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

    Neuronpedia

    INDEX
    Explanations

    decrease and drop

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nâng
    0.53
    增
    0.51
     올라
    0.48
     (++
    0.44
    升级
    0.44
    Upload
    0.43
     اٹھ
    0.43
    Lift
    0.43
     Upload
    0.43
    升
    0.43
    POSITIVE LOGITS
     decline
    2.66
     declining
    2.63
    下降
    2.52
     decreasing
    2.48
     declines
    2.47
     decrease
    2.36
     penurunan
    2.33
     dropping
    2.27
    decreasing
    2.27
     menurun
    2.27
    Act Density 0.237%

    No Known Activations