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

    Neuronpedia

    INDEX
    Explanations

    catastrophic forgetting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )|^{
    0.36
    갭
    0.35
    GPS
    0.35
     রাষ্ট্রদূত
    0.34
    តុ
    0.33
    PERATURE
    0.33
    âl
    0.33
     नाबाद
    0.33
    姦
    0.33
    र्चे
    0.32
    POSITIVE LOGITS
     forget
    4.38
     forgetting
    3.97
     forgot
    3.95
    forget
    3.89
     forgets
    3.88
     Forget
    3.83
    Forget
    3.78
     забы
    3.69
     forgotten
    3.64
    忘记
    3.58
    Act Density 0.044%

    No Known Activations