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

    Neuronpedia

    INDEX
    Explanations

    spatial directions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    เหมาะสม
    0.45
    限界
    0.45
     optimum
    0.44
    maximum
    0.41
     Period
    0.40
     максимум
    0.40
     maksimum
    0.38
    max
    0.38
    appropriate
    0.38
     maximum
    0.38
    POSITIVE LOGITS
     positions
    0.49
     endroit
    0.44
     endroits
    0.43
    ளுக்காக
    0.43
     sebelumnya
    0.42
    ăzi
    0.41
    positions
    0.41
     gehört
    0.40
     offsets
    0.40
     financières
    0.40
    Act Density 0.023%

    No Known Activations