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

    Neuronpedia

    INDEX
    Explanations

    but

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    및
    0.51
     மற்றும்
    0.51
    または
    0.49
    及其
    0.48
     или
    0.46
     Nhưng
    0.45
     แต่
    0.45
    หรือ
    0.44
     અથવા
    0.44
     หรือ
    0.43
    POSITIVE LOGITS
     importantly
    1.04
     crucially
    1.03
     interestingly
    0.93
     fortunately
    0.92
     thankfully
    0.92
     alas
    0.89
     conversely
    0.88
     luckily
    0.86
     let
    0.86
     hey
    0.86
    Act Density 0.488%

    No Known Activations