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

    Neuronpedia

    INDEX
    Explanations

    totals and sums

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    蛳
    0.39
     Ezekiel
    0.38
     الصور
    0.36
    乒
    0.36
    恻
    0.36
    鹛
    0.36
    沙发
    0.36
    ArchiveAction
    0.36
     Футбол
    0.35
     последних
    0.35
    POSITIVE LOGITS
     total
    0.51
    total
    0.50
     combined
    0.49
     combine
    0.46
    combined
    0.46
    合計
    0.45
     combiner
    0.44
     combining
    0.44
     комби
    0.43
     BOTH
    0.42
    Act Density 0.023%

    No Known Activations