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

    Neuronpedia

    INDEX
    Explanations

    text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     highlight
    0.83
     highlighting
    0.82
     republic
    0.80
     Republic
    0.80
     capitalizing
    0.78
     highlighted
    0.76
     uncompromising
    0.73
     nomination
    0.73
     enthusiasts
    0.71
     highlights
    0.71
    POSITIVE LOGITS
    spannung
    0.89
    esercizio
    0.88
     benutzer
    0.82
     cuentas
    0.82
     समांतर
    0.82
    cactus
    0.80
     cactus
    0.79
    функ
    0.79
     cacti
    0.79
     शुक्
    0.78
    Act Density 0.014%

    No Known Activations