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

    Neuronpedia

    INDEX
    Explanations

    chance of ending

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ných
    1.36
     וב
    1.31
    1.27
     dotycz
    1.19
    vät
    1.19
    presets
    1.18
    ্বিত
    1.15
    ᱲ
    1.14
    примеча
    1.13
    べき
    1.13
    POSITIVE LOGITS
    ль
    1.46
    r
    1.45
    at
    1.40
    l
    1.36
    イ
    1.32
    ী
    1.30
    ла
    1.26
    OL
    1.20
    k
    1.20
    IN
    1.19
    Act Density 0.156%

    No Known Activations