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

    Neuronpedia

    INDEX
    Explanations

    Numbers

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric expressions and measurement-related tokens (e.g. digits, decimals, and units).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     goggles
    -0.07
     ghost
    -0.07
     metrics
    -0.06
    onom
    -0.06
    最近
    -0.06
    nox
    -0.06
     sunscreen
    -0.06
     DataSet
    -0.06
    spd
    -0.06
     socks
    -0.06
    POSITIVE LOGITS
    _WHITE
    0.07
    SEN
    0.07
     ORM
    0.07
     MERCHANTABILITY
    0.07
     MART
    0.06
     Experimental
    0.06
    دار
    0.06
    OUS
    0.06
    FFE
    0.06
    َف
    0.06
    Act Density 0.055%

    No Known Activations