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

    Neuronpedia

    INDEX
    Explanations

    legal activity, bad endings, aluminum foil

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.79
     a
    0.68
    ，
    0.66
    。
    0.59
    ใน
    0.57
    ।
    0.53
     fertil
    0.50
    Pokud
    0.49
    you
    0.47
    いっぱい
    0.47
    POSITIVE LOGITS
    ik
    0.76
    o
    0.59
    and
    0.56
    us
    0.56
     had
    0.54
    '
    0.54
    ing
    0.54
    ud
    0.53
    ari
    0.53
    )
    0.51
    Act Density 3.530%

    No Known Activations