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

    Neuronpedia

    INDEX
    Explanations

    round

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of match round progress (e.g., “first round,” “second round,” “quarter-finals,” “semi-finals”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     trolls
    -0.07
    illo
    -0.07
    DFS
    -0.07
     Kazakhstan
    -0.07
    Aux
    -0.07
     nghiệ
    -0.06
    йом
    -0.06
    꾸
    -0.06
    .writer
    -0.06
    _three
    -0.06
    POSITIVE LOGITS
     //_
    0.06
    かに
    0.06
    
    0.06
    ág
    0.06
     loud
    0.06
     داده
    0.05
     THPT
    0.05
     verbal
    0.05
     principal
    0.05
    }$
    0.05
    Act Density 0.003%

    No Known Activations