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

    Neuronpedia

    INDEX
    Explanations

    standalone components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    荠
    0.47
     servis
    0.46
     자기
    0.44
     TimeUnit
    0.42
     Tells
    0.42
    禟
    0.42
     coisas
    0.42
     typographical
    0.41
     letras
    0.40
    imètres
    0.40
    POSITIVE LOGITS
    vec
    0.40
    Duc
    0.40
    SN
    0.38
    0.38
    타
    0.38
    Flare
    0.38
    Fast
    0.38
    Cancel
    0.38
    EX
    0.38
    Flex
    0.38
    Act Density 0.000%

    No Known Activations