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

    Neuronpedia

    INDEX
    Explanations

    non-english words and contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rank
    0.43
     less
    0.41
     WM
    0.38
     ro
    0.38
    ern
    0.37
     panor
    0.37
     FM
    0.36
     Ro
    0.36
     ranks
    0.36
     regimes
    0.36
    POSITIVE LOGITS
     ໃນ
    0.43
     volvió
    0.40
    ංග
    0.40
    ውስ
    0.40
     चाहि
    0.40
     توص
    0.40
     সত্ত্বেও
    0.39
     несмотря
    0.39
    راط
    0.39
    despite
    0.39
    Act Density 0.000%

    No Known Activations