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

    Neuronpedia

    INDEX
    Explanations

    certain conjunctions and descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cesa
    0.41
     innocence
    0.39
     mógł
    0.39
     pouvait
    0.38
     벡터
    0.38
    ரியா
    0.38
     optar
    0.38
    за
    0.37
    個人
    0.37
     천
    0.37
    POSITIVE LOGITS
     Firm
    0.38
    ackel
    0.38
     buttermilk
    0.37
    لە
    0.36
     Nā
    0.36
    卟
    0.36
    Subset
    0.36
     fæ
    0.35
     Abroad
    0.35
    Nickel
    0.35
    Act Density 0.000%

    No Known Activations