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

    Neuronpedia

    INDEX
    Explanations

    most common or best method

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    认识
    0.43
    彳
    0.41
    तावनी
    0.36
    wendung
    0.35
    ნენ
    0.35
    ydın
    0.35
    ীরে
    0.35
    ueur
    0.35
    annter
    0.35
    obut
    0.34
    POSITIVE LOGITS
     nhất
    0.54
     большинства
    0.51
    之一
    0.49
     большинство
    0.48
     boasts
    0.46
     अधिकांश
    0.46
     probablement
    0.46
     ever
    0.45
     (>
    0.45
     meeste
    0.44
    Act Density 0.237%

    No Known Activations