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

    Neuronpedia

    INDEX
    Explanations

    network protocols and parameters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reliable
    0.47
    Reli
    0.47
     reliability
    0.43
     reliably
    0.40
    reliability
    0.40
     Reliability
    0.38
     EXCHANGE
    0.38
    可靠
    0.38
    请求
    0.37
     peer
    0.37
    POSITIVE LOGITS
    ttl
    0.55
     IP
    0.48
     紙
    0.44
     マル
    0.44
     IC
    0.43
     DSC
    0.42
    ihl
    0.41
    IP
    0.41
    ิล
    0.41
     ttl
    0.40
    Act Density 0.021%

    No Known Activations