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

    Neuronpedia

    INDEX
    Explanations

    lists, math, and non-Latin scripts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ষিত
    0.67
                                   
    0.66
    axx
    0.66
    {\'
    0.66
    0.61
     تمام
    0.61
    ᚨ
    0.61
    <unused1989>
    0.59
    ారా
    0.59
    <unused309>
    0.58
    POSITIVE LOGITS
    Ბ
    1.25
     “,
    1.23
    Ს
    1.16
    🇧
    1.13
    Გ
    1.08
    Შ
    1.08
    “,
    1.06
    Დ
    1.05
     “.
    1.05
    Ე
    1.04
    Act Density 0.120%

    No Known Activations