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

    Neuronpedia

    INDEX
    Explanations

    not fully or not enough

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    そして
    1.11
     myös
    1.08
     גם
    1.06
     sowohl
    0.93
     ומ
    0.91
     különböző
    0.91
     וא
    0.89
     både
    0.89
     וש
    0.88
     різні
    0.88
    POSITIVE LOGITS
     fully
    1.83
     necessarily
    1.77
    necessarily
    1.54
     Fully
    1.53
     sepenuhnya
    1.46
    Fully
    1.39
     necessariamente
    1.35
     full
    1.34
     strictly
    1.27
     necesariamente
    1.26
    Act Density 1.593%

    No Known Activations