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 is sensitive to cardinal number words (like “two,” “five,” “three,” “hundred,” etc.) indicating quantities or counts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     QDateTime
    -0.07
    ための
    -0.06
     deceased
    -0.06
    poi
    -0.06
    _proto
    -0.06
    rowned
    -0.06
    unctuation
    -0.06
     motel
    -0.06
    flo
    -0.06
     compos
    -0.06
    POSITIVE LOGITS
    winner
    0.07
    rike
    0.06
     representation
    0.06
    arking
    0.06
    (TAG
    0.06
     Alic
    0.06
     handleClose
    0.06
     embassy
    0.06
     heute
    0.06
     exceptions
    0.06
    Act Density 0.043%

    No Known Activations