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

    Neuronpedia

    INDEX
    Explanations

    urban, rural, and regional variations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.77
     ).
    0.64
    á
    0.57
     meso
    0.55
     мои
    0.55
    ああ
    0.55
    削
    0.54
     acab
    0.53
    ).
    0.52
    ?).
    0.52
    POSITIVE LOGITS
     könnte
    0.56
    ünst
    0.55
    CGPoint
    0.54
    лан
    0.53
     pošt
    0.49
    OutputFile
    0.48
    rafpunkte
    0.47
    െയുള്ള
    0.47
    ﻏ
    0.47
     događ
    0.47
    Act Density 0.002%

    No Known Activations