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

    Neuronpedia

    INDEX
    Explanations

    adversarial examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    挼
    0.50
    觉得
    0.47
    赉
    0.46
     दुनिया
    0.45
    会社の
    0.45
     Misalnya
    0.44
     stvari
    0.43
    来到
    0.43
    谊
    0.43
    谀
    0.43
    POSITIVE LOGITS
     geometric
    0.47
     aforementioned
    0.47
     three
    0.46
     hepatic
    0.46
     bacterial
    0.45
     fungal
    0.45
     porcine
    0.45
     $\
    0.45
     weakly
    0.44
     two
    0.44
    Act Density 0.021%

    No Known Activations