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

    Neuronpedia

    INDEX
    Explanations

    instances where the described quantity or level is overly high or beyond acceptable limits

    oai_token-act-pair · gpt-4.1-nano
    New Auto-Interp
    Negative Logits
     wireType
    -0.59
    NegativeButton
    -0.57
    FXMLLoader
    -0.56
     Pair
    -0.55
    SupportActionBar
    -0.55
     AppCompatTheme
    -0.52
    inflater
    -0.52
     rootReducer
    -0.52
    Personendaten
    -0.51
    inut
    -0.50
    POSITIVE LOGITS
     excessive
    2.27
     Excessive
    2.27
    Excessive
    2.13
    cessive
    1.45
     excessively
    1.30
     exces
    0.99
     exager
    0.79
     exag
    0.74
     overly
    0.72
     exaggerated
    0.71
    Act Density 0.002%

    No Known Activations