0.70
    paying
    0.68
    weet
    0.67
    func
    0.66
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.