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

    Neuronpedia

    INDEX
    Explanations

    numerical values

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens (digits and numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    yasal
    -0.07
    .bit
    -0.07
    용
    -0.07
     sotto
    -0.06
     Horror
    -0.06
    -mode
    -0.06
     robbery
    -0.06
    ora
    -0.06
    _TWO
    -0.06
     sneakers
    -0.06
    POSITIVE LOGITS
    (elem
    0.06
     라
    0.06
    	prev
    0.06
    	sign
    0.06
    	BufferedReader
    0.06
    _GEN
    0.06
    ,name
    0.06
     ль
    0.06
    amount
    0.06
    Ell
    0.05
    Act Density 0.066%

    No Known Activations