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

    Neuronpedia

    INDEX
    Explanations

    impressed by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     moistur
    1.16
    )
    1.16
    า
    1.16
     It
    1.13
    ır
    1.07
     are
    1.05
    но
    1.05
     a
    1.02
    )،
    1.02
    та
    1.02
    POSITIVE LOGITS
    ד
    1.02
    GRect
    0.98
    جين
    0.96
    Arena
    0.92
    Impress
    0.90
    지
    0.88
    에
    0.85
    IS
    0.83
    ارو
    0.83
    ungsk
    0.83
    Act Density 0.009%

    No Known Activations