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

    Neuronpedia

    INDEX
    Explanations

    capital cities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Bavarian
    0.81
     Cornish
    0.79
     Scottish
    0.77
     Padres
    0.73
     Pons
    0.71
     Catalan
    0.70
     scottish
    0.68
     Polish
    0.68
     Heinz
    0.67
     Cheshire
    0.67
    POSITIVE LOGITS
     Abuja
    1.91
     Beijing
    1.77
     Jakarta
    1.76
     Islamabad
    1.76
    Beijing
    1.71
     Canberra
    1.69
     Nairobi
    1.69
     Bangkok
    1.69
    Jakarta
    1.68
     Kathmandu
    1.68
    Act Density 0.063%

    No Known Activations