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

    Neuronpedia

    INDEX
    Explanations

    starts or begins

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    瞎
    0.49
     美術
    0.47
    immä
    0.47
    hack
    0.45
    yp
    0.44
    نس
    0.44
    醫師
    0.44
    ي
    0.44
    planet
    0.44
     Bilder
    0.44
    POSITIVE LOGITS
     mobilise
    0.52
     mobil
    0.50
     carat
    0.49
     extensible
    0.49
     hair
    0.47
     mica
    0.47
     sexism
    0.47
     footwear
    0.47
     respeito
    0.46
     mobilize
    0.46
    Act Density 0.000%

    No Known Activations