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

    Neuronpedia

    INDEX
    Explanations

    ">"

    np_max-act · gemini-2.0-flash

    This neuron detects HTML metadata elements in the document’s head, especially `<meta>` tags (e.g. charset declarations).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    BEGIN
    -0.07
    liche
    -0.07
    vae
    -0.07
    ummer
    -0.06
     zaměř
    -0.06
    Rather
    -0.06
      
    -0.06
     Tob
    -0.06
    осп
    -0.06
     rightfully
    -0.06
    POSITIVE LOGITS
    рич
    0.07
     monstrous
    0.06
     ]);↵↵
    0.06
    作
    0.06
    ewriter
    0.06
    _gray
    0.06
     liner
    0.06
    -tests
    0.06
    	router
    0.06
     guarantees
    0.06
    Act Density 0.001%

    No Known Activations