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

    Neuronpedia

    INDEX
    Explanations

    arguments in code descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    bots
    0.45
     retorn
    0.44
    %,
    0.40
    transformer
    0.40
     смысла
    0.40
    ༧
    0.40
    Input
    0.40
     косми
    0.40
    د
    0.39
    ທີ່
    0.39
    POSITIVE LOGITS
     park
    0.47
    "":
    0.42
     FIRST
    0.42
     tailgate
    0.42
     '-':
    0.42
     LOT
    0.41
     വരുന്ന
    0.41
     parking
    0.40
    park
    0.40
    umet
    0.40
    Act Density 0.024%

    No Known Activations