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

    Neuronpedia

    INDEX
    Explanations

    protrusion

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of words describing protruding parts—e.g. “protruding,” “protrusion,” “protrude.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -make
    -0.06
    _snap
    -0.06
    Scheduler
    -0.06
     armies
    -0.06
    	ep
    -0.06
     lưu
    -0.06
     teamed
    -0.06
    ضم
    -0.06
    ามารถ
    -0.06
    _company
    -0.06
    POSITIVE LOGITS
     protr
    0.11
     bul
    0.09
     Cran
    0.07
    ��
    0.07
     cran
    0.07
     Bul
    0.07
    �이
    0.07
    rtle
    0.07
     recurring
    0.07
    _hat
    0.06
    Act Density 0.005%

    No Known Activations