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

    Neuronpedia

    INDEX
    Explanations

    numerical data

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens—especially measurements or statistical figures like percentages, decimal numbers, and quantities.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	w
    -0.07
    Seg
    -0.07
    複
    -0.06
     submissive
    -0.06
    postId
    -0.06
    	A
    -0.06
    elp
    -0.06
    �
    -0.06
     �
    -0.06
    paper
    -0.06
    POSITIVE LOGITS
     japan
    0.08
    (metrics
    0.07
     banc
    0.07
    ,ev
    0.07
    TRUE
    0.07
    	ev
    0.07
     Ngày
    0.07
     проблемы
    0.06
     inputs
    0.06
     russe
    0.06
    Act Density 0.035%

    No Known Activations