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

    Neuronpedia

    INDEX
    Explanations

    maintaining, prioritizing, strictly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Michael
    0.53
     Archibald
    0.53
    Michael
    0.52
    Eas
    0.50
    Buck
    0.47
     William
    0.46
    hem
    0.46
     Edward
    0.45
     James
    0.45
     Robert
    0.44
    POSITIVE LOGITS
     kệ
    0.47
    dl
    0.46
     ServicePolicy
    0.45
    陌生
    0.45
    .'</
    0.45
    ༦
    0.45
    ндай
    0.44
     ete
    0.44
    ieuses
    0.44
    GTEST
    0.44
    Act Density 0.002%

    No Known Activations