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

    Neuronpedia

    INDEX
    Explanations

    function returns or errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     daba
    1.21
     καὶ
    1.09
     yep
    1.07
     guess
    1.02
     mingle
    1.00
     Và
    0.99
     deje
    0.99
     haga
    0.97
     dejar
    0.97
     Mick
    0.97
    POSITIVE LOGITS
    シャフト
    1.13
    钯
    1.12
    基準
    1.09
    HOL
    1.06
    酉
    1.06
    髙
    1.05
    翅
    1.02
    罪
    1.00
    屡
    0.99
    烃
    0.99
    Act Density 0.000%

    No Known Activations