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

    Neuronpedia

    INDEX
    Explanations

    mistakes, incorrect, false, typo

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    7
    1.87
    5
    1.80
    2
    1.79
    8
    1.77
    9
    1.75
    6
    1.74
    0
    1.72
    il
    1.71
    3
    1.71
    it
    1.64
    POSITIVE LOGITS
     incorrect
    1.31
    ক
    1.17
    CurrentByte
    1.14
    ImageQueue
    1.11
    くない
    1.10
     mistaken
    1.06
     misleading
    1.05
    ным
    1.05
    ीकृत
    1.05
    พลาด
    1.03
    Act Density 0.589%

    No Known Activations