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

    Neuronpedia

    INDEX
    Explanations

    categories or labels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    件事
    0.99
    件事情
    0.89
     exorbit
    0.85
     dimensionality
    0.84
     bagel
    0.82
    िश्वत
    0.80
     backstory
    0.79
     videogame
    0.79
    倬
    0.78
    dimensionality
    0.78
    POSITIVE LOGITS
     vs
    1.08
     mode
    0.93
    -,
    0.91
     فقط
    0.90
     Mode
    0.88
    の場合
    0.88
     Only
    0.88
    のみ
    0.87
     ones
    0.87
     방식
    0.87
    Act Density 0.843%

    No Known Activations