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

    Neuronpedia

    INDEX
    Explanations

    )

    np_max-act · gemini-2.0-flash

    Explanation of neuron 4 behavior: the main thing this neuron does is find floating-point numeric literals.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     SHOW
    -0.07
    われる
    -0.06
    g
    -0.06
     American
    -0.06
     aracı
    -0.06
     camouflage
    -0.06
     […]
    -0.06
     Festival
    -0.06
     hWnd
    -0.06
    G
    -0.06
    POSITIVE LOGITS
    ेख
    0.07
    ограф
    0.06
    (Runtime
    0.06
    ней
    0.06
    .loggedIn
    0.06
     теп
    0.06
    ':[
    0.06
     سرم
    0.06
     onchange
    0.06
     ydk
    0.06
    Act Density 0.023%

    No Known Activations