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

    Neuronpedia

    INDEX
    Explanations

    graph neural networksSuperGlue represents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    atic
    0.54
    ih
    0.52
    ns
    0.50
    riving
    0.50
     Fitness
    0.50
     műkö
    0.50
    inal
    0.49
    adh
    0.48
    adjusted
    0.48
     compromet
    0.48
    POSITIVE LOGITS
    чества
    0.46
    лей
    0.46
     knives
    0.45
     disks
    0.44
    板
    0.44
    ཝ
    0.43
     beled
    0.43
    לות
    0.42
    硗
    0.42
     perils
    0.42
    Act Density 0.000%

    No Known Activations