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

    Neuronpedia

    INDEX
    Explanations

    Leo

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    PPE
    0.59
    小心
    0.56
    绰
    0.55
     trademark
    0.54
    ్వర
    0.53
    もら
    0.52
     Easton
    0.52
    islam
    0.51
    モリ
    0.51
     algod
    0.50
    POSITIVE LOGITS
     Leo
    0.85
    Leo
    0.75
     BLACK
    0.69
     Facts
    0.62
     Schema
    0.61
     Fact
    0.59
    BLACK
    0.58
     Cz
    0.58
     FACTS
    0.57
     Breaking
    0.57
    Act Density 0.001%

    No Known Activations