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

    Neuronpedia

    INDEX
    Explanations

    definitely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ش
    1.71
    ج
    1.48
    ق
    1.30
    ्ड
    1.23
    น
    1.22
     unrecognizable
    1.17
     autoload
    1.17
    ال
    1.16
    朮
    1.16
    ."]
    1.13
    POSITIVE LOGITS
    äure
    1.63
    ör
    1.42
    Đây
    1.37
    uştur
    1.35
    Después
    1.34
    În
    1.33
     définit
    1.29
    ма
    1.28
    이다
    1.27
    unun
    1.27
    Act Density 0.299%

    No Known Activations