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

    Neuronpedia

    INDEX
    Explanations

    Regel sollte

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     indicating
    0.37
    се
    0.35
     двой
    0.35
    Игра
    0.34
    Эк
    0.34
     anticon
    0.34
    称
    0.33
     функ
    0.33
     анг
    0.33
     senc
    0.33
    POSITIVE LOGITS
     jednot
    0.41
    يروس
    0.41
     vassals
    0.41
    囑
    0.41
     partisans
    0.40
     Tuc
    0.40
     Rog
    0.39
    varrho
    0.39
     SPDX
    0.39
     counselors
    0.39
    Act Density 0.000%

    No Known Activations