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

    Neuronpedia

    INDEX
    Explanations

    language prevention and protection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    可靠
    0.54
     સલા
    0.52
    emples
    0.50
    pessoa
    0.50
    に限
    0.49
    湋
    0.49
     Сте
    0.49
     бош
    0.49
    pairs
    0.48
    avnom
    0.48
    POSITIVE LOGITS
    呢
    0.48
     imag
    0.47
     didn
    0.47
     punished
    0.46
     }
    0.45
     let
    0.45
     me
    0.44
    innah
    0.43
     遊ん
    0.43
     ise
    0.43
    Act Density 0.001%

    No Known Activations