INDEX
    Explanations

    young people and generations

    New Auto-Interp
    Negative Logits
     primitive
    0.45
     purified
    0.40
     elegant
    0.39
     diabetic
    0.36
    Elegant
    0.36
     convex
    0.35
    readonly
    0.35
     improved
    0.35
     elegantly
    0.35
     manually
    0.34
    POSITIVE LOGITS
     younger
    2.28
    年轻人
    2.28
     young
    2.25
    2.23
     jovens
    2.19
    年輕
    2.19
     gençler
    2.17
     genç
    2.16
     jóvenes
    2.11
     jeunes
    2.06
    Act Density 0.018%

    No Known Activations