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

    Neuronpedia

    INDEX
    Explanations

    prohibited actions and conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Η
    0.39
    鐫
    0.38
    pository
    0.36
    oyen
    0.36
    बीर
    0.36
    وكان
    0.35
    тона
    0.35
    نګ
    0.35
     shin
    0.35
     وليس
    0.35
    POSITIVE LOGITS
     lest
    1.05
    以免
    1.04
    lest
    0.78
     too
    0.74
    否则
    0.73
     слишком
    0.70
     Too
    0.69
    Too
    0.68
     terlalu
    0.68
    too
    0.66
    Act Density 0.182%

    No Known Activations