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

    Neuronpedia

    INDEX
    Explanations

    code, memory addresses

    np_max-act · gemini-2.0-flash

    The neuron fires on long sequences of hexadecimal digits (often with “+” separators) as used in cheat‐code or memory‐address strings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	done
    -0.07
    (lp
    -0.07
     MDMA
    -0.07
    	ip
    -0.06
    @@
    -0.06
    мест
    -0.06
    	window
    -0.06
    收录
    -0.06
     onc
    -0.06
    血
    -0.06
    POSITIVE LOGITS
     spíše
    0.07
     chant
    0.06
         
    0.06
    ACIÓN
    0.06
    活動
    0.06
    }catch
    0.06
    ाहत
    0.06
     política
    0.06
     EE
    0.05
     underst
    0.05
    Act Density 0.007%

    No Known Activations