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

    Neuronpedia

    INDEX
    Explanations

    usage change development

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Quindi
    0.58
     Puede
    0.56
     Más
    0.55
     Não
    0.53
     Porque
    0.52
     Até
    0.51
     Según
    0.50
     Sobre
    0.50
     Oggi
    0.48
     Hacer
    0.48
    POSITIVE LOGITS
     penggunaan
    0.62
     berbagai
    0.61
     perubahan
    0.56
     perilaku
    0.55
     lingkungan
    0.55
     peningkatan
    0.55
     beberapa
    0.54
     sejumlah
    0.54
     pengembangan
    0.53
     dukungan
    0.53
    Act Density 0.004%

    No Known Activations