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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron activates on isolated numeric tokens (digits), especially those used as standalone reference or section numbers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     twig
    -0.07
     تحلیل
    -0.07
     parms
    -0.07
    avity
    -0.06
    NSSet
    -0.06
     hamburger
    -0.06
    Messages
    -0.06
     Lover
    -0.06
    Assertions
    -0.06
     数据
    -0.06
    POSITIVE LOGITS
    (InitializedTypeInfo
    0.07
     Δ
    0.06
     baktı
    0.06
     सक
    0.06
     getField
    0.06
     ALSO
    0.06
     Pall
    0.06
    ecz
    0.06
     jobId
    0.06
    513
    0.06
    Act Density 0.003%

    No Known Activations