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

    Neuronpedia

    INDEX
    Explanations

    code, programming

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric tokens—especially decimal or floating‐point numbers (and related units like “days” or “weeks”)—marking where numeric measurements appear.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Pix
    -0.06
    unctuation
    -0.06
    悉
    -0.06
    牙
    -0.06
    cour
    -0.06
    	store
    -0.06
     ApiController
    -0.06
    átek
    -0.06
    руз
    -0.06
    гів
    -0.06
    POSITIVE LOGITS
     máximo
    0.07
     guards
    0.07
     branch
    0.07
     marzo
    0.07
     nl
    0.07
     Cri
    0.06
     mechanisms
    0.06
     softball
    0.06
    _fm
    0.06
     vs
    0.06
    Act Density 0.142%

    No Known Activations