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

    Neuronpedia

    INDEX
    Explanations

    underlying abstract concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Sometimes
    0.91
     Sometimes
    0.85
    Although
    0.78
     Biasanya
    0.77
    Normally
    0.76
    因为
    0.75
    sometimes
    0.75
     sometimes
    0.75
    Being
    0.74
    因為
    0.73
    POSITIVE LOGITS
     aspects
    0.97
     notions
    0.93
     elements
    0.91
     آنچه
    0.90
     underlying
    0.89
     perceptions
    0.89
     reality
    0.88
     społecz
    0.86
     beneath
    0.85
     salient
    0.84
    Act Density 0.130%

    No Known Activations