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

    Neuronpedia

    INDEX
    Explanations

    or you

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Yankees
    0.47
     indistinct
    0.45
     Rita
    0.43
     ඇත
    0.43
    uyor
    0.43
     vietnam
    0.43
     yank
    0.43
     dukkh
    0.42
     katth
    0.42
    orney
    0.41
    POSITIVE LOGITS
     которую
    0.38
    ‐
    0.38
    ום
    0.38
    ᄀ
    0.38
    ᕈ
    0.38
     فِي
    0.37
    場合
    0.37
    َا
    0.36
    對
    0.36
    Guided
    0.36
    Act Density 0.003%

    No Known Activations