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

    Neuronpedia

    INDEX
    Explanations

    between entities or groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     önce
    1.03
     vilket
    1.03
    𝗯
    1.02
    minent
    1.00
    ри
    0.95
     arrivare
    0.94
     zumindest
    0.93
     letzt
    0.92
     subjected
    0.88
    Ни
    0.88
    POSITIVE LOGITS
     وبين
    2.75
     sexes
    1.35
    هایت
    1.28
    יים
    1.09
     genders
    1.08
    峨
    1.05
     الاقوامی
    1.03
     hemispheres
    1.02
     showers
    1.02
    ندگان
    1.02
    Act Density 0.092%

    No Known Activations