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

    Neuronpedia

    INDEX
    Explanations

    Rabbi

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that are parts of Jewish religious titles (e.g. “Rabbi,” “Rav,” “Rebbe,” “Posek”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _cats
    -0.07
     Hüs
    -0.07
    :NSLayout
    -0.06
    .STATE
    -0.06
    Gesture
    -0.06
    AllWindows
    -0.06
    	view
    -0.06
    享
    -0.06
    ("--------------------------------
    -0.06
    ={()=>
    -0.06
    POSITIVE LOGITS
     Rabbi
    0.08
     laboratories
    0.07
     abi
    0.07
    abi
    0.07
    ab
    0.07
     Prime
    0.07
    py
    0.07
    ummings
    0.07
     Jab
    0.07
    wr
    0.07
    Act Density 0.001%

    No Known Activations