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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nonostante
    0.49
     conquista
    0.43
    竟
    0.43
     mussten
    0.42
     Despite
    0.41
     brachte
    0.41
     Когда
    0.40
    ಾದ
    0.39
     mantener
    0.39
     осозна
    0.39
    POSITIVE LOGITS
    可能会
    0.52
     will
    0.50
    ன்ஹீ
    0.46
     depending
    0.45
     caution
    0.45
     might
    0.45
    会有
    0.45
     MAY
    0.45
     stadiums
    0.45
     moż
    0.44
    Act Density 0.152%

    No Known Activations