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

    Neuronpedia

    INDEX
    Explanations

    there is/are statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lateribus
    0.47
     simplesmente
    0.41
     суме
    0.41
    推广
    0.39
    事实上
    0.39
    гант
    0.39
     simplemente
    0.38
     rapidamente
    0.38
     пала
    0.37
     настоящий
    0.37
    POSITIVE LOGITS
     exist
    0.59
     exists
    0.59
     two
    0.55
     finns
    0.52
    exist
    0.51
     Existence
    0.49
     certain
    0.49
     terdapat
    0.48
    が存在
    0.48
     Exist
    0.48
    Act Density 0.005%

    No Known Activations