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

    Neuronpedia

    INDEX
    Explanations

    existence

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the word “exist” (and its morphological variants), flagging mentions of something’s existence.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     bezier
    -0.06
    ίλ
    -0.06
     dry
    -0.06
     closes
    -0.06
    Pag
    -0.06
     FOX
    -0.06
     initiated
    -0.06
     เม
    -0.06
    _enum
    -0.06
     ifade
    -0.06
    POSITIVE LOGITS
    .Exists
    0.08
     genu
    0.07
     necessity
    0.07
     existed
    0.07
     καθ
    0.07
     розвит
    0.06
    はい
    0.06
    회
    0.06
    ioned
    0.06
    _available
    0.06
    Act Density 0.061%

    No Known Activations