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

    Neuronpedia

    INDEX
    Explanations

    navy, white, cream, maroon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     charlas
    0.43
    ándo
    0.43
    ó
    0.41
    verständlich
    0.40
    setSelected
    0.40
    噙
    0.40
    σταν
    0.40
     chilled
    0.40
     merciful
    0.39
    otypic
    0.39
    POSITIVE LOGITS
    Luxury
    0.43
    马
    0.40
    ೫
    0.38
    Fox
    0.38
    Aut
    0.38
    亚
    0.37
     Vaugh
    0.37
    奢
    0.37
    よね
    0.36
    Ryan
    0.36
    Act Density 0.004%

    No Known Activations