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

    Neuronpedia

    INDEX
    Explanations

    explanation of the loop method

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     yayın
    0.52
    1
    0.51
    Gre
    0.47
     depo
    0.46
     nuevos
    0.46
     canals
    0.46
    Hydro
    0.44
     býval
    0.43
     luft
    0.43
     pusat
    0.43
    POSITIVE LOGITS
     か
    0.57
    வுடன்
    0.53
     эффективности
    0.48
     ಹೀ
    0.48
     እንዲሁ
    0.47
    ありがとう
    0.46
     acclaimed
    0.46
    画像の
    0.46
     amazingly
    0.45
    것도
    0.45
    Act Density 0.000%

    No Known Activations