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

    Neuronpedia

    INDEX
    Explanations

    instances of numerical data, especially in the context of scientific or statistical results

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    The neuron detects numeric and document-structure tokens — digits, numbers, numeric references (figures/sections), and related punctuation/labels.

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003
    New Auto-Interp
    Negative Logits
    uage
    -0.47
    essero
    -0.47
    isel
    -0.47
     mild
    -0.46
     từng
    -0.46
     bene
    -0.46
     recovered
    -0.45
    thood
    -0.45
     undamaged
    -0.45
    Recovered
    -0.45
    POSITIVE LOGITS
     GenerationType
    0.90
    AccessorTable
    0.89
     CreateTagHelper
    0.79
    FormTagHelper
    0.73
    UnsafeEnabled
    0.73
    بوابة
    0.70
    IsContent
    0.70
    ValueGenerated
    0.69
     NavController
    0.65
    AndEndTag
    0.65
    Act Density 0.603%

    No Known Activations