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

    Neuronpedia

    INDEX
    Explanations

    xml and html code declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Kf
    0.45
     、,
    0.43
    ())),
    0.43
     、
    0.43
    Mf
    0.43
    ('
    0.39
    fff
    0.39
     }),
    0.39
     hu
    0.38
     Bomb
    0.38
    POSITIVE LOGITS
     cientos
    0.40
    ього
    0.37
    FIGURE
    0.36
    اندان
    0.36
    "?
    0.36
    noch
    0.36
     прод
    0.35
    습니까
    0.35
    tall
    0.35
     پنج
    0.35
    Act Density 0.001%

    No Known Activations