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

    Neuronpedia

    INDEX
    Explanations

    can't go wrong

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Зак
    0.40
     लाहिड़ी
    0.39
     token
    0.38
     necessariamente
    0.37
    束
    0.37
     عفوا
    0.37
    饶
    0.37
     দিলাম
    0.36
    闭
    0.36
    حية
    0.36
    POSITIVE LOGITS
     wrong
    0.60
     Wrong
    0.55
     Bad
    0.55
    Wrong
    0.52
    Bad
    0.51
     bad
    0.50
    wrong
    0.50
     resisted
    0.50
     싫
    0.49
     BAD
    0.49
    Act Density 0.010%

    No Known Activations