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

    Neuronpedia

    INDEX
    Explanations

    early

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     olduk
    1.33
     gönd
    1.25
     fugit
    1.23
     zwią
    1.23
    भूति
    1.19
    ্রেট
    1.19
    茠
    1.19
    可能是
    1.16
     withstood
    1.16
     sasan
    1.16
    POSITIVE LOGITS
    و
    1.97
    о
    1.57
    ies
    1.56
    ו
    1.52
    ung
    1.45
    ר
    1.44
    ing
    1.42
    er
    1.41
    os
    1.41
    і
    1.41
    Act Density 0.013%

    No Known Activations