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

    Neuronpedia

    INDEX
    Explanations

    excess or degree

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    非常的
    0.46
     довольно
    0.44
     veoma
    0.41
     상당히
    0.40
    굉장
    0.39
    elter
    0.39
     quite
    0.39
    頗
    0.38
    URUK
    0.38
    非常
    0.37
    POSITIVE LOGITS
     overly
    0.79
     Ideally
    0.74
     excessively
    0.73
     excessive
    0.73
     Excessive
    0.70
    Ideally
    0.66
     fewer
    0.66
     ideally
    0.64
     слишком
    0.63
     exces
    0.63
    Act Density 0.069%

    No Known Activations