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

    Neuronpedia

    INDEX
    Explanations

    >

    np_max-act · gemini-2.0-flash

    The neuron responds to heading or title text in HTML (e.g. content inside `<title>`, `<h1>`, `<h2>`, nav links, or other section headers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     prz
    -0.07
     enormously
    -0.06
    Sexy
    -0.06
     vinc
    -0.06
     Qin
    -0.06
    954
    -0.06
    fx
    -0.06
     lions
    -0.06
    DX
    -0.06
     DX
    -0.06
    POSITIVE LOGITS
    Et
    0.08
    ett
    0.08
    et
    0.08
    >'
    0.08
    ret
    0.07
    set
    0.07
    	record
    0.07
    えた
    0.07
    aaa
    0.07
    ap
    0.07
    Act Density 0.030%

    No Known Activations