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

    Neuronpedia

    INDEX
    Explanations

    simply presenting concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unveils
    0.56
     (-\
    0.47
     ойноо
    0.45
     delves
    0.44
     ressources
    0.44
     restauración
    0.43
     provides
    0.42
     minimizes
    0.42
     comes
    0.42
     receives
    0.41
    POSITIVE LOGITS
    立即
    0.47
    abek
    0.47
    hormat
    0.47
    เป็น
    0.45
     jednostavno
    0.44
     શ
    0.44
     सिंपली
    0.44
    svp
    0.43
     lihtsalt
    0.42
     simply
    0.42
    Act Density 0.008%

    No Known Activations