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

    Neuronpedia

    INDEX
    Explanations

    surprisingly complex or tricky

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     obviously
    0.82
     évidemment
    0.82
     ovviamente
    0.79
    obviously
    0.77
     obviamente
    0.75
     oczywiście
    0.75
     tentunya
    0.74
     inevitably
    0.74
    Obviously
    0.71
     طبعا
    0.69
    POSITIVE LOGITS
     surprisingly
    1.97
     Surprisingly
    1.70
    Surprisingly
    1.66
    surprisingly
    1.54
     strangely
    1.39
     surprising
    1.38
     oddly
    1.35
     actually
    1.34
     Actually
    1.23
     unexpectedly
    1.15
    Act Density 0.046%

    No Known Activations