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

    Neuronpedia

    INDEX
    Explanations

    variety or range of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     της
    1.59
    ofthe
    1.59
     della
    1.52
     του
    1.52
     dell
    1.50
     ofthe
    1.46
    OfThe
    1.45
     DELLA
    1.43
     delle
    1.35
     degli
    1.25
    POSITIVE LOGITS
     de
    1.02
    eye
    0.72
     gimm
    0.62
     d
    0.60
     gồm
    0.57
    બંધ
    0.56
    ্বাস
    0.56
     qualifies
    0.56
     متنوع
    0.55
     eye
    0.55
    Act Density 0.469%

    No Known Activations