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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens containing digits, picking out version numbers, model numbers, numeric constants, and other digit-heavy identifiers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     interruptions
    -0.08
     setData
    -0.06
    restaurant
    -0.06
    intestinal
    -0.06
    .setHeader
    -0.06
     corrupt
    -0.06
    MatrixMode
    -0.06
     Medium
    -0.06
    TestingModule
    -0.06
    .setData
    -0.06
    POSITIVE LOGITS
    델
    0.07
     getClass
    0.06
     Opp
    0.06
     Ashley
    0.06
     cw
    0.06
    wert
    0.06
    /conf
    0.06
    ुबह
    0.06
     voc
    0.06
    exc
    0.06
    Act Density 0.012%

    No Known Activations