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

    Neuronpedia

    INDEX
    Explanations

    records of performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    꿋
    0.43
    分かり
    0.42
     اجر
    0.42
    ешься
    0.41
    рисо
    0.39
     désir
    0.39
    笺
    0.39
     말미암아
    0.39
    誊
    0.39
     proffered
    0.39
    POSITIVE LOGITS
     ≥
    0.52
     >=
    0.51
     youngest
    0.47
     HT
    0.46
     vs
    0.46
     速
    0.45
     DPI
    0.44
     quartile
    0.44
     leftmost
    0.44
     longest
    0.43
    Act Density 0.004%

    No Known Activations