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

    Neuronpedia

    INDEX
    Explanations

    success followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ت
    2.49
    ्स
    1.67
     Emoji
    1.61
    т
    1.59
     credence
    1.58
     Vedas
    1.57
     atrocities
    1.56
     destitute
    1.56
    なりません
    1.55
     unleashing
    1.53
    POSITIVE LOGITS
    я
    1.96
     पूर्वक
    1.71
    е
    1.46
    ఫ
    1.44
    ה
    1.36
    его
    1.34
     quả
    1.33
    <bos>
    1.30
    কারের
    1.30
    াঙ্গ
    1.29
    Act Density 0.382%

    No Known Activations