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

    Neuronpedia

    INDEX
    Explanations

    speaking/saying

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    和他
    0.50
    ⇌
    0.49
     देंगे
    0.48
    跟他
    0.48
    看不到
    0.47
     watchlist
    0.47
     kennen
    0.46
    ]-->
    0.46
    0.46
    เข้าร
    0.45
    POSITIVE LOGITS
     said
    3.17
    said
    2.81
     Said
    2.48
    Said
    2.34
     dijo
    2.33
     сказал
    2.31
     sagte
    2.16
     says
    2.14
     exclaimed
    2.06
     قال
    2.05
    Act Density 0.057%

    No Known Activations