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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron activates on code identifier tokens, especially PascalCase names and annotations (e.g. class, method, and attribute names).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (GTK
    -0.07
    (domain
    -0.07
    oggles
    -0.06
    	image
    -0.06
     유저
    -0.06
     pulses
    -0.06
     fic
    -0.06
    (original
    -0.06
    (org
    -0.06
     endeavor
    -0.06
    POSITIVE LOGITS
    长
    0.07
     Balanced
    0.06
    ATORS
    0.06
     LIFE
    0.06
     dangerous
    0.06
     lao
    0.06
    Creators
    0.06
    §
    0.06
     araya
    0.06
     Joint
    0.06
    Act Density 0.036%

    No Known Activations