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

    Neuronpedia

    INDEX
    Explanations

    negative opinions/situations

    np_max-act · gemini-2.0-flash

    This neuron activates on Dutch tokens expressing awareness (particularly the word “bewust” and the phrase “bewust van”), i.e. references to being conscious of something.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    can
    -0.07
     Fifty
    -0.06
    Can
    -0.06
    nak
    -0.06
     ";
↵
    -0.06
     consists
    -0.06
    _female
    -0.06
    _DEF
    -0.06
     nuit
    -0.06
    interop
    -0.06
    POSITIVE LOGITS
    践
    0.06
     compose
    0.06
    니다
    0.06
    опол
    0.06
     tote
    0.06
     другими
    0.06
     notion
    0.06
     thang
    0.06
     magnitude
    0.06
     Chef
    0.06
    Act Density 0.239%

    No Known Activations