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

    Neuronpedia

    INDEX
    Explanations

    documentation markers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     αλλά
    -2.42
    тися
    -2.39
    嫆
    -2.23
    та
    -2.17
    閹
    -2.17
    ва
    -2.14
    ܞ
    -2.13
     WITH
    -2.11
    varlak
    -2.11
    ρα
    -2.09
    POSITIVE LOGITS
    휀
    2.98
     was
    2.95
    毉
    2.67
    sandalia
    2.63
    蚡
    2.52
    ープン
    2.45
    甞
    2.42
    赑
    2.41
    ðsíða
    2.41
    ୮
    2.38
    Act Density 0.068%

    No Known Activations