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

    Neuronpedia

    INDEX
    Explanations

    commands and structured text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wooded
    0.48
     در
    0.42
     Milford
    0.41
     bittern
    0.41
     Coates
    0.40
     Middleton
    0.40
     bales
    0.40
     Lords
    0.39
    terbury
    0.39
     Breton
    0.39
    POSITIVE LOGITS
    ｧ
    0.47
     출력
    0.46
    This
    0.44
    ｱ
    0.44
     Metadata
    0.44
    데이터
    0.44
    があるので
    0.43
    코드
    0.43
    ８
    0.43
    Datos
    0.42
    Act Density 0.126%

    No Known Activations