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

    Neuronpedia

    INDEX
    Explanations

    performance optimization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     spirituality
    0.97
     제도
    0.96
    アレンジ
    0.94
     क्रांतिकार
    0.91
    介紹
    0.91
     обозна
    0.90
    指南
    0.88
     creencias
    0.88
     costruito
    0.87
     typology
    0.87
    POSITIVE LOGITS
     CPU
    1.93
    CPU
    1.75
     processing
    1.68
     throughput
    1.66
     latency
    1.55
     GPU
    1.55
     load
    1.53
     cpu
    1.52
     CPUs
    1.52
     slow
    1.47
    Act Density 1.215%

    No Known Activations