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

    Neuronpedia

    INDEX
    Explanations

    end of explanatory sentences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     semester
    0.45
    多
    0.43
     Turn
    0.42
     Sticks
    0.41
     Nature
    0.39
    กัน
    0.39
     unsol
    0.39
     harder
    0.38
    外交
    0.38
    ሑ
    0.38
    POSITIVE LOGITS
    Selon
    0.50
    alaikumsalam
    0.50
     каждом
    0.49
     joka
    0.46
    Según
    0.43
    🚍
    0.43
     dependiendo
    0.43
     Каждый
    0.43
    که
    0.42
    opts
    0.42
    Act Density 0.004%

    No Known Activations