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

    Neuronpedia

    INDEX
    Explanations

    code constructs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    лег
    0.47
    ochond
    0.43
     филосо
    0.41
    പര
    0.40
    ార
    0.39
     строго
    0.39
     szolg
    0.39
    ثم
    0.39
     сосу
    0.38
    баров
    0.38
    POSITIVE LOGITS
     बाबू
    0.40
     if
    0.40
    if
    0.39
     abelian
    0.39
     payback
    0.37
    ，
    0.37
    ,“
    0.36
    ఎల్
    0.36
    нский
    0.36
     Abelian
    0.36
    Act Density 0.002%

    No Known Activations