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

    Neuronpedia

    INDEX
    Explanations

    word endings and foreign words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ار
    0.44
    宻
    0.43
     الاسم
    0.42
     LANA
    0.41
     తయారు
    0.41
     انکار
    0.39
     هذ
    0.38
     اسمه
    0.38
     résultats
    0.38
    後の
    0.38
    POSITIVE LOGITS
    .
    0.56
    berg
    0.54
    umb
    0.53
    cek
    0.52
    ។
    0.50
    ante
    0.46
    rit
    0.46
    ndi
    0.46
    ümü
    0.46
    omb
    0.46
    Act Density 0.006%

    No Known Activations