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

    Neuronpedia

    INDEX
    Explanations

    privileged peers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hogs
    0.42
    馅
    0.36
     Höhen
    0.35
     locu
    0.35
     mus
    0.35
     cmd
    0.34
     absor
    0.34
     Cannon
    0.34
     ಗೊ
    0.34
     삽
    0.34
    POSITIVE LOGITS
     peer
    4.31
    Peer
    4.03
     Peer
    3.92
    peer
    3.91
     peers
    3.58
    peers
    3.28
     Peers
    3.28
     peered
    2.14
     peering
    1.91
    Pe
    1.82
    Act Density 0.038%

    No Known Activations