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

    Neuronpedia

    INDEX
    Explanations

    sequence and subsequent events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    药
    0.41
    Compress
    0.40
    比如
    0.40
    出错
    0.39
    场合
    0.39
    提出
    0.38
    眼睛
    0.38
    裙
    0.38
     제공
    0.38
    нии
    0.37
    POSITIVE LOGITS
     thereafter
    0.85
     Thereafter
    0.73
     onwards
    0.66
     coincided
    0.64
     이후
    0.60
     subsequently
    0.59
    以降
    0.59
    到现在
    0.58
     subsequent
    0.55
     сразу
    0.53
    Act Density 0.080%

    No Known Activations