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

    Neuronpedia

    INDEX
    Explanations

    Equation solving

    np_max-act · gemini-2.0-flash

    This neuron detects numeric multiplication terms involving single-letter variables (e.g., “*x”, “*y”, “*z”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     grac
    -0.07
     ander
    -0.07
     cham
    -0.06
     mesel
    -0.06
     trivia
    -0.06
     istem
    -0.06
     yaklaş
    -0.06
    Gün
    -0.06
     across
    -0.06
     собою
    -0.06
    POSITIVE LOGITS
    ":"","
    0.07
    _ED
    0.07
     TITLE
    0.06
     preaching
    0.06
    :before
    0.06
    Maker
    0.06
     storytelling
    0.06
     зуп
    0.06
    [__
    0.06
    \Component
    0.06
    Act Density 0.009%

    No Known Activations