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

    Neuronpedia

    INDEX
    Explanations

    articles

    np_max-act · gemini-2.0-flash

    This neuron consistently activates on numeric tokens—especially floating-point numbers—rather than on ordinary words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    where
    -0.07
     where
    -0.07
     crossings
    -0.07
     follows
    -0.07
    .A
    -0.07
     سبز
    -0.06
     frame
    -0.06
    included
    -0.06
     crossing
    -0.06
     labeled
    -0.06
    POSITIVE LOGITS
     lief
    0.07
    환
    0.06
     ListNode
    0.06
    arring
    0.06
    PerPixel
    0.06
    _pen
    0.06
    tsx
    0.06
    memiş
    0.06
     uninsured
    0.06
    漏
    0.06
    Act Density 0.018%

    No Known Activations