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

    Neuronpedia

    INDEX
    Explanations

    pauses and filler words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ಡೆ
    0.94
    قة
    0.91
    ׁ
    0.91
    ="
    0.86
    ं
    0.84
    Usage
    0.83
    ‍या
    0.83
    ட
    0.82
    бек
    0.82
    क्ता
    0.81
    POSITIVE LOGITS
     ähm
    1.51
     euh
    1.48
    etc
    1.47
     เอ่อ
    1.45
     ehm
    1.43
     .(
    1.42
     博文
    1.32
    呃
    1.29
     cavalry
    1.28
    嗯
    1.25
    Act Density 0.063%

    No Known Activations