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

    Neuronpedia

    INDEX
    Explanations

    comparisons

    np_max-act · gemini-2.0-flash

    The neuron activates on comparative and superlative cues (words signaling comparison, e.g. “more,” “than,” “superior”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     era
    -0.07
    .amount
    -0.06
     mixer
    -0.06
    ritch
    -0.06
     Bowling
    -0.06
    (E
    -0.06
     choir
    -0.06
     آینده
    -0.06
     процесу
    -0.06
     opción
    -0.06
    POSITIVE LOGITS
     Hedge
    0.07
    _peng
    0.07
    .JpaRepository
    0.06
    ＿＿＿
    0.06
    CppMethod
    0.06
     denies
    0.06
    (delegate
    0.06
    (figsize
    0.06
     NSMutableArray
    0.06
    _AUTO
    0.06
    Act Density 0.225%

    No Known Activations