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

    Neuronpedia

    INDEX
    Explanations

    physics/quantum mechanics

    np_max-act · gemini-2.0-flash

    This neuron primarily fires on numeric and mathematical-expression tokens (e.g. digits, fractions, brackets, and functions like “math.sqrt”) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    			↵
    -0.07
    WK
    -0.07
    (con
    -0.07
    _EXTENSION
    -0.06
     mixer
    -0.06
     Bayesian
    -0.06
     artery
    -0.06
     UPS
    -0.06
    .Spec
    -0.06
     카
    -0.06
    POSITIVE LOGITS
     nebezpeč
    0.06
    Avatar
    0.06
     recount
    0.06
     Michele
    0.06
     Egyptian
    0.06
    :date
    0.06
     Edwin
    0.06
     Hawai
    0.06
     recounts
    0.06
    derabad
    0.06
    Act Density 0.017%

    No Known Activations