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

    Neuronpedia

    INDEX
    Explanations

    exactness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ;$
    0.71
    +](=
    0.68
    基本的に
    0.65
    кон
    0.65
    ες
    0.64
    ጅም
    0.63
    ود
    0.63
    )[
    0.63
     العام
    0.62
     extravagant
    0.62
    POSITIVE LOGITS
    4
    0.79
    ↵
    0.77
    2
    0.77
     dokładnie
    0.75
    ne
    0.75
     정확
    0.74
    Accuracy
    0.74
    Exact
    0.73
    exactly
    0.71
     genau
    0.68
    Act Density 0.141%

    No Known Activations