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

    Neuronpedia

    INDEX
    Explanations

    four, complication, violate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     clase
    0.52
    티
    0.51
     direitos
    0.50
     higiene
    0.46
     jurisdict
    0.45
     terceros
    0.45
     cocon
    0.45
     chào
    0.45
     inexist
    0.45
     derechos
    0.45
    POSITIVE LOGITS
    を使
    0.46
    ukuran
    0.45
    size
    0.43
     ومد
    0.43
     حدی
    0.41
    を使って
    0.41
     معين
    0.40
    mim
    0.39
    使
    0.39
    ED
    0.38
    Act Density 0.002%

    No Known Activations