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

    Neuronpedia

    INDEX
    Explanations

    Golang, xlUp

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    으로
    5.10
    y
    3.69
    ה
    3.57
    াস
    3.57
    ein
    3.43
    ethanol
    3.42
    yen
    3.33
    elem
    3.29
    े
    3.28
    e
    3.26
    POSITIVE LOGITS
    ld
    3.50
    ties
    3.46
    ্ল
    3.46
    ্পনিক
    3.31
    to
    3.28
    ্ট
    3.25
    ਼
    3.10
    ts
    3.04
    ੍ਹ
    3.03
    ين
    3.03
    Act Density 0.575%

    No Known Activations