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

    Neuronpedia

    INDEX
    Explanations

    60

    np_max-act · gemini-2.0-flash

    The neuron never fires on any of the shown tokens—it’s looking for some other marker or token pattern (not present here) and therefore does not activate on any of these excerpts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Francisco
    -0.08
    Curr
    -0.07
    ヴィ
    -0.07
     Briggs
    -0.07
    ادت
    -0.07
    operation
    -0.07
     Gordon
    -0.06
    .GetInstance
    -0.06
     tự
    -0.06
    Christ
    -0.06
    POSITIVE LOGITS
    	mkdir
    0.07
     []↵
    0.07
     fread
    0.06
    bestos
    0.06
     ГО
    0.06
    =↵
    0.06
     GSM
    0.06
     rund
    0.06
    0.06
     �
    0.06
    Act Density 0.003%

    No Known Activations