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

    Neuronpedia

    INDEX
    Explanations

    way, way of, way back

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     состава
    1.11
    ات
    1.11
    蒡
    1.04
     вся
    1.03
    ों
    1.02
     представить
    0.99
     напротив
    0.99
     brengen
    0.99
     ventre
    0.98
     всего
    0.98
    POSITIVE LOGITS
    ward
    1.70
    laid
    1.34
    finding
    1.32
    yyyyyyyy
    1.30
    yyyy
    1.27
    WARD
    1.21
    laying
    1.20
    yy
    1.16
     obsol
    1.15
     ਨਾਲ
    1.13
    Act Density 0.107%

    No Known Activations