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

    Neuronpedia

    INDEX
    Explanations

    at the level

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ATON
    -1.05
     kapag
    -1.02
     habang
    -0.98
     nė
    -0.97
    ATT
    -0.93
     buvo
    -0.91
     vė
    -0.90
    Pageable
    -0.90
     hvilken
    -0.90
     kasama
    -0.88
    POSITIVE LOGITS
     at
    5.88
    ระดับ
    2.14
     ở
    2.11
     level
    1.80
     nível
    1.73
    At
    1.70
     уровне
    1.65
     tingkat
    1.59
     tại
    1.55
     niveau
    1.54
    Act Density 0.181%

    No Known Activations