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

    Neuronpedia

    INDEX
    Explanations

    if statements and approximate values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     observ
    0.39
    しょ
    0.37
     denunci
    0.36
     بُک
    0.36
    formul
    0.35
     ormai
    0.35
    ạt
    0.35
     füh
    0.35
    այի
    0.35
     dél
    0.35
    POSITIVE LOGITS
     cropping
    0.52
    __:
    0.47
     cropped
    0.46
     Aspect
    0.46
    +:
    0.46
     Specify
    0.45
    切割
    0.44
    ):
    0.44
    圆形
    0.44
    AspectRatio
    0.43
    Act Density 0.000%

    No Known Activations