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

    Neuronpedia

    INDEX
    Explanations

    typing

    np_max-act · gemini-2.0-flash

    This neuron activates on GUI‐navigation and instruction tokens in technical how-to steps—words like “click,” “type,” “Start,” “Next,” and other interface labels.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    optgroup
    -0.07
    -looking
    -0.07
     lives
    -0.06
    _pedido
    -0.06
    ivé
    -0.06
    اول
    -0.06
    _cert
    -0.06
    -this
    -0.06
     mul
    -0.06
    .cap
    -0.06
    POSITIVE LOGITS
    Ά
    0.07
    .nextToken
    0.07
    .JOptionPane
    0.07
     Audrey
    0.07
    ibilidad
    0.06
    TOKEN
    0.06
    $content
    0.06
     Of
    0.06
    .News
    0.06
    .ReadToEnd
    0.06
    Act Density 0.010%

    No Known Activations