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

    Neuronpedia

    INDEX
    Explanations

    AI assistant, ethics, transformer model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ong
    0.52
     اینڈ
    0.52
    ewnątrz
    0.52
     Coast
    0.52
    ib
    0.52
     αυτή
    0.51
    im
    0.50
    afety
    0.49
     கலை
    0.49
    ically
    0.48
    POSITIVE LOGITS
    用品
    0.59
     అంశ
    0.57
     נו
    0.52
     ಅಂಶ
    0.52
    氏
    0.52
     極
    0.51
    諾
    0.50
     시작
    0.50
    끼
    0.50
     모습
    0.50
    Act Density 0.364%

    No Known Activations