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

    Neuronpedia

    INDEX
    Explanations

    react component code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <bos>
    1.24
    യാ
    1.14
    ين
    1.09
    पूर्ण
    1.08
    စ
    1.04
    ket
    0.98
     собственного
    0.97
    atenated
    0.95
     createContext
    0.94
    suff
    0.94
    POSITIVE LOGITS
    ように
    1.27
    $-$,
    1.24
    不然
    1.23
    冫
    1.22
    1.17
    ائيل
    1.17
     solns
    1.11
    ປະກ
    1.10
    Ƹ
    1.09
    ┕
    1.09
    Act Density 0.003%

    No Known Activations