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

    Neuronpedia

    INDEX
    Explanations

    multi-language content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    0.52
     Meeting
    0.52
    oe
    0.50
    𝐞
    0.50
    ｅ
    0.50
    oc
    0.47
    orez
    0.47
    Ek
    0.47
    🧨
    0.47
    are
    0.46
    POSITIVE LOGITS
     hab
    0.42
     whoever
    0.42
    ጠበ
    0.41
     acara
    0.41
    ージー
    0.40
     whichever
    0.40
     radii
    0.40
    劇場
    0.40
     می‌ده
    0.39
     extern
    0.39
    Act Density 0.001%

    No Known Activations