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

    Neuronpedia

    INDEX
    Explanations

    programmed refusal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     glycer
    0.40
    霟
    0.38
     વૈ
    0.38
    紂
    0.37
    VK
    0.36
     Ghosh
    0.36
     ಕ್ಷೇತ್ರ
    0.36
    꺼
    0.36
     වෙ
    0.36
    Venezuela
    0.36
    POSITIVE LOGITS
     Song
    0.38
     Sono
    0.36
    Om
    0.36
     Bluff
    0.35
     augue
    0.34
     sün
    0.34
     unrelenting
    0.33
     Sadd
    0.33
    0.33
     readers
    0.32
    Act Density 0.020%

    No Known Activations