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

    Neuronpedia

    INDEX
    Explanations

    directional geographical regions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ﷺ
    0.44
     వివిధ
    0.42
     weltweit
    0.40
     personnelles
    0.40
    ၡ
    0.39
    खून
    0.38
     Wired
    0.38
     Ambedkar
    0.37
     Configure
    0.37
    ؤلاء
    0.37
    POSITIVE LOGITS
    most
    1.08
    MOST
    0.75
     hemisphere
    0.75
     regions
    0.68
     Hemisphere
    0.65
     portion
    0.64
     provinces
    0.64
     region
    0.62
     flank
    0.62
     part
    0.61
    Act Density 0.017%

    No Known Activations