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

    Neuronpedia

    INDEX
    Explanations

    foundational and original AI

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     magnification
    0.44
     currentLocation
    0.40
     currentState
    0.40
     current
    0.37
     glazing
    0.37
     확대
    0.37
     enhance
    0.36
     glazed
    0.36
     splashing
    0.36
     sponge
    0.36
    POSITIVE LOGITS
    original
    0.58
     original
    0.58
     originale
    0.56
     oryginal
    0.54
     ориги
    0.53
     оригина
    0.53
     originales
    0.52
    originally
    0.51
     orijinal
    0.51
     originally
    0.50
    Act Density 0.089%

    No Known Activations