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

    Neuronpedia

    INDEX
    Explanations

    ridiculous/absurd

    np_max-act · gemini-2.0-flash

    The neuron fires on words or morphemes that convey absurdity or ridicule (e.g., “ridiculous,” “absurdity,” “illogical”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Hao
    -0.06
    Fre
    -0.06
     verileri
    -0.06
    ynchronization
    -0.06
    Inner
    -0.06
     memories
    -0.06
    位置
    -0.06
     Dmit
    -0.06
     connection
    -0.06
     movers
    -0.06
    POSITIVE LOGITS
     absurd
    0.10
     ridiculous
    0.10
     silly
    0.10
     -(
    0.08
     ludicrous
    0.08
     ridiculously
    0.08
     підпис
    0.08
    оятель
    0.07
     ridicule
    0.07
    LAG
    0.07
    Act Density 0.011%

    No Known Activations