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

    Neuronpedia

    INDEX
    Explanations

    vans and cars

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens and quantities (e.g. digits, numbers, and numeric measures).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Bed
    -0.06
     proph
    -0.06
    .sw
    -0.06
    .Toolkit
    -0.06
     college
    -0.06
     salad
    -0.06
     Ultimately
    -0.06
     College
    -0.05
    imals
    -0.05
    essel
    -0.05
    POSITIVE LOGITS
    ií
    0.07
    .flex
    0.07
    %;↵
    0.07
     <$
    0.07
    350
    0.07
    >{$
    0.07
     GLfloat
    0.07
    ’à
    0.07
    ==(
    0.07
     　
    0.07
    Act Density 0.005%

    No Known Activations