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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of “walk-in closet” (references to spacious built-in closet storage).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    crafted
    -0.07
     stern
    -0.06
    家伙
    -0.06
    ゆ
    -0.06
    ちは
    -0.06
     Böylece
    -0.06
     embod
    -0.06
    세대
    -0.06
     Uttar
    -0.06
     freezes
    -0.06
    POSITIVE LOGITS
    кус
    0.07
     کند
    0.07
     Danny
    0.06
    -beta
    0.06
    manı
    0.06
     JOptionPane
    0.06
     아�
    0.06
     Slee
    0.06
    ="">↵
    0.06
     radical
    0.06
    Act Density 0.003%

    No Known Activations