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

    Neuronpedia

    INDEX
    Explanations

    $

    np_max-act · gemini-2.0-flash

    The neuron is triggered by numeric literals—particularly decimal numbers (e.g. “4.5078125,” “0.6953125”) often appearing in math‐mode contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     SCAN
    -0.07
    iesen
    -0.07
     toaster
    -0.07
    戒
    -0.07
     Bing
    -0.06
     VS
    -0.06
     alarm
    -0.06
     Scratch
    -0.06
     mensaje
    -0.06
    obutton
    -0.06
    POSITIVE LOGITS
    	sprintf
    0.07
    .Script
    0.07
    earned
    0.06
    toInt
    0.06
    hopefully
    0.06
    अ
    0.06
    .amazonaws
    0.06
     refund
    0.06
    as
    0.06
    .closest
    0.06
    Act Density 0.002%

    No Known Activations