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

    Neuronpedia

    INDEX
    Explanations

    a bank

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     only
    0.89
     and
    0.87
     equivalent
    0.86
     그
    0.85
     to
    0.82
     enough
    0.80
    only
    0.78
     및
    0.78
     that
    0.77
     এবং
    0.77
    POSITIVE LOGITS
     восприя
    1.06
     mujer
    1.00
    领域
    0.99
     género
    0.98
     sociedade
    0.97
     sueño
    0.96
     общество
    0.94
    mathcal
    0.94
     percepción
    0.93
     noite
    0.93
    Act Density 0.840%

    No Known Activations