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

    Neuronpedia

    INDEX
    Explanations

    is/that

    np_max-act · gemini-2.0-flash

    This neuron activates on copular/linking words (especially “is,” “this,” “that,” “most,” “what”) in explanatory or definitional statements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     زیست
    -0.07
    공부
    -0.07
     محصول
    -0.06
    yme
    -0.06
    ακ
    -0.06
    خوان
    -0.06
    全部
    -0.06
    �어
    -0.06
    ème
    -0.06
     shows
    -0.06
    POSITIVE LOGITS
     "/"↵
    0.07
     \↵↵
    0.07
    Deserialize
    0.06
     Birch
    0.06
    );\
    0.06
    (Self
    0.06
     Lie
    0.06
    _sprites
    0.06
     sockfd
    0.06
     NotImplemented
    0.06
    Act Density 0.046%

    No Known Activations