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

    Neuronpedia

    INDEX
    Explanations

    configurator or configuration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    1.91
    on
    1.88
    ل
    1.63
    ル
    1.52
    л
    1.48
    τή
    1.46
    ল
    1.44
    ด
    1.43
    ou
    1.39
    ul
    1.39
    POSITIVE LOGITS
    </h2>
    1.28
     service
    1.10
     musical
    1.10
     president
    1.09
     crumble
    1.09
     protein
    1.08
     secretary
    1.06
     suicide
    1.05
     restaurant
    1.04
     water
    1.03
    Act Density 0.007%

    No Known Activations