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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates specifically on decimal number tokens (floating‐point values) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    lık
    -0.07
     smart
    -0.07
     seeker
    -0.06
     Imperial
    -0.06
     tình
    -0.06
    era
    -0.06
    inding
    -0.06
    üt
    -0.06
    废
    -0.06
     Moody
    -0.06
    POSITIVE LOGITS
    _zero
    0.07
     undone
    0.06
    (after
    0.06
    *Math
    0.06
    getChild
    0.06
     klient
    0.06
     Друг
    0.06
     druh
    0.06
    lbrace
    0.06
    embedded
    0.06
    Act Density 0.022%

    No Known Activations