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

    Neuronpedia

    INDEX
    Explanations

    single digit number

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens—especially software version numbers or standalone digits.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    تى
    -0.07
    工
    -0.06
    -Am
    -0.06
    cher
    -0.06
     Their
    -0.06
     checkbox
    -0.06
     찬
    -0.06
    øy
    -0.06
     encour
    -0.06
    ことは
    -0.06
    POSITIVE LOGITS
     declaration
    0.08
    _STENCIL
    0.07
     civilizations
    0.06
    III
    0.06
    (Application
    0.06
    564
    0.06
     III
    0.06
    "text
    0.06
    _='
    0.06
    JWT
    0.06
    Act Density 0.025%

    No Known Activations