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

    Neuronpedia

    INDEX
    Explanations

    initialize or setup

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     volv
    0.72
    溶
    0.70
     tilbage
    0.69
     tredje
    0.68
     tilbake
    0.68
     contemplative
    0.68
     vuelto
    0.66
     volvió
    0.66
     voltou
    0.65
    꿩
    0.65
    POSITIVE LOGITS
     initialize
    3.26
     initializing
    3.11
     Initialize
    3.11
     initialization
    3.10
     initializes
    3.04
    初始化
    3.02
    Initialize
    2.94
     Initialization
    2.88
    initialize
    2.88
     初始化
    2.77
    Act Density 0.762%

    No Known Activations