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

    Neuronpedia

    INDEX
    Explanations

    South facing roofs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kilometre
    0.45
     kilometres
    0.44
    éreo
    0.43
    ières
    0.43
    urry
    0.43
    üsse
    0.42
    uç
    0.41
    верка
    0.41
     arrêté
    0.41
    single
    0.41
    POSITIVE LOGITS
    Wat
    0.41
     Wat
    0.40
     toward
    0.39
     לנו
    0.39
     slope
    0.38
    勺
    0.38
     양
    0.38
     Уи
    0.38
     웨
    0.37
    给
    0.37
    Act Density 0.003%

    No Known Activations