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

    Neuronpedia

    INDEX
    Explanations

    arrested

    np_max-act · gemini-2.0-flash

    This neuron selectively responds to tokens in scientific experimental passages that convey quantitative or measured results—i.e. numbers, measurements, and their surrounding context.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     shown
    -0.06
    (intent
    -0.06
    Bob
    -0.06
     spirited
    -0.06
     epub
    -0.06
    -0.06
    ordered
    -0.06
    ISTRIBUT
    -0.06
    الإنجليزية
    -0.06
     copyrighted
    -0.06
    POSITIVE LOGITS
     amnesty
    0.07
     militants
    0.07
     &(
    0.07
     виріш
    0.07
    .dest
    0.07
     الطب
    0.07
     چرا
    0.07
     EOF
    0.07
     geography
    0.06
     Ül
    0.06
    Act Density 0.055%

    No Known Activations