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

    Neuronpedia

    INDEX
    Explanations

    leaked document

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     renders
    0.45
     rendered
    0.43
    導致
    0.39
    ustainability
    0.39
    ibt
    0.39
    igans
    0.39
     rendering
    0.38
     render
    0.38
    rendered
    0.38
    usay
    0.37
    POSITIVE LOGITS
    范围
    0.40
     acesso
    0.37
     příst
    0.37
     Tension
    0.37
     напря
    0.37
    сом
    0.36
    Modules
    0.36
     зво
    0.35
     Access
    0.35
     добав
    0.35
    Act Density 0.000%

    No Known Activations