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

    Neuronpedia

    INDEX
    Explanations

    user actions and tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     akár
    0.63
    либо
    0.57
     अथवा
    0.54
     иных
    0.53
     quelconque
    0.51
     skillfully
    0.50
     또는
    0.48
    यदि
    0.46
     belirli
    0.46
     혹은
    0.46
    POSITIVE LOGITS
     myself
    1.08
    我自己
    0.79
     خودم
    0.77
     really
    0.71
    几个
    0.67
    幾個
    0.64
    但是我
    0.63
     my
    0.60
     pretty
    0.59
    really
    0.58
    Act Density 0.061%

    No Known Activations