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

    Neuronpedia

    INDEX
    Explanations

    lists and examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     faced
    0.41
     courage
    0.38
    相位
    0.38
    리어
    0.37
    کس
    0.37
    ဏ်
    0.37
     côté
    0.37
     teammates
    0.36
     profissional
    0.36
     parámetro
    0.36
    POSITIVE LOGITS
     छा
    0.39
    !==
    0.38
    чити
    0.38
     матери
    0.38
     свя
    0.38
    掴
    0.38
     Attic
    0.38
    諾
    0.38
     зака
    0.37
    voices
    0.37
    Act Density 0.001%

    No Known Activations