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

    Neuronpedia

    INDEX
    Explanations

    in the context of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    RE
    0.47
    да
    0.43
    以
    0.42
    LE
    0.42
     boasts
    0.41
     boldly
    0.40
     clearly
    0.40
    。",
    0.39
    得以
    0.38
    c
    0.38
    POSITIVE LOGITS
     ব্যাপারটা
    0.55
     xã
    0.47
    %]
    0.47
     tomto
    0.46
    后面的
    0.46
    這次
    0.45
    旁邊
    0.45
    這個
    0.44
    實驗
    0.44
     क्रमांका
    0.44
    Act Density 0.020%

    No Known Activations