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

    Neuronpedia

    INDEX
    Explanations

    weld width

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Welding
    0.70
     welding
    0.62
     свар
    0.56
    焊接
    0.54
     welder
    0.48
    銅
    0.48
    铜
    0.46
    鋁
    0.46
     welds
    0.46
     এয়ার
    0.45
    POSITIVE LOGITS
     widened
    0.43
     narrows
    0.42
    Width
    0.41
    PARENT
    0.40
     width
    0.38
     narrow
    0.38
     narrowed
    0.38
     worm
    0.38
     widen
    0.38
     broadened
    0.37
    Act Density 0.011%

    No Known Activations