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

    Neuronpedia

    INDEX
    Explanations

    adverbial and adjectival suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    1.19
     I
    0.92
    -
    0.83
    AT
    0.79
    스는
    0.79
     (
    0.78
    רים
    0.76
     Prothorax
    0.74
    ot
    0.73
     Burgund
    0.73
    POSITIVE LOGITS
     and
    1.30
     be
    1.27
     in
    1.21
    σ
    1.16
    of
    1.09
    ง
    1.09
    ق
    1.06
    ة
    1.03
     of
    1.00
     were
    0.99
    Act Density 0.745%

    No Known Activations