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

    Neuronpedia

    INDEX
    Explanations

    regular expression characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    semin
    0.44
    thur
    0.43
    auf
    0.42
    ahr
    0.41
    se
    0.40
    ơ
    0.40
    rest
    0.40
    Selon
    0.40
    keun
    0.39
    seur
    0.38
    POSITIVE LOGITS
    Ą
    0.49
     ಪೊಲೀ
    0.46
     વડે
    0.46
     matched
    0.46
     ಮುಂದೆ
    0.46
    nonsense
    0.44
     જગ
    0.43
     दादा
    0.43
     ಅಧಿಕಾರಿ
    0.43
     ۶
    0.42
    Act Density 0.001%

    No Known Activations