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

    Neuronpedia

    INDEX
    Explanations

    you are right

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     berupa
    0.39
     Instruction
    0.38
     almacen
    0.35
    Expect
    0.34
    дон
    0.33
    〔
    0.32
     unix
    0.32
    ங்களுக்கும்
    0.32
    ющихся
    0.32
     Vers
    0.32
    POSITIVE LOGITS
     right
    0.51
     understandably
    0.51
     राइट
    0.46
     wasting
    0.45
     RIGHT
    0.45
     impressively
    0.45
     fortunate
    0.45
     wisely
    0.44
     oike
    0.44
     lucky
    0.43
    Act Density 0.005%

    No Known Activations