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

    Neuronpedia

    INDEX
    Explanations

    Car/Cars/Coaches

    np_max-act · gemini-2.0-flash

    This neuron activates on numerical tokens—especially measurements, statistics, and decimal values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ए
    -0.06
     bolts
    -0.06
    edere
    -0.06
     Qgs
    -0.06
     referral
    -0.06
     Adding
    -0.06
    thesized
    -0.05
     coordinating
    -0.05
    theses
    -0.05
     gaining
    -0.05
    POSITIVE LOGITS
    .Character
    0.07
     √
    0.07
    教学
    0.07
    راف
    0.07
    {{$
    0.07
    ishes
    0.07
    恐
    0.07
    alties
    0.07
    -seeking
    0.06
     racked
    0.06
    Act Density 0.007%

    No Known Activations