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

    Neuronpedia

    INDEX
    Explanations

    Brackets

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens—especially digits and decimal numbers representing coefficients or matrix entries.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _START
    -0.07
     sauce
    -0.07
     buddies
    -0.07
    .floor
    -0.07
    IVERS
    -0.07
     EX
    -0.07
    ilerine
    -0.07
    ط
    -0.07
    ájem
    -0.07
    '})↵
    -0.07
    POSITIVE LOGITS
    。</
    0.06
     adorned
    0.06
     Founded
    0.06
     проф
    0.06
    _sched
    0.06
    .'
    0.06
    .IntPtr
    0.06
     overhead
    0.06
     Paramount
    0.05
    _]
    0.05
    Act Density 0.005%

    No Known Activations