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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens (counts, percentages, and other digit‐based figures).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    %x
    -0.07
    .JFrame
    -0.06
    Inventory
    -0.06
    Ex
    -0.06
    Sand
    -0.06
     oma
    -0.06
    _INPUT
    -0.06
    Input
    -0.06
    IN
    -0.06
     Bh
    -0.06
    POSITIVE LOGITS
     angi
    0.06
    pectrum
    0.06
    /features
    0.06
     tasting
    0.06
    醫
    0.06
    productId
    0.06
     téléphone
    0.06
    мот
    0.06
    πί
    0.06
    Exists
    0.06
    Act Density 0.030%

    No Known Activations