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

    Neuronpedia

    INDEX
    Explanations

    Rel

    np_max-act · gemini-2.0-flash

    The neuron consistently activates on occurrences of the word “relativity” (and its equivalents, e.g. 相對 in Chinese), flagging mentions of Einstein’s relativity theory.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    리아
    -0.08
     abbreviation
    -0.07
    iards
    -0.07
    (ad
    -0.06
    IMA
    -0.06
     раза
    -0.06
    標準
    -0.06
     abrupt
    -0.06
    iations
    -0.06
    _az
    -0.06
    POSITIVE LOGITS
     transferring
    0.07
    ��
    0.07
     Eu
    0.06
    nost
    0.06
     Kit
    0.06
    TCP
    0.06
     anarch
    0.06
    =$((
    0.06
    ným
    0.06
     Lyn
    0.06
    Act Density 0.009%

    No Known Activations