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

    Neuronpedia

    INDEX
    Explanations

    standard deduction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cabrio
    1.19
     Paola
    0.96
     señal
    0.96
     Existen
    0.95
     renormal
    0.95
     adecu
    0.94
     omdat
    0.93
     Tamaño
    0.93
     DGP
    0.92
     Campos
    0.91
    POSITIVE LOGITS
    ко
    0.82
    ма
    0.78
    вести
    0.77
    orus
    0.77
    izace
    0.75
     речи
    0.74
    ch
    0.73
     needed
    0.72
    inator
    0.71
    вод
    0.71
    Act Density 0.001%

    No Known Activations