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

    Neuronpedia

    INDEX
    Explanations

    sensitive topics and issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Entropy
    0.43
     malos
    0.43
    friendly
    0.41
     useless
    0.41
     autores
    0.39
     friendly
    0.38
     painters
    0.38
    purpose
    0.37
     machining
    0.37
     बदमाशों
    0.37
    POSITIVE LOGITS
     sensitive
    1.40
     Sensitive
    1.20
    Sensitive
    1.20
     controversial
    1.20
    敏感
    1.16
    sensitive
    1.15
     sensitively
    1.09
     حساس
    1.09
     topic
    1.08
     delicate
    1.07
    Act Density 0.019%

    No Known Activations