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

    Neuronpedia

    INDEX
    Explanations

    function and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ᓄ
    0.44
    μέν
    0.43
    езд
    0.41
     renforc
    0.41
     家
    0.41
     woven
    0.40
     contribu
    0.40
     lance
    0.40
    ői
    0.40
     contrô
    0.40
    POSITIVE LOGITS
     Accurate
    0.45
     Everyday
    0.44
     NEVER
    0.41
     HUOBI
    0.39
    oulli
    0.38
     DAILY
    0.38
     Burnu
    0.37
     ONLY
    0.37
     WHY
    0.37
    adilla
    0.37
    Act Density 6.199%

    No Known Activations