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

    Neuronpedia

    INDEX
    Explanations

    Placement or column

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     placement
    3.66
     Placement
    3.38
     placements
    3.34
    Placement
    3.22
     placed
    3.11
    placement
    2.89
    placed
    2.64
     placing
    2.55
     Placing
    2.31
    Placed
    2.30
    POSITIVE LOGITS
    Col
    1.01
     Col
    0.98
     col
    0.92
     ge
    0.88
     pen
    0.78
     COL
    0.76
     Pen
    0.74
     ánimo
    0.74
    Pen
    0.73
    Раз
    0.70
    Act Density 0.076%

    No Known Activations