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

    Neuronpedia

    INDEX
    Explanations

    question mark, quotation, parenthesis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ét
    0.46
    ᓱ
    0.44
    Distrib
    0.43
     consol
    0.42
    ልቅ
    0.42
     المقسوم
    0.42
    Isn
    0.41
    toISO
    0.41
    Saxon
    0.41
    دأ
    0.41
    POSITIVE LOGITS
    с
    0.49
     embrace
    0.42
     Boo
    0.42
    "></
    0.42
    '][
    0.41
     bounce
    0.41
     n
    0.41
    =[];
    0.40
     up
    0.40
    erville
    0.40
    Act Density 0.001%

    No Known Activations