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

    Neuronpedia

    INDEX
    Explanations

    address bar

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Barg
    0.48
     Barrow
    0.44
     triangles
    0.43
     circles
    0.42
     bottomRight
    0.41
     syringes
    0.41
     chessboard
    0.40
     squares
    0.39
     barrels
    0.39
     rectangles
    0.39
    POSITIVE LOGITS
     beam
    0.47
    vin
    0.43
     stem
    0.42
    beam
    0.42
    坚
    0.42
    velyan
    0.40
    Beam
    0.40
    Vin
    0.39
    BAL
    0.39
    各国
    0.39
    Act Density 0.005%

    No Known Activations