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

    Neuronpedia

    INDEX
    Explanations

    )

    np_max-act · gemini-2.0-flash

    This neuron specializes in spotting numeric literal tokens (especially decimal numbers) in mathematical expressions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    WithMany
    -0.07
     asi
    -0.07
     тебя
    -0.06
     DEAL
    -0.06
     Vak
    -0.06
    ีเม
    -0.06
     Petra
    -0.06
    $criteria
    -0.06
     vů
    -0.06
     stol
    -0.06
    POSITIVE LOGITS
    campaign
    0.07
    δό
    0.07
    _cats
    0.07
    ashington
    0.07
    ー
    0.07
    hydro
    0.06
    .branch
    0.06
     Hayes
    0.06
     neighborhoods
    0.06
     kwargs
    0.06
    Act Density 0.002%

    No Known Activations