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

    Neuronpedia

    INDEX
    Explanations

    require massive computational servers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hatched
    0.55
     underlie
    0.55
     underlies
    0.51
     basiert
    0.48
     erinnert
    0.48
     Wix
    0.48
     hatchery
    0.47
     correlates
    0.46
     Zentral
    0.46
     testifies
    0.46
    POSITIVE LOGITS
    as
    0.54
    E
    0.46
    en
    0.46
    in
    0.45
    ਸ
    0.45
    名
    0.45
    o
    0.45
    rov
    0.44
    𝐞
    0.44
    اس
    0.43
    Act Density 0.001%

    No Known Activations