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

    Neuronpedia

    INDEX
    Explanations

    Questions and Answers

    np_max-act · gemini-2.0-flash

    The neuron activates on words and phrases that offer help or assistance (e.g., “assist,” “help,” “provide,” “information,” “helpful,” “informative”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Только
    -0.08
    ित
    -0.07
    есс
    -0.07
    edi
    -0.07
    athers
    -0.06
    елем
    -0.06
    oes
    -0.06
     nhẹ
    -0.06
    QPushButton
    -0.06
    伯
    -0.06
    POSITIVE LOGITS
     vật
    0.06
    Dir
    0.06
     pretrained
    0.06
     exported
    0.06
    authenticated
    0.06
    ับผ
    0.06
    egrator
    0.06
     sneak
    0.06
    .Globalization
    0.06
     faded
    0.06
    Act Density 0.058%

    No Known Activations