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

    Neuronpedia

    INDEX
    Explanations

    lab, mat, mar contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ️
    2.61
    ive
    2.22
    ৬
    2.12
    ἱ
    2.12
    ︎
    1.96
    Glad
    1.92
    ంగ
    1.91
    ﺔ
    1.90
    clerc
    1.89
    urethane
    1.89
    POSITIVE LOGITS
    o
    4.35
    ে
    3.74
    ه
    3.65
    u
    3.58
    い
    3.37
    er
    3.27
    ו
    3.02
    dı
    2.89
    ம்
    2.88
    ্ট
    2.80
    Act Density 0.115%

    No Known Activations