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

    Neuronpedia

    INDEX
    Explanations

    attention mechanism

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    🚲
    0.47
    चाकी
    0.46
    谦
    0.44
    손
    0.43
     پیپلز
    0.43
    욀
    0.43
    بيب
    0.42
    ehicle
    0.42
     propiet
    0.42
    PROPERTY
    0.42
    POSITIVE LOGITS
    logits
    0.64
     processed
    0.59
     spatial
    0.52
     Tensor
    0.51
     logits
    0.50
     spatially
    0.49
     processing
    0.48
     punctuation
    0.48
     Linguistics
    0.48
     antennas
    0.47
    Act Density 0.188%

    No Known Activations