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

    Neuronpedia

    INDEX
    Explanations

    momentum or edge maps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    职责
    0.42
     spieg
    0.42
    责任
    0.39
     棉
    0.38
     เศษ
    0.38
    柇
    0.38
     कां
    0.38
     саме
    0.38
     відповіда
    0.37
     упро
    0.37
    POSITIVE LOGITS
    γρα
    0.39
    lina
    0.37
    Set
    0.36
     Rück
    0.36
    nicima
    0.36
    ]}
    0.35
     بن
    0.35
    路線
    0.35
    US
    0.35
    Objective
    0.35
    Act Density 0.000%

    No Known Activations