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

    Neuronpedia

    INDEX
    Explanations

    unique

    np_max-act · gemini-2.0-flash

    This neuron activates on the adjective “unique” (and close variants), i.e., it spotlights the word “unique” describing something distinctive.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Sharper
    -0.07
    Pages
    -0.07
     Fi
    -0.07
    .gif
    -0.06
    carousel
    -0.06
    发布
    -0.06
     Hard
    -0.06
    Thumbnail
    -0.06
     dense
    -0.06
    态
    -0.06
    POSITIVE LOGITS
     diese
    0.06
     venture
    0.06
     liberalism
    0.06
    elim
    0.06
    argin
    0.06
    ennessee
    0.06
    _ORIENTATION
    0.06
     heels
    0.06
    izont
    0.06
     awaited
    0.06
    Act Density 0.035%

    No Known Activations