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

    Neuronpedia

    INDEX
    Explanations

    =

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric literals—especially decimal numbers—present as data entries in table cells.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .Low
    -0.07
    manız
    -0.07
     EMP
    -0.07
     vin
    -0.06
     freq
    -0.06
    かって
    -0.06
     limbs
    -0.06
     acquisitions
    -0.06
    álním
    -0.06
    Sep
    -0.06
    POSITIVE LOGITS
    MODEL
    0.06
    replaceAll
    0.06
     VB
    0.06
    99
    0.06
    量
    0.06
    Xml
    0.06
     inch
    0.06
     Би
    0.06
     >&
    0.05
    navigator
    0.05
    Act Density 0.002%

    No Known Activations