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

    Neuronpedia

    INDEX
    Explanations

    national language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ሷ
    0.41
     של
    0.39
    ícul
    0.39
    ❿
    0.39
     νέ
    0.38
     उस
    0.38
    čne
    0.37
    ্দের
    0.37
    रील
    0.37
    RatingDiff
    0.37
    POSITIVE LOGITS
     instead
    0.53
    Instead
    0.50
    instead
    0.47
     Instead
    0.44
     literalmente
    0.41
     porque
    0.41
     invece
    0.39
     omdat
    0.38
    ではなく
    0.37
     capitalism
    0.37
    Act Density 0.037%

    No Known Activations