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

    Neuronpedia

    INDEX
    Explanations

    quantized states and specific quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ế
    0.40
     метод
    0.40
    どちら
    0.40
    ”?
    0.38
    ী
    0.38
    筁
    0.38
    Види
    0.37
    '\
    0.37
     chooser
    0.37
    க்
    0.37
    POSITIVE LOGITS
    levelup
    0.48
     outp
    0.44
     lineup
    0.40
     நன்ற
    0.39
     피해
    0.39
    出力
    0.38
     ከፍተኛ
    0.38
    先进
    0.37
     اعلی
    0.37
     ഉയർന്ന
    0.37
    Act Density 0.001%

    No Known Activations