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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    The neuron activates on quantifier and scope words indicating broad or approximate coverage—e.g. “world,” “nearly,” “any,” “mix.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     SHR
    -0.07
    ฟ
    -0.07
     rehe
    -0.06
    softmax
    -0.06
     deposited
    -0.06
     Mature
    -0.06
    련
    -0.06
     Golf
    -0.06
    /create
    -0.06
    	mem
    -0.06
    POSITIVE LOGITS
     puedes
    0.07
     come
    0.06
    convert
    0.06
     NK
    0.06
     konk
    0.06
    íc
    0.06
     кредит
    0.06
     você
    0.06
     elő
    0.06
    鎮
    0.06
    Act Density 0.145%

    No Known Activations