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

    Neuronpedia

    INDEX
    Explanations

    test methods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    つまり
    0.63
    an
    0.62
    Narr
    0.58
    바로
    0.57
    すなわち
    0.56
    kart
    0.54
    Earnings
    0.54
    रिक्
    0.54
    roscopic
    0.53
    ei
    0.53
    POSITIVE LOGITS
    idades
    0.66
     concentric
    0.63
     fadeIn
    0.63
    ܗ
    0.63
     Kirkland
    0.61
    主义
    0.60
     је
    0.60
     pang
    0.57
    ب
    0.57
    стью
    0.56
    Act Density 0.005%

    No Known Activations