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

    Neuronpedia

    INDEX
    Explanations

    layout directions

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric constants—especially floating-point literals (decimal numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cop
    -0.07
    .compare
    -0.07
     myth
    -0.06
     faire
    -0.06
     �
    -0.06
    _na
    -0.06
    -Mar
    -0.06
     بح
    -0.06
    кра
    -0.06
    pas
    -0.06
    POSITIVE LOGITS
     WebSocket
    0.07
    ел
    0.06
     allocator
    0.06
     bowling
    0.06
    (nullptr
    0.06
    .leadingAnchor
    0.06
    ADED
    0.06
    523
    0.06
    585
    0.06
    ｀ヽ
    0.06
    Act Density 0.002%

    No Known Activations