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

    Neuronpedia

    INDEX
    Explanations

    qualification or disclaimer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     или
    0.64
     অথবা
    0.63
    或是
    0.61
     hoặc
    0.60
    或者
    0.59
     eller
    0.59
    或
    0.59
     หรือ
    0.58
    หรือ
    0.57
    または
    0.55
    POSITIVE LOGITS
    あくまで
    0.97
     only
    0.70
     lediglich
    0.68
     jedynie
    0.67
     tikai
    0.67
     tylko
    0.66
     ONLY
    0.65
     толькі
    0.64
    only
    0.64
    仅
    0.64
    Act Density 0.012%

    No Known Activations