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

    Neuronpedia

    INDEX
    Explanations

    Planck, builds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     jeopard
    0.41
    ية
    0.40
    কে
    0.34
     comprise
    0.34
     and
    0.33
     doua
    0.33
    يا
    0.32
     മറ്റു
    0.32
    the
    0.32
     ಸಮಸ್ಯೆ
    0.32
    POSITIVE LOGITS
    :
    0.50
    ьте
    0.32
    ار
    0.31
    бами
    0.31
    .
    0.30
    you
    0.30
    uterine
    0.29
     aches
    0.29
    ول
    0.28
     علاوه
    0.28
    Act Density 0.132%

    No Known Activations