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

    Neuronpedia

    INDEX
    Explanations

    facial expressions

    np_max-act · gemini-2.0-flash

    The neuron consistently lights up on words that describe facial expressions or emotional demeanor (e.g. “stern,” “expression,” “face,” “serious,” “firm”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    illustr
    -0.07
    gpio
    -0.06
     sadness
    -0.06
    980
    -0.06
    bio
    -0.06
     pb
    -0.06
     breathtaking
    -0.06
    آمد
    -0.06
    vp
    -0.06
     clone
    -0.06
    POSITIVE LOGITS
     tế
    0.07
     Ceiling
    0.07
     meant
    0.07
     progen
    0.06
    _CARD
    0.06
     Harley
    0.06
    崇
    0.06
     Ocak
    0.06
    어서
    0.06
     시
    0.06
    Act Density 0.013%

    No Known Activations