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

    Neuronpedia

    INDEX
    Explanations

    competitive analysis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     But
    0.75
    كن
    0.71
    in
    0.63
     It
    0.61
    ان
    0.61
     لكن
    0.61
    But
    0.59
    لي
    0.58
     A
    0.57
     लेकिन
    0.57
    POSITIVE LOGITS
     competitors
    0.75
    ۰
    0.74
    amai
    0.70
     benchmarks
    0.66
    w
    0.66
    ot
    0.65
     compét
    0.65
     jakie
    0.65
    μοι
    0.65
     рік
    0.63
    Act Density 0.052%

    No Known Activations