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

    Neuronpedia

    INDEX
    Explanations

    fast chat

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cards
    1.04
     Dung
    1.01
     Tamil
    0.99
     Card
    0.99
    Dart
    0.99
     detoxification
    0.98
    蚌
    0.98
     Airport
    0.98
    school
    0.98
     Airports
    0.98
    POSITIVE LOGITS
     soc
    0.69
     ress
    0.66
     incentives
    0.63
     sociaux
    0.63
    teddy
    0.61
     Reeves
    0.61
     lus
    0.60
     mirror
    0.59
    鹿
    0.58
    jap
    0.58
    Act Density 0.960%

    No Known Activations