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

    Neuronpedia

    INDEX
    Explanations

    today, November

    np_acts-logits-general · gemini-2.5-flash-lite

    the neuron detects floating-point numeric tokens (numbers with decimals).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31

    The neuron activates on numeric tokens representing floating‐point values (e.g. “650.4088”)

    oai_token-act-pair · o4-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    و
    1.32
    to
    1.28
    ${
    1.27
    MSFontService
    1.25
    ó
    1.25
    icolored
    1.23
    かった
    1.23
    sman
    1.21
    ship
    1.20
    ⃣
    1.20
    POSITIVE LOGITS
    い
    1.32
    부터
    1.27
     provoke
    1.27
    вна
    1.26
    и
    1.23
     hostilities
    1.23
     spines
    1.20
    י
    1.17
     DAY
    1.16
     fees
    1.16
    Act Density 0.029%

    No Known Activations