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

    Neuronpedia

    INDEX
    Explanations

    mathematical notation

    np_max-act · gemini-2.0-flash

    This neuron activates on LaTeX‐style mathematical tokens—subscripts, superscripts, equation labels, and related numeric/math symbols.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Breath
    -0.07
    Enemy
    -0.07
    *p
    -0.06
    ainty
    -0.06
     vlád
    -0.06
     imap
    -0.06
     deceit
    -0.06
    _chunk
    -0.06
    eday
    -0.06
    <Image
    -0.06
    POSITIVE LOGITS
     eliminates
    0.07
     PRESS
    0.06
     cardboard
    0.06
    _RESERVED
    0.06
    _basic
    0.06
     reduce
    0.06
     discrimin
    0.06
    _idx
    0.06
     cardiovascular
    0.06
    /documents
    0.06
    Act Density 0.002%

    No Known Activations