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

    Neuronpedia

    INDEX
    Explanations

    energy/environment comparisons

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens (digits and decimals), i.e. it detects quantities and measurements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     titles
    -0.07
     ist
    -0.06
     blasted
    -0.06
     angles
    -0.06
     Yong
    -0.06
     tiện
    -0.06
     Walls
    -0.06
    _ratio
    -0.06
    ürk
    -0.06
    edicine
    -0.06
    POSITIVE LOGITS
    平成
    0.08
     نسخ
    0.07
     iCloud
    0.07
     errorMessage
    0.06
    ,const
    0.06
    ACL
    0.06
    setWidth
    0.06
    MeshPro
    0.06
     multitude
    0.06
     amnesty
    0.06
    Act Density 0.008%

    No Known Activations