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

    Neuronpedia

    INDEX
    Explanations

    non-latin characters and sequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ください
    0.81
     dotycz
    0.78
    i
    0.75
     sorte
    0.67
     Kudos
    0.65
     plupart
    0.65
    Neces
    0.63
    march
    0.63
    Neben
    0.62
    jenigen
    0.62
    POSITIVE LOGITS
    𝐞
    1.06
    𝐢
    0.92
    𝐲
    0.90
    𝐝
    0.87
    𝐥
    0.85
    𝐬
    0.83
    𝐫
    0.78
    𝐡
    0.78
    ित
    0.75
     الَّذ
    0.75
    Act Density 0.428%

    No Known Activations