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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     방법에
    0.43
    ).</
    0.41
     de
    0.41
    .'),
    0.40
    .").
    0.39
     to
    0.37
    ."),
    0.36
    ."],
    0.36
    城的
    0.36
     प्रकारचे
    0.36
    POSITIVE LOGITS
    ：
    0.73
    ؟
    0.61
    :
    0.57
    ?:
    0.54
    ？
    0.54
    ៖
    0.53
    】
    0.51
    ":
    0.50
    》
    0.50
     ：
    0.49
    Act Density 1.132%

    No Known Activations